-
-Inflect-Nano-v2 is designed for local applications where the entire speech stack must fit below **4M parameters**. The count includes text encoding, duration modeling, latent generation, and waveform decoding. It does not hide a separately downloaded vocoder.
+Inflect-Nano-v2 is the portability-focused member of Inflect v2. The complete deployable system has **3,966,721 parameters**; there is no second vocoder download and no hidden large language model at inference time.
## Listen First
-All samples below use unseen evaluation text, not training-audio reconstructions.
+Every clip below uses held-out text rather than a training-audio reconstruction.
-| Prompt | Male voice |
+| Prompt type | Audio |
| --- | --- |
-| **Conversational** - It wasn't until later that I realized what had actually happened. | |
-| **Punctuation** - First, close the window; second, turn off the lamp; finally, lock the door. | |
-| **Numbers** - The package weighs twelve point six kilograms and arrived on July twenty-first. | |
-| **Names Places** - Gwendolyn photographed the eucalyptus trees outside Ljubljana. | |
-| **Technical** - The system runs on three core components that all have to stay in sync. | |
+| Conversational | |
+| Punctuation | |
+| Numbers | |
+| Names and places | |
+| Technical | |
-## Model Facts
+## At a Glance
| | Inflect-Nano-v2 |
| --- | ---: |
-| Deployable inference parameters | **3,966,721** |
-| PyTorch checkpoint size | **15.97 MB** |
+| Complete deployable parameters | **3,966,721** |
+| FP32 model weights | **15.97 MB** |
| Sample rate | **24 kHz** |
-| Included voices | **1 male voice** |
+| Voice | **1 fixed English male voice** |
| External vocoder | **None** |
-| Inference package | **PyTorch** |
+| Release format | **PyTorch FP32** |
+| Long text | **Punctuation-aware chunking** |
+
+Inflect v2 is released in two deliberately different sizes. **Micro** spends its larger budget on waveform capacity and hidden width; **Nano** prioritizes footprint while preserving the same public runtime and text frontend. Both numbers above count the complete deployable generator, including the waveform decoder.
-## Training Data
+## Human Listening
-The deployable Inflect weights were trained from scratch on a single-speaker synthetic English corpus produced with Qwen3-TTS. Qwen3-TTS is a training-time teacher only and is not downloaded or used for inference. The corpus itself is not included in this repository.
+
+
+Inflect-Nano-v2 recorded a **63.9% preference rate** (22 wins, 12 losses, 2 ties) in the final anonymous community study. The full study contains **147 pairwise decisions** across the published lineup. Ties count as half a win. This is descriptive community evidence, not a formal MOS study; pair coverage and vote counts differ by system.
## Evaluation
-
+### Quality versus footprint
-| System | UTMOS22 higher is better | Whisper-large-v3 semantic WER lower is better |
-| --- | ---: | ---: |
-| Inflect-Micro-v2 | 4.406 | 1.23% |
-| Inflect-Nano-v2 | 4.392 | 1.53% |
-| Piper Ryan Low | 4.256 | 1.23% |
-| Kitten Nano Bruno | 4.246 | 0.96% |
+
-The selected Inflect-Nano-v2 checkpoint scored **4.392 UTMOS22** and **1.53% semantic WER** on the matched 200-prompt release set. Automated metrics are diagnostics, not substitutes for listening.
+### Intelligibility
-
-Evaluation protocol and raw reports
+
+
+### Cross-ASR robustness
+
+
+
+### CPU throughput
+
+
+
+### Quality and speed
+
+
+
+### Error analysis
+
+
+
+### Complete model footprint
+
+
+
+| System | Weights | UTMOS22 ↑ | Semantic WER ↓ |
+| --- | ---: | ---: | ---: |
+| **Inflect-Micro-v2** | 37.5 MB | 4.395 | 1.93% |
+| **Inflect-Nano-v2** | 16.0 MB | 4.386 | 2.48% |
+| KittenTTS Nano · Bruno | 56.8 MB | 4.217 | 0.97% |
+| KittenTTS Nano · Hugo | 56.8 MB | 4.191 | 1.15% |
+| Piper · Ryan Low | 63.1 MB | 4.289 | 2.33% |
+| Piper · Danny Low | 63.1 MB | 4.195 | 1.81% |
+| Supertonic 3 · James · 3-step | 398.1 MB | 2.471 | 692.82% |
+| Supertonic 3 · James · 8-step | 398.1 MB | 4.295 | 622.52% |
-- All four systems synthesize the same 200 unseen English prompts.
-- Semantic WER uses Whisper-large-v3 and the disclosed English text normalizer.
-- UTMOS22 uses `tarepan/SpeechMOS` v1.2.0 with 5,000 paired bootstrap samples.
-- NISQA dimensions, signal diagnostics, categorized ASR failures, speaker consistency, and multi-seed robustness are under `evaluation/`.
-- `evaluation/asr_robustness.json` separates sentence failures, semantic character errors, tail WER, and category-level errors.
-- Competitor results apply only to the named checkpoints and voices, not every configuration of those projects.
+On the 500-prompt matched set, Inflect-Nano-v2 measured **4.386 UTMOS22** (95% bootstrap CI 4.372–4.399) and **2.48% semantic WER** (95% CI 1.95–3.07). UTMOS22 is an automated quality predictor, not human MOS. WER measures transcript recoverability, not naturalness.
+
+Under the prespecified Whisper-large-v3 protocol, Supertonic produced insertion-heavy hypotheses and landed off-scale at **692.82%** (3-step) and **622.52%** (8-step). The independent wav2vec2 pass scored the same clips at **13.47%** and **3.76%**, respectively. This large disagreement is reported as ASR-family sensitivity, not collapsed into an aggregate or interpreted as a direct measure of Supertonic intelligibility. Exact hypotheses and per-clip errors ship in the raw reports.
+
+
+Protocol and raw evidence
+
+- 500 identical unseen English prompts per system; 4,000 generated clips total.
+- The set combines a 200-prompt stress corpus with 300 deterministic CMU Arctic phonetic-balance sentences.
+- Exact-text exclusion was run against 87,362 training transcripts.
+- Headline semantic WER uses Whisper-large-v3 and the disclosed English normalizer; wav2vec2 LV-60K is included as an independent ASR-sensitivity check.
+- UTMOS22 uses `tarepan/SpeechMOS` v1.2.0.
+- Headline intervals use 10,000 bootstrap samples.
+- Kitten and Piper voices are reported separately; equal-weight family averages are in `evaluation/final/evidence-summary.json`.
+- Supertonic 3-step and 8-step are never pooled.
+- Per-clip outputs, category breakdowns, runtime rows, and signal diagnostics are under `evaluation/final/`.
## Quickstart
```bash
-pip install torch numpy scipy soundfile phonemizer espeakng-loader num2words Unidecode huggingface_hub
-hf download owensong/Inflect-Nano-v2 --local-dir Inflect-Nano-v2
-python Inflect-Nano-v2/inference.py --model-dir Inflect-Nano-v2 --device cpu --text "Small models should still sound alive." --output sample.wav
+pip install -r requirements.txt
+python inference.py \
+ --model-dir . \
+ --device cpu \
+ --text "A small voice can still have something meaningful to say." \
+ --output sample.wav
```
-Python:
-
```python
import sys
from huggingface_hub import snapshot_download
@@ -92,58 +134,94 @@ sys.path.insert(0, model_dir)
from inference import InflectTTS
tts = InflectTTS(model_dir, device="cpu")
-tts.save("The complete model runs locally.", "sample.wav", seed=7)
+tts.save(
+ "The complete speech model runs locally.",
+ "sample.wav",
+ speed=1.0,
+ variation=0.667,
+ seed=7,
+)
```
-Long text is split at punctuation-aware boundaries by the packaged runtime. `speed` ranges from `0.5` to `2.0`; `variation` ranges from `0.0` to `1.0`.
+The packaged runtime accepts `speed` from `0.5` to `2.0`, `variation` from `0.0` to `1.0`, and a deterministic integer seed. Long input is split at punctuation-aware boundaries and concatenated with controlled pauses.
+
+### Controls
+
+| Control | Default | Effect |
+| --- | ---: | --- |
+| `speed` | `1.0` | Speech rate; supported range `0.5`–`2.0` |
+| `variation` | `0.667` | Stochastic latent variation; lower is steadier, higher is more variable |
+| `seed` | `0` | Reproducible sampling seed; incremented per long-text chunk |
+
+Use a fixed seed when comparing systems. A different seed is a different stochastic sample, not a different checkpoint.
## Architecture
-Inflect v2 is a size-specific compact VITS-derived system:
+Inflect v2 is a **parameter-efficient VITS-family, end-to-end text-to-waveform speech generator** with a phoneme frontend, monotonic alignment, stochastic latent synthesis, and an integrated neural waveform decoder.
-- English normalization and stress-marked eSpeak phonemes with punctuation retained
-- compact Transformer text encoder and deterministic duration predictor
-- monotonic alignment, posterior encoder, and residual coupling flow during training
-- integrated adversarial waveform decoder
-- multi-resolution STFT, adversarial, feature, mel, duration, and KL supervision
+- punctuation-preserving English normalization and stress-marked phonemes
+- compact Transformer text encoder and duration model
+- latent prior and residual coupling flow
+- integrated adversarial waveform decoder at 24 kHz
+- inference-only checkpoint with training discriminators and optimizer state removed
-| Component setting | Value |
+| Component | Value |
| --- | ---: |
| Latent channels | 128 |
| Text hidden channels | 72 |
| Text encoder layers / heads | 3 / 2 |
| Feed-forward channels | 384 |
-| Posterior encoder layers | 2 |
+| Flow coupling blocks | 4 |
| Initial decoder channels | 192 |
| Upsample rates | 8, 8, 2, 2 |
| Training segment | 16,384 samples |
-The downloadable checkpoint is inference-only. Training discriminators and optimizer state are excluded from both the package and deployable parameter count.
+The model card describes the deployable architecture. The private corpus-construction and optimization recipe is not part of this open-weight release.
## Runtime
-End-to-end warm inference, including the text frontend:
+Warm end-to-end CPU inference on **AMD EPYC 7K62 48-Core Processor**, with **192 logical CPUs visible** to the isolated process and runtime-default thread settings:
-| Device | RTF lower is better | Audio / wall time | p95 utterance latency | Load time |
-| --- | ---: | ---: | ---: | ---: |
-| CPU | 0.5578 | 1.8x | 3.571 s | 0.91 s |
-| CUDA | 0.0161 | 62.0x | 0.083 s | 0.97 s |
+| RTF ↓ | Audio / wall time ↑ | Median latency | p95 latency | Cold load |
+| ---: | ---: | ---: | ---: | ---: |
+| 0.5400 | 1.85× | 2.199 s | 3.850 s | 0.82 s |
-Runtime depends strongly on CPU, PyTorch build, thread settings, and text length. Reproduce it from the JSON reports under `evaluation/runtime/`.
+Runtime varies with processor, PyTorch build, thread policy, text length, and operating system. Use the raw report rather than extrapolating these measurements to untested devices.
+
+## Fine-tuning
+
+The initial release is inference-first. The checkpoint is structurally compatible with voice and language adaptation, but the private training recipe and corpus-generation pipeline are not included. A tested public adaptation toolkit should be treated as a separate release rather than inferred from the inference package. See `docs/FINETUNING.md` for the supported status and data contract.
+
+## Repository Layout
+
+| Path | Purpose |
+| --- | --- |
+| `model.pth` | Inference-only generator checkpoint |
+| `config.json` | Deployable architecture and audio configuration |
+| `inference.py` | Public Python API and CLI |
+| `inflect_vits_frontend.py` | Normalization, phonemization, and punctuation frontend |
+| `runtime/` | Self-contained model implementation required for inference |
+| `samples/` | Held-out example generations |
+| `evaluation/final/` | Final matched benchmark reports and protocol artifacts |
+| `release_manifest.json` | File sizes and SHA-256 integrity hashes |
## Limitations
-- English only, with one fixed male voice; this release does not support voice cloning.
+- English only and one fixed male voice. This is not a zero-shot voice-cloning model.
- Unfamiliar phrasing can become flatter or less expressive.
-- Numbers and uncommon names are the weakest intelligibility categories in the release evaluation; verify critical verbalizations.
-- Homograph pronunciation depends on context available to the frontend.
-- Long passages are synthesized in chunks rather than as one unbounded sequence.
-- This model is not validated for medical, legal, emergency, or accessibility-critical use.
-- Synthetic training speech can transfer teacher cadence and biases.
+- Numbers, abbreviations, homographs, and uncommon names depend strongly on frontend normalization and context.
+- Long passages are synthesized in chunks, so transitions can differ from a single long-form model pass.
+- Stochastic variation can change pronunciation and timing; use a fixed seed for reproducibility.
+- Learned MOS predictors can reward or penalize spectral traits differently from human listeners.
+- Not validated for medical, legal, emergency, or accessibility-critical communication.
+
+## Responsible Use
+
+Do not use the included voice to impersonate a real person, deceive listeners, or create fraudulent content. Disclose synthetic speech where the context could otherwise mislead. Users are responsible for applicable laws and the license.
## Integrity and License
-`release_manifest.json` records the source checkpoint, deployable parameter count, byte sizes, and SHA-256 hashes. Original Inflect work is licensed under Apache-2.0; bundled third-party portions retain the notices in `THIRD_PARTY_NOTICES.md`. Do not use the included voice to impersonate a real person or deceive listeners.
+`release_manifest.json` records file sizes and SHA-256 hashes. Original Inflect code and weights in this repository are licensed under Apache-2.0; bundled third-party components retain their own notices in `THIRD_PARTY_NOTICES.md`.
## Citation
diff --git a/assets/evidence/asr-robustness.svg b/assets/evidence/asr-robustness.svg
new file mode 100644
index 0000000000000000000000000000000000000000..343d6e2941d6b39d05bc38fe9b0a0c68a2bc0cac
--- /dev/null
+++ b/assets/evidence/asr-robustness.svg
@@ -0,0 +1,65 @@
+
diff --git a/assets/evidence/category-semantic-wer.svg b/assets/evidence/category-semantic-wer.svg
new file mode 100644
index 0000000000000000000000000000000000000000..a2dbc941e42e5162d515727103bc62037ddc264f
--- /dev/null
+++ b/assets/evidence/category-semantic-wer.svg
@@ -0,0 +1,94 @@
+
diff --git a/assets/evidence/cpu-throughput.svg b/assets/evidence/cpu-throughput.svg
new file mode 100644
index 0000000000000000000000000000000000000000..eca0a79e7bd0045f6fbc32c2230f9eccc86a43ad
--- /dev/null
+++ b/assets/evidence/cpu-throughput.svg
@@ -0,0 +1,51 @@
+
diff --git a/assets/evidence/evidence-summary.json b/assets/evidence/evidence-summary.json
new file mode 100644
index 0000000000000000000000000000000000000000..957e6d0d7d59cfe98a048d24929393cb15c7b356
--- /dev/null
+++ b/assets/evidence/evidence-summary.json
@@ -0,0 +1,1990 @@
+{
+ "format": "inflect_v2_release_evidence_v1",
+ "systems": {
+ "inflect-micro-v2": {
+ "label": "Inflect-Micro-v2",
+ "weight_mb": 37.529995,
+ "semantic_wer": {
+ "wer": 0.019293218720152817,
+ "ci": [
+ 0.014622733476311172,
+ 0.024179280518969924
+ ],
+ "categories": {
+ "conversational": {
+ "raw": {
+ "utterances": 25,
+ "errors": 0,
+ "words": 296,
+ "wer": 0.0
+ },
+ "semantic": {
+ "utterances": 25,
+ "errors": 0,
+ "words": 309,
+ "wer": 0.0
+ }
+ },
+ "descriptive": {
+ "raw": {
+ "utterances": 16,
+ "errors": 2,
+ "words": 217,
+ "wer": 0.009216589861751152
+ },
+ "semantic": {
+ "utterances": 16,
+ "errors": 1,
+ "words": 226,
+ "wer": 0.004424778761061947
+ }
+ },
+ "emotional": {
+ "raw": {
+ "utterances": 24,
+ "errors": 0,
+ "words": 314,
+ "wer": 0.0
+ },
+ "semantic": {
+ "utterances": 24,
+ "errors": 0,
+ "words": 329,
+ "wer": 0.0
+ }
+ },
+ "homographs": {
+ "raw": {
+ "utterances": 6,
+ "errors": 2,
+ "words": 70,
+ "wer": 0.02857142857142857
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 2,
+ "words": 70,
+ "wer": 0.02857142857142857
+ }
+ },
+ "long": {
+ "raw": {
+ "utterances": 15,
+ "errors": 3,
+ "words": 412,
+ "wer": 0.007281553398058253
+ },
+ "semantic": {
+ "utterances": 15,
+ "errors": 3,
+ "words": 432,
+ "wer": 0.006944444444444444
+ }
+ },
+ "long_range": {
+ "raw": {
+ "utterances": 6,
+ "errors": 0,
+ "words": 121,
+ "wer": 0.0
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 0,
+ "words": 121,
+ "wer": 0.0
+ }
+ },
+ "names_places": {
+ "raw": {
+ "utterances": 6,
+ "errors": 4,
+ "words": 57,
+ "wer": 0.07017543859649122
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 4,
+ "words": 57,
+ "wer": 0.07017543859649122
+ }
+ },
+ "narrative": {
+ "raw": {
+ "utterances": 24,
+ "errors": 2,
+ "words": 342,
+ "wer": 0.005847953216374269
+ },
+ "semantic": {
+ "utterances": 24,
+ "errors": 1,
+ "words": 352,
+ "wer": 0.002840909090909091
+ }
+ },
+ "numbers": {
+ "raw": {
+ "utterances": 6,
+ "errors": 33,
+ "words": 81,
+ "wer": 0.4074074074074074
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 8,
+ "words": 64,
+ "wer": 0.125
+ }
+ },
+ "phonetic_balance": {
+ "raw": {
+ "utterances": 300,
+ "errors": 82,
+ "words": 2605,
+ "wer": 0.03147792706333973
+ },
+ "semantic": {
+ "utterances": 300,
+ "errors": 68,
+ "words": 2627,
+ "wer": 0.02588503996954701
+ }
+ },
+ "punctuation": {
+ "raw": {
+ "utterances": 6,
+ "errors": 0,
+ "words": 65,
+ "wer": 0.0
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 0,
+ "words": 65,
+ "wer": 0.0
+ }
+ },
+ "question": {
+ "raw": {
+ "utterances": 15,
+ "errors": 0,
+ "words": 163,
+ "wer": 0.0
+ },
+ "semantic": {
+ "utterances": 15,
+ "errors": 0,
+ "words": 165,
+ "wer": 0.0
+ }
+ },
+ "rare_words": {
+ "raw": {
+ "utterances": 6,
+ "errors": 3,
+ "words": 57,
+ "wer": 0.05263157894736842
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 3,
+ "words": 57,
+ "wer": 0.05263157894736842
+ }
+ },
+ "short": {
+ "raw": {
+ "utterances": 24,
+ "errors": 3,
+ "words": 98,
+ "wer": 0.030612244897959183
+ },
+ "semantic": {
+ "utterances": 24,
+ "errors": 3,
+ "words": 105,
+ "wer": 0.02857142857142857
+ }
+ },
+ "technical": {
+ "raw": {
+ "utterances": 21,
+ "errors": 12,
+ "words": 251,
+ "wer": 0.04780876494023904
+ },
+ "semantic": {
+ "utterances": 21,
+ "errors": 8,
+ "words": 256,
+ "wer": 0.03125
+ }
+ }
+ }
+ },
+ "utmos22": {
+ "mean": 4.395012595653534,
+ "ci": [
+ 4.3812509111166,
+ 4.4079569326162344
+ ]
+ }
+ },
+ "inflect-nano-v2": {
+ "label": "Inflect-Nano-v2",
+ "weight_mb": 15.971083,
+ "semantic_wer": {
+ "wer": 0.024832855778414518,
+ "ci": [
+ 0.019525267993874426,
+ 0.030727866333781446
+ ],
+ "categories": {
+ "conversational": {
+ "raw": {
+ "utterances": 25,
+ "errors": 5,
+ "words": 296,
+ "wer": 0.016891891891891893
+ },
+ "semantic": {
+ "utterances": 25,
+ "errors": 3,
+ "words": 309,
+ "wer": 0.009708737864077669
+ }
+ },
+ "descriptive": {
+ "raw": {
+ "utterances": 16,
+ "errors": 2,
+ "words": 217,
+ "wer": 0.009216589861751152
+ },
+ "semantic": {
+ "utterances": 16,
+ "errors": 1,
+ "words": 226,
+ "wer": 0.004424778761061947
+ }
+ },
+ "emotional": {
+ "raw": {
+ "utterances": 24,
+ "errors": 1,
+ "words": 314,
+ "wer": 0.0031847133757961785
+ },
+ "semantic": {
+ "utterances": 24,
+ "errors": 1,
+ "words": 329,
+ "wer": 0.00303951367781155
+ }
+ },
+ "homographs": {
+ "raw": {
+ "utterances": 6,
+ "errors": 3,
+ "words": 70,
+ "wer": 0.04285714285714286
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 3,
+ "words": 70,
+ "wer": 0.04285714285714286
+ }
+ },
+ "long": {
+ "raw": {
+ "utterances": 15,
+ "errors": 3,
+ "words": 412,
+ "wer": 0.007281553398058253
+ },
+ "semantic": {
+ "utterances": 15,
+ "errors": 3,
+ "words": 432,
+ "wer": 0.006944444444444444
+ }
+ },
+ "long_range": {
+ "raw": {
+ "utterances": 6,
+ "errors": 1,
+ "words": 121,
+ "wer": 0.008264462809917356
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 1,
+ "words": 121,
+ "wer": 0.008264462809917356
+ }
+ },
+ "names_places": {
+ "raw": {
+ "utterances": 6,
+ "errors": 3,
+ "words": 57,
+ "wer": 0.05263157894736842
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 3,
+ "words": 57,
+ "wer": 0.05263157894736842
+ }
+ },
+ "narrative": {
+ "raw": {
+ "utterances": 24,
+ "errors": 5,
+ "words": 342,
+ "wer": 0.014619883040935672
+ },
+ "semantic": {
+ "utterances": 24,
+ "errors": 4,
+ "words": 352,
+ "wer": 0.011363636363636364
+ }
+ },
+ "numbers": {
+ "raw": {
+ "utterances": 6,
+ "errors": 33,
+ "words": 81,
+ "wer": 0.4074074074074074
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 6,
+ "words": 64,
+ "wer": 0.09375
+ }
+ },
+ "phonetic_balance": {
+ "raw": {
+ "utterances": 300,
+ "errors": 96,
+ "words": 2605,
+ "wer": 0.03685220729366603
+ },
+ "semantic": {
+ "utterances": 300,
+ "errors": 83,
+ "words": 2627,
+ "wer": 0.031594975256947086
+ }
+ },
+ "punctuation": {
+ "raw": {
+ "utterances": 6,
+ "errors": 0,
+ "words": 65,
+ "wer": 0.0
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 0,
+ "words": 65,
+ "wer": 0.0
+ }
+ },
+ "question": {
+ "raw": {
+ "utterances": 15,
+ "errors": 1,
+ "words": 163,
+ "wer": 0.006134969325153374
+ },
+ "semantic": {
+ "utterances": 15,
+ "errors": 1,
+ "words": 165,
+ "wer": 0.006060606060606061
+ }
+ },
+ "rare_words": {
+ "raw": {
+ "utterances": 6,
+ "errors": 5,
+ "words": 57,
+ "wer": 0.08771929824561403
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 5,
+ "words": 57,
+ "wer": 0.08771929824561403
+ }
+ },
+ "short": {
+ "raw": {
+ "utterances": 24,
+ "errors": 8,
+ "words": 98,
+ "wer": 0.08163265306122448
+ },
+ "semantic": {
+ "utterances": 24,
+ "errors": 8,
+ "words": 105,
+ "wer": 0.0761904761904762
+ }
+ },
+ "technical": {
+ "raw": {
+ "utterances": 21,
+ "errors": 12,
+ "words": 251,
+ "wer": 0.04780876494023904
+ },
+ "semantic": {
+ "utterances": 21,
+ "errors": 8,
+ "words": 256,
+ "wer": 0.03125
+ }
+ }
+ }
+ },
+ "utmos22": {
+ "mean": 4.3861008615493775,
+ "ci": [
+ 4.372184220826625,
+ 4.3987462292552
+ ]
+ }
+ },
+ "kitten-nano-bruno": {
+ "label": "KittenTTS Nano \u00b7 Bruno",
+ "weight_mb": 56.767095,
+ "semantic_wer": {
+ "wer": 0.009742120343839542,
+ "ci": [
+ 0.006622516556291391,
+ 0.013268998793727383
+ ],
+ "categories": {
+ "conversational": {
+ "raw": {
+ "utterances": 25,
+ "errors": 0,
+ "words": 296,
+ "wer": 0.0
+ },
+ "semantic": {
+ "utterances": 25,
+ "errors": 0,
+ "words": 309,
+ "wer": 0.0
+ }
+ },
+ "descriptive": {
+ "raw": {
+ "utterances": 16,
+ "errors": 1,
+ "words": 217,
+ "wer": 0.004608294930875576
+ },
+ "semantic": {
+ "utterances": 16,
+ "errors": 0,
+ "words": 226,
+ "wer": 0.0
+ }
+ },
+ "emotional": {
+ "raw": {
+ "utterances": 24,
+ "errors": 0,
+ "words": 314,
+ "wer": 0.0
+ },
+ "semantic": {
+ "utterances": 24,
+ "errors": 0,
+ "words": 329,
+ "wer": 0.0
+ }
+ },
+ "homographs": {
+ "raw": {
+ "utterances": 6,
+ "errors": 1,
+ "words": 70,
+ "wer": 0.014285714285714285
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 1,
+ "words": 70,
+ "wer": 0.014285714285714285
+ }
+ },
+ "long": {
+ "raw": {
+ "utterances": 15,
+ "errors": 2,
+ "words": 412,
+ "wer": 0.0048543689320388345
+ },
+ "semantic": {
+ "utterances": 15,
+ "errors": 2,
+ "words": 432,
+ "wer": 0.004629629629629629
+ }
+ },
+ "long_range": {
+ "raw": {
+ "utterances": 6,
+ "errors": 0,
+ "words": 121,
+ "wer": 0.0
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 0,
+ "words": 121,
+ "wer": 0.0
+ }
+ },
+ "names_places": {
+ "raw": {
+ "utterances": 6,
+ "errors": 2,
+ "words": 57,
+ "wer": 0.03508771929824561
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 2,
+ "words": 57,
+ "wer": 0.03508771929824561
+ }
+ },
+ "narrative": {
+ "raw": {
+ "utterances": 24,
+ "errors": 3,
+ "words": 342,
+ "wer": 0.008771929824561403
+ },
+ "semantic": {
+ "utterances": 24,
+ "errors": 2,
+ "words": 352,
+ "wer": 0.005681818181818182
+ }
+ },
+ "numbers": {
+ "raw": {
+ "utterances": 6,
+ "errors": 32,
+ "words": 81,
+ "wer": 0.3950617283950617
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 5,
+ "words": 64,
+ "wer": 0.078125
+ }
+ },
+ "phonetic_balance": {
+ "raw": {
+ "utterances": 300,
+ "errors": 39,
+ "words": 2605,
+ "wer": 0.014971209213051824
+ },
+ "semantic": {
+ "utterances": 300,
+ "errors": 30,
+ "words": 2627,
+ "wer": 0.011419870574800152
+ }
+ },
+ "punctuation": {
+ "raw": {
+ "utterances": 6,
+ "errors": 0,
+ "words": 65,
+ "wer": 0.0
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 0,
+ "words": 65,
+ "wer": 0.0
+ }
+ },
+ "question": {
+ "raw": {
+ "utterances": 15,
+ "errors": 0,
+ "words": 163,
+ "wer": 0.0
+ },
+ "semantic": {
+ "utterances": 15,
+ "errors": 0,
+ "words": 165,
+ "wer": 0.0
+ }
+ },
+ "rare_words": {
+ "raw": {
+ "utterances": 6,
+ "errors": 2,
+ "words": 57,
+ "wer": 0.03508771929824561
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 2,
+ "words": 57,
+ "wer": 0.03508771929824561
+ }
+ },
+ "short": {
+ "raw": {
+ "utterances": 24,
+ "errors": 2,
+ "words": 98,
+ "wer": 0.02040816326530612
+ },
+ "semantic": {
+ "utterances": 24,
+ "errors": 2,
+ "words": 105,
+ "wer": 0.01904761904761905
+ }
+ },
+ "technical": {
+ "raw": {
+ "utterances": 21,
+ "errors": 9,
+ "words": 251,
+ "wer": 0.035856573705179286
+ },
+ "semantic": {
+ "utterances": 21,
+ "errors": 5,
+ "words": 256,
+ "wer": 0.01953125
+ }
+ }
+ }
+ },
+ "utmos22": {
+ "mean": 4.21684304523468,
+ "ci": [
+ 4.193502563333512,
+ 4.239324071252346
+ ]
+ }
+ },
+ "kitten-nano-hugo": {
+ "label": "KittenTTS Nano \u00b7 Hugo",
+ "weight_mb": 56.767095,
+ "semantic_wer": {
+ "wer": 0.011461318051575931,
+ "ci": [
+ 0.007995319812792511,
+ 0.01523625843780135
+ ],
+ "categories": {
+ "conversational": {
+ "raw": {
+ "utterances": 25,
+ "errors": 1,
+ "words": 296,
+ "wer": 0.0033783783783783786
+ },
+ "semantic": {
+ "utterances": 25,
+ "errors": 1,
+ "words": 309,
+ "wer": 0.003236245954692557
+ }
+ },
+ "descriptive": {
+ "raw": {
+ "utterances": 16,
+ "errors": 1,
+ "words": 217,
+ "wer": 0.004608294930875576
+ },
+ "semantic": {
+ "utterances": 16,
+ "errors": 0,
+ "words": 226,
+ "wer": 0.0
+ }
+ },
+ "emotional": {
+ "raw": {
+ "utterances": 24,
+ "errors": 0,
+ "words": 314,
+ "wer": 0.0
+ },
+ "semantic": {
+ "utterances": 24,
+ "errors": 0,
+ "words": 329,
+ "wer": 0.0
+ }
+ },
+ "homographs": {
+ "raw": {
+ "utterances": 6,
+ "errors": 1,
+ "words": 70,
+ "wer": 0.014285714285714285
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 1,
+ "words": 70,
+ "wer": 0.014285714285714285
+ }
+ },
+ "long": {
+ "raw": {
+ "utterances": 15,
+ "errors": 3,
+ "words": 412,
+ "wer": 0.007281553398058253
+ },
+ "semantic": {
+ "utterances": 15,
+ "errors": 3,
+ "words": 432,
+ "wer": 0.006944444444444444
+ }
+ },
+ "long_range": {
+ "raw": {
+ "utterances": 6,
+ "errors": 0,
+ "words": 121,
+ "wer": 0.0
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 0,
+ "words": 121,
+ "wer": 0.0
+ }
+ },
+ "names_places": {
+ "raw": {
+ "utterances": 6,
+ "errors": 2,
+ "words": 57,
+ "wer": 0.03508771929824561
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 2,
+ "words": 57,
+ "wer": 0.03508771929824561
+ }
+ },
+ "narrative": {
+ "raw": {
+ "utterances": 24,
+ "errors": 6,
+ "words": 342,
+ "wer": 0.017543859649122806
+ },
+ "semantic": {
+ "utterances": 24,
+ "errors": 5,
+ "words": 352,
+ "wer": 0.014204545454545454
+ }
+ },
+ "numbers": {
+ "raw": {
+ "utterances": 6,
+ "errors": 32,
+ "words": 81,
+ "wer": 0.3950617283950617
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 5,
+ "words": 64,
+ "wer": 0.078125
+ }
+ },
+ "phonetic_balance": {
+ "raw": {
+ "utterances": 300,
+ "errors": 39,
+ "words": 2605,
+ "wer": 0.014971209213051824
+ },
+ "semantic": {
+ "utterances": 300,
+ "errors": 30,
+ "words": 2627,
+ "wer": 0.011419870574800152
+ }
+ },
+ "punctuation": {
+ "raw": {
+ "utterances": 6,
+ "errors": 0,
+ "words": 65,
+ "wer": 0.0
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 0,
+ "words": 65,
+ "wer": 0.0
+ }
+ },
+ "question": {
+ "raw": {
+ "utterances": 15,
+ "errors": 0,
+ "words": 163,
+ "wer": 0.0
+ },
+ "semantic": {
+ "utterances": 15,
+ "errors": 0,
+ "words": 165,
+ "wer": 0.0
+ }
+ },
+ "rare_words": {
+ "raw": {
+ "utterances": 6,
+ "errors": 3,
+ "words": 57,
+ "wer": 0.05263157894736842
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 3,
+ "words": 57,
+ "wer": 0.05263157894736842
+ }
+ },
+ "short": {
+ "raw": {
+ "utterances": 24,
+ "errors": 4,
+ "words": 98,
+ "wer": 0.04081632653061224
+ },
+ "semantic": {
+ "utterances": 24,
+ "errors": 4,
+ "words": 105,
+ "wer": 0.0380952380952381
+ }
+ },
+ "technical": {
+ "raw": {
+ "utterances": 21,
+ "errors": 12,
+ "words": 251,
+ "wer": 0.04780876494023904
+ },
+ "semantic": {
+ "utterances": 21,
+ "errors": 6,
+ "words": 256,
+ "wer": 0.0234375
+ }
+ }
+ }
+ },
+ "utmos22": {
+ "mean": 4.191078673839569,
+ "ci": [
+ 4.168552933418751,
+ 4.212353882300854
+ ]
+ }
+ },
+ "piper-ryan-low": {
+ "label": "Piper \u00b7 Ryan Low",
+ "weight_mb": 63.104526,
+ "semantic_wer": {
+ "wer": 0.023304680038204392,
+ "ci": [
+ 0.018028169014084508,
+ 0.02901655306718598
+ ],
+ "categories": {
+ "conversational": {
+ "raw": {
+ "utterances": 25,
+ "errors": 8,
+ "words": 296,
+ "wer": 0.02702702702702703
+ },
+ "semantic": {
+ "utterances": 25,
+ "errors": 5,
+ "words": 309,
+ "wer": 0.016181229773462782
+ }
+ },
+ "descriptive": {
+ "raw": {
+ "utterances": 16,
+ "errors": 3,
+ "words": 217,
+ "wer": 0.013824884792626729
+ },
+ "semantic": {
+ "utterances": 16,
+ "errors": 1,
+ "words": 226,
+ "wer": 0.004424778761061947
+ }
+ },
+ "emotional": {
+ "raw": {
+ "utterances": 24,
+ "errors": 1,
+ "words": 314,
+ "wer": 0.0031847133757961785
+ },
+ "semantic": {
+ "utterances": 24,
+ "errors": 1,
+ "words": 329,
+ "wer": 0.00303951367781155
+ }
+ },
+ "homographs": {
+ "raw": {
+ "utterances": 6,
+ "errors": 1,
+ "words": 70,
+ "wer": 0.014285714285714285
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 1,
+ "words": 70,
+ "wer": 0.014285714285714285
+ }
+ },
+ "long": {
+ "raw": {
+ "utterances": 15,
+ "errors": 3,
+ "words": 412,
+ "wer": 0.007281553398058253
+ },
+ "semantic": {
+ "utterances": 15,
+ "errors": 3,
+ "words": 432,
+ "wer": 0.006944444444444444
+ }
+ },
+ "long_range": {
+ "raw": {
+ "utterances": 6,
+ "errors": 1,
+ "words": 121,
+ "wer": 0.008264462809917356
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 1,
+ "words": 121,
+ "wer": 0.008264462809917356
+ }
+ },
+ "names_places": {
+ "raw": {
+ "utterances": 6,
+ "errors": 2,
+ "words": 57,
+ "wer": 0.03508771929824561
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 2,
+ "words": 57,
+ "wer": 0.03508771929824561
+ }
+ },
+ "narrative": {
+ "raw": {
+ "utterances": 24,
+ "errors": 1,
+ "words": 342,
+ "wer": 0.0029239766081871343
+ },
+ "semantic": {
+ "utterances": 24,
+ "errors": 0,
+ "words": 352,
+ "wer": 0.0
+ }
+ },
+ "numbers": {
+ "raw": {
+ "utterances": 6,
+ "errors": 32,
+ "words": 81,
+ "wer": 0.3950617283950617
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 6,
+ "words": 64,
+ "wer": 0.09375
+ }
+ },
+ "phonetic_balance": {
+ "raw": {
+ "utterances": 300,
+ "errors": 94,
+ "words": 2605,
+ "wer": 0.03608445297504798
+ },
+ "semantic": {
+ "utterances": 300,
+ "errors": 83,
+ "words": 2627,
+ "wer": 0.031594975256947086
+ }
+ },
+ "punctuation": {
+ "raw": {
+ "utterances": 6,
+ "errors": 1,
+ "words": 65,
+ "wer": 0.015384615384615385
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 1,
+ "words": 65,
+ "wer": 0.015384615384615385
+ }
+ },
+ "question": {
+ "raw": {
+ "utterances": 15,
+ "errors": 0,
+ "words": 163,
+ "wer": 0.0
+ },
+ "semantic": {
+ "utterances": 15,
+ "errors": 0,
+ "words": 165,
+ "wer": 0.0
+ }
+ },
+ "rare_words": {
+ "raw": {
+ "utterances": 6,
+ "errors": 4,
+ "words": 57,
+ "wer": 0.07017543859649122
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 4,
+ "words": 57,
+ "wer": 0.07017543859649122
+ }
+ },
+ "short": {
+ "raw": {
+ "utterances": 24,
+ "errors": 7,
+ "words": 98,
+ "wer": 0.07142857142857142
+ },
+ "semantic": {
+ "utterances": 24,
+ "errors": 8,
+ "words": 105,
+ "wer": 0.0761904761904762
+ }
+ },
+ "technical": {
+ "raw": {
+ "utterances": 21,
+ "errors": 10,
+ "words": 251,
+ "wer": 0.0398406374501992
+ },
+ "semantic": {
+ "utterances": 21,
+ "errors": 6,
+ "words": 256,
+ "wer": 0.0234375
+ }
+ }
+ }
+ },
+ "utmos22": {
+ "mean": 4.289378658294678,
+ "ci": [
+ 4.271461564397812,
+ 4.306235534155369
+ ]
+ }
+ },
+ "piper-danny-low": {
+ "label": "Piper \u00b7 Danny Low",
+ "weight_mb": 63.104526,
+ "semantic_wer": {
+ "wer": 0.018147086914995225,
+ "ci": [
+ 0.013549618320610687,
+ 0.022975077881619937
+ ],
+ "categories": {
+ "conversational": {
+ "raw": {
+ "utterances": 25,
+ "errors": 2,
+ "words": 296,
+ "wer": 0.006756756756756757
+ },
+ "semantic": {
+ "utterances": 25,
+ "errors": 2,
+ "words": 309,
+ "wer": 0.006472491909385114
+ }
+ },
+ "descriptive": {
+ "raw": {
+ "utterances": 16,
+ "errors": 2,
+ "words": 217,
+ "wer": 0.009216589861751152
+ },
+ "semantic": {
+ "utterances": 16,
+ "errors": 1,
+ "words": 226,
+ "wer": 0.004424778761061947
+ }
+ },
+ "emotional": {
+ "raw": {
+ "utterances": 24,
+ "errors": 1,
+ "words": 314,
+ "wer": 0.0031847133757961785
+ },
+ "semantic": {
+ "utterances": 24,
+ "errors": 1,
+ "words": 329,
+ "wer": 0.00303951367781155
+ }
+ },
+ "homographs": {
+ "raw": {
+ "utterances": 6,
+ "errors": 1,
+ "words": 70,
+ "wer": 0.014285714285714285
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 1,
+ "words": 70,
+ "wer": 0.014285714285714285
+ }
+ },
+ "long": {
+ "raw": {
+ "utterances": 15,
+ "errors": 7,
+ "words": 412,
+ "wer": 0.01699029126213592
+ },
+ "semantic": {
+ "utterances": 15,
+ "errors": 7,
+ "words": 432,
+ "wer": 0.016203703703703703
+ }
+ },
+ "long_range": {
+ "raw": {
+ "utterances": 6,
+ "errors": 0,
+ "words": 121,
+ "wer": 0.0
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 0,
+ "words": 121,
+ "wer": 0.0
+ }
+ },
+ "names_places": {
+ "raw": {
+ "utterances": 6,
+ "errors": 3,
+ "words": 57,
+ "wer": 0.05263157894736842
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 3,
+ "words": 57,
+ "wer": 0.05263157894736842
+ }
+ },
+ "narrative": {
+ "raw": {
+ "utterances": 24,
+ "errors": 1,
+ "words": 342,
+ "wer": 0.0029239766081871343
+ },
+ "semantic": {
+ "utterances": 24,
+ "errors": 0,
+ "words": 352,
+ "wer": 0.0
+ }
+ },
+ "numbers": {
+ "raw": {
+ "utterances": 6,
+ "errors": 33,
+ "words": 81,
+ "wer": 0.4074074074074074
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 7,
+ "words": 64,
+ "wer": 0.109375
+ }
+ },
+ "phonetic_balance": {
+ "raw": {
+ "utterances": 300,
+ "errors": 73,
+ "words": 2605,
+ "wer": 0.02802303262955854
+ },
+ "semantic": {
+ "utterances": 300,
+ "errors": 61,
+ "words": 2627,
+ "wer": 0.02322040350209364
+ }
+ },
+ "punctuation": {
+ "raw": {
+ "utterances": 6,
+ "errors": 0,
+ "words": 65,
+ "wer": 0.0
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 0,
+ "words": 65,
+ "wer": 0.0
+ }
+ },
+ "question": {
+ "raw": {
+ "utterances": 15,
+ "errors": 2,
+ "words": 163,
+ "wer": 0.012269938650306749
+ },
+ "semantic": {
+ "utterances": 15,
+ "errors": 2,
+ "words": 165,
+ "wer": 0.012121212121212121
+ }
+ },
+ "rare_words": {
+ "raw": {
+ "utterances": 6,
+ "errors": 2,
+ "words": 57,
+ "wer": 0.03508771929824561
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 2,
+ "words": 57,
+ "wer": 0.03508771929824561
+ }
+ },
+ "short": {
+ "raw": {
+ "utterances": 24,
+ "errors": 3,
+ "words": 98,
+ "wer": 0.030612244897959183
+ },
+ "semantic": {
+ "utterances": 24,
+ "errors": 3,
+ "words": 105,
+ "wer": 0.02857142857142857
+ }
+ },
+ "technical": {
+ "raw": {
+ "utterances": 21,
+ "errors": 9,
+ "words": 251,
+ "wer": 0.035856573705179286
+ },
+ "semantic": {
+ "utterances": 21,
+ "errors": 5,
+ "words": 256,
+ "wer": 0.01953125
+ }
+ }
+ }
+ },
+ "utmos22": {
+ "mean": 4.194989757061005,
+ "ci": [
+ 4.179823832964897,
+ 4.20924334000349
+ ]
+ }
+ },
+ "supertonic3-james-3step": {
+ "label": "Supertonic 3 \u00b7 James \u00b7 3-step",
+ "weight_mb": 398.075273,
+ "semantic_wer": {
+ "wer": 6.928175740210124,
+ "ci": [
+ 5.861435406698565,
+ 8.043254987161761
+ ],
+ "categories": {
+ "conversational": {
+ "raw": {
+ "utterances": 25,
+ "errors": 1295,
+ "words": 296,
+ "wer": 4.375
+ },
+ "semantic": {
+ "utterances": 25,
+ "errors": 1449,
+ "words": 309,
+ "wer": 4.689320388349515
+ }
+ },
+ "descriptive": {
+ "raw": {
+ "utterances": 16,
+ "errors": 902,
+ "words": 217,
+ "wer": 4.15668202764977
+ },
+ "semantic": {
+ "utterances": 16,
+ "errors": 998,
+ "words": 226,
+ "wer": 4.415929203539823
+ }
+ },
+ "emotional": {
+ "raw": {
+ "utterances": 24,
+ "errors": 2189,
+ "words": 314,
+ "wer": 6.971337579617835
+ },
+ "semantic": {
+ "utterances": 24,
+ "errors": 2306,
+ "words": 329,
+ "wer": 7.009118541033435
+ }
+ },
+ "homographs": {
+ "raw": {
+ "utterances": 6,
+ "errors": 70,
+ "words": 70,
+ "wer": 1.0
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 73,
+ "words": 70,
+ "wer": 1.042857142857143
+ }
+ },
+ "long": {
+ "raw": {
+ "utterances": 15,
+ "errors": 1113,
+ "words": 412,
+ "wer": 2.7014563106796117
+ },
+ "semantic": {
+ "utterances": 15,
+ "errors": 1190,
+ "words": 432,
+ "wer": 2.7546296296296298
+ }
+ },
+ "long_range": {
+ "raw": {
+ "utterances": 6,
+ "errors": 153,
+ "words": 121,
+ "wer": 1.2644628099173554
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 159,
+ "words": 121,
+ "wer": 1.3140495867768596
+ }
+ },
+ "names_places": {
+ "raw": {
+ "utterances": 6,
+ "errors": 666,
+ "words": 57,
+ "wer": 11.68421052631579
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 828,
+ "words": 57,
+ "wer": 14.526315789473685
+ }
+ },
+ "narrative": {
+ "raw": {
+ "utterances": 24,
+ "errors": 2415,
+ "words": 342,
+ "wer": 7.06140350877193
+ },
+ "semantic": {
+ "utterances": 24,
+ "errors": 2562,
+ "words": 352,
+ "wer": 7.278409090909091
+ }
+ },
+ "numbers": {
+ "raw": {
+ "utterances": 6,
+ "errors": 94,
+ "words": 81,
+ "wer": 1.1604938271604939
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 91,
+ "words": 64,
+ "wer": 1.421875
+ }
+ },
+ "phonetic_balance": {
+ "raw": {
+ "utterances": 300,
+ "errors": 19668,
+ "words": 2605,
+ "wer": 7.550095969289827
+ },
+ "semantic": {
+ "utterances": 300,
+ "errors": 21028,
+ "words": 2627,
+ "wer": 8.00456794822992
+ }
+ },
+ "punctuation": {
+ "raw": {
+ "utterances": 6,
+ "errors": 803,
+ "words": 65,
+ "wer": 12.353846153846154
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 807,
+ "words": 65,
+ "wer": 12.415384615384616
+ }
+ },
+ "question": {
+ "raw": {
+ "utterances": 15,
+ "errors": 1412,
+ "words": 163,
+ "wer": 8.662576687116564
+ },
+ "semantic": {
+ "utterances": 15,
+ "errors": 1529,
+ "words": 165,
+ "wer": 9.266666666666667
+ }
+ },
+ "rare_words": {
+ "raw": {
+ "utterances": 6,
+ "errors": 819,
+ "words": 57,
+ "wer": 14.368421052631579
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 820,
+ "words": 57,
+ "wer": 14.385964912280702
+ }
+ },
+ "short": {
+ "raw": {
+ "utterances": 24,
+ "errors": 436,
+ "words": 98,
+ "wer": 4.448979591836735
+ },
+ "semantic": {
+ "utterances": 24,
+ "errors": 447,
+ "words": 105,
+ "wer": 4.257142857142857
+ }
+ },
+ "technical": {
+ "raw": {
+ "utterances": 21,
+ "errors": 1976,
+ "words": 251,
+ "wer": 7.872509960159363
+ },
+ "semantic": {
+ "utterances": 21,
+ "errors": 1982,
+ "words": 256,
+ "wer": 7.7421875
+ }
+ }
+ }
+ },
+ "utmos22": {
+ "mean": 2.4712263617515564,
+ "ci": [
+ 2.412739716243744,
+ 2.5290928330957887
+ ]
+ }
+ },
+ "supertonic3-james-8step": {
+ "label": "Supertonic 3 \u00b7 James \u00b7 8-step",
+ "weight_mb": 398.075273,
+ "semantic_wer": {
+ "wer": 6.225214899713467,
+ "ci": [
+ 5.2127497621313035,
+ 7.2831739961759085
+ ],
+ "categories": {
+ "conversational": {
+ "raw": {
+ "utterances": 25,
+ "errors": 1746,
+ "words": 296,
+ "wer": 5.898648648648648
+ },
+ "semantic": {
+ "utterances": 25,
+ "errors": 1812,
+ "words": 309,
+ "wer": 5.864077669902913
+ }
+ },
+ "descriptive": {
+ "raw": {
+ "utterances": 16,
+ "errors": 1230,
+ "words": 217,
+ "wer": 5.668202764976958
+ },
+ "semantic": {
+ "utterances": 16,
+ "errors": 1234,
+ "words": 226,
+ "wer": 5.460176991150442
+ }
+ },
+ "emotional": {
+ "raw": {
+ "utterances": 24,
+ "errors": 1592,
+ "words": 314,
+ "wer": 5.070063694267516
+ },
+ "semantic": {
+ "utterances": 24,
+ "errors": 1670,
+ "words": 329,
+ "wer": 5.075987841945289
+ }
+ },
+ "homographs": {
+ "raw": {
+ "utterances": 6,
+ "errors": 326,
+ "words": 70,
+ "wer": 4.6571428571428575
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 413,
+ "words": 70,
+ "wer": 5.9
+ }
+ },
+ "long": {
+ "raw": {
+ "utterances": 15,
+ "errors": 1459,
+ "words": 412,
+ "wer": 3.54126213592233
+ },
+ "semantic": {
+ "utterances": 15,
+ "errors": 1468,
+ "words": 432,
+ "wer": 3.3981481481481484
+ }
+ },
+ "long_range": {
+ "raw": {
+ "utterances": 6,
+ "errors": 113,
+ "words": 121,
+ "wer": 0.9338842975206612
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 113,
+ "words": 121,
+ "wer": 0.9338842975206612
+ }
+ },
+ "names_places": {
+ "raw": {
+ "utterances": 6,
+ "errors": 381,
+ "words": 57,
+ "wer": 6.684210526315789
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 476,
+ "words": 57,
+ "wer": 8.350877192982455
+ }
+ },
+ "narrative": {
+ "raw": {
+ "utterances": 24,
+ "errors": 1338,
+ "words": 342,
+ "wer": 3.912280701754386
+ },
+ "semantic": {
+ "utterances": 24,
+ "errors": 1344,
+ "words": 352,
+ "wer": 3.8181818181818183
+ }
+ },
+ "numbers": {
+ "raw": {
+ "utterances": 6,
+ "errors": 90,
+ "words": 81,
+ "wer": 1.1111111111111112
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 87,
+ "words": 64,
+ "wer": 1.359375
+ }
+ },
+ "phonetic_balance": {
+ "raw": {
+ "utterances": 300,
+ "errors": 18274,
+ "words": 2605,
+ "wer": 7.014971209213051
+ },
+ "semantic": {
+ "utterances": 300,
+ "errors": 19044,
+ "words": 2627,
+ "wer": 7.2493338408831365
+ }
+ },
+ "punctuation": {
+ "raw": {
+ "utterances": 6,
+ "errors": 361,
+ "words": 65,
+ "wer": 5.553846153846154
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 438,
+ "words": 65,
+ "wer": 6.7384615384615385
+ }
+ },
+ "question": {
+ "raw": {
+ "utterances": 15,
+ "errors": 1153,
+ "words": 163,
+ "wer": 7.07361963190184
+ },
+ "semantic": {
+ "utterances": 15,
+ "errors": 1267,
+ "words": 165,
+ "wer": 7.678787878787879
+ }
+ },
+ "rare_words": {
+ "raw": {
+ "utterances": 6,
+ "errors": 438,
+ "words": 57,
+ "wer": 7.684210526315789
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 444,
+ "words": 57,
+ "wer": 7.7894736842105265
+ }
+ },
+ "short": {
+ "raw": {
+ "utterances": 24,
+ "errors": 1005,
+ "words": 98,
+ "wer": 10.255102040816327
+ },
+ "semantic": {
+ "utterances": 24,
+ "errors": 1057,
+ "words": 105,
+ "wer": 10.066666666666666
+ }
+ },
+ "technical": {
+ "raw": {
+ "utterances": 21,
+ "errors": 1585,
+ "words": 251,
+ "wer": 6.314741035856573
+ },
+ "semantic": {
+ "utterances": 21,
+ "errors": 1722,
+ "words": 256,
+ "wer": 6.7265625
+ }
+ }
+ }
+ },
+ "utmos22": {
+ "mean": 4.29470673418045,
+ "ci": [
+ 4.281224336981774,
+ 4.307890736854076
+ ]
+ }
+ }
+ },
+ "asr_robustness": {
+ "inflect-micro-v2": {
+ "whisper-large-v3": {
+ "wer": 0.019293218720152817,
+ "ci": [
+ 0.014622733476311172,
+ 0.024179280518969924
+ ]
+ },
+ "wav2vec2-large-lv60k-960h": {
+ "wer": 0.048137535816618914,
+ "ci": [
+ 0.03997704667176741,
+ 0.05712124077926991
+ ]
+ }
+ },
+ "inflect-nano-v2": {
+ "whisper-large-v3": {
+ "wer": 0.024832855778414518,
+ "ci": [
+ 0.019525267993874426,
+ 0.030727866333781446
+ ]
+ },
+ "wav2vec2-large-lv60k-960h": {
+ "wer": 0.05673352435530086,
+ "ci": [
+ 0.04819277108433735,
+ 0.06602129719264278
+ ]
+ }
+ },
+ "kitten-nano-bruno": {
+ "whisper-large-v3": {
+ "wer": 0.009742120343839542,
+ "ci": [
+ 0.006622516556291391,
+ 0.013268998793727383
+ ]
+ },
+ "wav2vec2-large-lv60k-960h": {
+ "wer": 0.03553008595988539,
+ "ci": [
+ 0.028724626579854463,
+ 0.04309006211180124
+ ]
+ }
+ },
+ "kitten-nano-hugo": {
+ "whisper-large-v3": {
+ "wer": 0.011461318051575931,
+ "ci": [
+ 0.007995319812792511,
+ 0.01523625843780135
+ ]
+ },
+ "wav2vec2-large-lv60k-960h": {
+ "wer": 0.03629417382999045,
+ "ci": [
+ 0.029451395572666024,
+ 0.043798449612403104
+ ]
+ }
+ },
+ "piper-ryan-low": {
+ "whisper-large-v3": {
+ "wer": 0.023304680038204392,
+ "ci": [
+ 0.018028169014084508,
+ 0.02901655306718598
+ ]
+ },
+ "wav2vec2-large-lv60k-960h": {
+ "wer": 0.0670487106017192,
+ "ci": [
+ 0.05838601710673113,
+ 0.07638619620007756
+ ]
+ }
+ },
+ "piper-danny-low": {
+ "whisper-large-v3": {
+ "wer": 0.018147086914995225,
+ "ci": [
+ 0.013549618320610687,
+ 0.022975077881619937
+ ]
+ },
+ "wav2vec2-large-lv60k-960h": {
+ "wer": 0.053104106972301814,
+ "ci": [
+ 0.04440632742519535,
+ 0.06227176628867961
+ ]
+ }
+ },
+ "supertonic3-james-3step": {
+ "whisper-large-v3": {
+ "wer": 6.928175740210124,
+ "ci": [
+ 5.861435406698565,
+ 8.043254987161761
+ ]
+ },
+ "wav2vec2-large-lv60k-960h": {
+ "wer": 0.1346704871060172,
+ "ci": [
+ 0.11719324026451139,
+ 0.15343099460292983
+ ]
+ }
+ },
+ "supertonic3-james-8step": {
+ "whisper-large-v3": {
+ "wer": 6.225214899713467,
+ "ci": [
+ 5.2127497621313035,
+ 7.2831739961759085
+ ]
+ },
+ "wav2vec2-large-lv60k-960h": {
+ "wer": 0.037631327602674305,
+ "ci": [
+ 0.030763314747164008,
+ 0.044854881266490766
+ ]
+ }
+ }
+ },
+ "family_averages": {
+ "kitten": {
+ "semantic_wer": 0.010601719197707736,
+ "utmos22": 4.2039608595371245
+ },
+ "piper": {
+ "semantic_wer": 0.02072588347659981,
+ "utmos22": 4.242184207677841
+ }
+ },
+ "notes": [
+ "Family averages are equal-weight macro averages over the two named voices.",
+ "Supertonic 3-step and 8-step remain separate.",
+ "UTMOS22 and WER are not combined into one score."
+ ]
+}
diff --git a/assets/evidence/human-preference.svg b/assets/evidence/human-preference.svg
new file mode 100644
index 0000000000000000000000000000000000000000..ce8f76ec6c351d955865f56720f69cf1784c67b7
--- /dev/null
+++ b/assets/evidence/human-preference.svg
@@ -0,0 +1,62 @@
+
diff --git a/assets/evidence/model-footprint.svg b/assets/evidence/model-footprint.svg
new file mode 100644
index 0000000000000000000000000000000000000000..4b4cb2db8b8f1dbb8ae1fcd94a016d3bb1e65780
--- /dev/null
+++ b/assets/evidence/model-footprint.svg
@@ -0,0 +1,41 @@
+
diff --git a/assets/evidence/quality-vs-cpu-speed.svg b/assets/evidence/quality-vs-cpu-speed.svg
new file mode 100644
index 0000000000000000000000000000000000000000..4933c2ddf00d15e2d6190ca0d7b7e703a1df9f4f
--- /dev/null
+++ b/assets/evidence/quality-vs-cpu-speed.svg
@@ -0,0 +1,55 @@
+
diff --git a/assets/evidence/quality-vs-footprint.svg b/assets/evidence/quality-vs-footprint.svg
new file mode 100644
index 0000000000000000000000000000000000000000..d9a8a8e791543c0ff1cf3d1fe21389c27b9dbffd
--- /dev/null
+++ b/assets/evidence/quality-vs-footprint.svg
@@ -0,0 +1,63 @@
+
diff --git a/assets/evidence/semantic-wer.svg b/assets/evidence/semantic-wer.svg
new file mode 100644
index 0000000000000000000000000000000000000000..ac1da836c1d9ecbbfa5c4910a36c12f4e92106e9
--- /dev/null
+++ b/assets/evidence/semantic-wer.svg
@@ -0,0 +1,63 @@
+
diff --git a/assets/inflect-v2-repository-hero.png b/assets/inflect-v2-repository-hero.png
new file mode 100644
index 0000000000000000000000000000000000000000..a7c38313926ebca70e7a48ea4ac6acee1332df51
--- /dev/null
+++ b/assets/inflect-v2-repository-hero.png
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0b33f6acc09095b13274654b196e1d6f522323a4108cbf6d71862edd86472de3
+size 1609634
diff --git a/assets/matched-evaluation.svg b/assets/matched-evaluation.svg
deleted file mode 100644
index 0050b12bb580ca31e671a6a0b4f95525f130fc85..0000000000000000000000000000000000000000
--- a/assets/matched-evaluation.svg
+++ /dev/null
@@ -1,32 +0,0 @@
-
diff --git a/config.json b/config.json
index f6d664d35aaf810fe53e7007c48453ede7031c25..e3c049121673683eaf8ebd9f189ded5fdfa7cffa 100644
--- a/config.json
+++ b/config.json
@@ -1,49 +1,9 @@
{
+ "format": "inflect_v2_inference_config_v1",
"train": {
- "log_interval": 25,
- "eval_interval": 4000,
- "max_steps": 156000,
- "seed": 20260720,
- "epochs": 1000,
- "learning_rate": 1e-05,
- "resume_learning_rate": 1e-05,
- "betas": [
- 0.8,
- 0.99
- ],
- "eps": 1e-09,
- "batch_size": 64,
- "num_workers": 12,
- "fp16_run": true,
- "lr_decay": 0.999875,
- "segment_size": 16384,
- "init_lr_ratio": 1,
- "warmup_epochs": 0,
- "c_mel": 45,
- "c_kl": 1.0,
- "c_stft": 1.5,
- "stft_resolutions": [
- [
- 1024,
- 256,
- 1024
- ],
- [
- 512,
- 128,
- 512
- ],
- [
- 256,
- 64,
- 256
- ]
- ],
- "freeze_linguistic": true
+ "segment_size": 16384
},
"data": {
- "training_files": "filelists/inflect_qwenfull_train_inflect_frontend.txt",
- "validation_files": "filelists/inflect_qwenfull_dev_inflect_frontend.txt",
"text_cleaners": [],
"max_wav_value": 32768.0,
"sampling_rate": 24000,
@@ -103,6 +63,7 @@
],
"n_layers_q": 2,
"use_spectral_norm": false,
- "use_sdp": false
+ "use_sdp": false,
+ "inference_only": true
}
}
diff --git a/docs/API.md b/docs/API.md
new file mode 100644
index 0000000000000000000000000000000000000000..30c6e943aa2ac868eb967ccfb9d3c0a3bb7a3986
--- /dev/null
+++ b/docs/API.md
@@ -0,0 +1,23 @@
+# Python API
+
+## `InflectTTS(model_dir, device="cpu")`
+
+Loads one released checkpoint and its configuration. Construct the engine once and reuse it for multiple utterances.
+
+## `synthesize(text, *, speed=1.0, variation=0.667, seed=0)`
+
+Returns a mono `numpy.float32` waveform at 24 kHz. Empty text raises `ValueError`. `speed` must be between `0.5` and `2.0`; `variation` must be between `0.0` and `1.0`. Long input is split at punctuation-aware boundaries, and each chunk uses `seed + chunk_index`.
+
+## `save(text, output, **kwargs)`
+
+Synthesizes and writes a WAV file, creating parent directories as needed. It returns the output `Path`.
+
+```python
+from inference import InflectTTS
+
+tts = InflectTTS(".", device="cpu")
+waveform = tts.synthesize("A compact model can still speak clearly.", seed=7)
+tts.save("This is written directly to disk.", "out.wav", seed=7)
+```
+
+The API is intentionally small. It does not expose speaker cloning, streaming state, SSML, or language switching in the v2 launch package.
diff --git a/docs/DEPLOYMENT.md b/docs/DEPLOYMENT.md
new file mode 100644
index 0000000000000000000000000000000000000000..d4efdcda803eca8de95e08fac50871b840260fb6
--- /dev/null
+++ b/docs/DEPLOYMENT.md
@@ -0,0 +1,29 @@
+# Deployment notes
+
+## Installation
+
+Use Python 3.10 or newer in a clean virtual environment, install `requirements.txt`, then run `inference.py` from the model repository. The package is tested with PyTorch CPU and CUDA execution; device-specific performance depends on the local PyTorch build.
+
+## Reproducibility
+
+Keep the checkpoint, `config.json`, frontend files, and `runtime/` directory together. Record `Inflect-Nano-v2`, the release revision, `speed`, `variation`, and `seed` with generated artifacts. Verify files against `release_manifest.json` when moving the package between machines.
+
+## Long text
+
+The runtime splits long text at sentence punctuation and then at safe intra-sentence boundaries. Chunks are synthesized independently and joined with punctuation-dependent silence. This prevents unbounded inference but means a long paragraph is not one globally planned prosody pass.
+
+## Production boundaries
+
+- Instantiate `InflectTTS` once and reuse it; loading per request wastes time.
+- Serialize calls to one model instance unless the surrounding application explicitly manages device memory and random seeds.
+- Validate empty or untrusted input before synthesis.
+- Set a fixed seed for deterministic caching and tests.
+- Measure on the actual target device before claiming real-time operation.
+
+## Troubleshooting
+
+- Missing phonemizer errors: reinstall `phonemizer` and `espeakng-loader` from `requirements.txt`.
+- Unexpected pronunciation: spell out ambiguous abbreviations or add context; the release frontend is English-only.
+- Flat or unstable delivery: keep `variation` near the default before changing speed.
+- Long pauses: inspect source punctuation because chunk boundaries intentionally follow punctuation.
+- Slow first request: model loading is measured separately from warm synthesis and should not happen per utterance.
diff --git a/docs/EVALUATION.md b/docs/EVALUATION.md
new file mode 100644
index 0000000000000000000000000000000000000000..7efbb7e2f5a2d86ae04888305e3fbc92f2388b93
--- /dev/null
+++ b/docs/EVALUATION.md
@@ -0,0 +1,34 @@
+# Final evaluation protocol
+
+## Matched corpus
+
+All automated headline results use 500 identical unseen English prompts per system. The set combines 200 manually audited stress prompts with 300 deterministic CMU Arctic sentences. Exact normalized text was excluded against 87,362 training transcripts. The frozen prompt manifest and exclusion report are included under `evaluation/final/raw/`.
+
+## Systems
+
+- Inflect-Micro-v2 and Inflect-Nano-v2, fixed release checkpoints.
+- KittenTTS Nano, Bruno and Hugo reported separately.
+- Piper low, Ryan and Danny reported separately.
+- Supertonic 3 James at 3 and 8 steps, reported separately.
+
+Family averages for Kitten and Piper are equal-weight macro averages of the two named voices. The two Supertonic step counts are never pooled.
+
+## Intelligibility
+
+Headline semantic WER uses Whisper-large-v3. A second complete pass with wav2vec2 Large LV-60K tests whether conclusions are sensitive to one ASR family. References and hypotheses are normalized into equivalent spoken English forms so formatting differences such as `12.6` versus `twelve point six` are not automatically counted as speech errors. Both reports include per-clip rows, prompt-category breakdowns, and 10,000-sample bootstrap intervals. ASR-family results are never averaged into a single headline value.
+
+## Predicted quality
+
+UTMOS22 is run over every generated clip with 10,000-sample bootstrap intervals. It is a learned quality predictor, not human MOS, and is reported separately from WER and human preference.
+
+## Runtime
+
+Runtime reports use one named host, the same 50-prompt subset, warm end-to-end synthesis, one isolated process per system, each public runtime's default thread/provider behavior, and separate model-load timing. RTF, audio-seconds per wall-second, median latency, p95 latency, and per-prompt rows are retained. Cross-machine numbers are not mixed.
+
+## Human evidence
+
+The community study was blind, randomized left/right order, and stored pairwise preferences rather than user identity. Preference rate is `(wins + 0.5 × ties) / appearances`. It is descriptive community evidence, not a formal laboratory MOS study.
+
+## Reporting rules
+
+No aggregate score combines WER, UTMOS22, speed, size, and human preference. Confidence intervals, raw reports, model weight footprint, and known limitations accompany headline values.
diff --git a/docs/FINETUNING.md b/docs/FINETUNING.md
new file mode 100644
index 0000000000000000000000000000000000000000..fa6755b20524d74a0ed4d9dc2579194d4c3841e0
--- /dev/null
+++ b/docs/FINETUNING.md
@@ -0,0 +1,27 @@
+# Fine-tuning and adaptation status
+
+`Inflect-Nano-v2` is an open-weight, inference-first release. The repository does not ship the private corpus-construction recipe, teacher setup, discriminator checkpoints, optimizer state, or a supported training command.
+
+## What is supported
+
+- Local inference from the released generator checkpoint.
+- Deterministic regression testing through a fixed text, seed, speed, and variation value.
+- Inspection of the complete deployable architecture and frontend.
+- Research adaptation in a separate VITS-compatible training environment, at the user's own risk.
+
+## What is not yet supported
+
+- A one-command new-voice trainer.
+- Zero-shot voice cloning or speaker embeddings.
+- Language adaptation without rebuilding the symbol inventory and frontend.
+- Resuming the release checkpoint as if it contained discriminators or optimizer state.
+
+## Data contract for future adapters
+
+A defensible voice-adaptation toolkit must require a UTF-8 manifest with one audio path and one verified transcript per row; mono, consistently sampled, unclipped speech; held-out prompts that never enter training; speaker-identity and transcript audits; and an inference-only export that removes all training-only state. Duration, silence, clipping, and ASR checks must run before training rather than after a failed model.
+
+## Evaluation contract
+
+Any adapted checkpoint should be compared with the frozen release on identical unseen prompts, a fixed seed policy, semantic WER, signal diagnostics, a learned quality predictor, and a blinded human listening test. Improving speaker similarity while reducing intelligibility is not a successful adaptation.
+
+This file intentionally does not pretend that an unvalidated training recipe is a supported feature. A public adapter will be documented separately if it passes the same package and evaluation gates as the release models.
diff --git a/evaluation/asr.json b/evaluation/asr.json
deleted file mode 100644
index 0f8217c56198332d60241777327127472d6cef05..0000000000000000000000000000000000000000
--- a/evaluation/asr.json
+++ /dev/null
@@ -1,19319 +0,0 @@
-{
- "format": "inflect_tts_multi_asr_wer_v1",
- "manifest": "release_polish_overnight_20260720/final_large200/combined_manifest.jsonl",
- "asr_models": {
- "whisper-large-v3": "openai/whisper-large-v3"
- },
- "normalization": {
- "raw": "lowercase_punctuation_stripped",
- "semantic": "transformers_whisper_EnglishTextNormalizer_then_lexical"
- },
- "systems": {
- "inflect-micro-v2-finalist": {
- "asr": {
- "whisper-large-v3": {
- "raw": {
- "utterances": 200,
- "errors": 65,
- "words": 2544,
- "wer": 0.025550314465408806
- },
- "raw_95ci": [
- 0.013276064037485357,
- 0.0397168698387731
- ],
- "semantic": {
- "utterances": 200,
- "errors": 32,
- "words": 2608,
- "wer": 0.012269938650306749
- },
- "semantic_95ci": [
- 0.006958657388456815,
- 0.018608113137327874
- ],
- "categories": {
- "conversational": {
- "raw": {
- "utterances": 25,
- "errors": 1,
- "words": 296,
- "wer": 0.0033783783783783786
- },
- "semantic": {
- "utterances": 25,
- "errors": 0,
- "words": 309,
- "wer": 0.0
- }
- },
- "descriptive": {
- "raw": {
- "utterances": 16,
- "errors": 2,
- "words": 217,
- "wer": 0.009216589861751152
- },
- "semantic": {
- "utterances": 16,
- "errors": 0,
- "words": 226,
- "wer": 0.0
- }
- },
- "emotional": {
- "raw": {
- "utterances": 24,
- "errors": 0,
- "words": 314,
- "wer": 0.0
- },
- "semantic": {
- "utterances": 24,
- "errors": 0,
- "words": 329,
- "wer": 0.0
- }
- },
- "homographs": {
- "raw": {
- "utterances": 6,
- "errors": 1,
- "words": 70,
- "wer": 0.014285714285714285
- },
- "semantic": {
- "utterances": 6,
- "errors": 1,
- "words": 70,
- "wer": 0.014285714285714285
- }
- },
- "long": {
- "raw": {
- "utterances": 15,
- "errors": 1,
- "words": 412,
- "wer": 0.0024271844660194173
- },
- "semantic": {
- "utterances": 15,
- "errors": 1,
- "words": 432,
- "wer": 0.0023148148148148147
- }
- },
- "long_range": {
- "raw": {
- "utterances": 6,
- "errors": 0,
- "words": 121,
- "wer": 0.0
- },
- "semantic": {
- "utterances": 6,
- "errors": 0,
- "words": 121,
- "wer": 0.0
- }
- },
- "names_places": {
- "raw": {
- "utterances": 6,
- "errors": 4,
- "words": 57,
- "wer": 0.07017543859649122
- },
- "semantic": {
- "utterances": 6,
- "errors": 4,
- "words": 57,
- "wer": 0.07017543859649122
- }
- },
- "narrative": {
- "raw": {
- "utterances": 24,
- "errors": 2,
- "words": 342,
- "wer": 0.005847953216374269
- },
- "semantic": {
- "utterances": 24,
- "errors": 1,
- "words": 352,
- "wer": 0.002840909090909091
- }
- },
- "numbers": {
- "raw": {
- "utterances": 6,
- "errors": 35,
- "words": 81,
- "wer": 0.43209876543209874
- },
- "semantic": {
- "utterances": 6,
- "errors": 10,
- "words": 64,
- "wer": 0.15625
- }
- },
- "punctuation": {
- "raw": {
- "utterances": 6,
- "errors": 0,
- "words": 65,
- "wer": 0.0
- },
- "semantic": {
- "utterances": 6,
- "errors": 0,
- "words": 65,
- "wer": 0.0
- }
- },
- "question": {
- "raw": {
- "utterances": 15,
- "errors": 0,
- "words": 163,
- "wer": 0.0
- },
- "semantic": {
- "utterances": 15,
- "errors": 0,
- "words": 165,
- "wer": 0.0
- }
- },
- "rare_words": {
- "raw": {
- "utterances": 6,
- "errors": 3,
- "words": 57,
- "wer": 0.05263157894736842
- },
- "semantic": {
- "utterances": 6,
- "errors": 3,
- "words": 57,
- "wer": 0.05263157894736842
- }
- },
- "short": {
- "raw": {
- "utterances": 24,
- "errors": 4,
- "words": 98,
- "wer": 0.04081632653061224
- },
- "semantic": {
- "utterances": 24,
- "errors": 4,
- "words": 105,
- "wer": 0.0380952380952381
- }
- },
- "technical": {
- "raw": {
- "utterances": 21,
- "errors": 12,
- "words": 251,
- "wer": 0.04780876494023904
- },
- "semantic": {
- "utterances": 21,
- "errors": 8,
- "words": 256,
- "wer": 0.03125
- }
- }
- }
- }
- },
- "cross_asr_mean_semantic_wer": 0.012269938650306749,
- "cross_asr_worst_semantic_wer": 0.012269938650306749
- },
- "inflect-nano-v2-finalist": {
- "asr": {
- "whisper-large-v3": {
- "raw": {
- "utterances": 200,
- "errors": 75,
- "words": 2544,
- "wer": 0.0294811320754717
- },
- "raw_95ci": [
- 0.016894609814963796,
- 0.04385618332674832
- ],
- "semantic": {
- "utterances": 200,
- "errors": 40,
- "words": 2608,
- "wer": 0.015337423312883436
- },
- "semantic_95ci": [
- 0.009104704097116844,
- 0.022605363984674328
- ],
- "categories": {
- "conversational": {
- "raw": {
- "utterances": 25,
- "errors": 4,
- "words": 296,
- "wer": 0.013513513513513514
- },
- "semantic": {
- "utterances": 25,
- "errors": 1,
- "words": 309,
- "wer": 0.003236245954692557
- }
- },
- "descriptive": {
- "raw": {
- "utterances": 16,
- "errors": 1,
- "words": 217,
- "wer": 0.004608294930875576
- },
- "semantic": {
- "utterances": 16,
- "errors": 0,
- "words": 226,
- "wer": 0.0
- }
- },
- "emotional": {
- "raw": {
- "utterances": 24,
- "errors": 0,
- "words": 314,
- "wer": 0.0
- },
- "semantic": {
- "utterances": 24,
- "errors": 0,
- "words": 329,
- "wer": 0.0
- }
- },
- "homographs": {
- "raw": {
- "utterances": 6,
- "errors": 4,
- "words": 70,
- "wer": 0.05714285714285714
- },
- "semantic": {
- "utterances": 6,
- "errors": 4,
- "words": 70,
- "wer": 0.05714285714285714
- }
- },
- "long": {
- "raw": {
- "utterances": 15,
- "errors": 2,
- "words": 412,
- "wer": 0.0048543689320388345
- },
- "semantic": {
- "utterances": 15,
- "errors": 2,
- "words": 432,
- "wer": 0.004629629629629629
- }
- },
- "long_range": {
- "raw": {
- "utterances": 6,
- "errors": 1,
- "words": 121,
- "wer": 0.008264462809917356
- },
- "semantic": {
- "utterances": 6,
- "errors": 1,
- "words": 121,
- "wer": 0.008264462809917356
- }
- },
- "names_places": {
- "raw": {
- "utterances": 6,
- "errors": 5,
- "words": 57,
- "wer": 0.08771929824561403
- },
- "semantic": {
- "utterances": 6,
- "errors": 5,
- "words": 57,
- "wer": 0.08771929824561403
- }
- },
- "narrative": {
- "raw": {
- "utterances": 24,
- "errors": 5,
- "words": 342,
- "wer": 0.014619883040935672
- },
- "semantic": {
- "utterances": 24,
- "errors": 4,
- "words": 352,
- "wer": 0.011363636363636364
- }
- },
- "numbers": {
- "raw": {
- "utterances": 6,
- "errors": 34,
- "words": 81,
- "wer": 0.41975308641975306
- },
- "semantic": {
- "utterances": 6,
- "errors": 8,
- "words": 64,
- "wer": 0.125
- }
- },
- "punctuation": {
- "raw": {
- "utterances": 6,
- "errors": 0,
- "words": 65,
- "wer": 0.0
- },
- "semantic": {
- "utterances": 6,
- "errors": 0,
- "words": 65,
- "wer": 0.0
- }
- },
- "question": {
- "raw": {
- "utterances": 15,
- "errors": 0,
- "words": 163,
- "wer": 0.0
- },
- "semantic": {
- "utterances": 15,
- "errors": 0,
- "words": 165,
- "wer": 0.0
- }
- },
- "rare_words": {
- "raw": {
- "utterances": 6,
- "errors": 2,
- "words": 57,
- "wer": 0.03508771929824561
- },
- "semantic": {
- "utterances": 6,
- "errors": 2,
- "words": 57,
- "wer": 0.03508771929824561
- }
- },
- "short": {
- "raw": {
- "utterances": 24,
- "errors": 5,
- "words": 98,
- "wer": 0.05102040816326531
- },
- "semantic": {
- "utterances": 24,
- "errors": 5,
- "words": 105,
- "wer": 0.047619047619047616
- }
- },
- "technical": {
- "raw": {
- "utterances": 21,
- "errors": 12,
- "words": 251,
- "wer": 0.04780876494023904
- },
- "semantic": {
- "utterances": 21,
- "errors": 8,
- "words": 256,
- "wer": 0.03125
- }
- }
- }
- }
- },
- "cross_asr_mean_semantic_wer": 0.015337423312883436,
- "cross_asr_worst_semantic_wer": 0.015337423312883436
- },
- "kitten-nano-bruno": {
- "asr": {
- "whisper-large-v3": {
- "raw": {
- "utterances": 200,
- "errors": 57,
- "words": 2544,
- "wer": 0.02240566037735849
- },
- "raw_95ci": [
- 0.011168727562824093,
- 0.03543788187372709
- ],
- "semantic": {
- "utterances": 200,
- "errors": 25,
- "words": 2608,
- "wer": 0.009585889570552147
- },
- "semantic_95ci": [
- 0.004838709677419355,
- 0.01526433358153388
- ],
- "categories": {
- "conversational": {
- "raw": {
- "utterances": 25,
- "errors": 1,
- "words": 296,
- "wer": 0.0033783783783783786
- },
- "semantic": {
- "utterances": 25,
- "errors": 1,
- "words": 309,
- "wer": 0.003236245954692557
- }
- },
- "descriptive": {
- "raw": {
- "utterances": 16,
- "errors": 1,
- "words": 217,
- "wer": 0.004608294930875576
- },
- "semantic": {
- "utterances": 16,
- "errors": 0,
- "words": 226,
- "wer": 0.0
- }
- },
- "emotional": {
- "raw": {
- "utterances": 24,
- "errors": 0,
- "words": 314,
- "wer": 0.0
- },
- "semantic": {
- "utterances": 24,
- "errors": 0,
- "words": 329,
- "wer": 0.0
- }
- },
- "homographs": {
- "raw": {
- "utterances": 6,
- "errors": 1,
- "words": 70,
- "wer": 0.014285714285714285
- },
- "semantic": {
- "utterances": 6,
- "errors": 1,
- "words": 70,
- "wer": 0.014285714285714285
- }
- },
- "long": {
- "raw": {
- "utterances": 15,
- "errors": 1,
- "words": 412,
- "wer": 0.0024271844660194173
- },
- "semantic": {
- "utterances": 15,
- "errors": 1,
- "words": 432,
- "wer": 0.0023148148148148147
- }
- },
- "long_range": {
- "raw": {
- "utterances": 6,
- "errors": 0,
- "words": 121,
- "wer": 0.0
- },
- "semantic": {
- "utterances": 6,
- "errors": 0,
- "words": 121,
- "wer": 0.0
- }
- },
- "names_places": {
- "raw": {
- "utterances": 6,
- "errors": 3,
- "words": 57,
- "wer": 0.05263157894736842
- },
- "semantic": {
- "utterances": 6,
- "errors": 3,
- "words": 57,
- "wer": 0.05263157894736842
- }
- },
- "narrative": {
- "raw": {
- "utterances": 24,
- "errors": 5,
- "words": 342,
- "wer": 0.014619883040935672
- },
- "semantic": {
- "utterances": 24,
- "errors": 4,
- "words": 352,
- "wer": 0.011363636363636364
- }
- },
- "numbers": {
- "raw": {
- "utterances": 6,
- "errors": 32,
- "words": 81,
- "wer": 0.3950617283950617
- },
- "semantic": {
- "utterances": 6,
- "errors": 6,
- "words": 64,
- "wer": 0.09375
- }
- },
- "punctuation": {
- "raw": {
- "utterances": 6,
- "errors": 0,
- "words": 65,
- "wer": 0.0
- },
- "semantic": {
- "utterances": 6,
- "errors": 0,
- "words": 65,
- "wer": 0.0
- }
- },
- "question": {
- "raw": {
- "utterances": 15,
- "errors": 0,
- "words": 163,
- "wer": 0.0
- },
- "semantic": {
- "utterances": 15,
- "errors": 0,
- "words": 165,
- "wer": 0.0
- }
- },
- "rare_words": {
- "raw": {
- "utterances": 6,
- "errors": 2,
- "words": 57,
- "wer": 0.03508771929824561
- },
- "semantic": {
- "utterances": 6,
- "errors": 2,
- "words": 57,
- "wer": 0.03508771929824561
- }
- },
- "short": {
- "raw": {
- "utterances": 24,
- "errors": 2,
- "words": 98,
- "wer": 0.02040816326530612
- },
- "semantic": {
- "utterances": 24,
- "errors": 2,
- "words": 105,
- "wer": 0.01904761904761905
- }
- },
- "technical": {
- "raw": {
- "utterances": 21,
- "errors": 9,
- "words": 251,
- "wer": 0.035856573705179286
- },
- "semantic": {
- "utterances": 21,
- "errors": 5,
- "words": 256,
- "wer": 0.01953125
- }
- }
- }
- }
- },
- "cross_asr_mean_semantic_wer": 0.009585889570552147,
- "cross_asr_worst_semantic_wer": 0.009585889570552147
- },
- "piper-ryan-low": {
- "asr": {
- "whisper-large-v3": {
- "raw": {
- "utterances": 200,
- "errors": 73,
- "words": 2544,
- "wer": 0.02869496855345912
- },
- "raw_95ci": [
- 0.016779864163004393,
- 0.042346542346542344
- ],
- "semantic": {
- "utterances": 200,
- "errors": 32,
- "words": 2608,
- "wer": 0.012269938650306749
- },
- "semantic_95ci": [
- 0.006775607811877242,
- 0.01875225387666787
- ],
- "categories": {
- "conversational": {
- "raw": {
- "utterances": 25,
- "errors": 5,
- "words": 296,
- "wer": 0.016891891891891893
- },
- "semantic": {
- "utterances": 25,
- "errors": 2,
- "words": 309,
- "wer": 0.006472491909385114
- }
- },
- "descriptive": {
- "raw": {
- "utterances": 16,
- "errors": 4,
- "words": 217,
- "wer": 0.018433179723502304
- },
- "semantic": {
- "utterances": 16,
- "errors": 2,
- "words": 226,
- "wer": 0.008849557522123894
- }
- },
- "emotional": {
- "raw": {
- "utterances": 24,
- "errors": 0,
- "words": 314,
- "wer": 0.0
- },
- "semantic": {
- "utterances": 24,
- "errors": 0,
- "words": 329,
- "wer": 0.0
- }
- },
- "homographs": {
- "raw": {
- "utterances": 6,
- "errors": 1,
- "words": 70,
- "wer": 0.014285714285714285
- },
- "semantic": {
- "utterances": 6,
- "errors": 1,
- "words": 70,
- "wer": 0.014285714285714285
- }
- },
- "long": {
- "raw": {
- "utterances": 15,
- "errors": 5,
- "words": 412,
- "wer": 0.012135922330097087
- },
- "semantic": {
- "utterances": 15,
- "errors": 2,
- "words": 432,
- "wer": 0.004629629629629629
- }
- },
- "long_range": {
- "raw": {
- "utterances": 6,
- "errors": 0,
- "words": 121,
- "wer": 0.0
- },
- "semantic": {
- "utterances": 6,
- "errors": 0,
- "words": 121,
- "wer": 0.0
- }
- },
- "names_places": {
- "raw": {
- "utterances": 6,
- "errors": 4,
- "words": 57,
- "wer": 0.07017543859649122
- },
- "semantic": {
- "utterances": 6,
- "errors": 4,
- "words": 57,
- "wer": 0.07017543859649122
- }
- },
- "narrative": {
- "raw": {
- "utterances": 24,
- "errors": 1,
- "words": 342,
- "wer": 0.0029239766081871343
- },
- "semantic": {
- "utterances": 24,
- "errors": 0,
- "words": 352,
- "wer": 0.0
- }
- },
- "numbers": {
- "raw": {
- "utterances": 6,
- "errors": 32,
- "words": 81,
- "wer": 0.3950617283950617
- },
- "semantic": {
- "utterances": 6,
- "errors": 5,
- "words": 64,
- "wer": 0.078125
- }
- },
- "punctuation": {
- "raw": {
- "utterances": 6,
- "errors": 0,
- "words": 65,
- "wer": 0.0
- },
- "semantic": {
- "utterances": 6,
- "errors": 0,
- "words": 65,
- "wer": 0.0
- }
- },
- "question": {
- "raw": {
- "utterances": 15,
- "errors": 0,
- "words": 163,
- "wer": 0.0
- },
- "semantic": {
- "utterances": 15,
- "errors": 0,
- "words": 165,
- "wer": 0.0
- }
- },
- "rare_words": {
- "raw": {
- "utterances": 6,
- "errors": 5,
- "words": 57,
- "wer": 0.08771929824561403
- },
- "semantic": {
- "utterances": 6,
- "errors": 5,
- "words": 57,
- "wer": 0.08771929824561403
- }
- },
- "short": {
- "raw": {
- "utterances": 24,
- "errors": 5,
- "words": 98,
- "wer": 0.05102040816326531
- },
- "semantic": {
- "utterances": 24,
- "errors": 4,
- "words": 105,
- "wer": 0.0380952380952381
- }
- },
- "technical": {
- "raw": {
- "utterances": 21,
- "errors": 11,
- "words": 251,
- "wer": 0.043824701195219126
- },
- "semantic": {
- "utterances": 21,
- "errors": 7,
- "words": 256,
- "wer": 0.02734375
- }
- }
- }
- }
- },
- "cross_asr_mean_semantic_wer": 0.012269938650306749,
- "cross_asr_worst_semantic_wer": 0.012269938650306749
- }
- },
- "rows": [
- {
- "id": "utmos200_001",
- "category": "conversational",
- "text": "It wasn't until later that I realized what had actually happened.",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_001.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "It wasn't until later that I realized what had actually happened.",
- "raw": {
- "normalized_reference": "it wasn't until later that i realized what had actually happened.",
- "normalized_hypothesis": "it wasn't until later that i realized what had actually happened.",
- "errors": 0,
- "words": 11,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "it was not until later that i realized what had actually happened",
- "normalized_hypothesis": "it was not until later that i realized what had actually happened",
- "errors": 0,
- "words": 12,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_002",
- "category": "descriptive",
- "text": "The light at that time of afternoon is perfect for about twenty minutes.",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_002.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "The light at that time of afternoon is perfect for about 20 minutes.",
- "raw": {
- "normalized_reference": "the light at that time of afternoon is perfect for about twenty minutes.",
- "normalized_hypothesis": "the light at that time of afternoon is perfect for about 20 minutes.",
- "errors": 1,
- "words": 13,
- "wer": 0.07692307692307693
- },
- "semantic": {
- "normalized_reference": "the light at that time of afternoon is perfect for about 20 minutes",
- "normalized_hypothesis": "the light at that time of afternoon is perfect for about 20 minutes",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_003",
- "category": "emotional",
- "text": "That was the scariest thing I've ever done, and also maybe the best.",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_003.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "That was the scariest thing I've ever done, and also maybe the best.",
- "raw": {
- "normalized_reference": "that was the scariest thing i've ever done, and also maybe the best.",
- "normalized_hypothesis": "that was the scariest thing i've ever done, and also maybe the best.",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "that was the scariest thing i have ever done and also maybe the best",
- "normalized_hypothesis": "that was the scariest thing i have ever done and also maybe the best",
- "errors": 0,
- "words": 14,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_004",
- "category": "homographs",
- "text": "After I record the record, please present the present to the bass player.",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_004.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "After I record the record, please present the present to the bass player.",
- "raw": {
- "normalized_reference": "after i record the record, please present the present to the bass player.",
- "normalized_hypothesis": "after i record the record, please present the present to the bass player.",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "after i record the record please present the present to the bass player",
- "normalized_hypothesis": "after i record the record please present the present to the bass player",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_005",
- "category": "long",
- "text": "If I'm being completely honest, I had no idea what I was doing for most of it, and I think that might be the most normal thing about the whole experience.",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_005.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "If I'm being completely honest, I had no idea what I was doing for most of it, and I think that might be the most normal thing about the whole experience.",
- "raw": {
- "normalized_reference": "if i'm being completely honest, i had no idea what i was doing for most of it, and i think that might be the most normal thing about the whole experience.",
- "normalized_hypothesis": "if i'm being completely honest, i had no idea what i was doing for most of it, and i think that might be the most normal thing about the whole experience.",
- "errors": 0,
- "words": 31,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "if i am being completely honest i had no idea what i was doing for most of it and i think that might be the most normal thing about the whole experience",
- "normalized_hypothesis": "if i am being completely honest i had no idea what i was doing for most of it and i think that might be the most normal thing about the whole experience",
- "errors": 0,
- "words": 32,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_006",
- "category": "long_range",
- "text": "The committee postponed its decision after discovering that three pages of the report had been printed in the wrong order.",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_006.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "The committee postponed its decision after discovering that three pages of the report had been printed in the wrong order.",
- "raw": {
- "normalized_reference": "the committee postponed its decision after discovering that three pages of the report had been printed in the wrong order.",
- "normalized_hypothesis": "the committee postponed its decision after discovering that three pages of the report had been printed in the wrong order.",
- "errors": 0,
- "words": 20,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "the committee postponed its decision after discovering that 3 pages of the report had been printed in the wrong order",
- "normalized_hypothesis": "the committee postponed its decision after discovering that 3 pages of the report had been printed in the wrong order",
- "errors": 0,
- "words": 20,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_007",
- "category": "names_places",
- "text": "Gwendolyn photographed the eucalyptus trees outside Ljubljana.",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_007.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Gwendolyn photographed the eucalyptus trees outside Ljubljana.",
- "raw": {
- "normalized_reference": "gwendolyn photographed the eucalyptus trees outside ljubljana.",
- "normalized_hypothesis": "gwendolyn photographed the eucalyptus trees outside ljubljana.",
- "errors": 0,
- "words": 7,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "gwendolyn photographed the eucalyptus trees outside ljubljana",
- "normalized_hypothesis": "gwendolyn photographed the eucalyptus trees outside ljubljana",
- "errors": 0,
- "words": 7,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_008",
- "category": "narrative",
- "text": "He had this theory, and the more time passes, the more I think he was right.",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_008.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "He had this theory, and the more time passes, the more I think he was right.",
- "raw": {
- "normalized_reference": "he had this theory, and the more time passes, the more i think he was right.",
- "normalized_hypothesis": "he had this theory, and the more time passes, the more i think he was right.",
- "errors": 0,
- "words": 16,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "he had this theory and the more time passes the more i think he was right",
- "normalized_hypothesis": "he had this theory and the more time passes the more i think he was right",
- "errors": 0,
- "words": 16,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_009",
- "category": "numbers",
- "text": "The package weighs twelve point six kilograms and arrived on July twenty-first.",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_009.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "The package weighs 12.6kg and arrived on July 21.",
- "raw": {
- "normalized_reference": "the package weighs twelve point six kilograms and arrived on july twenty-first.",
- "normalized_hypothesis": "the package weighs 12.6kg and arrived on july 21.",
- "errors": 6,
- "words": 13,
- "wer": 0.46153846153846156
- },
- "semantic": {
- "normalized_reference": "the package weighs 12.6 kilograms and arrived on july 21st",
- "normalized_hypothesis": "the package weighs 12.6 kg and arrived on july 21",
- "errors": 2,
- "words": 11,
- "wer": 0.18181818181818182
- }
- },
- {
- "id": "utmos200_010",
- "category": "punctuation",
- "text": "First, close the window; second, turn off the lamp; finally, lock the door.",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_010.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "First, close the window. Second, turn off the lamp. Finally, lock the door.",
- "raw": {
- "normalized_reference": "first, close the window; second, turn off the lamp; finally, lock the door.",
- "normalized_hypothesis": "first, close the window. second, turn off the lamp. finally, lock the door.",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "1st close the window 2nd turn off the lamp finally lock the door",
- "normalized_hypothesis": "1st close the window 2nd turn off the lamp finally lock the door",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_011",
- "category": "question",
- "text": "When did you first realize something was wrong?",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_011.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "When did you first realize something was wrong?",
- "raw": {
- "normalized_reference": "when did you first realize something was wrong?",
- "normalized_hypothesis": "when did you first realize something was wrong?",
- "errors": 0,
- "words": 8,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "when did you 1st realize something was wrong",
- "normalized_hypothesis": "when did you 1st realize something was wrong",
- "errors": 0,
- "words": 8,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_012",
- "category": "rare_words",
- "text": "The archaeologist catalogued a Byzantine reliquary and several anachronistic fragments.",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_012.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "the archaeologist catalogued a Byzantine reliquary and several anachronistic fragments.",
- "raw": {
- "normalized_reference": "the archaeologist catalogued a byzantine reliquary and several anachronistic fragments.",
- "normalized_hypothesis": "the archaeologist catalogued a byzantine reliquary and several anachronistic fragments.",
- "errors": 0,
- "words": 10,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "the archaeologist catalogued a byzantine reliquary and several anachronistic fragments",
- "normalized_hypothesis": "the archaeologist catalogued a byzantine reliquary and several anachronistic fragments",
- "errors": 0,
- "words": 10,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_013",
- "category": "short",
- "text": "I have no idea.",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_013.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "I have no idea.",
- "raw": {
- "normalized_reference": "i have no idea.",
- "normalized_hypothesis": "i have no idea.",
- "errors": 0,
- "words": 4,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "i have no idea",
- "normalized_hypothesis": "i have no idea",
- "errors": 0,
- "words": 4,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_014",
- "category": "technical",
- "text": "The system runs on three core components that all have to stay in sync.",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_014.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "The system runs on three core components that all have to stay in sync.",
- "raw": {
- "normalized_reference": "the system runs on three core components that all have to stay in sync.",
- "normalized_hypothesis": "the system runs on three core components that all have to stay in sync.",
- "errors": 0,
- "words": 14,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "the system runs on 3 core components that all have to stay in sync",
- "normalized_hypothesis": "the system runs on 3 core components that all have to stay in sync",
- "errors": 0,
- "words": 14,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_015",
- "category": "conversational",
- "text": "We improvised the entire second half and somehow nobody noticed.",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_015.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "We improvised the entire second half and somehow nobody noticed.",
- "raw": {
- "normalized_reference": "we improvised the entire second half and somehow nobody noticed.",
- "normalized_hypothesis": "we improvised the entire second half and somehow nobody noticed.",
- "errors": 0,
- "words": 10,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "we improvised the entire 2nd half and somehow nobody noticed",
- "normalized_hypothesis": "we improvised the entire 2nd half and somehow nobody noticed",
- "errors": 0,
- "words": 10,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_016",
- "category": "descriptive",
- "text": "There's a smell in old buildings that's impossible to describe but immediately familiar.",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_016.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "There's a smell in old buildings that's impossible to describe but immediately familiar.",
- "raw": {
- "normalized_reference": "there's a smell in old buildings that's impossible to describe but immediately familiar.",
- "normalized_hypothesis": "there's a smell in old buildings that's impossible to describe but immediately familiar.",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "there is a smell in old buildings that is impossible to describe but immediately familiar",
- "normalized_hypothesis": "there is a smell in old buildings that is impossible to describe but immediately familiar",
- "errors": 0,
- "words": 15,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_017",
- "category": "emotional",
- "text": "I specifically asked for one thing. One thing. And it still didn't happen.",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_017.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "I specifically asked for one thing, one thing, and it still didn't happen.",
- "raw": {
- "normalized_reference": "i specifically asked for one thing. one thing. and it still didn't happen.",
- "normalized_hypothesis": "i specifically asked for one thing, one thing, and it still didn't happen.",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "i specifically asked for one thing one thing and it still did not happen",
- "normalized_hypothesis": "i specifically asked for one thing one thing and it still did not happen",
- "errors": 0,
- "words": 14,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_018",
- "category": "homographs",
- "text": "They refuse to place the refuse beside the produce aisle.",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_018.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "They refused to place the refuse beside the produce aisle.",
- "raw": {
- "normalized_reference": "they refuse to place the refuse beside the produce aisle.",
- "normalized_hypothesis": "they refused to place the refuse beside the produce aisle.",
- "errors": 1,
- "words": 10,
- "wer": 0.1
- },
- "semantic": {
- "normalized_reference": "they refuse to place the refuse beside the produce aisle",
- "normalized_hypothesis": "they refused to place the refuse beside the produce aisle",
- "errors": 1,
- "words": 10,
- "wer": 0.1
- }
- },
- {
- "id": "utmos200_019",
- "category": "long",
- "text": "There were days last year when I genuinely didn't know how I was going to get through it, and somehow I did, and I still don't fully understand how.",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_019.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "There were days last year when I genuinely didn't know how I was going to get through it, and somehow I did, and I still don't fully understand how.",
- "raw": {
- "normalized_reference": "there were days last year when i genuinely didn't know how i was going to get through it, and somehow i did, and i still don't fully understand how.",
- "normalized_hypothesis": "there were days last year when i genuinely didn't know how i was going to get through it, and somehow i did, and i still don't fully understand how.",
- "errors": 0,
- "words": 29,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "there were days last year when i genuinely did not know how i was going to get through it and somehow i did and i still do not fully understand how",
- "normalized_hypothesis": "there were days last year when i genuinely did not know how i was going to get through it and somehow i did and i still do not fully understand how",
- "errors": 0,
- "words": 31,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_020",
- "category": "long_range",
- "text": "Before announcing the winner, the judge thanked the volunteers who had worked through the storm to repair the damaged stage.",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_020.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Before announcing the winner, the judge thanked the volunteers who had worked through the storm to repair the damaged stage.",
- "raw": {
- "normalized_reference": "before announcing the winner, the judge thanked the volunteers who had worked through the storm to repair the damaged stage.",
- "normalized_hypothesis": "before announcing the winner, the judge thanked the volunteers who had worked through the storm to repair the damaged stage.",
- "errors": 0,
- "words": 20,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "before announcing the winner the judge thanked the volunteers who had worked through the storm to repair the damaged stage",
- "normalized_hypothesis": "before announcing the winner the judge thanked the volunteers who had worked through the storm to repair the damaged stage",
- "errors": 0,
- "words": 20,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_021",
- "category": "names_places",
- "text": "Professor Nguyen mailed the fluorescent poster to Reykjavik on Thursday.",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_021.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Professor Nguyen mailed the fluorescent poster to Reykjavik on Thursday.",
- "raw": {
- "normalized_reference": "professor nguyen mailed the fluorescent poster to reykjavik on thursday.",
- "normalized_hypothesis": "professor nguyen mailed the fluorescent poster to reykjavik on thursday.",
- "errors": 0,
- "words": 10,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "professor nguyen mailed the fluorescent poster to reykjavik on thursday",
- "normalized_hypothesis": "professor nguyen mailed the fluorescent poster to reykjavik on thursday",
- "errors": 0,
- "words": 10,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_022",
- "category": "narrative",
- "text": "The plan was simple. The execution was anything but.",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_022.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "The plan was simple. The execution was anything but.",
- "raw": {
- "normalized_reference": "the plan was simple. the execution was anything but.",
- "normalized_hypothesis": "the plan was simple. the execution was anything but.",
- "errors": 0,
- "words": 9,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "the plan was simple the execution was anything but",
- "normalized_hypothesis": "the plan was simple the execution was anything but",
- "errors": 0,
- "words": 9,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_023",
- "category": "numbers",
- "text": "Her score improved from eighty-two percent to ninety-seven percent in two weeks.",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_023.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "GERS score improved from 82% to 97% in two weeks.",
- "raw": {
- "normalized_reference": "her score improved from eighty-two percent to ninety-seven percent in two weeks.",
- "normalized_hypothesis": "gers score improved from 82% to 97% in two weeks.",
- "errors": 7,
- "words": 14,
- "wer": 0.5
- },
- "semantic": {
- "normalized_reference": "her score improved from 82% to 97% in 2 weeks",
- "normalized_hypothesis": "gers score improved from 82% to 97% in 2 weeks",
- "errors": 1,
- "words": 10,
- "wer": 0.1
- }
- },
- {
- "id": "utmos200_024",
- "category": "punctuation",
- "text": "Why would anyone pack spoons, batteries, and a violin for one short trip?",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_024.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Why would anyone pack spoons, batteries, and a violin for one short trip?",
- "raw": {
- "normalized_reference": "why would anyone pack spoons, batteries, and a violin for one short trip?",
- "normalized_hypothesis": "why would anyone pack spoons, batteries, and a violin for one short trip?",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "why would anyone pack spoons batteries and a violin for one short trip",
- "normalized_hypothesis": "why would anyone pack spoons batteries and a violin for one short trip",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_025",
- "category": "question",
- "text": "How long have you been sitting on this?",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_025.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "How long have you been sitting on this?",
- "raw": {
- "normalized_reference": "how long have you been sitting on this?",
- "normalized_hypothesis": "how long have you been sitting on this?",
- "errors": 0,
- "words": 8,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "how long have you been sitting on this",
- "normalized_hypothesis": "how long have you been sitting on this",
- "errors": 0,
- "words": 8,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_026",
- "category": "rare_words",
- "text": "A mischievous chameleon rested beside the chrysanthemum and the bougainvillea.",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_026.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "A mischievous chameleon rested beside the Chrysanthemum and the Baugainvillia.",
- "raw": {
- "normalized_reference": "a mischievous chameleon rested beside the chrysanthemum and the bougainvillea.",
- "normalized_hypothesis": "a mischievous chameleon rested beside the chrysanthemum and the baugainvillia.",
- "errors": 1,
- "words": 10,
- "wer": 0.1
- },
- "semantic": {
- "normalized_reference": "a mischievous chameleon rested beside the chrysanthemum and the bougainvillea",
- "normalized_hypothesis": "a mischievous chameleon rested beside the chrysanthemum and the baugainvillia",
- "errors": 1,
- "words": 10,
- "wer": 0.1
- }
- },
- {
- "id": "utmos200_027",
- "category": "short",
- "text": "We don't talk about that.",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_027.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "We don't talk about that.",
- "raw": {
- "normalized_reference": "we don't talk about that.",
- "normalized_hypothesis": "we don't talk about that.",
- "errors": 0,
- "words": 5,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "we do not talk about that",
- "normalized_hypothesis": "we do not talk about that",
- "errors": 0,
- "words": 6,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_028",
- "category": "technical",
- "text": "Hash the UTF eight payload with S H A two fifty-six before opening the checkpoint.",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_028.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Cache the UTF-8 payload with SH-256 before opening the checkpoint.",
- "raw": {
- "normalized_reference": "hash the utf eight payload with s h a two fifty-six before opening the checkpoint.",
- "normalized_hypothesis": "cache the utf-8 payload with sh-256 before opening the checkpoint.",
- "errors": 8,
- "words": 16,
- "wer": 0.5
- },
- "semantic": {
- "normalized_reference": "hash the utf 8 payload with s h a 256 before opening the checkpoint",
- "normalized_hypothesis": "cache the utf 8 payload with sh 256 before opening the checkpoint",
- "errors": 4,
- "words": 14,
- "wer": 0.2857142857142857
- }
- },
- {
- "id": "utmos200_029",
- "category": "conversational",
- "text": "I thought I had it all figured out, and then this happened.",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_029.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "I thought I had it all figured out, and then this happened.",
- "raw": {
- "normalized_reference": "i thought i had it all figured out, and then this happened.",
- "normalized_hypothesis": "i thought i had it all figured out, and then this happened.",
- "errors": 0,
- "words": 12,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "i thought i had it all figured out and then this happened",
- "normalized_hypothesis": "i thought i had it all figured out and then this happened",
- "errors": 0,
- "words": 12,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_030",
- "category": "descriptive",
- "text": "The kind of quiet you only get at three in the morning in a city.",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_030.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "the kind of quiet you only get at 3 in the morning in a city.",
- "raw": {
- "normalized_reference": "the kind of quiet you only get at three in the morning in a city.",
- "normalized_hypothesis": "the kind of quiet you only get at 3 in the morning in a city.",
- "errors": 1,
- "words": 15,
- "wer": 0.06666666666666667
- },
- "semantic": {
- "normalized_reference": "the kind of quiet you only get at 3 in the morning in a city",
- "normalized_hypothesis": "the kind of quiet you only get at 3 in the morning in a city",
- "errors": 0,
- "words": 15,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_031",
- "category": "emotional",
- "text": "The joy of it was almost overwhelming, like too much good all at once.",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_031.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "The joy of it was almost overwhelming, like too much good all at once.",
- "raw": {
- "normalized_reference": "the joy of it was almost overwhelming, like too much good all at once.",
- "normalized_hypothesis": "the joy of it was almost overwhelming, like too much good all at once.",
- "errors": 0,
- "words": 14,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "the joy of it was almost overwhelming like too much good all at once",
- "normalized_hypothesis": "the joy of it was almost overwhelming like too much good all at once",
- "errors": 0,
- "words": 14,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_032",
- "category": "homographs",
- "text": "Please close the door while I stand close to the microphone.",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_032.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Please close the door while I stand close to the microphone.",
- "raw": {
- "normalized_reference": "please close the door while i stand close to the microphone.",
- "normalized_hypothesis": "please close the door while i stand close to the microphone.",
- "errors": 0,
- "words": 11,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "please close the door while i stand close to the microphone",
- "normalized_hypothesis": "please close the door while i stand close to the microphone",
- "errors": 0,
- "words": 11,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_033",
- "category": "long",
- "text": "I think the most underrated skill in the world is knowing when to stop explaining yourself and just let people draw their own conclusions.",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_033.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "I think the most underrated skill in the world is knowing when to stop explaining yourself and just let people draw their own conclusions.",
- "raw": {
- "normalized_reference": "i think the most underrated skill in the world is knowing when to stop explaining yourself and just let people draw their own conclusions.",
- "normalized_hypothesis": "i think the most underrated skill in the world is knowing when to stop explaining yourself and just let people draw their own conclusions.",
- "errors": 0,
- "words": 24,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "i think the most underrated skill in the world is knowing when to stop explaining yourself and just let people draw their own conclusions",
- "normalized_hypothesis": "i think the most underrated skill in the world is knowing when to stop explaining yourself and just let people draw their own conclusions",
- "errors": 0,
- "words": 24,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_034",
- "category": "long_range",
- "text": "When the final bus disappeared around the corner, Nadia remembered that her umbrella, notebook, and lunch were still inside the quiet office.",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_034.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "When the final bus disappeared around the corner, Nadia remembered that her umbrella, notebook, and lunch were still inside the quiet office.",
- "raw": {
- "normalized_reference": "when the final bus disappeared around the corner, nadia remembered that her umbrella, notebook, and lunch were still inside the quiet office.",
- "normalized_hypothesis": "when the final bus disappeared around the corner, nadia remembered that her umbrella, notebook, and lunch were still inside the quiet office.",
- "errors": 0,
- "words": 22,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "when the final bus disappeared around the corner nadia remembered that her umbrella notebook and lunch were still inside the quiet office",
- "normalized_hypothesis": "when the final bus disappeared around the corner nadia remembered that her umbrella notebook and lunch were still inside the quiet office",
- "errors": 0,
- "words": 22,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_035",
- "category": "names_places",
- "text": "Aurelia and Joaquin crossed the Schuylkill River before noon.",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_035.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Aurelia and Joachim crossed the Schalke river before noon.",
- "raw": {
- "normalized_reference": "aurelia and joaquin crossed the schuylkill river before noon.",
- "normalized_hypothesis": "aurelia and joachim crossed the schalke river before noon.",
- "errors": 2,
- "words": 9,
- "wer": 0.2222222222222222
- },
- "semantic": {
- "normalized_reference": "aurelia and joaquin crossed the schuylkill river before noon",
- "normalized_hypothesis": "aurelia and joachim crossed the schalke river before noon",
- "errors": 2,
- "words": 9,
- "wer": 0.2222222222222222
- }
- },
- {
- "id": "utmos200_036",
- "category": "narrative",
- "text": "The village had a name, but nobody used it. Everyone just called it home.",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_036.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "The village had a name, but nobody used it. Everyone just called it home.",
- "raw": {
- "normalized_reference": "the village had a name, but nobody used it. everyone just called it home.",
- "normalized_hypothesis": "the village had a name, but nobody used it. everyone just called it home.",
- "errors": 0,
- "words": 14,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "the village had a name but nobody used it everyone just called it home",
- "normalized_hypothesis": "the village had a name but nobody used it everyone just called it home",
- "errors": 0,
- "words": 14,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_037",
- "category": "numbers",
- "text": "Increase the gain by three decibels, wait five seconds, then reduce it by one.",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_037.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Increase the gain by 3 decibels, wait 5 seconds, then reduce it by 1.",
- "raw": {
- "normalized_reference": "increase the gain by three decibels, wait five seconds, then reduce it by one.",
- "normalized_hypothesis": "increase the gain by 3 decibels, wait 5 seconds, then reduce it by 1.",
- "errors": 3,
- "words": 14,
- "wer": 0.21428571428571427
- },
- "semantic": {
- "normalized_reference": "increase the gain by 3 decibels wait 5 seconds then reduce it by one",
- "normalized_hypothesis": "increase the gain by 3 decibels wait 5 seconds then reduce it by one",
- "errors": 0,
- "words": 14,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_038",
- "category": "punctuation",
- "text": "The result was unexpected: clear, stable, and surprisingly natural.",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_038.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "The result was unexpected, clear, stable, and surprisingly natural.",
- "raw": {
- "normalized_reference": "the result was unexpected: clear, stable, and surprisingly natural.",
- "normalized_hypothesis": "the result was unexpected, clear, stable, and surprisingly natural.",
- "errors": 0,
- "words": 9,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "the result was unexpected clear stable and surprisingly natural",
- "normalized_hypothesis": "the result was unexpected clear stable and surprisingly natural",
- "errors": 0,
- "words": 9,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_039",
- "category": "question",
- "text": "Who told you that was the right way to do it?",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_039.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Who told you that was the right way to do it?",
- "raw": {
- "normalized_reference": "who told you that was the right way to do it?",
- "normalized_hypothesis": "who told you that was the right way to do it?",
- "errors": 0,
- "words": 11,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "who told you that was the right way to do it",
- "normalized_hypothesis": "who told you that was the right way to do it",
- "errors": 0,
- "words": 11,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_040",
- "category": "rare_words",
- "text": "The meticulous cartographer examined an iridescent archipelago through binoculars.",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_040.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "The meticulous cartographer examined an iridescent archipelago through binoculars.",
- "raw": {
- "normalized_reference": "the meticulous cartographer examined an iridescent archipelago through binoculars.",
- "normalized_hypothesis": "the meticulous cartographer examined an iridescent archipelago through binoculars.",
- "errors": 0,
- "words": 9,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "the meticulous cartographer examined an iridescent archipelago through binoculars",
- "normalized_hypothesis": "the meticulous cartographer examined an iridescent archipelago through binoculars",
- "errors": 0,
- "words": 9,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_041",
- "category": "short",
- "text": "I mean. Yeah. Kind of.",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_041.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "I mean, yeah, I kind of.",
- "raw": {
- "normalized_reference": "i mean. yeah. kind of.",
- "normalized_hypothesis": "i mean, yeah, i kind of.",
- "errors": 1,
- "words": 5,
- "wer": 0.2
- },
- "semantic": {
- "normalized_reference": "i mean yeah kind of",
- "normalized_hypothesis": "i mean yeah i kind of",
- "errors": 1,
- "words": 5,
- "wer": 0.2
- }
- },
- {
- "id": "utmos200_042",
- "category": "technical",
- "text": "The logs are your best friend when something behaves unexpectedly.",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_042.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "The logs are your best friend when something behaves unexpectedly.",
- "raw": {
- "normalized_reference": "the logs are your best friend when something behaves unexpectedly.",
- "normalized_hypothesis": "the logs are your best friend when something behaves unexpectedly.",
- "errors": 0,
- "words": 10,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "the logs are your best friend when something behaves unexpectedly",
- "normalized_hypothesis": "the logs are your best friend when something behaves unexpectedly",
- "errors": 0,
- "words": 10,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_043",
- "category": "conversational",
- "text": "He kept talking and I kept nodding and neither of us was listening.",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_043.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "He kept talking, and I kept nodding, and neither of us was listening.",
- "raw": {
- "normalized_reference": "he kept talking and i kept nodding and neither of us was listening.",
- "normalized_hypothesis": "he kept talking, and i kept nodding, and neither of us was listening.",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "he kept talking and i kept nodding and neither of us was listening",
- "normalized_hypothesis": "he kept talking and i kept nodding and neither of us was listening",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_044",
- "category": "descriptive",
- "text": "It's one of those places where everyone seems to know each other already.",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_044.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "It's one of those places where everyone seems to know each other already.",
- "raw": {
- "normalized_reference": "it's one of those places where everyone seems to know each other already.",
- "normalized_hypothesis": "it's one of those places where everyone seems to know each other already.",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "it is one of those places where everyone seems to know each other already",
- "normalized_hypothesis": "it is one of those places where everyone seems to know each other already",
- "errors": 0,
- "words": 14,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_045",
- "category": "emotional",
- "text": "He said he was proud of me and nobody had ever said that to me before.",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_045.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "He said he was proud of me and nobody had ever said that to me before.",
- "raw": {
- "normalized_reference": "he said he was proud of me and nobody had ever said that to me before.",
- "normalized_hypothesis": "he said he was proud of me and nobody had ever said that to me before.",
- "errors": 0,
- "words": 16,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "he said he was proud of me and nobody had ever said that to me before",
- "normalized_hypothesis": "he said he was proud of me and nobody had ever said that to me before",
- "errors": 0,
- "words": 16,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_046",
- "category": "homographs",
- "text": "The minute detail took a minute to explain clearly.",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_046.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "The minute detail took a minute to explain clearly.",
- "raw": {
- "normalized_reference": "the minute detail took a minute to explain clearly.",
- "normalized_hypothesis": "the minute detail took a minute to explain clearly.",
- "errors": 0,
- "words": 9,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "the minute detail took a minute to explain clearly",
- "normalized_hypothesis": "the minute detail took a minute to explain clearly",
- "errors": 0,
- "words": 9,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_047",
- "category": "long",
- "text": "The strange thing about really wanting something is that getting it never quite feels the way you imagined it would, and not getting it never quite feels as bad.",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_047.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "The strange thing about really wanting something is that getting it never quite feels the way you imagined it would, and not getting it never quite feels as bad.",
- "raw": {
- "normalized_reference": "the strange thing about really wanting something is that getting it never quite feels the way you imagined it would, and not getting it never quite feels as bad.",
- "normalized_hypothesis": "the strange thing about really wanting something is that getting it never quite feels the way you imagined it would, and not getting it never quite feels as bad.",
- "errors": 0,
- "words": 29,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "the strange thing about really wanting something is that getting it never quite feels the way you imagined it would and not getting it never quite feels as bad",
- "normalized_hypothesis": "the strange thing about really wanting something is that getting it never quite feels the way you imagined it would and not getting it never quite feels as bad",
- "errors": 0,
- "words": 29,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_048",
- "category": "long_range",
- "text": "Even though the recipe looked simple at first, preparing the sauce required patience, careful timing, and a much larger pan.",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_048.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Even though the recipe looked simple at first, preparing the sauce required patience, careful timing, and a much larger pan.",
- "raw": {
- "normalized_reference": "even though the recipe looked simple at first, preparing the sauce required patience, careful timing, and a much larger pan.",
- "normalized_hypothesis": "even though the recipe looked simple at first, preparing the sauce required patience, careful timing, and a much larger pan.",
- "errors": 0,
- "words": 20,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "even though the recipe looked simple at 1st preparing the sauce required patience careful timing and a much larger pan",
- "normalized_hypothesis": "even though the recipe looked simple at 1st preparing the sauce required patience careful timing and a much larger pan",
- "errors": 0,
- "words": 20,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_049",
- "category": "names_places",
- "text": "Xiomara met Siobhan near the old library in Worcestershire.",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_049.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Zamara met Siobhan near the old library in Worcestershire.",
- "raw": {
- "normalized_reference": "xiomara met siobhan near the old library in worcestershire.",
- "normalized_hypothesis": "zamara met siobhan near the old library in worcestershire.",
- "errors": 1,
- "words": 9,
- "wer": 0.1111111111111111
- },
- "semantic": {
- "normalized_reference": "xiomara met siobhan near the old library in worcestershire",
- "normalized_hypothesis": "zamara met siobhan near the old library in worcestershire",
- "errors": 1,
- "words": 9,
- "wer": 0.1111111111111111
- }
- },
- {
- "id": "utmos200_050",
- "category": "narrative",
- "text": "She walked into the room and everyone went quiet, not because she was loud, but because she wasn't.",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_050.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "She walked into the room and everyone went quiet, not because she was loud, but because she wasn't.",
- "raw": {
- "normalized_reference": "she walked into the room and everyone went quiet, not because she was loud, but because she wasn't.",
- "normalized_hypothesis": "she walked into the room and everyone went quiet, not because she was loud, but because she wasn't.",
- "errors": 0,
- "words": 18,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "she walked into the room and everyone went quiet not because she was loud but because she was not",
- "normalized_hypothesis": "she walked into the room and everyone went quiet not because she was loud but because she was not",
- "errors": 0,
- "words": 19,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_051",
- "category": "numbers",
- "text": "At exactly seven forty-five, Morgan placed eighteen dollars and fifty cents beside the notebook.",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_051.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "At exactly $7.45, Morton placed $18.50 beside the notebook.",
- "raw": {
- "normalized_reference": "at exactly seven forty-five, morgan placed eighteen dollars and fifty cents beside the notebook.",
- "normalized_hypothesis": "at exactly $7.45, morton placed $18.50 beside the notebook.",
- "errors": 9,
- "words": 15,
- "wer": 0.6
- },
- "semantic": {
- "normalized_reference": "at exactly 745 morgan placed $18.50 beside the notebook",
- "normalized_hypothesis": "at exactly $7.45 morton placed $18.50 beside the notebook",
- "errors": 3,
- "words": 10,
- "wer": 0.3
- }
- },
- {
- "id": "utmos200_052",
- "category": "punctuation",
- "text": "Does punctuation work? Yes: commas, semicolons, and periods should create sensible boundaries.",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_052.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Does punctuation work? Yes. Commas, semicolons, and periods should create sensible boundaries.",
- "raw": {
- "normalized_reference": "does punctuation work? yes: commas, semicolons, and periods should create sensible boundaries.",
- "normalized_hypothesis": "does punctuation work? yes. commas, semicolons, and periods should create sensible boundaries.",
- "errors": 0,
- "words": 12,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "does punctuation work yes commas semicolons and periods should create sensible boundaries",
- "normalized_hypothesis": "does punctuation work yes commas semicolons and periods should create sensible boundaries",
- "errors": 0,
- "words": 12,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_053",
- "category": "question",
- "text": "Have you ever just stopped and thought about how strange all of this is?",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_053.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Have you ever just stopped and thought about how strange all of this is?",
- "raw": {
- "normalized_reference": "have you ever just stopped and thought about how strange all of this is?",
- "normalized_hypothesis": "have you ever just stopped and thought about how strange all of this is?",
- "errors": 0,
- "words": 14,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "have you ever just stopped and thought about how strange all of this is",
- "normalized_hypothesis": "have you ever just stopped and thought about how strange all of this is",
- "errors": 0,
- "words": 14,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_054",
- "category": "rare_words",
- "text": "Her idiosyncratic explanation sounded plausible but remained fundamentally ambiguous.",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_054.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Her idiosyncratic explanation sounded plausible but remained fundamentally ambiguous.",
- "raw": {
- "normalized_reference": "her idiosyncratic explanation sounded plausible but remained fundamentally ambiguous.",
- "normalized_hypothesis": "her idiosyncratic explanation sounded plausible but remained fundamentally ambiguous.",
- "errors": 0,
- "words": 9,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "her idiosyncratic explanation sounded plausible but remained fundamentally ambiguous",
- "normalized_hypothesis": "her idiosyncratic explanation sounded plausible but remained fundamentally ambiguous",
- "errors": 0,
- "words": 9,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_055",
- "category": "short",
- "text": "That tracks.",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_055.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "That tracks.",
- "raw": {
- "normalized_reference": "that tracks.",
- "normalized_hypothesis": "that tracks.",
- "errors": 0,
- "words": 2,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "that tracks",
- "normalized_hypothesis": "that tracks",
- "errors": 0,
- "words": 2,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_056",
- "category": "technical",
- "text": "The key thing to remember is that order matters here more than you'd think.",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_056.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "The key thing to remember is that order matters here more than you'd think.",
- "raw": {
- "normalized_reference": "the key thing to remember is that order matters here more than you'd think.",
- "normalized_hypothesis": "the key thing to remember is that order matters here more than you'd think.",
- "errors": 0,
- "words": 14,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "the key thing to remember is that order matters here more than you would think",
- "normalized_hypothesis": "the key thing to remember is that order matters here more than you would think",
- "errors": 0,
- "words": 15,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_057",
- "category": "conversational",
- "text": "I would not say that again in this room if I were you.",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_057.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "I would not say that again in this room if I were you.",
- "raw": {
- "normalized_reference": "i would not say that again in this room if i were you.",
- "normalized_hypothesis": "i would not say that again in this room if i were you.",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "i would not say that again in this room if i were you",
- "normalized_hypothesis": "i would not say that again in this room if i were you",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_058",
- "category": "descriptive",
- "text": "The room was filled with the quiet that means something is about to change.",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_058.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "The room was filled with the quiet that means something is about to change.",
- "raw": {
- "normalized_reference": "the room was filled with the quiet that means something is about to change.",
- "normalized_hypothesis": "the room was filled with the quiet that means something is about to change.",
- "errors": 0,
- "words": 14,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "the room was filled with the quiet that means something is about to change",
- "normalized_hypothesis": "the room was filled with the quiet that means something is about to change",
- "errors": 0,
- "words": 14,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_059",
- "category": "emotional",
- "text": "I didn't expect to feel that way about it, but I did.",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_059.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "I didn't expect to feel that way about it, but I did.",
- "raw": {
- "normalized_reference": "i didn't expect to feel that way about it, but i did.",
- "normalized_hypothesis": "i didn't expect to feel that way about it, but i did.",
- "errors": 0,
- "words": 12,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "i did not expect to feel that way about it but i did",
- "normalized_hypothesis": "i did not expect to feel that way about it but i did",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_060",
- "category": "homographs",
- "text": "I object to the strange object that was left near the entrance.",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_060.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "I object to the strange object that was left near the entrance.",
- "raw": {
- "normalized_reference": "i object to the strange object that was left near the entrance.",
- "normalized_hypothesis": "i object to the strange object that was left near the entrance.",
- "errors": 0,
- "words": 12,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "i object to the strange object that was left near the entrance",
- "normalized_hypothesis": "i object to the strange object that was left near the entrance",
- "errors": 0,
- "words": 12,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_061",
- "category": "long",
- "text": "I've learned more from the things that went wrong than from the things that worked, which is a frustrating way to have to learn things but apparently the only way that stuck.",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_061.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "I've learned more from the things that went wrong than from the things that worked, which is a frustrating way to have to learn things but apparently the only way that stuck.",
- "raw": {
- "normalized_reference": "i've learned more from the things that went wrong than from the things that worked, which is a frustrating way to have to learn things but apparently the only way that stuck.",
- "normalized_hypothesis": "i've learned more from the things that went wrong than from the things that worked, which is a frustrating way to have to learn things but apparently the only way that stuck.",
- "errors": 0,
- "words": 32,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "i have learned more from the things that went wrong than from the things that worked which is a frustrating way to have to learn things but apparently the only way that stuck",
- "normalized_hypothesis": "i have learned more from the things that went wrong than from the things that worked which is a frustrating way to have to learn things but apparently the only way that stuck",
- "errors": 0,
- "words": 33,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_062",
- "category": "long_range",
- "text": "Because every nearby hotel was full, the exhausted hikers followed the narrow road until they found a farmhouse with one empty room.",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_062.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Because every nearby hotel was full, the exhausted hikers followed the narrow road until they found a farmhouse with one empty room.",
- "raw": {
- "normalized_reference": "because every nearby hotel was full, the exhausted hikers followed the narrow road until they found a farmhouse with one empty room.",
- "normalized_hypothesis": "because every nearby hotel was full, the exhausted hikers followed the narrow road until they found a farmhouse with one empty room.",
- "errors": 0,
- "words": 22,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "because every nearby hotel was full the exhausted hikers followed the narrow road until they found a farmhouse with one empty room",
- "normalized_hypothesis": "because every nearby hotel was full the exhausted hikers followed the narrow road until they found a farmhouse with one empty room",
- "errors": 0,
- "words": 22,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_063",
- "category": "names_places",
- "text": "The itinerary sends Beatrice from Albuquerque to Des Moines and then to Boise.",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_063.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "The itinerary sends Beatrice from Albuquerque to Des Moines and then to Boise.",
- "raw": {
- "normalized_reference": "the itinerary sends beatrice from albuquerque to des moines and then to boise.",
- "normalized_hypothesis": "the itinerary sends beatrice from albuquerque to des moines and then to boise.",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "the itinerary sends beatrice from albuquerque to des moines and then to boise",
- "normalized_hypothesis": "the itinerary sends beatrice from albuquerque to des moines and then to boise",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_064",
- "category": "narrative",
- "text": "We had been driving for six hours when we finally saw the sign for the town.",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_064.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "We had been driving for six hours when we finally saw the sign for the town.",
- "raw": {
- "normalized_reference": "we had been driving for six hours when we finally saw the sign for the town.",
- "normalized_hypothesis": "we had been driving for six hours when we finally saw the sign for the town.",
- "errors": 0,
- "words": 16,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "we had been driving for 6 hours when we finally saw the sign for the town",
- "normalized_hypothesis": "we had been driving for 6 hours when we finally saw the sign for the town",
- "errors": 0,
- "words": 16,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_065",
- "category": "numbers",
- "text": "Call extension zero four nine after eight thirty in the morning.",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_065.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Call extension 049 after 8.30 in the morning.",
- "raw": {
- "normalized_reference": "call extension zero four nine after eight thirty in the morning.",
- "normalized_hypothesis": "call extension 049 after 8.30 in the morning.",
- "errors": 5,
- "words": 11,
- "wer": 0.45454545454545453
- },
- "semantic": {
- "normalized_reference": "call extension 049 after 830 in the morning",
- "normalized_hypothesis": "call extension 49 after 8.30 in the morning",
- "errors": 3,
- "words": 8,
- "wer": 0.375
- }
- },
- {
- "id": "utmos200_066",
- "category": "punctuation",
- "text": "Stop - listen carefully - then answer the question without guessing.",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_066.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Stop listen carefully then answer the question without guessing.",
- "raw": {
- "normalized_reference": "stop - listen carefully - then answer the question without guessing.",
- "normalized_hypothesis": "stop listen carefully then answer the question without guessing.",
- "errors": 0,
- "words": 9,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "stop listen carefully then answer the question without guessing",
- "normalized_hypothesis": "stop listen carefully then answer the question without guessing",
- "errors": 0,
- "words": 9,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_067",
- "category": "question",
- "text": "Does it matter at this point, honestly?",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_067.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Does it matter at this point, honestly?",
- "raw": {
- "normalized_reference": "does it matter at this point, honestly?",
- "normalized_hypothesis": "does it matter at this point, honestly?",
- "errors": 0,
- "words": 7,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "does it matter at this point honestly",
- "normalized_hypothesis": "does it matter at this point honestly",
- "errors": 0,
- "words": 7,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_068",
- "category": "rare_words",
- "text": "An entrepreneurial meteorologist described the unprecedented atmospheric phenomenon.",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_068.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "An entrepreneurial meteorologist described the unprecedented atmospheric phenomenon.",
- "raw": {
- "normalized_reference": "an entrepreneurial meteorologist described the unprecedented atmospheric phenomenon.",
- "normalized_hypothesis": "an entrepreneurial meteorologist described the unprecedented atmospheric phenomenon.",
- "errors": 0,
- "words": 8,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "an entrepreneurial meteorologist described the unprecedented atmospheric phenomenon",
- "normalized_hypothesis": "an entrepreneurial meteorologist described the unprecedented atmospheric phenomenon",
- "errors": 0,
- "words": 8,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_069",
- "category": "short",
- "text": "That's not what I expected.",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_069.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "That's not what I expected.",
- "raw": {
- "normalized_reference": "that's not what i expected.",
- "normalized_hypothesis": "that's not what i expected.",
- "errors": 0,
- "words": 5,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "that is not what i expected",
- "normalized_hypothesis": "that is not what i expected",
- "errors": 0,
- "words": 6,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_070",
- "category": "technical",
- "text": "There's a known issue with that version that causes failures under load.",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_070.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "There's a known issue with that version that causes failures under load.",
- "raw": {
- "normalized_reference": "there's a known issue with that version that causes failures under load.",
- "normalized_hypothesis": "there's a known issue with that version that causes failures under load.",
- "errors": 0,
- "words": 12,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "there is a known issue with that version that causes failures under load",
- "normalized_hypothesis": "there is a known issue with that version that causes failures under load",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_071",
- "category": "conversational",
- "text": "Don't look now but I think that's him over there.",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_071.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Don't look now but I think that's him over there.",
- "raw": {
- "normalized_reference": "don't look now but i think that's him over there.",
- "normalized_hypothesis": "don't look now but i think that's him over there.",
- "errors": 0,
- "words": 10,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "do not look now but i think that is him over there",
- "normalized_hypothesis": "do not look now but i think that is him over there",
- "errors": 0,
- "words": 12,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_072",
- "category": "descriptive",
- "text": "The city at night looks completely different from how it feels during the day.",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_072.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "The city at night looks completely different from how it feels during the day.",
- "raw": {
- "normalized_reference": "the city at night looks completely different from how it feels during the day.",
- "normalized_hypothesis": "the city at night looks completely different from how it feels during the day.",
- "errors": 0,
- "words": 14,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "the city at night looks completely different from how it feels during the day",
- "normalized_hypothesis": "the city at night looks completely different from how it feels during the day",
- "errors": 0,
- "words": 14,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_073",
- "category": "emotional",
- "text": "It's the kind of happy that makes you nervous because you know it can't last.",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_073.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "It's the kind of happy that makes you nervous because you know it can't last.",
- "raw": {
- "normalized_reference": "it's the kind of happy that makes you nervous because you know it can't last.",
- "normalized_hypothesis": "it's the kind of happy that makes you nervous because you know it can't last.",
- "errors": 0,
- "words": 15,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "it is the kind of happy that makes you nervous because you know it can not last",
- "normalized_hypothesis": "it is the kind of happy that makes you nervous because you know it can not last",
- "errors": 0,
- "words": 17,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_074",
- "category": "homographs",
- "text": "The dove dove beneath the branch while the wind began to wind the loose cable.",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_074.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "The dove dove beneath the branch while the wind began to wind the loose cable.",
- "raw": {
- "normalized_reference": "the dove dove beneath the branch while the wind began to wind the loose cable.",
- "normalized_hypothesis": "the dove dove beneath the branch while the wind began to wind the loose cable.",
- "errors": 0,
- "words": 15,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "the dove dove beneath the branch while the wind began to wind the loose cable",
- "normalized_hypothesis": "the dove dove beneath the branch while the wind began to wind the loose cable",
- "errors": 0,
- "words": 15,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_075",
- "category": "long",
- "text": "I've thought a lot about what I would do differently if I could go back, and the honest answer is probably not as much as I'd like to think.",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_075.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "I've thought a lot about what I would do differently if I could go back, and the honest answer is probably not as much as I'd like to think.",
- "raw": {
- "normalized_reference": "i've thought a lot about what i would do differently if i could go back, and the honest answer is probably not as much as i'd like to think.",
- "normalized_hypothesis": "i've thought a lot about what i would do differently if i could go back, and the honest answer is probably not as much as i'd like to think.",
- "errors": 0,
- "words": 29,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "i have thought a lot about what i would do differently if i could go back and the honest answer is probably not as much as i would like to think",
- "normalized_hypothesis": "i have thought a lot about what i would do differently if i could go back and the honest answer is probably not as much as i would like to think",
- "errors": 0,
- "words": 31,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_076",
- "category": "long_range",
- "text": "Although the weather changed without warning, the musicians finished packing before anyone opened the enormous glass door.",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_076.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Although the weather changed without warning, the musicians finished packing before anyone opened the enormous glass door.",
- "raw": {
- "normalized_reference": "although the weather changed without warning, the musicians finished packing before anyone opened the enormous glass door.",
- "normalized_hypothesis": "although the weather changed without warning, the musicians finished packing before anyone opened the enormous glass door.",
- "errors": 0,
- "words": 17,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "although the weather changed without warning the musicians finished packing before anyone opened the enormous glass door",
- "normalized_hypothesis": "although the weather changed without warning the musicians finished packing before anyone opened the enormous glass door",
- "errors": 0,
- "words": 17,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_077",
- "category": "names_places",
- "text": "Saskatchewan looked unusually bright beneath the pale February sunrise.",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_077.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Saskatchewan looked unusually bright beneath that pale February sunrise.",
- "raw": {
- "normalized_reference": "saskatchewan looked unusually bright beneath the pale february sunrise.",
- "normalized_hypothesis": "saskatchewan looked unusually bright beneath that pale february sunrise.",
- "errors": 1,
- "words": 9,
- "wer": 0.1111111111111111
- },
- "semantic": {
- "normalized_reference": "saskatchewan looked unusually bright beneath the pale february sunrise",
- "normalized_hypothesis": "saskatchewan looked unusually bright beneath that pale february sunrise",
- "errors": 1,
- "words": 9,
- "wer": 0.1111111111111111
- }
- },
- {
- "id": "utmos200_078",
- "category": "narrative",
- "text": "She handled it with a grace I genuinely did not expect.",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_078.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "She handled it with a grace I genuinely did not expect.",
- "raw": {
- "normalized_reference": "she handled it with a grace i genuinely did not expect.",
- "normalized_hypothesis": "she handled it with a grace i genuinely did not expect.",
- "errors": 0,
- "words": 11,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "she handled it with a grace i genuinely did not expect",
- "normalized_hypothesis": "she handled it with a grace i genuinely did not expect",
- "errors": 0,
- "words": 11,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_079",
- "category": "numbers",
- "text": "Apartment four hundred seven is across from room twenty-nine on the third floor.",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_079.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Apartment 47 is across from Room 29 on the third floor.",
- "raw": {
- "normalized_reference": "apartment four hundred seven is across from room twenty-nine on the third floor.",
- "normalized_hypothesis": "apartment 47 is across from room 29 on the third floor.",
- "errors": 5,
- "words": 14,
- "wer": 0.35714285714285715
- },
- "semantic": {
- "normalized_reference": "apartment 407 is across from room 29 on the 3rd floor",
- "normalized_hypothesis": "apartment 47 is across from room 29 on the 3rd floor",
- "errors": 1,
- "words": 11,
- "wer": 0.09090909090909091
- }
- },
- {
- "id": "utmos200_080",
- "category": "punctuation",
- "text": "\"Not yet,\" Rowan whispered. \"Give me another minute... please.\"",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_080.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Not yet, Rowan whispered. Give me another minute, please.",
- "raw": {
- "normalized_reference": "\"not yet,\" rowan whispered. \"give me another minute... please.\"",
- "normalized_hypothesis": "not yet, rowan whispered. give me another minute, please.",
- "errors": 0,
- "words": 9,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "not yet rowan whispered give me another minute please",
- "normalized_hypothesis": "not yet rowan whispered give me another minute please",
- "errors": 0,
- "words": 9,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_081",
- "category": "question",
- "text": "What are we actually trying to accomplish here?",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_081.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "What are we actually trying to accomplish here?",
- "raw": {
- "normalized_reference": "what are we actually trying to accomplish here?",
- "normalized_hypothesis": "what are we actually trying to accomplish here?",
- "errors": 0,
- "words": 8,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "what are we actually trying to accomplish here",
- "normalized_hypothesis": "what are we actually trying to accomplish here",
- "errors": 0,
- "words": 8,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_082",
- "category": "rare_words",
- "text": "The physician recommended a noninvasive procedure for the recurring neurological symptoms.",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_082.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "The physician recommended a non-invasive procedure for the recurring neurological symptoms.",
- "raw": {
- "normalized_reference": "the physician recommended a noninvasive procedure for the recurring neurological symptoms.",
- "normalized_hypothesis": "the physician recommended a non-invasive procedure for the recurring neurological symptoms.",
- "errors": 2,
- "words": 11,
- "wer": 0.18181818181818182
- },
- "semantic": {
- "normalized_reference": "the physician recommended a noninvasive procedure for the recurring neurological symptoms",
- "normalized_hypothesis": "the physician recommended a non invasive procedure for the recurring neurological symptoms",
- "errors": 2,
- "words": 11,
- "wer": 0.18181818181818182
- }
- },
- {
- "id": "utmos200_083",
- "category": "short",
- "text": "Are you serious right now?",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_083.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Are you serious right now?",
- "raw": {
- "normalized_reference": "are you serious right now?",
- "normalized_hypothesis": "are you serious right now?",
- "errors": 0,
- "words": 5,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "are you serious right now",
- "normalized_hypothesis": "are you serious right now",
- "errors": 0,
- "words": 5,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_084",
- "category": "technical",
- "text": "The performance improvement is real, but it comes with tradeoffs.",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_084.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "The performance improvement is real, but it comes with trade-offs.",
- "raw": {
- "normalized_reference": "the performance improvement is real, but it comes with tradeoffs.",
- "normalized_hypothesis": "the performance improvement is real, but it comes with trade-offs.",
- "errors": 2,
- "words": 10,
- "wer": 0.2
- },
- "semantic": {
- "normalized_reference": "the performance improvement is real but it comes with tradeoffs",
- "normalized_hypothesis": "the performance improvement is real but it comes with trade offs",
- "errors": 2,
- "words": 10,
- "wer": 0.2
- }
- },
- {
- "id": "utmos200_085",
- "category": "conversational",
- "text": "I don't know. It's just been one of those weeks, you know?",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_085.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "I don't know, it's just been one of those weeks, you know?",
- "raw": {
- "normalized_reference": "i don't know. it's just been one of those weeks, you know?",
- "normalized_hypothesis": "i don't know, it's just been one of those weeks, you know?",
- "errors": 0,
- "words": 12,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "i do not know it is just been one of those weeks you know",
- "normalized_hypothesis": "i do not know it is just been one of those weeks you know",
- "errors": 0,
- "words": 14,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_086",
- "category": "descriptive",
- "text": "The moment you stop expecting perfection, things get a lot more interesting.",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_086.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "The moment you stop expecting perfection, things get a lot more interesting.",
- "raw": {
- "normalized_reference": "the moment you stop expecting perfection, things get a lot more interesting.",
- "normalized_hypothesis": "the moment you stop expecting perfection, things get a lot more interesting.",
- "errors": 0,
- "words": 12,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "the moment you stop expecting perfection things get a lot more interesting",
- "normalized_hypothesis": "the moment you stop expecting perfection things get a lot more interesting",
- "errors": 0,
- "words": 12,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_087",
- "category": "emotional",
- "text": "It's the kind of thing that stays with you even years later.",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_087.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "It's the kind of thing that stays with you even years later.",
- "raw": {
- "normalized_reference": "it's the kind of thing that stays with you even years later.",
- "normalized_hypothesis": "it's the kind of thing that stays with you even years later.",
- "errors": 0,
- "words": 12,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "it is the kind of thing that stays with you even years later",
- "normalized_hypothesis": "it is the kind of thing that stays with you even years later",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_088",
- "category": "long",
- "text": "She told me once that the most important skill she'd developed was learning to disagree with someone without making them feel they were wrong.",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_088.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "She told me once that the most important skill she'd developed was learning to disagree with someone without making them feel they were wrong.",
- "raw": {
- "normalized_reference": "she told me once that the most important skill she'd developed was learning to disagree with someone without making them feel they were wrong.",
- "normalized_hypothesis": "she told me once that the most important skill she'd developed was learning to disagree with someone without making them feel they were wrong.",
- "errors": 0,
- "words": 24,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "she told me once that the most important skill she would developed was learning to disagree with someone without making them feel they were wrong",
- "normalized_hypothesis": "she told me once that the most important skill she would developed was learning to disagree with someone without making them feel they were wrong",
- "errors": 0,
- "words": 25,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_089",
- "category": "narrative",
- "text": "He stood at the edge of the cliff and wondered, not for the first time, how he'd gotten here.",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_089.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "He stood at the edge of the cliff and wondered, not for the first time, how he'd gotten here.",
- "raw": {
- "normalized_reference": "he stood at the edge of the cliff and wondered, not for the first time, how he'd gotten here.",
- "normalized_hypothesis": "he stood at the edge of the cliff and wondered, not for the first time, how he'd gotten here.",
- "errors": 0,
- "words": 19,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "he stood at the edge of the cliff and wondered not for the 1st time how he would gotten here",
- "normalized_hypothesis": "he stood at the edge of the cliff and wondered not for the 1st time how he would gotten here",
- "errors": 0,
- "words": 20,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_090",
- "category": "question",
- "text": "Why does asking for help feel so hard when you actually need it?",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_090.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Why does asking for help feel so hard when you actually need it?",
- "raw": {
- "normalized_reference": "why does asking for help feel so hard when you actually need it?",
- "normalized_hypothesis": "why does asking for help feel so hard when you actually need it?",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "why does asking for help feel so hard when you actually need it",
- "normalized_hypothesis": "why does asking for help feel so hard when you actually need it",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_091",
- "category": "short",
- "text": "That's fair.",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_091.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "That's fair.",
- "raw": {
- "normalized_reference": "that's fair.",
- "normalized_hypothesis": "that's fair.",
- "errors": 0,
- "words": 2,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "that is fair",
- "normalized_hypothesis": "that is fair",
- "errors": 0,
- "words": 3,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_092",
- "category": "technical",
- "text": "If you run into that error, the fix is usually just restarting the service.",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_092.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "If you run into that error, the fix is usually just restarting this service.",
- "raw": {
- "normalized_reference": "if you run into that error, the fix is usually just restarting the service.",
- "normalized_hypothesis": "if you run into that error, the fix is usually just restarting this service.",
- "errors": 1,
- "words": 14,
- "wer": 0.07142857142857142
- },
- "semantic": {
- "normalized_reference": "if you run into that error the fix is usually just restarting the service",
- "normalized_hypothesis": "if you run into that error the fix is usually just restarting this service",
- "errors": 1,
- "words": 14,
- "wer": 0.07142857142857142
- }
- },
- {
- "id": "utmos200_093",
- "category": "conversational",
- "text": "It was a disaster in the best possible way.",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_093.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "It was a disaster in the best possible way.",
- "raw": {
- "normalized_reference": "it was a disaster in the best possible way.",
- "normalized_hypothesis": "it was a disaster in the best possible way.",
- "errors": 0,
- "words": 9,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "it was a disaster in the best possible way",
- "normalized_hypothesis": "it was a disaster in the best possible way",
- "errors": 0,
- "words": 9,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_094",
- "category": "descriptive",
- "text": "It had the look of a place that used to mean something to someone.",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_094.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "It had the look of a place that used to mean something to someone.",
- "raw": {
- "normalized_reference": "it had the look of a place that used to mean something to someone.",
- "normalized_hypothesis": "it had the look of a place that used to mean something to someone.",
- "errors": 0,
- "words": 14,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "it had the look of a place that used to mean something to someone",
- "normalized_hypothesis": "it had the look of a place that used to mean something to someone",
- "errors": 0,
- "words": 14,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_095",
- "category": "emotional",
- "text": "There's this feeling you get when something finally works after weeks of failure.",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_095.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "There's this feeling you get when something finally works after weeks of failure.",
- "raw": {
- "normalized_reference": "there's this feeling you get when something finally works after weeks of failure.",
- "normalized_hypothesis": "there's this feeling you get when something finally works after weeks of failure.",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "there is this feeling you get when something finally works after weeks of failure",
- "normalized_hypothesis": "there is this feeling you get when something finally works after weeks of failure",
- "errors": 0,
- "words": 14,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_096",
- "category": "long",
- "text": "He had a theory that most arguments aren't actually about what they seem to be about, and the longer I live the more I think he was onto something.",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_096.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "He had a theory that most arguments aren't actually about what they seem to be about, and the longer I live the more I think he was onto something.",
- "raw": {
- "normalized_reference": "he had a theory that most arguments aren't actually about what they seem to be about, and the longer i live the more i think he was onto something.",
- "normalized_hypothesis": "he had a theory that most arguments aren't actually about what they seem to be about, and the longer i live the more i think he was onto something.",
- "errors": 0,
- "words": 29,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "he had a theory that most arguments are not actually about what they seem to be about and the longer i live the more i think he was onto something",
- "normalized_hypothesis": "he had a theory that most arguments are not actually about what they seem to be about and the longer i live the more i think he was onto something",
- "errors": 0,
- "words": 30,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_097",
- "category": "narrative",
- "text": "The thing nobody tells you about starting over is how ordinary most of it feels.",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_097.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "The thing nobody tells you about starting overt is how ordinary most of it feels.",
- "raw": {
- "normalized_reference": "the thing nobody tells you about starting over is how ordinary most of it feels.",
- "normalized_hypothesis": "the thing nobody tells you about starting overt is how ordinary most of it feels.",
- "errors": 1,
- "words": 15,
- "wer": 0.06666666666666667
- },
- "semantic": {
- "normalized_reference": "the thing nobody tells you about starting over is how ordinary most of it feels",
- "normalized_hypothesis": "the thing nobody tells you about starting overt is how ordinary most of it feels",
- "errors": 1,
- "words": 15,
- "wer": 0.06666666666666667
- }
- },
- {
- "id": "utmos200_098",
- "category": "question",
- "text": "What exactly are we waiting for at this point?",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_098.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "What exactly are we waiting for at this point?",
- "raw": {
- "normalized_reference": "what exactly are we waiting for at this point?",
- "normalized_hypothesis": "what exactly are we waiting for at this point?",
- "errors": 0,
- "words": 9,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "what exactly are we waiting for at this point",
- "normalized_hypothesis": "what exactly are we waiting for at this point",
- "errors": 0,
- "words": 9,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_099",
- "category": "short",
- "text": "I stand by what I said.",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_099.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "I stand by what I said.",
- "raw": {
- "normalized_reference": "i stand by what i said.",
- "normalized_hypothesis": "i stand by what i said.",
- "errors": 0,
- "words": 6,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "i stand by what i said",
- "normalized_hypothesis": "i stand by what i said",
- "errors": 0,
- "words": 6,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_100",
- "category": "technical",
- "text": "Double-check that value before you push anything to production.",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_100.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Double-check that value before you push anything to production.",
- "raw": {
- "normalized_reference": "double-check that value before you push anything to production.",
- "normalized_hypothesis": "double-check that value before you push anything to production.",
- "errors": 0,
- "words": 10,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "double check that value before you push anything to production",
- "normalized_hypothesis": "double check that value before you push anything to production",
- "errors": 0,
- "words": 10,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_101",
- "category": "conversational",
- "text": "I understand your point; I just do not agree with the conclusion.",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_101.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "I understand your point. I just do not agree with the conclusion.",
- "raw": {
- "normalized_reference": "i understand your point; i just do not agree with the conclusion.",
- "normalized_hypothesis": "i understand your point. i just do not agree with the conclusion.",
- "errors": 0,
- "words": 12,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "i understand your point i just do not agree with the conclusion",
- "normalized_hypothesis": "i understand your point i just do not agree with the conclusion",
- "errors": 0,
- "words": 12,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_102",
- "category": "descriptive",
- "text": "The sky that day was a strange shade of orange I'd never seen before.",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_102.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "The sky that day was a strange shade of orange I'd never seen before.",
- "raw": {
- "normalized_reference": "the sky that day was a strange shade of orange i'd never seen before.",
- "normalized_hypothesis": "the sky that day was a strange shade of orange i'd never seen before.",
- "errors": 0,
- "words": 14,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "the sky that day was a strange shade of orange i would never seen before",
- "normalized_hypothesis": "the sky that day was a strange shade of orange i would never seen before",
- "errors": 0,
- "words": 15,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_103",
- "category": "emotional",
- "text": "It hurt more than I expected, and I thought I was prepared.",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_103.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "It hurt more than I expected, and I thought I was prepared.",
- "raw": {
- "normalized_reference": "it hurt more than i expected, and i thought i was prepared.",
- "normalized_hypothesis": "it hurt more than i expected, and i thought i was prepared.",
- "errors": 0,
- "words": 12,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "it hurt more than i expected and i thought i was prepared",
- "normalized_hypothesis": "it hurt more than i expected and i thought i was prepared",
- "errors": 0,
- "words": 12,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_104",
- "category": "long",
- "text": "The hardest part wasn't making the decision, it was sitting with the uncertainty of not knowing whether it was the right one for months afterward.",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_104.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "The hardest part wasn't making the decision, it was sitting with the uncertainty of not knowing whether it was the right one for months afterward.",
- "raw": {
- "normalized_reference": "the hardest part wasn't making the decision, it was sitting with the uncertainty of not knowing whether it was the right one for months afterward.",
- "normalized_hypothesis": "the hardest part wasn't making the decision, it was sitting with the uncertainty of not knowing whether it was the right one for months afterward.",
- "errors": 0,
- "words": 25,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "the hardest part was not making the decision it was sitting with the uncertainty of not knowing whether it was the right one for months afterward",
- "normalized_hypothesis": "the hardest part was not making the decision it was sitting with the uncertainty of not knowing whether it was the right one for months afterward",
- "errors": 0,
- "words": 26,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_105",
- "category": "narrative",
- "text": "There was this one day last summer when everything just clicked into place.",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_105.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "There was this one day last summer when everything just clicked into place.",
- "raw": {
- "normalized_reference": "there was this one day last summer when everything just clicked into place.",
- "normalized_hypothesis": "there was this one day last summer when everything just clicked into place.",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "there was this one day last summer when everything just clicked into place",
- "normalized_hypothesis": "there was this one day last summer when everything just clicked into place",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_106",
- "category": "question",
- "text": "Have you ever thought about why we keep doing things the same way when they stop working?",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_106.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Have you ever thought about why we keep doing things the same way when they stop working?",
- "raw": {
- "normalized_reference": "have you ever thought about why we keep doing things the same way when they stop working?",
- "normalized_hypothesis": "have you ever thought about why we keep doing things the same way when they stop working?",
- "errors": 0,
- "words": 17,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "have you ever thought about why we keep doing things the same way when they stop working",
- "normalized_hypothesis": "have you ever thought about why we keep doing things the same way when they stop working",
- "errors": 0,
- "words": 17,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_107",
- "category": "short",
- "text": "I can't believe you just said that.",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_107.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "I can't believe you just said that.",
- "raw": {
- "normalized_reference": "i can't believe you just said that.",
- "normalized_hypothesis": "i can't believe you just said that.",
- "errors": 0,
- "words": 7,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "i can not believe you just said that",
- "normalized_hypothesis": "i can not believe you just said that",
- "errors": 0,
- "words": 8,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_108",
- "category": "technical",
- "text": "You'll want to isolate that variable before drawing any conclusions.",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_108.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "you'll want to isolate that variable before drawing any conclusions.",
- "raw": {
- "normalized_reference": "you'll want to isolate that variable before drawing any conclusions.",
- "normalized_hypothesis": "you'll want to isolate that variable before drawing any conclusions.",
- "errors": 0,
- "words": 10,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "you will want to isolate that variable before drawing any conclusions",
- "normalized_hypothesis": "you will want to isolate that variable before drawing any conclusions",
- "errors": 0,
- "words": 11,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_109",
- "category": "conversational",
- "text": "The meeting that could have been an email lasted two hours.",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_109.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "The meeting that could have been an email lasted two hours.",
- "raw": {
- "normalized_reference": "the meeting that could have been an email lasted two hours.",
- "normalized_hypothesis": "the meeting that could have been an email lasted two hours.",
- "errors": 0,
- "words": 11,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "the meeting that could have been an email lasted 2 hours",
- "normalized_hypothesis": "the meeting that could have been an email lasted 2 hours",
- "errors": 0,
- "words": 11,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_110",
- "category": "descriptive",
- "text": "There's a specific kind of tired that sleep doesn't fix.",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_110.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "There's a specific kind of tired that sleep doesn't fix.",
- "raw": {
- "normalized_reference": "there's a specific kind of tired that sleep doesn't fix.",
- "normalized_hypothesis": "there's a specific kind of tired that sleep doesn't fix.",
- "errors": 0,
- "words": 10,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "there is a specific kind of tired that sleep does not fix",
- "normalized_hypothesis": "there is a specific kind of tired that sleep does not fix",
- "errors": 0,
- "words": 12,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_111",
- "category": "emotional",
- "text": "Something about that moment broke something open in me that I hadn't realized was closed.",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_111.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Something about that moment broke something open in me that I hadn't realized was closed.",
- "raw": {
- "normalized_reference": "something about that moment broke something open in me that i hadn't realized was closed.",
- "normalized_hypothesis": "something about that moment broke something open in me that i hadn't realized was closed.",
- "errors": 0,
- "words": 15,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "something about that moment broke something open in me that i had not realized was closed",
- "normalized_hypothesis": "something about that moment broke something open in me that i had not realized was closed",
- "errors": 0,
- "words": 16,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_112",
- "category": "long",
- "text": "The thing about making something good is that you don't usually know it's good while you're making it. You just know you're not done yet.",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_112.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "The thing about making something good is that you don't usually know it's good while you're making it. You just know you're not done yet.",
- "raw": {
- "normalized_reference": "the thing about making something good is that you don't usually know it's good while you're making it. you just know you're not done yet.",
- "normalized_hypothesis": "the thing about making something good is that you don't usually know it's good while you're making it. you just know you're not done yet.",
- "errors": 0,
- "words": 25,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "the thing about making something good is that you do not usually know it is good while you are making it you just know you are not done yet",
- "normalized_hypothesis": "the thing about making something good is that you do not usually know it is good while you are making it you just know you are not done yet",
- "errors": 0,
- "words": 29,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_113",
- "category": "narrative",
- "text": "She opened the door slowly, as if she expected to find something she wasn't ready for.",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_113.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "She opened the door slowly, as if she expected to find something she wasn't ready for.",
- "raw": {
- "normalized_reference": "she opened the door slowly, as if she expected to find something she wasn't ready for.",
- "normalized_hypothesis": "she opened the door slowly, as if she expected to find something she wasn't ready for.",
- "errors": 0,
- "words": 16,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "she opened the door slowly as if she expected to find something she was not ready for",
- "normalized_hypothesis": "she opened the door slowly as if she expected to find something she was not ready for",
- "errors": 0,
- "words": 17,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_114",
- "category": "question",
- "text": "What would you do differently if you could go back?",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_114.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "What would you do differently if you could go back?",
- "raw": {
- "normalized_reference": "what would you do differently if you could go back?",
- "normalized_hypothesis": "what would you do differently if you could go back?",
- "errors": 0,
- "words": 10,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "what would you do differently if you could go back",
- "normalized_hypothesis": "what would you do differently if you could go back",
- "errors": 0,
- "words": 10,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_115",
- "category": "short",
- "text": "Okay. Okay okay okay.",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_115.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Okay, okay, okay, okay.",
- "raw": {
- "normalized_reference": "okay. okay okay okay.",
- "normalized_hypothesis": "okay, okay, okay, okay.",
- "errors": 0,
- "words": 4,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "okay okay okay okay",
- "normalized_hypothesis": "okay okay okay okay",
- "errors": 0,
- "words": 4,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_116",
- "category": "technical",
- "text": "The latency goes up significantly at scale, which is expected.",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_116.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "The latency goes up significantly at scale, which is expected,",
- "raw": {
- "normalized_reference": "the latency goes up significantly at scale, which is expected.",
- "normalized_hypothesis": "the latency goes up significantly at scale, which is expected,",
- "errors": 0,
- "words": 10,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "the latency goes up significantly at scale which is expected",
- "normalized_hypothesis": "the latency goes up significantly at scale which is expected",
- "errors": 0,
- "words": 10,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_117",
- "category": "conversational",
- "text": "We ended up staying until two in the morning talking about nothing.",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_117.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "We ended up staying until 2 in the morning talking about nothing.",
- "raw": {
- "normalized_reference": "we ended up staying until two in the morning talking about nothing.",
- "normalized_hypothesis": "we ended up staying until 2 in the morning talking about nothing.",
- "errors": 1,
- "words": 12,
- "wer": 0.08333333333333333
- },
- "semantic": {
- "normalized_reference": "we ended up staying until 2 in the morning talking about nothing",
- "normalized_hypothesis": "we ended up staying until 2 in the morning talking about nothing",
- "errors": 0,
- "words": 12,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_118",
- "category": "descriptive",
- "text": "The rain started slowly, and then all at once, the way it usually does.",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_118.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "The rain started slowly, and then all at once, the way it usually does.",
- "raw": {
- "normalized_reference": "the rain started slowly, and then all at once, the way it usually does.",
- "normalized_hypothesis": "the rain started slowly, and then all at once, the way it usually does.",
- "errors": 0,
- "words": 14,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "the rain started slowly and then all at once the way it usually does",
- "normalized_hypothesis": "the rain started slowly and then all at once the way it usually does",
- "errors": 0,
- "words": 14,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_119",
- "category": "emotional",
- "text": "I honestly did not think I was going to make it through that year.",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_119.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "I honestly did not think I was going to make it through that year.",
- "raw": {
- "normalized_reference": "i honestly did not think i was going to make it through that year.",
- "normalized_hypothesis": "i honestly did not think i was going to make it through that year.",
- "errors": 0,
- "words": 14,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "i honestly did not think i was going to make it through that year",
- "normalized_hypothesis": "i honestly did not think i was going to make it through that year",
- "errors": 0,
- "words": 14,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_120",
- "category": "long",
- "text": "There's an assumption that people who seem confident all the time actually feel confident, and in my experience that's almost never true.",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_120.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "There's an assumption that people who seem confident all the time actually feel confident, and in my experience that's almost never true.",
- "raw": {
- "normalized_reference": "there's an assumption that people who seem confident all the time actually feel confident, and in my experience that's almost never true.",
- "normalized_hypothesis": "there's an assumption that people who seem confident all the time actually feel confident, and in my experience that's almost never true.",
- "errors": 0,
- "words": 22,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "there is an assumption that people who seem confident all the time actually feel confident and in my experience that is almost never true",
- "normalized_hypothesis": "there is an assumption that people who seem confident all the time actually feel confident and in my experience that is almost never true",
- "errors": 0,
- "words": 24,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_121",
- "category": "narrative",
- "text": "The moment I opened the door I knew something was different, even if I couldn't say what.",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_121.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "The moment I opened the door I knew something was different, even if I couldn't say what.",
- "raw": {
- "normalized_reference": "the moment i opened the door i knew something was different, even if i couldn't say what.",
- "normalized_hypothesis": "the moment i opened the door i knew something was different, even if i couldn't say what.",
- "errors": 0,
- "words": 17,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "the moment i opened the door i knew something was different even if i could not say what",
- "normalized_hypothesis": "the moment i opened the door i knew something was different even if i could not say what",
- "errors": 0,
- "words": 18,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_122",
- "category": "question",
- "text": "How do you explain something that doesn't make sense even to you?",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_122.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "How do you explain something that doesn't make sense even to you?",
- "raw": {
- "normalized_reference": "how do you explain something that doesn't make sense even to you?",
- "normalized_hypothesis": "how do you explain something that doesn't make sense even to you?",
- "errors": 0,
- "words": 12,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "how do you explain something that does not make sense even to you",
- "normalized_hypothesis": "how do you explain something that does not make sense even to you",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_123",
- "category": "short",
- "text": "Hold on a second.",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_123.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Hold on a second.",
- "raw": {
- "normalized_reference": "hold on a second.",
- "normalized_hypothesis": "hold on a second.",
- "errors": 0,
- "words": 4,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "hold on a 2nd",
- "normalized_hypothesis": "hold on a 2nd",
- "errors": 0,
- "words": 4,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_124",
- "category": "technical",
- "text": "A deterministic split prevents duplicate normalized text from leaking into evaluation.",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_124.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "A deterministic split prevents duplicate normalized text from leaking into evaluation.",
- "raw": {
- "normalized_reference": "a deterministic split prevents duplicate normalized text from leaking into evaluation.",
- "normalized_hypothesis": "a deterministic split prevents duplicate normalized text from leaking into evaluation.",
- "errors": 0,
- "words": 11,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "a deterministic split prevents duplicate normalized text from leaking into evaluation",
- "normalized_hypothesis": "a deterministic split prevents duplicate normalized text from leaking into evaluation",
- "errors": 0,
- "words": 11,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_125",
- "category": "conversational",
- "text": "Honestly, that was much better than I expected, but the ending still felt rushed.",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_125.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Honestly, that was much better than I expected, but the ending still felt rushed.",
- "raw": {
- "normalized_reference": "honestly, that was much better than i expected, but the ending still felt rushed.",
- "normalized_hypothesis": "honestly, that was much better than i expected, but the ending still felt rushed.",
- "errors": 0,
- "words": 14,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "honestly that was much better than i expected but the ending still felt rushed",
- "normalized_hypothesis": "honestly that was much better than i expected but the ending still felt rushed",
- "errors": 0,
- "words": 14,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_126",
- "category": "descriptive",
- "text": "She had a habit of tilting her head slightly when thinking hard about something.",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_126.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "she had a habit of tilting her head slightly when thinking hard about something.",
- "raw": {
- "normalized_reference": "she had a habit of tilting her head slightly when thinking hard about something.",
- "normalized_hypothesis": "she had a habit of tilting her head slightly when thinking hard about something.",
- "errors": 0,
- "words": 14,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "she had a habit of tilting her head slightly when thinking hard about something",
- "normalized_hypothesis": "she had a habit of tilting her head slightly when thinking hard about something",
- "errors": 0,
- "words": 14,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_127",
- "category": "emotional",
- "text": "I cannot stress enough how much I did not want to be right about this.",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_127.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "I cannot stress enough how much I did not want to be right about this.",
- "raw": {
- "normalized_reference": "i cannot stress enough how much i did not want to be right about this.",
- "normalized_hypothesis": "i cannot stress enough how much i did not want to be right about this.",
- "errors": 0,
- "words": 15,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "i cannot stress enough how much i did not want to be right about this",
- "normalized_hypothesis": "i cannot stress enough how much i did not want to be right about this",
- "errors": 0,
- "words": 15,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_128",
- "category": "long",
- "text": "I used to think the goal was to get to a point where nothing was hard anymore, and now I think that point doesn't exist and I'm not sure I'd want it to.",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_128.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "I used to think the goal was to get to a point where nothing was hard anymore, and now I think that point doesn't exist and I'm not sure I want it to.",
- "raw": {
- "normalized_reference": "i used to think the goal was to get to a point where nothing was hard anymore, and now i think that point doesn't exist and i'm not sure i'd want it to.",
- "normalized_hypothesis": "i used to think the goal was to get to a point where nothing was hard anymore, and now i think that point doesn't exist and i'm not sure i want it to.",
- "errors": 1,
- "words": 33,
- "wer": 0.030303030303030304
- },
- "semantic": {
- "normalized_reference": "i used to think the goal was to get to a point where nothing was hard anymore and now i think that point does not exist and i am not sure i would want it to",
- "normalized_hypothesis": "i used to think the goal was to get to a point where nothing was hard anymore and now i think that point does not exist and i am not sure i want it to",
- "errors": 1,
- "words": 36,
- "wer": 0.027777777777777776
- }
- },
- {
- "id": "utmos200_129",
- "category": "narrative",
- "text": "So this is the part of the story where everything goes completely sideways.",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_129.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "So this is the part of the story where everything goes completely sideways.",
- "raw": {
- "normalized_reference": "so this is the part of the story where everything goes completely sideways.",
- "normalized_hypothesis": "so this is the part of the story where everything goes completely sideways.",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "so this is the part of the story where everything goes completely sideways",
- "normalized_hypothesis": "so this is the part of the story where everything goes completely sideways",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_130",
- "category": "question",
- "text": "Is there a version of this where it actually works out okay?",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_130.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Is there a version of this where it actually works out okay?",
- "raw": {
- "normalized_reference": "is there a version of this where it actually works out okay?",
- "normalized_hypothesis": "is there a version of this where it actually works out okay?",
- "errors": 0,
- "words": 12,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "is there a version of this where it actually works out okay",
- "normalized_hypothesis": "is there a version of this where it actually works out okay",
- "errors": 0,
- "words": 12,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_131",
- "category": "short",
- "text": "Oh no. Oh no no no.",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_131.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Oh no, oh no no no.",
- "raw": {
- "normalized_reference": "oh no. oh no no no.",
- "normalized_hypothesis": "oh no, oh no no no.",
- "errors": 0,
- "words": 6,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "0 no 0 no no no",
- "normalized_hypothesis": "0 no 0 no no no",
- "errors": 0,
- "words": 6,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_132",
- "category": "technical",
- "text": "Always test on a copy before modifying the original.",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_132.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Always test on a copy before modifying the original.",
- "raw": {
- "normalized_reference": "always test on a copy before modifying the original.",
- "normalized_hypothesis": "always test on a copy before modifying the original.",
- "errors": 0,
- "words": 9,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "always test on a copy before modifying the original",
- "normalized_hypothesis": "always test on a copy before modifying the original",
- "errors": 0,
- "words": 9,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_133",
- "category": "conversational",
- "text": "I wasn't even supposed to be there. That's the wild part.",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_133.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "I wasn't even supposed to be there. That's the wild part.",
- "raw": {
- "normalized_reference": "i wasn't even supposed to be there. that's the wild part.",
- "normalized_hypothesis": "i wasn't even supposed to be there. that's the wild part.",
- "errors": 0,
- "words": 11,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "i was not even supposed to be there that is the wild part",
- "normalized_hypothesis": "i was not even supposed to be there that is the wild part",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_134",
- "category": "descriptive",
- "text": "It's a small town. Not the kind you leave. The kind you leave and come back to.",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_134.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "It's a small town, not the kind you leave, the kind you leave and come back to.",
- "raw": {
- "normalized_reference": "it's a small town. not the kind you leave. the kind you leave and come back to.",
- "normalized_hypothesis": "it's a small town, not the kind you leave, the kind you leave and come back to.",
- "errors": 0,
- "words": 17,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "it is a small town not the kind you leave the kind you leave and come back to",
- "normalized_hypothesis": "it is a small town not the kind you leave the kind you leave and come back to",
- "errors": 0,
- "words": 18,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_135",
- "category": "emotional",
- "text": "She cried. I cried. We all just sort of stood there crying.",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_135.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "She cried, I cried, we all just sort of stood there crying.",
- "raw": {
- "normalized_reference": "she cried. i cried. we all just sort of stood there crying.",
- "normalized_hypothesis": "she cried, i cried, we all just sort of stood there crying.",
- "errors": 0,
- "words": 12,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "she cried i cried we all just sort of stood there crying",
- "normalized_hypothesis": "she cried i cried we all just sort of stood there crying",
- "errors": 0,
- "words": 12,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_136",
- "category": "long",
- "text": "There's a version of being busy that's actually just a very elaborate way of avoiding the one thing you know you should be doing.",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_136.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "There's a version of being busy that's actually just a very elaborate way of avoiding the one thing you know you should be doing.",
- "raw": {
- "normalized_reference": "there's a version of being busy that's actually just a very elaborate way of avoiding the one thing you know you should be doing.",
- "normalized_hypothesis": "there's a version of being busy that's actually just a very elaborate way of avoiding the one thing you know you should be doing.",
- "errors": 0,
- "words": 24,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "there is a version of being busy that is actually just a very elaborate way of avoiding the one thing you know you should be doing",
- "normalized_hypothesis": "there is a version of being busy that is actually just a very elaborate way of avoiding the one thing you know you should be doing",
- "errors": 0,
- "words": 26,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_137",
- "category": "narrative",
- "text": "The letter arrived on a Tuesday, which felt appropriate somehow.",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_137.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "The letter arrived on a Tuesday, which felt appropriate somehow.",
- "raw": {
- "normalized_reference": "the letter arrived on a tuesday, which felt appropriate somehow.",
- "normalized_hypothesis": "the letter arrived on a tuesday, which felt appropriate somehow.",
- "errors": 0,
- "words": 10,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "the letter arrived on a tuesday which felt appropriate somehow",
- "normalized_hypothesis": "the letter arrived on a tuesday which felt appropriate somehow",
- "errors": 0,
- "words": 10,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_138",
- "category": "question",
- "text": "What exactly did she say when you told her?",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_138.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "What exactly did she say when you told her?",
- "raw": {
- "normalized_reference": "what exactly did she say when you told her?",
- "normalized_hypothesis": "what exactly did she say when you told her?",
- "errors": 0,
- "words": 9,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "what exactly did she say when you told her",
- "normalized_hypothesis": "what exactly did she say when you told her",
- "errors": 0,
- "words": 9,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_139",
- "category": "short",
- "text": "No way. Absolutely not.",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_139.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "No wait, absolutely not.",
- "raw": {
- "normalized_reference": "no way. absolutely not.",
- "normalized_hypothesis": "no wait, absolutely not.",
- "errors": 1,
- "words": 4,
- "wer": 0.25
- },
- "semantic": {
- "normalized_reference": "no way absolutely not",
- "normalized_hypothesis": "no wait absolutely not",
- "errors": 1,
- "words": 4,
- "wer": 0.25
- }
- },
- {
- "id": "utmos200_140",
- "category": "technical",
- "text": "The first thing you want to do is make sure everything is backed up.",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_140.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "The first thing you want to do is make sure everything is backed up.",
- "raw": {
- "normalized_reference": "the first thing you want to do is make sure everything is backed up.",
- "normalized_hypothesis": "the first thing you want to do is make sure everything is backed up.",
- "errors": 0,
- "words": 14,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "the 1st thing you want to do is make sure everything is backed up",
- "normalized_hypothesis": "the 1st thing you want to do is make sure everything is backed up",
- "errors": 0,
- "words": 14,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_141",
- "category": "conversational",
- "text": "I have the most insane idea and you need to just hear me out.",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_141.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "I have the most insane idea and you need to just hear me out.",
- "raw": {
- "normalized_reference": "i have the most insane idea and you need to just hear me out.",
- "normalized_hypothesis": "i have the most insane idea and you need to just hear me out.",
- "errors": 0,
- "words": 14,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "i have the most insane idea and you need to just hear me out",
- "normalized_hypothesis": "i have the most insane idea and you need to just hear me out",
- "errors": 0,
- "words": 14,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_142",
- "category": "descriptive",
- "text": "There's a coffee shop on the corner that's been there longer than anyone can remember.",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_142.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "There's a coffee shop on the corner that's been there longer than anyone can remember.",
- "raw": {
- "normalized_reference": "there's a coffee shop on the corner that's been there longer than anyone can remember.",
- "normalized_hypothesis": "there's a coffee shop on the corner that's been there longer than anyone can remember.",
- "errors": 0,
- "words": 15,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "there is a coffee shop on the corner that has been there longer than anyone can remember",
- "normalized_hypothesis": "there is a coffee shop on the corner that has been there longer than anyone can remember",
- "errors": 0,
- "words": 17,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_143",
- "category": "emotional",
- "text": "I kept it together until I got to the car, and then I completely fell apart.",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_143.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "I kept it together until I got to the car, and then I completely fell apart.",
- "raw": {
- "normalized_reference": "i kept it together until i got to the car, and then i completely fell apart.",
- "normalized_hypothesis": "i kept it together until i got to the car, and then i completely fell apart.",
- "errors": 0,
- "words": 16,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "i kept it together until i got to the car and then i completely fell apart",
- "normalized_hypothesis": "i kept it together until i got to the car and then i completely fell apart",
- "errors": 0,
- "words": 16,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_144",
- "category": "long",
- "text": "The older I get, the more I think that most of what we call wisdom is just pattern recognition built from having been wrong enough times to start noticing.",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_144.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "The older I get, the more I think that most of what we call wisdom is just pattern recognition built from having been wrong enough times to start noticing.",
- "raw": {
- "normalized_reference": "the older i get, the more i think that most of what we call wisdom is just pattern recognition built from having been wrong enough times to start noticing.",
- "normalized_hypothesis": "the older i get, the more i think that most of what we call wisdom is just pattern recognition built from having been wrong enough times to start noticing.",
- "errors": 0,
- "words": 29,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "the older i get the more i think that most of what we call wisdom is just pattern recognition built from having been wrong enough times to start noticing",
- "normalized_hypothesis": "the older i get the more i think that most of what we call wisdom is just pattern recognition built from having been wrong enough times to start noticing",
- "errors": 0,
- "words": 29,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_145",
- "category": "narrative",
- "text": "I remember the exact moment I realized I had made a huge mistake.",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_145.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "I remember the exact moment I realized I had made a huge mistake.",
- "raw": {
- "normalized_reference": "i remember the exact moment i realized i had made a huge mistake.",
- "normalized_hypothesis": "i remember the exact moment i realized i had made a huge mistake.",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "i remember the exact moment i realized i had made a huge mistake",
- "normalized_hypothesis": "i remember the exact moment i realized i had made a huge mistake",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_146",
- "category": "question",
- "text": "Do you think it would have gone differently if we'd left earlier?",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_146.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Do you think it would have gone differently if we'd left earlier?",
- "raw": {
- "normalized_reference": "do you think it would have gone differently if we'd left earlier?",
- "normalized_hypothesis": "do you think it would have gone differently if we'd left earlier?",
- "errors": 0,
- "words": 12,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "do you think it would have gone differently if we would left earlier",
- "normalized_hypothesis": "do you think it would have gone differently if we would left earlier",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_147",
- "category": "short",
- "text": "You have to be kidding me.",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_147.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "You have to be kidding me.",
- "raw": {
- "normalized_reference": "you have to be kidding me.",
- "normalized_hypothesis": "you have to be kidding me.",
- "errors": 0,
- "words": 6,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "you have to be kidding me",
- "normalized_hypothesis": "you have to be kidding me",
- "errors": 0,
- "words": 6,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_148",
- "category": "technical",
- "text": "If the checksum fails, restart the worker process without deleting the cached tensors.",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_148.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "If the checksum fails, restart the worker process without deleting the cache tensors.",
- "raw": {
- "normalized_reference": "if the checksum fails, restart the worker process without deleting the cached tensors.",
- "normalized_hypothesis": "if the checksum fails, restart the worker process without deleting the cache tensors.",
- "errors": 1,
- "words": 13,
- "wer": 0.07692307692307693
- },
- "semantic": {
- "normalized_reference": "if the checksum fails restart the worker process without deleting the cached tensors",
- "normalized_hypothesis": "if the checksum fails restart the worker process without deleting the cache tensors",
- "errors": 1,
- "words": 13,
- "wer": 0.07692307692307693
- }
- },
- {
- "id": "utmos200_149",
- "category": "conversational",
- "text": "I need you to understand how close that was to not working.",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_149.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "I need you to understand how close that was to not working.",
- "raw": {
- "normalized_reference": "i need you to understand how close that was to not working.",
- "normalized_hypothesis": "i need you to understand how close that was to not working.",
- "errors": 0,
- "words": 12,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "i need you to understand how close that was to not working",
- "normalized_hypothesis": "i need you to understand how close that was to not working",
- "errors": 0,
- "words": 12,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_150",
- "category": "descriptive",
- "text": "It was the sort of morning where the whole world looks rinsed clean.",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_150.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "It was the sort of morning where the whole world looks rinsed clean.",
- "raw": {
- "normalized_reference": "it was the sort of morning where the whole world looks rinsed clean.",
- "normalized_hypothesis": "it was the sort of morning where the whole world looks rinsed clean.",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "it was the sort of morning where the whole world looks rinsed clean",
- "normalized_hypothesis": "it was the sort of morning where the whole world looks rinsed clean",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_151",
- "category": "emotional",
- "text": "She started laughing and I started laughing and we couldn't stop for a full minute.",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_151.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "She started laughing, and I started laughing, and we couldn't stop for a full minute.",
- "raw": {
- "normalized_reference": "she started laughing and i started laughing and we couldn't stop for a full minute.",
- "normalized_hypothesis": "she started laughing, and i started laughing, and we couldn't stop for a full minute.",
- "errors": 0,
- "words": 15,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "she started laughing and i started laughing and we could not stop for a full minute",
- "normalized_hypothesis": "she started laughing and i started laughing and we could not stop for a full minute",
- "errors": 0,
- "words": 16,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_152",
- "category": "long",
- "text": "We spent three years building something we were genuinely proud of, and then the market shifted and none of it mattered the way we thought it would.",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_152.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "We spent three years building something we were genuinely proud of, and then the market shifted and none of it mattered the way we thought it would.",
- "raw": {
- "normalized_reference": "we spent three years building something we were genuinely proud of, and then the market shifted and none of it mattered the way we thought it would.",
- "normalized_hypothesis": "we spent three years building something we were genuinely proud of, and then the market shifted and none of it mattered the way we thought it would.",
- "errors": 0,
- "words": 27,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "we spent 3 years building something we were genuinely proud of and then the market shifted and none of it mattered the way we thought it would",
- "normalized_hypothesis": "we spent 3 years building something we were genuinely proud of and then the market shifted and none of it mattered the way we thought it would",
- "errors": 0,
- "words": 27,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_153",
- "category": "narrative",
- "text": "And somehow, despite all of that, it actually turned out okay.",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_153.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "And somehow, despite all of that, it actually turned out okay.",
- "raw": {
- "normalized_reference": "and somehow, despite all of that, it actually turned out okay.",
- "normalized_hypothesis": "and somehow, despite all of that, it actually turned out okay.",
- "errors": 0,
- "words": 11,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "and somehow despite all of that it actually turned out okay",
- "normalized_hypothesis": "and somehow despite all of that it actually turned out okay",
- "errors": 0,
- "words": 11,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_154",
- "category": "question",
- "text": "Is that something you actually want, or something you think you should want?",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_154.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Is that something you actually want or something you think you should want?",
- "raw": {
- "normalized_reference": "is that something you actually want, or something you think you should want?",
- "normalized_hypothesis": "is that something you actually want or something you think you should want?",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "is that something you actually want or something you think you should want",
- "normalized_hypothesis": "is that something you actually want or something you think you should want",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_155",
- "category": "short",
- "text": "This is fine.",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_155.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "This is fine.",
- "raw": {
- "normalized_reference": "this is fine.",
- "normalized_hypothesis": "this is fine.",
- "errors": 0,
- "words": 3,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "this is fine",
- "normalized_hypothesis": "this is fine",
- "errors": 0,
- "words": 3,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_156",
- "category": "technical",
- "text": "The tricky part isn't the setup, it's making sure all dependencies line up.",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_156.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "The tricky part isn't the setup, it's making sure all dependencies line up.",
- "raw": {
- "normalized_reference": "the tricky part isn't the setup, it's making sure all dependencies line up.",
- "normalized_hypothesis": "the tricky part isn't the setup, it's making sure all dependencies line up.",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "the tricky part is not the setup it is making sure all dependencies line up",
- "normalized_hypothesis": "the tricky part is not the setup it is making sure all dependencies line up",
- "errors": 0,
- "words": 15,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_157",
- "category": "conversational",
- "text": "He called it a minor setback. I called it a disaster.",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_157.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "He called it a minor setback. I called it a disaster.",
- "raw": {
- "normalized_reference": "he called it a minor setback. i called it a disaster.",
- "normalized_hypothesis": "he called it a minor setback. i called it a disaster.",
- "errors": 0,
- "words": 11,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "he called it a minor setback i called it a disaster",
- "normalized_hypothesis": "he called it a minor setback i called it a disaster",
- "errors": 0,
- "words": 11,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_158",
- "category": "descriptive",
- "text": "The whole place smelled like pine and old wood and cold air.",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_158.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "The whole place smelled like pine and old wood and cold air.",
- "raw": {
- "normalized_reference": "the whole place smelled like pine and old wood and cold air.",
- "normalized_hypothesis": "the whole place smelled like pine and old wood and cold air.",
- "errors": 0,
- "words": 12,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "the whole place smelled like pine and old wood and cold air",
- "normalized_hypothesis": "the whole place smelled like pine and old wood and cold air",
- "errors": 0,
- "words": 12,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_159",
- "category": "emotional",
- "text": "I've never been so relieved in my entire life.",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_159.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "I've never been so relieved in my entire life.",
- "raw": {
- "normalized_reference": "i've never been so relieved in my entire life.",
- "normalized_hypothesis": "i've never been so relieved in my entire life.",
- "errors": 0,
- "words": 9,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "i have never been so relieved in my entire life",
- "normalized_hypothesis": "i have never been so relieved in my entire life",
- "errors": 0,
- "words": 10,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_160",
- "category": "narrative",
- "text": "I wrote it all down that night so I wouldn't forget, and I still haven't looked at it.",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_160.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "I wrote it all down that night so I wouldn't forget, and I still haven't looked at it.",
- "raw": {
- "normalized_reference": "i wrote it all down that night so i wouldn't forget, and i still haven't looked at it.",
- "normalized_hypothesis": "i wrote it all down that night so i wouldn't forget, and i still haven't looked at it.",
- "errors": 0,
- "words": 18,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "i wrote it all down that night so i would not forget and i still have not looked at it",
- "normalized_hypothesis": "i wrote it all down that night so i would not forget and i still have not looked at it",
- "errors": 0,
- "words": 20,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_161",
- "category": "short",
- "text": "That's kind of wild actually.",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_161.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "That's kind of wild actually.",
- "raw": {
- "normalized_reference": "that's kind of wild actually.",
- "normalized_hypothesis": "that's kind of wild actually.",
- "errors": 0,
- "words": 5,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "that is kind of wild actually",
- "normalized_hypothesis": "that is kind of wild actually",
- "errors": 0,
- "words": 6,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_162",
- "category": "technical",
- "text": "Version pinning will save you a lot of pain later down the line.",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_162.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Version pinning will save you a lot of pain later down the line.",
- "raw": {
- "normalized_reference": "version pinning will save you a lot of pain later down the line.",
- "normalized_hypothesis": "version pinning will save you a lot of pain later down the line.",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "version pinning will save you a lot of pain later down the line",
- "normalized_hypothesis": "version pinning will save you a lot of pain later down the line",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_163",
- "category": "conversational",
- "text": "At some point you just have to commit and see what happens.",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_163.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "At some point you just have to commit and see what happens.",
- "raw": {
- "normalized_reference": "at some point you just have to commit and see what happens.",
- "normalized_hypothesis": "at some point you just have to commit and see what happens.",
- "errors": 0,
- "words": 12,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "at some point you just have to commit and see what happens",
- "normalized_hypothesis": "at some point you just have to commit and see what happens",
- "errors": 0,
- "words": 12,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_164",
- "category": "emotional",
- "text": "Nobody told me it was going to be this hard, and honestly, I'm grateful.",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_164.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Nobody told me it was going to be this hard, and honestly, I'm grateful.",
- "raw": {
- "normalized_reference": "nobody told me it was going to be this hard, and honestly, i'm grateful.",
- "normalized_hypothesis": "nobody told me it was going to be this hard, and honestly, i'm grateful.",
- "errors": 0,
- "words": 14,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "nobody told me it was going to be this hard and honestly i am grateful",
- "normalized_hypothesis": "nobody told me it was going to be this hard and honestly i am grateful",
- "errors": 0,
- "words": 15,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_165",
- "category": "narrative",
- "text": "Chapter seven. In which everything that could go wrong did, and a few things that couldn't.",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_165.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Chapter 7, in which everything that could go wrong did, and a few things that couldn't.",
- "raw": {
- "normalized_reference": "chapter seven. in which everything that could go wrong did, and a few things that couldn't.",
- "normalized_hypothesis": "chapter 7, in which everything that could go wrong did, and a few things that couldn't.",
- "errors": 1,
- "words": 16,
- "wer": 0.0625
- },
- "semantic": {
- "normalized_reference": "chapter 7 in which everything that could go wrong did and a few things that could not",
- "normalized_hypothesis": "chapter 7 in which everything that could go wrong did and a few things that could not",
- "errors": 0,
- "words": 17,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_166",
- "category": "short",
- "text": "Wait, what?",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_166.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Wait, what?",
- "raw": {
- "normalized_reference": "wait, what?",
- "normalized_hypothesis": "wait, what?",
- "errors": 0,
- "words": 2,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "wait what",
- "normalized_hypothesis": "wait what",
- "errors": 0,
- "words": 2,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_167",
- "category": "technical",
- "text": "The graphics processor remained idle because the data loader exhausted every central processing core.",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_167.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "The graphics processor remained idle because the data loader exhausted every central processing core.",
- "raw": {
- "normalized_reference": "the graphics processor remained idle because the data loader exhausted every central processing core.",
- "normalized_hypothesis": "the graphics processor remained idle because the data loader exhausted every central processing core.",
- "errors": 0,
- "words": 14,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "the graphics processor remained idle because the data loader exhausted every central processing core",
- "normalized_hypothesis": "the graphics processor remained idle because the data loader exhausted every central processing core",
- "errors": 0,
- "words": 14,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_168",
- "category": "conversational",
- "text": "Could you speak a little more slowly? I am trying to write every detail down.",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_168.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Could you speak a little more slowly? I am trying to write every detail down.",
- "raw": {
- "normalized_reference": "could you speak a little more slowly? i am trying to write every detail down.",
- "normalized_hypothesis": "could you speak a little more slowly? i am trying to write every detail down.",
- "errors": 0,
- "words": 15,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "could you speak a little more slowly i am trying to write every detail down",
- "normalized_hypothesis": "could you speak a little more slowly i am trying to write every detail down",
- "errors": 0,
- "words": 15,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_169",
- "category": "emotional",
- "text": "I was so excited I could barely form a complete sentence.",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_169.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "I was so excited I could barely form a complete sentence.",
- "raw": {
- "normalized_reference": "i was so excited i could barely form a complete sentence.",
- "normalized_hypothesis": "i was so excited i could barely form a complete sentence.",
- "errors": 0,
- "words": 11,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "i was so excited i could barely form a complete sentence",
- "normalized_hypothesis": "i was so excited i could barely form a complete sentence",
- "errors": 0,
- "words": 11,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_170",
- "category": "narrative",
- "text": "He looked at me like I had said something in a language he didn't speak.",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_170.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "He looked at me like I had said something in a language he didn't speak.",
- "raw": {
- "normalized_reference": "he looked at me like i had said something in a language he didn't speak.",
- "normalized_hypothesis": "he looked at me like i had said something in a language he didn't speak.",
- "errors": 0,
- "words": 15,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "he looked at me like i had said something in a language he did not speak",
- "normalized_hypothesis": "he looked at me like i had said something in a language he did not speak",
- "errors": 0,
- "words": 16,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_171",
- "category": "short",
- "text": "That's actually hilarious.",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_171.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "That's actually hilarious.",
- "raw": {
- "normalized_reference": "that's actually hilarious.",
- "normalized_hypothesis": "that's actually hilarious.",
- "errors": 0,
- "words": 3,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "that is actually hilarious",
- "normalized_hypothesis": "that is actually hilarious",
- "errors": 0,
- "words": 4,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_172",
- "category": "technical",
- "text": "The encoder maps phonemes, stress, and phrase boundaries into a shared hidden representation.",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_172.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "The encoder maps phonemes, stress, and phrase boundaries into a shared hidden representation.",
- "raw": {
- "normalized_reference": "the encoder maps phonemes, stress, and phrase boundaries into a shared hidden representation.",
- "normalized_hypothesis": "the encoder maps phonemes, stress, and phrase boundaries into a shared hidden representation.",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "the encoder maps phonemes stress and phrase boundaries into a shared hidden representation",
- "normalized_hypothesis": "the encoder maps phonemes stress and phrase boundaries into a shared hidden representation",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_173",
- "category": "conversational",
- "text": "She showed up three hours late and acted like nothing had happened.",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_173.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "She showed up three hours late and acted like nothing had happened.",
- "raw": {
- "normalized_reference": "she showed up three hours late and acted like nothing had happened.",
- "normalized_hypothesis": "she showed up three hours late and acted like nothing had happened.",
- "errors": 0,
- "words": 12,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "she showed up 3 hours late and acted like nothing had happened",
- "normalized_hypothesis": "she showed up 3 hours late and acted like nothing had happened",
- "errors": 0,
- "words": 12,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_174",
- "category": "emotional",
- "text": "Oh my god. This is the best thing I've ever tasted.",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_174.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Oh my god, this is the best thing I've ever tasted.",
- "raw": {
- "normalized_reference": "oh my god. this is the best thing i've ever tasted.",
- "normalized_hypothesis": "oh my god, this is the best thing i've ever tasted.",
- "errors": 0,
- "words": 11,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "0 my god this is the best thing i have ever tasted",
- "normalized_hypothesis": "0 my god this is the best thing i have ever tasted",
- "errors": 0,
- "words": 12,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_175",
- "category": "narrative",
- "text": "The note was short, which was somehow more frightening than a long one would have been.",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_175.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "The note was short, which was somehow more frightening than a long one would have been.",
- "raw": {
- "normalized_reference": "the note was short, which was somehow more frightening than a long one would have been.",
- "normalized_hypothesis": "the note was short, which was somehow more frightening than a long one would have been.",
- "errors": 0,
- "words": 16,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "the note was short which was somehow more frightening than a long one would have been",
- "normalized_hypothesis": "the note was short which was somehow more frightening than a long one would have been",
- "errors": 0,
- "words": 16,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_176",
- "category": "short",
- "text": "Noted.",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_176.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Noid.",
- "raw": {
- "normalized_reference": "noted.",
- "normalized_hypothesis": "noid.",
- "errors": 1,
- "words": 1,
- "wer": 1.0
- },
- "semantic": {
- "normalized_reference": "noted",
- "normalized_hypothesis": "noid",
- "errors": 1,
- "words": 1,
- "wer": 1.0
- }
- },
- {
- "id": "utmos200_177",
- "category": "technical",
- "text": "It's a simple process once you understand the underlying structure.",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_177.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "It's a simple process once you understand the underlying structure.",
- "raw": {
- "normalized_reference": "it's a simple process once you understand the underlying structure.",
- "normalized_hypothesis": "it's a simple process once you understand the underlying structure.",
- "errors": 0,
- "words": 10,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "it is a simple process once you understand the underlying structure",
- "normalized_hypothesis": "it is a simple process once you understand the underlying structure",
- "errors": 0,
- "words": 11,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_178",
- "category": "conversational",
- "text": "You have to hear this story because it's almost too good to be true.",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_178.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "You have to hear this story because it's almost too good to be true.",
- "raw": {
- "normalized_reference": "you have to hear this story because it's almost too good to be true.",
- "normalized_hypothesis": "you have to hear this story because it's almost too good to be true.",
- "errors": 0,
- "words": 14,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "you have to hear this story because it is almost too good to be true",
- "normalized_hypothesis": "you have to hear this story because it is almost too good to be true",
- "errors": 0,
- "words": 15,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_179",
- "category": "emotional",
- "text": "I have never felt more seen in my entire life than in that moment.",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_179.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "I have never felt more seen in my entire life than in that moment.",
- "raw": {
- "normalized_reference": "i have never felt more seen in my entire life than in that moment.",
- "normalized_hypothesis": "i have never felt more seen in my entire life than in that moment.",
- "errors": 0,
- "words": 14,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "i have never felt more seen in my entire life than in that moment",
- "normalized_hypothesis": "i have never felt more seen in my entire life than in that moment",
- "errors": 0,
- "words": 14,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_180",
- "category": "narrative",
- "text": "She knew. She had always known. She just waited for the right moment.",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_180.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "She knew, she had always known, she just waited for the right moment.",
- "raw": {
- "normalized_reference": "she knew. she had always known. she just waited for the right moment.",
- "normalized_hypothesis": "she knew, she had always known, she just waited for the right moment.",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "she knew she had always known she just waited for the right moment",
- "normalized_hypothesis": "she knew she had always known she just waited for the right moment",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_181",
- "category": "short",
- "text": "Yeah, no. Hard pass.",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_181.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Yeah, no but hard pass.",
- "raw": {
- "normalized_reference": "yeah, no. hard pass.",
- "normalized_hypothesis": "yeah, no but hard pass.",
- "errors": 1,
- "words": 4,
- "wer": 0.25
- },
- "semantic": {
- "normalized_reference": "yeah no hard pass",
- "normalized_hypothesis": "yeah no but hard pass",
- "errors": 1,
- "words": 4,
- "wer": 0.25
- }
- },
- {
- "id": "utmos200_182",
- "category": "technical",
- "text": "Please verify the sample rate, hop length, and mel filter configuration before inference.",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_182.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Please verify the sample rate, hop length, and MEL filter configuration before inference.",
- "raw": {
- "normalized_reference": "please verify the sample rate, hop length, and mel filter configuration before inference.",
- "normalized_hypothesis": "please verify the sample rate, hop length, and mel filter configuration before inference.",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "please verify the sample rate hop length and mel filter configuration before inference",
- "normalized_hypothesis": "please verify the sample rate hop length and mel filter configuration before inference",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_183",
- "category": "conversational",
- "text": "That's the worst idea I've ever heard and I'm completely in.",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_183.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "That's the worst idea I've ever heard and I'm completely in.",
- "raw": {
- "normalized_reference": "that's the worst idea i've ever heard and i'm completely in.",
- "normalized_hypothesis": "that's the worst idea i've ever heard and i'm completely in.",
- "errors": 0,
- "words": 11,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "that is the worst idea i have ever heard and i am completely in",
- "normalized_hypothesis": "that is the worst idea i have ever heard and i am completely in",
- "errors": 0,
- "words": 14,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_184",
- "category": "emotional",
- "text": "I remember thinking: this is it. This is the thing I've been waiting for.",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_184.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "I remember thinking, this is it, this is the thing I've been waiting for.",
- "raw": {
- "normalized_reference": "i remember thinking: this is it. this is the thing i've been waiting for.",
- "normalized_hypothesis": "i remember thinking, this is it, this is the thing i've been waiting for.",
- "errors": 0,
- "words": 14,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "i remember thinking this is it this is the thing i have been waiting for",
- "normalized_hypothesis": "i remember thinking this is it this is the thing i have been waiting for",
- "errors": 0,
- "words": 15,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_185",
- "category": "narrative",
- "text": "We never talked about it again, which I think was the right call.",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_185.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "We never talked about it again, which I think was the right call.",
- "raw": {
- "normalized_reference": "we never talked about it again, which i think was the right call.",
- "normalized_hypothesis": "we never talked about it again, which i think was the right call.",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "we never talked about it again which i think was the right call",
- "normalized_hypothesis": "we never talked about it again which i think was the right call",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_186",
- "category": "short",
- "text": "In hindsight, obvious.",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_186.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "In hindsight, obvious.",
- "raw": {
- "normalized_reference": "in hindsight, obvious.",
- "normalized_hypothesis": "in hindsight, obvious.",
- "errors": 0,
- "words": 3,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "in hindsight obvious",
- "normalized_hypothesis": "in hindsight obvious",
- "errors": 0,
- "words": 3,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_187",
- "category": "technical",
- "text": "Once it's configured correctly, it basically runs itself.",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_187.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Once it's configured correctly, it basically runs itself.",
- "raw": {
- "normalized_reference": "once it's configured correctly, it basically runs itself.",
- "normalized_hypothesis": "once it's configured correctly, it basically runs itself.",
- "errors": 0,
- "words": 8,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "once it is configured correctly it basically runs itself",
- "normalized_hypothesis": "once it is configured correctly it basically runs itself",
- "errors": 0,
- "words": 9,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_188",
- "category": "conversational",
- "text": "You got engaged? When? How? Tell me everything right now.",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_188.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "You got engaged? When? How? Tell me everything right now.",
- "raw": {
- "normalized_reference": "you got engaged? when? how? tell me everything right now.",
- "normalized_hypothesis": "you got engaged? when? how? tell me everything right now.",
- "errors": 0,
- "words": 10,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "you got engaged when how tell me everything right now",
- "normalized_hypothesis": "you got engaged when how tell me everything right now",
- "errors": 0,
- "words": 10,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_189",
- "category": "emotional",
- "text": "It's hard to explain how much that moment meant to me.",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_189.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "It's hard to explain how much that moment meant to me.",
- "raw": {
- "normalized_reference": "it's hard to explain how much that moment meant to me.",
- "normalized_hypothesis": "it's hard to explain how much that moment meant to me.",
- "errors": 0,
- "words": 11,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "it is hard to explain how much that moment meant to me",
- "normalized_hypothesis": "it is hard to explain how much that moment meant to me",
- "errors": 0,
- "words": 12,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_190",
- "category": "narrative",
- "text": "He told the story three different times that night and somehow it got better each time.",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_190.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "He told the story three different times that night and somehow it got better each time.",
- "raw": {
- "normalized_reference": "he told the story three different times that night and somehow it got better each time.",
- "normalized_hypothesis": "he told the story three different times that night and somehow it got better each time.",
- "errors": 0,
- "words": 16,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "he told the story 3 different times that night and somehow it got better each time",
- "normalized_hypothesis": "he told the story 3 different times that night and somehow it got better each time",
- "errors": 0,
- "words": 16,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_191",
- "category": "short",
- "text": "Right. Moving on.",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_191.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Right, moving on.",
- "raw": {
- "normalized_reference": "right. moving on.",
- "normalized_hypothesis": "right, moving on.",
- "errors": 0,
- "words": 3,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "right moving on",
- "normalized_hypothesis": "right moving on",
- "errors": 0,
- "words": 3,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_192",
- "category": "conversational",
- "text": "No, I did not move your keys; check beside the small ceramic bowl.",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_192.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "No, I did not move your keys. Check beside the small ceramic bowl.",
- "raw": {
- "normalized_reference": "no, i did not move your keys; check beside the small ceramic bowl.",
- "normalized_hypothesis": "no, i did not move your keys. check beside the small ceramic bowl.",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "no i did not move your keys check beside the small ceramic bowl",
- "normalized_hypothesis": "no i did not move your keys check beside the small ceramic bowl",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_193",
- "category": "emotional",
- "text": "I was devastated. I'm not going to pretend otherwise.",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_193.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "I was devastated. I'm not going to pretend otherwise.",
- "raw": {
- "normalized_reference": "i was devastated. i'm not going to pretend otherwise.",
- "normalized_hypothesis": "i was devastated. i'm not going to pretend otherwise.",
- "errors": 0,
- "words": 9,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "i was devastated i am not going to pretend otherwise",
- "normalized_hypothesis": "i was devastated i am not going to pretend otherwise",
- "errors": 0,
- "words": 10,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_194",
- "category": "narrative",
- "text": "That was the last summer before everything changed, though we didn't know it yet.",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_194.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "That was the last summer before everything changed, though we didn't know it yet.",
- "raw": {
- "normalized_reference": "that was the last summer before everything changed, though we didn't know it yet.",
- "normalized_hypothesis": "that was the last summer before everything changed, though we didn't know it yet.",
- "errors": 0,
- "words": 14,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "that was the last summer before everything changed though we did not know it yet",
- "normalized_hypothesis": "that was the last summer before everything changed though we did not know it yet",
- "errors": 0,
- "words": 15,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_195",
- "category": "short",
- "text": "Bold of you to assume.",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_195.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Bold of you to assume.",
- "raw": {
- "normalized_reference": "bold of you to assume.",
- "normalized_hypothesis": "bold of you to assume.",
- "errors": 0,
- "words": 5,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "bold of you to assume",
- "normalized_hypothesis": "bold of you to assume",
- "errors": 0,
- "words": 5,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_196",
- "category": "conversational",
- "text": "It's one of those decisions that makes total sense in the moment.",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_196.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "It's one of those decisions that makes total sense in the moment.",
- "raw": {
- "normalized_reference": "it's one of those decisions that makes total sense in the moment.",
- "normalized_hypothesis": "it's one of those decisions that makes total sense in the moment.",
- "errors": 0,
- "words": 12,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "it is one of those decisions that makes total sense in the moment",
- "normalized_hypothesis": "it is one of those decisions that makes total sense in the moment",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_197",
- "category": "emotional",
- "text": "I was furious. Not the loud kind. The quiet kind that lasts for days.",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_197.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "I was furious, not the loud kind, the quiet kind that lasts for days.",
- "raw": {
- "normalized_reference": "i was furious. not the loud kind. the quiet kind that lasts for days.",
- "normalized_hypothesis": "i was furious, not the loud kind, the quiet kind that lasts for days.",
- "errors": 0,
- "words": 14,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "i was furious not the loud kind the quiet kind that lasts for days",
- "normalized_hypothesis": "i was furious not the loud kind the quiet kind that lasts for days",
- "errors": 0,
- "words": 14,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_198",
- "category": "narrative",
- "text": "That's the part of the story I usually leave out.",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_198.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "That's the part of the story I usually leave out.",
- "raw": {
- "normalized_reference": "that's the part of the story i usually leave out.",
- "normalized_hypothesis": "that's the part of the story i usually leave out.",
- "errors": 0,
- "words": 10,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "that is the part of the story i usually leave out",
- "normalized_hypothesis": "that is the part of the story i usually leave out",
- "errors": 0,
- "words": 11,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_199",
- "category": "short",
- "text": "I've said too much.",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_199.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "I've said too much.",
- "raw": {
- "normalized_reference": "i've said too much.",
- "normalized_hypothesis": "i've said too much.",
- "errors": 0,
- "words": 4,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "i have said too much",
- "normalized_hypothesis": "i have said too much",
- "errors": 0,
- "words": 5,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_200",
- "category": "conversational",
- "text": "I've been faking that I understood this the entire time.",
- "model": "inflect-micro-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_200.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "I've been faking that I understood this the entire time.",
- "raw": {
- "normalized_reference": "i've been faking that i understood this the entire time.",
- "normalized_hypothesis": "i've been faking that i understood this the entire time.",
- "errors": 0,
- "words": 10,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "i have been faking that i understood this the entire time",
- "normalized_hypothesis": "i have been faking that i understood this the entire time",
- "errors": 0,
- "words": 11,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_001",
- "category": "conversational",
- "text": "It wasn't until later that I realized what had actually happened.",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_001.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "It wasn't until later that I realized what had actually happened.",
- "raw": {
- "normalized_reference": "it wasn't until later that i realized what had actually happened.",
- "normalized_hypothesis": "it wasn't until later that i realized what had actually happened.",
- "errors": 0,
- "words": 11,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "it was not until later that i realized what had actually happened",
- "normalized_hypothesis": "it was not until later that i realized what had actually happened",
- "errors": 0,
- "words": 12,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_002",
- "category": "descriptive",
- "text": "The light at that time of afternoon is perfect for about twenty minutes.",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_002.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "The light at that time of afternoon is perfect for about 20 minutes.",
- "raw": {
- "normalized_reference": "the light at that time of afternoon is perfect for about twenty minutes.",
- "normalized_hypothesis": "the light at that time of afternoon is perfect for about 20 minutes.",
- "errors": 1,
- "words": 13,
- "wer": 0.07692307692307693
- },
- "semantic": {
- "normalized_reference": "the light at that time of afternoon is perfect for about 20 minutes",
- "normalized_hypothesis": "the light at that time of afternoon is perfect for about 20 minutes",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_003",
- "category": "emotional",
- "text": "That was the scariest thing I've ever done, and also maybe the best.",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_003.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "That was the scariest thing I've ever done, and also maybe the best.",
- "raw": {
- "normalized_reference": "that was the scariest thing i've ever done, and also maybe the best.",
- "normalized_hypothesis": "that was the scariest thing i've ever done, and also maybe the best.",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "that was the scariest thing i have ever done and also maybe the best",
- "normalized_hypothesis": "that was the scariest thing i have ever done and also maybe the best",
- "errors": 0,
- "words": 14,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_004",
- "category": "homographs",
- "text": "After I record the record, please present the present to the bass player.",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_004.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "After I record the record, please present the present to the bass player.",
- "raw": {
- "normalized_reference": "after i record the record, please present the present to the bass player.",
- "normalized_hypothesis": "after i record the record, please present the present to the bass player.",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "after i record the record please present the present to the bass player",
- "normalized_hypothesis": "after i record the record please present the present to the bass player",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_005",
- "category": "long",
- "text": "If I'm being completely honest, I had no idea what I was doing for most of it, and I think that might be the most normal thing about the whole experience.",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_005.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "If I'm being completely honest, I had no idea what I was doing for most of it, and I think that might be the most normal thing about the whole experience.",
- "raw": {
- "normalized_reference": "if i'm being completely honest, i had no idea what i was doing for most of it, and i think that might be the most normal thing about the whole experience.",
- "normalized_hypothesis": "if i'm being completely honest, i had no idea what i was doing for most of it, and i think that might be the most normal thing about the whole experience.",
- "errors": 0,
- "words": 31,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "if i am being completely honest i had no idea what i was doing for most of it and i think that might be the most normal thing about the whole experience",
- "normalized_hypothesis": "if i am being completely honest i had no idea what i was doing for most of it and i think that might be the most normal thing about the whole experience",
- "errors": 0,
- "words": 32,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_006",
- "category": "long_range",
- "text": "The committee postponed its decision after discovering that three pages of the report had been printed in the wrong order.",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_006.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "The committee postponed its decision after discovering that three pages of the report had been printed in the wrong order.",
- "raw": {
- "normalized_reference": "the committee postponed its decision after discovering that three pages of the report had been printed in the wrong order.",
- "normalized_hypothesis": "the committee postponed its decision after discovering that three pages of the report had been printed in the wrong order.",
- "errors": 0,
- "words": 20,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "the committee postponed its decision after discovering that 3 pages of the report had been printed in the wrong order",
- "normalized_hypothesis": "the committee postponed its decision after discovering that 3 pages of the report had been printed in the wrong order",
- "errors": 0,
- "words": 20,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_007",
- "category": "names_places",
- "text": "Gwendolyn photographed the eucalyptus trees outside Ljubljana.",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_007.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Gwendolyn photographed the eucalyptus trees outside Ljubljana.",
- "raw": {
- "normalized_reference": "gwendolyn photographed the eucalyptus trees outside ljubljana.",
- "normalized_hypothesis": "gwendolyn photographed the eucalyptus trees outside ljubljana.",
- "errors": 0,
- "words": 7,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "gwendolyn photographed the eucalyptus trees outside ljubljana",
- "normalized_hypothesis": "gwendolyn photographed the eucalyptus trees outside ljubljana",
- "errors": 0,
- "words": 7,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_008",
- "category": "narrative",
- "text": "He had this theory, and the more time passes, the more I think he was right.",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_008.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "He had this theory, and the more time passes, the more I think he was right.",
- "raw": {
- "normalized_reference": "he had this theory, and the more time passes, the more i think he was right.",
- "normalized_hypothesis": "he had this theory, and the more time passes, the more i think he was right.",
- "errors": 0,
- "words": 16,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "he had this theory and the more time passes the more i think he was right",
- "normalized_hypothesis": "he had this theory and the more time passes the more i think he was right",
- "errors": 0,
- "words": 16,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_009",
- "category": "numbers",
- "text": "The package weighs twelve point six kilograms and arrived on July twenty-first.",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_009.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "The package weighs 12.6kg and arrived on July 21.",
- "raw": {
- "normalized_reference": "the package weighs twelve point six kilograms and arrived on july twenty-first.",
- "normalized_hypothesis": "the package weighs 12.6kg and arrived on july 21.",
- "errors": 6,
- "words": 13,
- "wer": 0.46153846153846156
- },
- "semantic": {
- "normalized_reference": "the package weighs 12.6 kilograms and arrived on july 21st",
- "normalized_hypothesis": "the package weighs 12.6 kg and arrived on july 21",
- "errors": 2,
- "words": 11,
- "wer": 0.18181818181818182
- }
- },
- {
- "id": "utmos200_010",
- "category": "punctuation",
- "text": "First, close the window; second, turn off the lamp; finally, lock the door.",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_010.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "First, close the window. Second, turn off the lamp. Finally, lock the door.",
- "raw": {
- "normalized_reference": "first, close the window; second, turn off the lamp; finally, lock the door.",
- "normalized_hypothesis": "first, close the window. second, turn off the lamp. finally, lock the door.",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "1st close the window 2nd turn off the lamp finally lock the door",
- "normalized_hypothesis": "1st close the window 2nd turn off the lamp finally lock the door",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_011",
- "category": "question",
- "text": "When did you first realize something was wrong?",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_011.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "When did you first realize something was wrong?",
- "raw": {
- "normalized_reference": "when did you first realize something was wrong?",
- "normalized_hypothesis": "when did you first realize something was wrong?",
- "errors": 0,
- "words": 8,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "when did you 1st realize something was wrong",
- "normalized_hypothesis": "when did you 1st realize something was wrong",
- "errors": 0,
- "words": 8,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_012",
- "category": "rare_words",
- "text": "The archaeologist catalogued a Byzantine reliquary and several anachronistic fragments.",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_012.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "the archaeologist catalogued a Byzantine reliquary and several anachronistic fragments.",
- "raw": {
- "normalized_reference": "the archaeologist catalogued a byzantine reliquary and several anachronistic fragments.",
- "normalized_hypothesis": "the archaeologist catalogued a byzantine reliquary and several anachronistic fragments.",
- "errors": 0,
- "words": 10,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "the archaeologist catalogued a byzantine reliquary and several anachronistic fragments",
- "normalized_hypothesis": "the archaeologist catalogued a byzantine reliquary and several anachronistic fragments",
- "errors": 0,
- "words": 10,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_013",
- "category": "short",
- "text": "I have no idea.",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_013.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "I have no idea.",
- "raw": {
- "normalized_reference": "i have no idea.",
- "normalized_hypothesis": "i have no idea.",
- "errors": 0,
- "words": 4,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "i have no idea",
- "normalized_hypothesis": "i have no idea",
- "errors": 0,
- "words": 4,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_014",
- "category": "technical",
- "text": "The system runs on three core components that all have to stay in sync.",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_014.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "The system runs on three core components that all have to stay in sync.",
- "raw": {
- "normalized_reference": "the system runs on three core components that all have to stay in sync.",
- "normalized_hypothesis": "the system runs on three core components that all have to stay in sync.",
- "errors": 0,
- "words": 14,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "the system runs on 3 core components that all have to stay in sync",
- "normalized_hypothesis": "the system runs on 3 core components that all have to stay in sync",
- "errors": 0,
- "words": 14,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_015",
- "category": "conversational",
- "text": "We improvised the entire second half and somehow nobody noticed.",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_015.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "We improvised the entire second half and somehow nobody noticed.",
- "raw": {
- "normalized_reference": "we improvised the entire second half and somehow nobody noticed.",
- "normalized_hypothesis": "we improvised the entire second half and somehow nobody noticed.",
- "errors": 0,
- "words": 10,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "we improvised the entire 2nd half and somehow nobody noticed",
- "normalized_hypothesis": "we improvised the entire 2nd half and somehow nobody noticed",
- "errors": 0,
- "words": 10,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_016",
- "category": "descriptive",
- "text": "There's a smell in old buildings that's impossible to describe but immediately familiar.",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_016.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "There's a smell in old buildings that's impossible to describe but immediately familiar.",
- "raw": {
- "normalized_reference": "there's a smell in old buildings that's impossible to describe but immediately familiar.",
- "normalized_hypothesis": "there's a smell in old buildings that's impossible to describe but immediately familiar.",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "there is a smell in old buildings that is impossible to describe but immediately familiar",
- "normalized_hypothesis": "there is a smell in old buildings that is impossible to describe but immediately familiar",
- "errors": 0,
- "words": 15,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_017",
- "category": "emotional",
- "text": "I specifically asked for one thing. One thing. And it still didn't happen.",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_017.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "I specifically asked for one thing, one thing, and it still didn't happen.",
- "raw": {
- "normalized_reference": "i specifically asked for one thing. one thing. and it still didn't happen.",
- "normalized_hypothesis": "i specifically asked for one thing, one thing, and it still didn't happen.",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "i specifically asked for one thing one thing and it still did not happen",
- "normalized_hypothesis": "i specifically asked for one thing one thing and it still did not happen",
- "errors": 0,
- "words": 14,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_018",
- "category": "homographs",
- "text": "They refuse to place the refuse beside the produce aisle.",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_018.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "They refused to place the refuse beside the produce aisle.",
- "raw": {
- "normalized_reference": "they refuse to place the refuse beside the produce aisle.",
- "normalized_hypothesis": "they refused to place the refuse beside the produce aisle.",
- "errors": 1,
- "words": 10,
- "wer": 0.1
- },
- "semantic": {
- "normalized_reference": "they refuse to place the refuse beside the produce aisle",
- "normalized_hypothesis": "they refused to place the refuse beside the produce aisle",
- "errors": 1,
- "words": 10,
- "wer": 0.1
- }
- },
- {
- "id": "utmos200_019",
- "category": "long",
- "text": "There were days last year when I genuinely didn't know how I was going to get through it, and somehow I did, and I still don't fully understand how.",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_019.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "There were days last year when I genuinely didn't know how I was going to get through it, and somehow I did, and I still don't fully understand how.",
- "raw": {
- "normalized_reference": "there were days last year when i genuinely didn't know how i was going to get through it, and somehow i did, and i still don't fully understand how.",
- "normalized_hypothesis": "there were days last year when i genuinely didn't know how i was going to get through it, and somehow i did, and i still don't fully understand how.",
- "errors": 0,
- "words": 29,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "there were days last year when i genuinely did not know how i was going to get through it and somehow i did and i still do not fully understand how",
- "normalized_hypothesis": "there were days last year when i genuinely did not know how i was going to get through it and somehow i did and i still do not fully understand how",
- "errors": 0,
- "words": 31,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_020",
- "category": "long_range",
- "text": "Before announcing the winner, the judge thanked the volunteers who had worked through the storm to repair the damaged stage.",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_020.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Before announcing the winner, the judge thanked the volunteers who had worked through this storm to repair the damaged stage.",
- "raw": {
- "normalized_reference": "before announcing the winner, the judge thanked the volunteers who had worked through the storm to repair the damaged stage.",
- "normalized_hypothesis": "before announcing the winner, the judge thanked the volunteers who had worked through this storm to repair the damaged stage.",
- "errors": 1,
- "words": 20,
- "wer": 0.05
- },
- "semantic": {
- "normalized_reference": "before announcing the winner the judge thanked the volunteers who had worked through the storm to repair the damaged stage",
- "normalized_hypothesis": "before announcing the winner the judge thanked the volunteers who had worked through this storm to repair the damaged stage",
- "errors": 1,
- "words": 20,
- "wer": 0.05
- }
- },
- {
- "id": "utmos200_021",
- "category": "names_places",
- "text": "Professor Nguyen mailed the fluorescent poster to Reykjavik on Thursday.",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_021.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Professor Neumann mailed the fluorescent poster to Reykjavik on Thursday.",
- "raw": {
- "normalized_reference": "professor nguyen mailed the fluorescent poster to reykjavik on thursday.",
- "normalized_hypothesis": "professor neumann mailed the fluorescent poster to reykjavik on thursday.",
- "errors": 1,
- "words": 10,
- "wer": 0.1
- },
- "semantic": {
- "normalized_reference": "professor nguyen mailed the fluorescent poster to reykjavik on thursday",
- "normalized_hypothesis": "professor neumann mailed the fluorescent poster to reykjavik on thursday",
- "errors": 1,
- "words": 10,
- "wer": 0.1
- }
- },
- {
- "id": "utmos200_022",
- "category": "narrative",
- "text": "The plan was simple. The execution was anything but.",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_022.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "The plan was simple. The execution was anything but perfect.",
- "raw": {
- "normalized_reference": "the plan was simple. the execution was anything but.",
- "normalized_hypothesis": "the plan was simple. the execution was anything but perfect.",
- "errors": 1,
- "words": 9,
- "wer": 0.1111111111111111
- },
- "semantic": {
- "normalized_reference": "the plan was simple the execution was anything but",
- "normalized_hypothesis": "the plan was simple the execution was anything but perfect",
- "errors": 1,
- "words": 9,
- "wer": 0.1111111111111111
- }
- },
- {
- "id": "utmos200_023",
- "category": "numbers",
- "text": "Her score improved from eighty-two percent to ninety-seven percent in two weeks.",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_023.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Her score improved from 82% to 97% in two weeks.",
- "raw": {
- "normalized_reference": "her score improved from eighty-two percent to ninety-seven percent in two weeks.",
- "normalized_hypothesis": "her score improved from 82% to 97% in two weeks.",
- "errors": 6,
- "words": 14,
- "wer": 0.42857142857142855
- },
- "semantic": {
- "normalized_reference": "her score improved from 82% to 97% in 2 weeks",
- "normalized_hypothesis": "her score improved from 82% to 97% in 2 weeks",
- "errors": 0,
- "words": 10,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_024",
- "category": "punctuation",
- "text": "Why would anyone pack spoons, batteries, and a violin for one short trip?",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_024.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Why would anyone pack spoons, batteries, and a violin for one short trip?",
- "raw": {
- "normalized_reference": "why would anyone pack spoons, batteries, and a violin for one short trip?",
- "normalized_hypothesis": "why would anyone pack spoons, batteries, and a violin for one short trip?",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "why would anyone pack spoons batteries and a violin for one short trip",
- "normalized_hypothesis": "why would anyone pack spoons batteries and a violin for one short trip",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_025",
- "category": "question",
- "text": "How long have you been sitting on this?",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_025.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "How long have you been sitting on this?",
- "raw": {
- "normalized_reference": "how long have you been sitting on this?",
- "normalized_hypothesis": "how long have you been sitting on this?",
- "errors": 0,
- "words": 8,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "how long have you been sitting on this",
- "normalized_hypothesis": "how long have you been sitting on this",
- "errors": 0,
- "words": 8,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_026",
- "category": "rare_words",
- "text": "A mischievous chameleon rested beside the chrysanthemum and the bougainvillea.",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_026.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "A mischievous chameleon rested beside the Chrysanthemum and the Bougainvillea.",
- "raw": {
- "normalized_reference": "a mischievous chameleon rested beside the chrysanthemum and the bougainvillea.",
- "normalized_hypothesis": "a mischievous chameleon rested beside the chrysanthemum and the bougainvillea.",
- "errors": 0,
- "words": 10,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "a mischievous chameleon rested beside the chrysanthemum and the bougainvillea",
- "normalized_hypothesis": "a mischievous chameleon rested beside the chrysanthemum and the bougainvillea",
- "errors": 0,
- "words": 10,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_027",
- "category": "short",
- "text": "We don't talk about that.",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_027.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "We don't talk about that.",
- "raw": {
- "normalized_reference": "we don't talk about that.",
- "normalized_hypothesis": "we don't talk about that.",
- "errors": 0,
- "words": 5,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "we do not talk about that",
- "normalized_hypothesis": "we do not talk about that",
- "errors": 0,
- "words": 6,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_028",
- "category": "technical",
- "text": "Hash the UTF eight payload with S H A two fifty-six before opening the checkpoint.",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_028.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Ash the UTF-8 payload with SHA-256 before opening the checkpoint.",
- "raw": {
- "normalized_reference": "hash the utf eight payload with s h a two fifty-six before opening the checkpoint.",
- "normalized_hypothesis": "ash the utf-8 payload with sha-256 before opening the checkpoint.",
- "errors": 8,
- "words": 16,
- "wer": 0.5
- },
- "semantic": {
- "normalized_reference": "hash the utf 8 payload with s h a 256 before opening the checkpoint",
- "normalized_hypothesis": "ash the utf 8 payload with sha 256 before opening the checkpoint",
- "errors": 4,
- "words": 14,
- "wer": 0.2857142857142857
- }
- },
- {
- "id": "utmos200_029",
- "category": "conversational",
- "text": "I thought I had it all figured out, and then this happened.",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_029.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "I thought I had it all figured out, and then this happened.",
- "raw": {
- "normalized_reference": "i thought i had it all figured out, and then this happened.",
- "normalized_hypothesis": "i thought i had it all figured out, and then this happened.",
- "errors": 0,
- "words": 12,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "i thought i had it all figured out and then this happened",
- "normalized_hypothesis": "i thought i had it all figured out and then this happened",
- "errors": 0,
- "words": 12,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_030",
- "category": "descriptive",
- "text": "The kind of quiet you only get at three in the morning in a city.",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_030.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "the kind of quiet you only get at three in the morning in a city.",
- "raw": {
- "normalized_reference": "the kind of quiet you only get at three in the morning in a city.",
- "normalized_hypothesis": "the kind of quiet you only get at three in the morning in a city.",
- "errors": 0,
- "words": 15,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "the kind of quiet you only get at 3 in the morning in a city",
- "normalized_hypothesis": "the kind of quiet you only get at 3 in the morning in a city",
- "errors": 0,
- "words": 15,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_031",
- "category": "emotional",
- "text": "The joy of it was almost overwhelming, like too much good all at once.",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_031.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "The joy of it was almost overwhelming, like too much good all at once.",
- "raw": {
- "normalized_reference": "the joy of it was almost overwhelming, like too much good all at once.",
- "normalized_hypothesis": "the joy of it was almost overwhelming, like too much good all at once.",
- "errors": 0,
- "words": 14,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "the joy of it was almost overwhelming like too much good all at once",
- "normalized_hypothesis": "the joy of it was almost overwhelming like too much good all at once",
- "errors": 0,
- "words": 14,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_032",
- "category": "homographs",
- "text": "Please close the door while I stand close to the microphone.",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_032.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Please close the door while I stand close to the microphone.",
- "raw": {
- "normalized_reference": "please close the door while i stand close to the microphone.",
- "normalized_hypothesis": "please close the door while i stand close to the microphone.",
- "errors": 0,
- "words": 11,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "please close the door while i stand close to the microphone",
- "normalized_hypothesis": "please close the door while i stand close to the microphone",
- "errors": 0,
- "words": 11,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_033",
- "category": "long",
- "text": "I think the most underrated skill in the world is knowing when to stop explaining yourself and just let people draw their own conclusions.",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_033.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "I think the most underrated skill in the world is knowing when to stop explaining yourself and just let people draw their own conclusions.",
- "raw": {
- "normalized_reference": "i think the most underrated skill in the world is knowing when to stop explaining yourself and just let people draw their own conclusions.",
- "normalized_hypothesis": "i think the most underrated skill in the world is knowing when to stop explaining yourself and just let people draw their own conclusions.",
- "errors": 0,
- "words": 24,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "i think the most underrated skill in the world is knowing when to stop explaining yourself and just let people draw their own conclusions",
- "normalized_hypothesis": "i think the most underrated skill in the world is knowing when to stop explaining yourself and just let people draw their own conclusions",
- "errors": 0,
- "words": 24,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_034",
- "category": "long_range",
- "text": "When the final bus disappeared around the corner, Nadia remembered that her umbrella, notebook, and lunch were still inside the quiet office.",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_034.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "When the final bus disappeared around the corner, Nadia remembered that her umbrella, notebook, and lunch were still inside the quiet office.",
- "raw": {
- "normalized_reference": "when the final bus disappeared around the corner, nadia remembered that her umbrella, notebook, and lunch were still inside the quiet office.",
- "normalized_hypothesis": "when the final bus disappeared around the corner, nadia remembered that her umbrella, notebook, and lunch were still inside the quiet office.",
- "errors": 0,
- "words": 22,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "when the final bus disappeared around the corner nadia remembered that her umbrella notebook and lunch were still inside the quiet office",
- "normalized_hypothesis": "when the final bus disappeared around the corner nadia remembered that her umbrella notebook and lunch were still inside the quiet office",
- "errors": 0,
- "words": 22,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_035",
- "category": "names_places",
- "text": "Aurelia and Joaquin crossed the Schuylkill River before noon.",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_035.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Uralaya and Jokwing crossed the Sialkal River before noon.",
- "raw": {
- "normalized_reference": "aurelia and joaquin crossed the schuylkill river before noon.",
- "normalized_hypothesis": "uralaya and jokwing crossed the sialkal river before noon.",
- "errors": 3,
- "words": 9,
- "wer": 0.3333333333333333
- },
- "semantic": {
- "normalized_reference": "aurelia and joaquin crossed the schuylkill river before noon",
- "normalized_hypothesis": "uralaya and jokwing crossed the sialkal river before noon",
- "errors": 3,
- "words": 9,
- "wer": 0.3333333333333333
- }
- },
- {
- "id": "utmos200_036",
- "category": "narrative",
- "text": "The village had a name, but nobody used it. Everyone just called it home.",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_036.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "The village had a name, but nobody used it. Everyone just called it home.",
- "raw": {
- "normalized_reference": "the village had a name, but nobody used it. everyone just called it home.",
- "normalized_hypothesis": "the village had a name, but nobody used it. everyone just called it home.",
- "errors": 0,
- "words": 14,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "the village had a name but nobody used it everyone just called it home",
- "normalized_hypothesis": "the village had a name but nobody used it everyone just called it home",
- "errors": 0,
- "words": 14,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_037",
- "category": "numbers",
- "text": "Increase the gain by three decibels, wait five seconds, then reduce it by one.",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_037.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Increase the gain by 3 decibels, wait 5 seconds, then reduce it by 1.",
- "raw": {
- "normalized_reference": "increase the gain by three decibels, wait five seconds, then reduce it by one.",
- "normalized_hypothesis": "increase the gain by 3 decibels, wait 5 seconds, then reduce it by 1.",
- "errors": 3,
- "words": 14,
- "wer": 0.21428571428571427
- },
- "semantic": {
- "normalized_reference": "increase the gain by 3 decibels wait 5 seconds then reduce it by one",
- "normalized_hypothesis": "increase the gain by 3 decibels wait 5 seconds then reduce it by one",
- "errors": 0,
- "words": 14,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_038",
- "category": "punctuation",
- "text": "The result was unexpected: clear, stable, and surprisingly natural.",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_038.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "The result was unexpected, clear, stable, and surprisingly natural.",
- "raw": {
- "normalized_reference": "the result was unexpected: clear, stable, and surprisingly natural.",
- "normalized_hypothesis": "the result was unexpected, clear, stable, and surprisingly natural.",
- "errors": 0,
- "words": 9,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "the result was unexpected clear stable and surprisingly natural",
- "normalized_hypothesis": "the result was unexpected clear stable and surprisingly natural",
- "errors": 0,
- "words": 9,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_039",
- "category": "question",
- "text": "Who told you that was the right way to do it?",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_039.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Who told you that was the right way to do it?",
- "raw": {
- "normalized_reference": "who told you that was the right way to do it?",
- "normalized_hypothesis": "who told you that was the right way to do it?",
- "errors": 0,
- "words": 11,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "who told you that was the right way to do it",
- "normalized_hypothesis": "who told you that was the right way to do it",
- "errors": 0,
- "words": 11,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_040",
- "category": "rare_words",
- "text": "The meticulous cartographer examined an iridescent archipelago through binoculars.",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_040.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "The meticulous cartographer examined an iridescent archipelago through binoculars.",
- "raw": {
- "normalized_reference": "the meticulous cartographer examined an iridescent archipelago through binoculars.",
- "normalized_hypothesis": "the meticulous cartographer examined an iridescent archipelago through binoculars.",
- "errors": 0,
- "words": 9,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "the meticulous cartographer examined an iridescent archipelago through binoculars",
- "normalized_hypothesis": "the meticulous cartographer examined an iridescent archipelago through binoculars",
- "errors": 0,
- "words": 9,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_041",
- "category": "short",
- "text": "I mean. Yeah. Kind of.",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_041.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "I mean that, yeah, that kind of,",
- "raw": {
- "normalized_reference": "i mean. yeah. kind of.",
- "normalized_hypothesis": "i mean that, yeah, that kind of,",
- "errors": 2,
- "words": 5,
- "wer": 0.4
- },
- "semantic": {
- "normalized_reference": "i mean yeah kind of",
- "normalized_hypothesis": "i mean that yeah that kind of",
- "errors": 2,
- "words": 5,
- "wer": 0.4
- }
- },
- {
- "id": "utmos200_042",
- "category": "technical",
- "text": "The logs are your best friend when something behaves unexpectedly.",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_042.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "The logs are your best friend when something behaves unexpectedly.",
- "raw": {
- "normalized_reference": "the logs are your best friend when something behaves unexpectedly.",
- "normalized_hypothesis": "the logs are your best friend when something behaves unexpectedly.",
- "errors": 0,
- "words": 10,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "the logs are your best friend when something behaves unexpectedly",
- "normalized_hypothesis": "the logs are your best friend when something behaves unexpectedly",
- "errors": 0,
- "words": 10,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_043",
- "category": "conversational",
- "text": "He kept talking and I kept nodding and neither of us was listening.",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_043.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "He kept talking, and I kept nodding, and neither of us was listening.",
- "raw": {
- "normalized_reference": "he kept talking and i kept nodding and neither of us was listening.",
- "normalized_hypothesis": "he kept talking, and i kept nodding, and neither of us was listening.",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "he kept talking and i kept nodding and neither of us was listening",
- "normalized_hypothesis": "he kept talking and i kept nodding and neither of us was listening",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_044",
- "category": "descriptive",
- "text": "It's one of those places where everyone seems to know each other already.",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_044.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "It's one of those places where everyone seems to know each other already.",
- "raw": {
- "normalized_reference": "it's one of those places where everyone seems to know each other already.",
- "normalized_hypothesis": "it's one of those places where everyone seems to know each other already.",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "it is one of those places where everyone seems to know each other already",
- "normalized_hypothesis": "it is one of those places where everyone seems to know each other already",
- "errors": 0,
- "words": 14,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_045",
- "category": "emotional",
- "text": "He said he was proud of me and nobody had ever said that to me before.",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_045.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "He said he was proud of me and nobody had ever said that to me before.",
- "raw": {
- "normalized_reference": "he said he was proud of me and nobody had ever said that to me before.",
- "normalized_hypothesis": "he said he was proud of me and nobody had ever said that to me before.",
- "errors": 0,
- "words": 16,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "he said he was proud of me and nobody had ever said that to me before",
- "normalized_hypothesis": "he said he was proud of me and nobody had ever said that to me before",
- "errors": 0,
- "words": 16,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_046",
- "category": "homographs",
- "text": "The minute detail took a minute to explain clearly.",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_046.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "The minute detail took a minute to explain clearly.",
- "raw": {
- "normalized_reference": "the minute detail took a minute to explain clearly.",
- "normalized_hypothesis": "the minute detail took a minute to explain clearly.",
- "errors": 0,
- "words": 9,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "the minute detail took a minute to explain clearly",
- "normalized_hypothesis": "the minute detail took a minute to explain clearly",
- "errors": 0,
- "words": 9,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_047",
- "category": "long",
- "text": "The strange thing about really wanting something is that getting it never quite feels the way you imagined it would, and not getting it never quite feels as bad.",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_047.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "The strange thing about really wanting something is that getting it never quite feels the way you imagined it would, and not getting it never quite feels as bad.",
- "raw": {
- "normalized_reference": "the strange thing about really wanting something is that getting it never quite feels the way you imagined it would, and not getting it never quite feels as bad.",
- "normalized_hypothesis": "the strange thing about really wanting something is that getting it never quite feels the way you imagined it would, and not getting it never quite feels as bad.",
- "errors": 0,
- "words": 29,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "the strange thing about really wanting something is that getting it never quite feels the way you imagined it would and not getting it never quite feels as bad",
- "normalized_hypothesis": "the strange thing about really wanting something is that getting it never quite feels the way you imagined it would and not getting it never quite feels as bad",
- "errors": 0,
- "words": 29,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_048",
- "category": "long_range",
- "text": "Even though the recipe looked simple at first, preparing the sauce required patience, careful timing, and a much larger pan.",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_048.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Even though the recipe looked simple at first, preparing the sauce required patience, careful timing, and a much larger pan.",
- "raw": {
- "normalized_reference": "even though the recipe looked simple at first, preparing the sauce required patience, careful timing, and a much larger pan.",
- "normalized_hypothesis": "even though the recipe looked simple at first, preparing the sauce required patience, careful timing, and a much larger pan.",
- "errors": 0,
- "words": 20,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "even though the recipe looked simple at 1st preparing the sauce required patience careful timing and a much larger pan",
- "normalized_hypothesis": "even though the recipe looked simple at 1st preparing the sauce required patience careful timing and a much larger pan",
- "errors": 0,
- "words": 20,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_049",
- "category": "names_places",
- "text": "Xiomara met Siobhan near the old library in Worcestershire.",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_049.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Damara met Siobhan near the old library in Worcestershire.",
- "raw": {
- "normalized_reference": "xiomara met siobhan near the old library in worcestershire.",
- "normalized_hypothesis": "damara met siobhan near the old library in worcestershire.",
- "errors": 1,
- "words": 9,
- "wer": 0.1111111111111111
- },
- "semantic": {
- "normalized_reference": "xiomara met siobhan near the old library in worcestershire",
- "normalized_hypothesis": "damara met siobhan near the old library in worcestershire",
- "errors": 1,
- "words": 9,
- "wer": 0.1111111111111111
- }
- },
- {
- "id": "utmos200_050",
- "category": "narrative",
- "text": "She walked into the room and everyone went quiet, not because she was loud, but because she wasn't.",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_050.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "She walked into the room and everyone went quiet, not because she was loud, but because she wasn't.",
- "raw": {
- "normalized_reference": "she walked into the room and everyone went quiet, not because she was loud, but because she wasn't.",
- "normalized_hypothesis": "she walked into the room and everyone went quiet, not because she was loud, but because she wasn't.",
- "errors": 0,
- "words": 18,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "she walked into the room and everyone went quiet not because she was loud but because she was not",
- "normalized_hypothesis": "she walked into the room and everyone went quiet not because she was loud but because she was not",
- "errors": 0,
- "words": 19,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_051",
- "category": "numbers",
- "text": "At exactly seven forty-five, Morgan placed eighteen dollars and fifty cents beside the notebook.",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_051.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Exactly $7.45, Morgan placed $18.50 beside the notebook.",
- "raw": {
- "normalized_reference": "at exactly seven forty-five, morgan placed eighteen dollars and fifty cents beside the notebook.",
- "normalized_hypothesis": "exactly $7.45, morgan placed $18.50 beside the notebook.",
- "errors": 9,
- "words": 15,
- "wer": 0.6
- },
- "semantic": {
- "normalized_reference": "at exactly 745 morgan placed $18.50 beside the notebook",
- "normalized_hypothesis": "exactly $7.45 morgan placed $18.50 beside the notebook",
- "errors": 3,
- "words": 10,
- "wer": 0.3
- }
- },
- {
- "id": "utmos200_052",
- "category": "punctuation",
- "text": "Does punctuation work? Yes: commas, semicolons, and periods should create sensible boundaries.",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_052.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Does punctuation work? Yes, commas, semicolons, and periods should create sensible boundaries.",
- "raw": {
- "normalized_reference": "does punctuation work? yes: commas, semicolons, and periods should create sensible boundaries.",
- "normalized_hypothesis": "does punctuation work? yes, commas, semicolons, and periods should create sensible boundaries.",
- "errors": 0,
- "words": 12,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "does punctuation work yes commas semicolons and periods should create sensible boundaries",
- "normalized_hypothesis": "does punctuation work yes commas semicolons and periods should create sensible boundaries",
- "errors": 0,
- "words": 12,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_053",
- "category": "question",
- "text": "Have you ever just stopped and thought about how strange all of this is?",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_053.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Have you ever just stopped and thought about how strange all of this is?",
- "raw": {
- "normalized_reference": "have you ever just stopped and thought about how strange all of this is?",
- "normalized_hypothesis": "have you ever just stopped and thought about how strange all of this is?",
- "errors": 0,
- "words": 14,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "have you ever just stopped and thought about how strange all of this is",
- "normalized_hypothesis": "have you ever just stopped and thought about how strange all of this is",
- "errors": 0,
- "words": 14,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_054",
- "category": "rare_words",
- "text": "Her idiosyncratic explanation sounded plausible but remained fundamentally ambiguous.",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_054.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Her idiosyncratic explanation sounded plausible but remained fundamentally ambiguous.",
- "raw": {
- "normalized_reference": "her idiosyncratic explanation sounded plausible but remained fundamentally ambiguous.",
- "normalized_hypothesis": "her idiosyncratic explanation sounded plausible but remained fundamentally ambiguous.",
- "errors": 0,
- "words": 9,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "her idiosyncratic explanation sounded plausible but remained fundamentally ambiguous",
- "normalized_hypothesis": "her idiosyncratic explanation sounded plausible but remained fundamentally ambiguous",
- "errors": 0,
- "words": 9,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_055",
- "category": "short",
- "text": "That tracks.",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_055.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "that tracks",
- "raw": {
- "normalized_reference": "that tracks.",
- "normalized_hypothesis": "that tracks",
- "errors": 0,
- "words": 2,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "that tracks",
- "normalized_hypothesis": "that tracks",
- "errors": 0,
- "words": 2,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_056",
- "category": "technical",
- "text": "The key thing to remember is that order matters here more than you'd think.",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_056.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "The key thing to remember is that order matters here more than you'd think.",
- "raw": {
- "normalized_reference": "the key thing to remember is that order matters here more than you'd think.",
- "normalized_hypothesis": "the key thing to remember is that order matters here more than you'd think.",
- "errors": 0,
- "words": 14,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "the key thing to remember is that order matters here more than you would think",
- "normalized_hypothesis": "the key thing to remember is that order matters here more than you would think",
- "errors": 0,
- "words": 15,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_057",
- "category": "conversational",
- "text": "I would not say that again in this room if I were you.",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_057.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "I would not say that again in this room if I were you.",
- "raw": {
- "normalized_reference": "i would not say that again in this room if i were you.",
- "normalized_hypothesis": "i would not say that again in this room if i were you.",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "i would not say that again in this room if i were you",
- "normalized_hypothesis": "i would not say that again in this room if i were you",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_058",
- "category": "descriptive",
- "text": "The room was filled with the quiet that means something is about to change.",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_058.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "The room was filled with the quiet that means something is about to change.",
- "raw": {
- "normalized_reference": "the room was filled with the quiet that means something is about to change.",
- "normalized_hypothesis": "the room was filled with the quiet that means something is about to change.",
- "errors": 0,
- "words": 14,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "the room was filled with the quiet that means something is about to change",
- "normalized_hypothesis": "the room was filled with the quiet that means something is about to change",
- "errors": 0,
- "words": 14,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_059",
- "category": "emotional",
- "text": "I didn't expect to feel that way about it, but I did.",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_059.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "I didn't expect to feel that way about it, but I did.",
- "raw": {
- "normalized_reference": "i didn't expect to feel that way about it, but i did.",
- "normalized_hypothesis": "i didn't expect to feel that way about it, but i did.",
- "errors": 0,
- "words": 12,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "i did not expect to feel that way about it but i did",
- "normalized_hypothesis": "i did not expect to feel that way about it but i did",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_060",
- "category": "homographs",
- "text": "I object to the strange object that was left near the entrance.",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_060.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "I object to the strange object that was left near the entrance.",
- "raw": {
- "normalized_reference": "i object to the strange object that was left near the entrance.",
- "normalized_hypothesis": "i object to the strange object that was left near the entrance.",
- "errors": 0,
- "words": 12,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "i object to the strange object that was left near the entrance",
- "normalized_hypothesis": "i object to the strange object that was left near the entrance",
- "errors": 0,
- "words": 12,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_061",
- "category": "long",
- "text": "I've learned more from the things that went wrong than from the things that worked, which is a frustrating way to have to learn things but apparently the only way that stuck.",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_061.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "I've learned more from the things that went wrong than from the things that worked, which is a frustrating way to have to learn things but apparently the only way that's to",
- "raw": {
- "normalized_reference": "i've learned more from the things that went wrong than from the things that worked, which is a frustrating way to have to learn things but apparently the only way that stuck.",
- "normalized_hypothesis": "i've learned more from the things that went wrong than from the things that worked, which is a frustrating way to have to learn things but apparently the only way that's to",
- "errors": 2,
- "words": 32,
- "wer": 0.0625
- },
- "semantic": {
- "normalized_reference": "i have learned more from the things that went wrong than from the things that worked which is a frustrating way to have to learn things but apparently the only way that stuck",
- "normalized_hypothesis": "i have learned more from the things that went wrong than from the things that worked which is a frustrating way to have to learn things but apparently the only way that is to",
- "errors": 2,
- "words": 33,
- "wer": 0.06060606060606061
- }
- },
- {
- "id": "utmos200_062",
- "category": "long_range",
- "text": "Because every nearby hotel was full, the exhausted hikers followed the narrow road until they found a farmhouse with one empty room.",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_062.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Because every nearby hotel was full, the exhausted hikers followed the narrow road until they found a farmhouse with one empty room.",
- "raw": {
- "normalized_reference": "because every nearby hotel was full, the exhausted hikers followed the narrow road until they found a farmhouse with one empty room.",
- "normalized_hypothesis": "because every nearby hotel was full, the exhausted hikers followed the narrow road until they found a farmhouse with one empty room.",
- "errors": 0,
- "words": 22,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "because every nearby hotel was full the exhausted hikers followed the narrow road until they found a farmhouse with one empty room",
- "normalized_hypothesis": "because every nearby hotel was full the exhausted hikers followed the narrow road until they found a farmhouse with one empty room",
- "errors": 0,
- "words": 22,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_063",
- "category": "names_places",
- "text": "The itinerary sends Beatrice from Albuquerque to Des Moines and then to Boise.",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_063.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "The itinerary sends Beatrice from Albuquerque to Des Moines and then to Boise,",
- "raw": {
- "normalized_reference": "the itinerary sends beatrice from albuquerque to des moines and then to boise.",
- "normalized_hypothesis": "the itinerary sends beatrice from albuquerque to des moines and then to boise,",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "the itinerary sends beatrice from albuquerque to des moines and then to boise",
- "normalized_hypothesis": "the itinerary sends beatrice from albuquerque to des moines and then to boise",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_064",
- "category": "narrative",
- "text": "We had been driving for six hours when we finally saw the sign for the town.",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_064.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "We had been driving for six hours when we finally saw the sign for the town.",
- "raw": {
- "normalized_reference": "we had been driving for six hours when we finally saw the sign for the town.",
- "normalized_hypothesis": "we had been driving for six hours when we finally saw the sign for the town.",
- "errors": 0,
- "words": 16,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "we had been driving for 6 hours when we finally saw the sign for the town",
- "normalized_hypothesis": "we had been driving for 6 hours when we finally saw the sign for the town",
- "errors": 0,
- "words": 16,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_065",
- "category": "numbers",
- "text": "Call extension zero four nine after eight thirty in the morning.",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_065.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Call extension 049 after 8.30 in the morning.",
- "raw": {
- "normalized_reference": "call extension zero four nine after eight thirty in the morning.",
- "normalized_hypothesis": "call extension 049 after 8.30 in the morning.",
- "errors": 5,
- "words": 11,
- "wer": 0.45454545454545453
- },
- "semantic": {
- "normalized_reference": "call extension 049 after 830 in the morning",
- "normalized_hypothesis": "call extension 49 after 8.30 in the morning",
- "errors": 3,
- "words": 8,
- "wer": 0.375
- }
- },
- {
- "id": "utmos200_066",
- "category": "punctuation",
- "text": "Stop - listen carefully - then answer the question without guessing.",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_066.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Stop listen carefully then answer the question without guessing.",
- "raw": {
- "normalized_reference": "stop - listen carefully - then answer the question without guessing.",
- "normalized_hypothesis": "stop listen carefully then answer the question without guessing.",
- "errors": 0,
- "words": 9,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "stop listen carefully then answer the question without guessing",
- "normalized_hypothesis": "stop listen carefully then answer the question without guessing",
- "errors": 0,
- "words": 9,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_067",
- "category": "question",
- "text": "Does it matter at this point, honestly?",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_067.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Does it matter at this point? Honestly,",
- "raw": {
- "normalized_reference": "does it matter at this point, honestly?",
- "normalized_hypothesis": "does it matter at this point? honestly,",
- "errors": 0,
- "words": 7,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "does it matter at this point honestly",
- "normalized_hypothesis": "does it matter at this point honestly",
- "errors": 0,
- "words": 7,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_068",
- "category": "rare_words",
- "text": "An entrepreneurial meteorologist described the unprecedented atmospheric phenomenon.",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_068.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "An entrepreneurial meteorologist described the unprecedented atmospheric phenomenon.",
- "raw": {
- "normalized_reference": "an entrepreneurial meteorologist described the unprecedented atmospheric phenomenon.",
- "normalized_hypothesis": "an entrepreneurial meteorologist described the unprecedented atmospheric phenomenon.",
- "errors": 0,
- "words": 8,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "an entrepreneurial meteorologist described the unprecedented atmospheric phenomenon",
- "normalized_hypothesis": "an entrepreneurial meteorologist described the unprecedented atmospheric phenomenon",
- "errors": 0,
- "words": 8,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_069",
- "category": "short",
- "text": "That's not what I expected.",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_069.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "That's not what I expected.",
- "raw": {
- "normalized_reference": "that's not what i expected.",
- "normalized_hypothesis": "that's not what i expected.",
- "errors": 0,
- "words": 5,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "that is not what i expected",
- "normalized_hypothesis": "that is not what i expected",
- "errors": 0,
- "words": 6,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_070",
- "category": "technical",
- "text": "There's a known issue with that version that causes failures under load.",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_070.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "There's a known issue with that version that causes failures under load.",
- "raw": {
- "normalized_reference": "there's a known issue with that version that causes failures under load.",
- "normalized_hypothesis": "there's a known issue with that version that causes failures under load.",
- "errors": 0,
- "words": 12,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "there is a known issue with that version that causes failures under load",
- "normalized_hypothesis": "there is a known issue with that version that causes failures under load",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_071",
- "category": "conversational",
- "text": "Don't look now but I think that's him over there.",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_071.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Don't look now but I think that's him over there.",
- "raw": {
- "normalized_reference": "don't look now but i think that's him over there.",
- "normalized_hypothesis": "don't look now but i think that's him over there.",
- "errors": 0,
- "words": 10,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "do not look now but i think that is him over there",
- "normalized_hypothesis": "do not look now but i think that is him over there",
- "errors": 0,
- "words": 12,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_072",
- "category": "descriptive",
- "text": "The city at night looks completely different from how it feels during the day.",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_072.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "The city at night looks completely different from how it feels during the day.",
- "raw": {
- "normalized_reference": "the city at night looks completely different from how it feels during the day.",
- "normalized_hypothesis": "the city at night looks completely different from how it feels during the day.",
- "errors": 0,
- "words": 14,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "the city at night looks completely different from how it feels during the day",
- "normalized_hypothesis": "the city at night looks completely different from how it feels during the day",
- "errors": 0,
- "words": 14,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_073",
- "category": "emotional",
- "text": "It's the kind of happy that makes you nervous because you know it can't last.",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_073.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "It's the kind of happy that makes you nervous because you know it can't last.",
- "raw": {
- "normalized_reference": "it's the kind of happy that makes you nervous because you know it can't last.",
- "normalized_hypothesis": "it's the kind of happy that makes you nervous because you know it can't last.",
- "errors": 0,
- "words": 15,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "it is the kind of happy that makes you nervous because you know it can not last",
- "normalized_hypothesis": "it is the kind of happy that makes you nervous because you know it can not last",
- "errors": 0,
- "words": 17,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_074",
- "category": "homographs",
- "text": "The dove dove beneath the branch while the wind began to wind the loose cable.",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_074.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "The dub-dub beneath the branch wall the wind began to wind the loose cable.",
- "raw": {
- "normalized_reference": "the dove dove beneath the branch while the wind began to wind the loose cable.",
- "normalized_hypothesis": "the dub-dub beneath the branch wall the wind began to wind the loose cable.",
- "errors": 3,
- "words": 15,
- "wer": 0.2
- },
- "semantic": {
- "normalized_reference": "the dove dove beneath the branch while the wind began to wind the loose cable",
- "normalized_hypothesis": "the dub dub beneath the branch wall the wind began to wind the loose cable",
- "errors": 3,
- "words": 15,
- "wer": 0.2
- }
- },
- {
- "id": "utmos200_075",
- "category": "long",
- "text": "I've thought a lot about what I would do differently if I could go back, and the honest answer is probably not as much as I'd like to think.",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_075.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "I've thought a lot about what I would do differently if I could go back, and the honest answer is probably not as much as I'd like to think.",
- "raw": {
- "normalized_reference": "i've thought a lot about what i would do differently if i could go back, and the honest answer is probably not as much as i'd like to think.",
- "normalized_hypothesis": "i've thought a lot about what i would do differently if i could go back, and the honest answer is probably not as much as i'd like to think.",
- "errors": 0,
- "words": 29,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "i have thought a lot about what i would do differently if i could go back and the honest answer is probably not as much as i would like to think",
- "normalized_hypothesis": "i have thought a lot about what i would do differently if i could go back and the honest answer is probably not as much as i would like to think",
- "errors": 0,
- "words": 31,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_076",
- "category": "long_range",
- "text": "Although the weather changed without warning, the musicians finished packing before anyone opened the enormous glass door.",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_076.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Although the weather changed without warning, the musicians finished packing before anyone opened the enormous glass door.",
- "raw": {
- "normalized_reference": "although the weather changed without warning, the musicians finished packing before anyone opened the enormous glass door.",
- "normalized_hypothesis": "although the weather changed without warning, the musicians finished packing before anyone opened the enormous glass door.",
- "errors": 0,
- "words": 17,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "although the weather changed without warning the musicians finished packing before anyone opened the enormous glass door",
- "normalized_hypothesis": "although the weather changed without warning the musicians finished packing before anyone opened the enormous glass door",
- "errors": 0,
- "words": 17,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_077",
- "category": "names_places",
- "text": "Saskatchewan looked unusually bright beneath the pale February sunrise.",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_077.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Saskatchewan looked unusually bright beneath the pale February sunrise.",
- "raw": {
- "normalized_reference": "saskatchewan looked unusually bright beneath the pale february sunrise.",
- "normalized_hypothesis": "saskatchewan looked unusually bright beneath the pale february sunrise.",
- "errors": 0,
- "words": 9,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "saskatchewan looked unusually bright beneath the pale february sunrise",
- "normalized_hypothesis": "saskatchewan looked unusually bright beneath the pale february sunrise",
- "errors": 0,
- "words": 9,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_078",
- "category": "narrative",
- "text": "She handled it with a grace I genuinely did not expect.",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_078.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "She handled it with a grace I genuinely did not expect.",
- "raw": {
- "normalized_reference": "she handled it with a grace i genuinely did not expect.",
- "normalized_hypothesis": "she handled it with a grace i genuinely did not expect.",
- "errors": 0,
- "words": 11,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "she handled it with a grace i genuinely did not expect",
- "normalized_hypothesis": "she handled it with a grace i genuinely did not expect",
- "errors": 0,
- "words": 11,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_079",
- "category": "numbers",
- "text": "Apartment four hundred seven is across from room twenty-nine on the third floor.",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_079.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Apartment 407 is across from Room 29 on the third floor.",
- "raw": {
- "normalized_reference": "apartment four hundred seven is across from room twenty-nine on the third floor.",
- "normalized_hypothesis": "apartment 407 is across from room 29 on the third floor.",
- "errors": 5,
- "words": 14,
- "wer": 0.35714285714285715
- },
- "semantic": {
- "normalized_reference": "apartment 407 is across from room 29 on the 3rd floor",
- "normalized_hypothesis": "apartment 407 is across from room 29 on the 3rd floor",
- "errors": 0,
- "words": 11,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_080",
- "category": "punctuation",
- "text": "\"Not yet,\" Rowan whispered. \"Give me another minute... please.\"",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_080.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Not yet, Rowan whispered. Give me another minute, please.",
- "raw": {
- "normalized_reference": "\"not yet,\" rowan whispered. \"give me another minute... please.\"",
- "normalized_hypothesis": "not yet, rowan whispered. give me another minute, please.",
- "errors": 0,
- "words": 9,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "not yet rowan whispered give me another minute please",
- "normalized_hypothesis": "not yet rowan whispered give me another minute please",
- "errors": 0,
- "words": 9,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_081",
- "category": "question",
- "text": "What are we actually trying to accomplish here?",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_081.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "What are we actually trying to accomplish here?",
- "raw": {
- "normalized_reference": "what are we actually trying to accomplish here?",
- "normalized_hypothesis": "what are we actually trying to accomplish here?",
- "errors": 0,
- "words": 8,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "what are we actually trying to accomplish here",
- "normalized_hypothesis": "what are we actually trying to accomplish here",
- "errors": 0,
- "words": 8,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_082",
- "category": "rare_words",
- "text": "The physician recommended a noninvasive procedure for the recurring neurological symptoms.",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_082.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "The physician recommended a non-invasive procedure for the recurring neurological symptoms.",
- "raw": {
- "normalized_reference": "the physician recommended a noninvasive procedure for the recurring neurological symptoms.",
- "normalized_hypothesis": "the physician recommended a non-invasive procedure for the recurring neurological symptoms.",
- "errors": 2,
- "words": 11,
- "wer": 0.18181818181818182
- },
- "semantic": {
- "normalized_reference": "the physician recommended a noninvasive procedure for the recurring neurological symptoms",
- "normalized_hypothesis": "the physician recommended a non invasive procedure for the recurring neurological symptoms",
- "errors": 2,
- "words": 11,
- "wer": 0.18181818181818182
- }
- },
- {
- "id": "utmos200_083",
- "category": "short",
- "text": "Are you serious right now?",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_083.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Are you serious right now?",
- "raw": {
- "normalized_reference": "are you serious right now?",
- "normalized_hypothesis": "are you serious right now?",
- "errors": 0,
- "words": 5,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "are you serious right now",
- "normalized_hypothesis": "are you serious right now",
- "errors": 0,
- "words": 5,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_084",
- "category": "technical",
- "text": "The performance improvement is real, but it comes with tradeoffs.",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_084.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "The performance improvement is real, but it comes with trade-offs.",
- "raw": {
- "normalized_reference": "the performance improvement is real, but it comes with tradeoffs.",
- "normalized_hypothesis": "the performance improvement is real, but it comes with trade-offs.",
- "errors": 2,
- "words": 10,
- "wer": 0.2
- },
- "semantic": {
- "normalized_reference": "the performance improvement is real but it comes with tradeoffs",
- "normalized_hypothesis": "the performance improvement is real but it comes with trade offs",
- "errors": 2,
- "words": 10,
- "wer": 0.2
- }
- },
- {
- "id": "utmos200_085",
- "category": "conversational",
- "text": "I don't know. It's just been one of those weeks, you know?",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_085.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "I don't know, it's just been one of those weeks, you know?",
- "raw": {
- "normalized_reference": "i don't know. it's just been one of those weeks, you know?",
- "normalized_hypothesis": "i don't know, it's just been one of those weeks, you know?",
- "errors": 0,
- "words": 12,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "i do not know it is just been one of those weeks you know",
- "normalized_hypothesis": "i do not know it is just been one of those weeks you know",
- "errors": 0,
- "words": 14,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_086",
- "category": "descriptive",
- "text": "The moment you stop expecting perfection, things get a lot more interesting.",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_086.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "The moment you stop expecting perfection, things get a lot more interesting.",
- "raw": {
- "normalized_reference": "the moment you stop expecting perfection, things get a lot more interesting.",
- "normalized_hypothesis": "the moment you stop expecting perfection, things get a lot more interesting.",
- "errors": 0,
- "words": 12,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "the moment you stop expecting perfection things get a lot more interesting",
- "normalized_hypothesis": "the moment you stop expecting perfection things get a lot more interesting",
- "errors": 0,
- "words": 12,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_087",
- "category": "emotional",
- "text": "It's the kind of thing that stays with you even years later.",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_087.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "It's the kind of thing that stays with you even years later.",
- "raw": {
- "normalized_reference": "it's the kind of thing that stays with you even years later.",
- "normalized_hypothesis": "it's the kind of thing that stays with you even years later.",
- "errors": 0,
- "words": 12,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "it is the kind of thing that stays with you even years later",
- "normalized_hypothesis": "it is the kind of thing that stays with you even years later",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_088",
- "category": "long",
- "text": "She told me once that the most important skill she'd developed was learning to disagree with someone without making them feel they were wrong.",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_088.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "She told me once that the most important skill she'd developed was learning to disagree with someone without making them feel they were wrong.",
- "raw": {
- "normalized_reference": "she told me once that the most important skill she'd developed was learning to disagree with someone without making them feel they were wrong.",
- "normalized_hypothesis": "she told me once that the most important skill she'd developed was learning to disagree with someone without making them feel they were wrong.",
- "errors": 0,
- "words": 24,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "she told me once that the most important skill she would developed was learning to disagree with someone without making them feel they were wrong",
- "normalized_hypothesis": "she told me once that the most important skill she would developed was learning to disagree with someone without making them feel they were wrong",
- "errors": 0,
- "words": 25,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_089",
- "category": "narrative",
- "text": "He stood at the edge of the cliff and wondered, not for the first time, how he'd gotten here.",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_089.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Keyes stood at the edge of the cliff and wondered, not for the first time, how he'd gotten here.",
- "raw": {
- "normalized_reference": "he stood at the edge of the cliff and wondered, not for the first time, how he'd gotten here.",
- "normalized_hypothesis": "keyes stood at the edge of the cliff and wondered, not for the first time, how he'd gotten here.",
- "errors": 1,
- "words": 19,
- "wer": 0.05263157894736842
- },
- "semantic": {
- "normalized_reference": "he stood at the edge of the cliff and wondered not for the 1st time how he would gotten here",
- "normalized_hypothesis": "keyes stood at the edge of the cliff and wondered not for the 1st time how he would gotten here",
- "errors": 1,
- "words": 20,
- "wer": 0.05
- }
- },
- {
- "id": "utmos200_090",
- "category": "question",
- "text": "Why does asking for help feel so hard when you actually need it?",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_090.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Why does asking for help feel so hard when you actually need it?",
- "raw": {
- "normalized_reference": "why does asking for help feel so hard when you actually need it?",
- "normalized_hypothesis": "why does asking for help feel so hard when you actually need it?",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "why does asking for help feel so hard when you actually need it",
- "normalized_hypothesis": "why does asking for help feel so hard when you actually need it",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_091",
- "category": "short",
- "text": "That's fair.",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_091.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "That's fair.",
- "raw": {
- "normalized_reference": "that's fair.",
- "normalized_hypothesis": "that's fair.",
- "errors": 0,
- "words": 2,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "that is fair",
- "normalized_hypothesis": "that is fair",
- "errors": 0,
- "words": 3,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_092",
- "category": "technical",
- "text": "If you run into that error, the fix is usually just restarting the service.",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_092.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "If you run into that error, the fix is usually just restarting the service.",
- "raw": {
- "normalized_reference": "if you run into that error, the fix is usually just restarting the service.",
- "normalized_hypothesis": "if you run into that error, the fix is usually just restarting the service.",
- "errors": 0,
- "words": 14,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "if you run into that error the fix is usually just restarting the service",
- "normalized_hypothesis": "if you run into that error the fix is usually just restarting the service",
- "errors": 0,
- "words": 14,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_093",
- "category": "conversational",
- "text": "It was a disaster in the best possible way.",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_093.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "It was a disaster in the best possible way.",
- "raw": {
- "normalized_reference": "it was a disaster in the best possible way.",
- "normalized_hypothesis": "it was a disaster in the best possible way.",
- "errors": 0,
- "words": 9,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "it was a disaster in the best possible way",
- "normalized_hypothesis": "it was a disaster in the best possible way",
- "errors": 0,
- "words": 9,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_094",
- "category": "descriptive",
- "text": "It had the look of a place that used to mean something to someone.",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_094.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "It had the look of a place that used to mean something to someone.",
- "raw": {
- "normalized_reference": "it had the look of a place that used to mean something to someone.",
- "normalized_hypothesis": "it had the look of a place that used to mean something to someone.",
- "errors": 0,
- "words": 14,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "it had the look of a place that used to mean something to someone",
- "normalized_hypothesis": "it had the look of a place that used to mean something to someone",
- "errors": 0,
- "words": 14,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_095",
- "category": "emotional",
- "text": "There's this feeling you get when something finally works after weeks of failure.",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_095.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "There's this feeling you get when something finally works after weeks of failure.",
- "raw": {
- "normalized_reference": "there's this feeling you get when something finally works after weeks of failure.",
- "normalized_hypothesis": "there's this feeling you get when something finally works after weeks of failure.",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "there is this feeling you get when something finally works after weeks of failure",
- "normalized_hypothesis": "there is this feeling you get when something finally works after weeks of failure",
- "errors": 0,
- "words": 14,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_096",
- "category": "long",
- "text": "He had a theory that most arguments aren't actually about what they seem to be about, and the longer I live the more I think he was onto something.",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_096.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "He had a theory that most arguments aren't actually about what they seem to be about, and the longer I live the more I think he was onto something.",
- "raw": {
- "normalized_reference": "he had a theory that most arguments aren't actually about what they seem to be about, and the longer i live the more i think he was onto something.",
- "normalized_hypothesis": "he had a theory that most arguments aren't actually about what they seem to be about, and the longer i live the more i think he was onto something.",
- "errors": 0,
- "words": 29,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "he had a theory that most arguments are not actually about what they seem to be about and the longer i live the more i think he was onto something",
- "normalized_hypothesis": "he had a theory that most arguments are not actually about what they seem to be about and the longer i live the more i think he was onto something",
- "errors": 0,
- "words": 30,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_097",
- "category": "narrative",
- "text": "The thing nobody tells you about starting over is how ordinary most of it feels.",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_097.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "The thing nobody tells you about starting over is how ordinary most of it feels.",
- "raw": {
- "normalized_reference": "the thing nobody tells you about starting over is how ordinary most of it feels.",
- "normalized_hypothesis": "the thing nobody tells you about starting over is how ordinary most of it feels.",
- "errors": 0,
- "words": 15,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "the thing nobody tells you about starting over is how ordinary most of it feels",
- "normalized_hypothesis": "the thing nobody tells you about starting over is how ordinary most of it feels",
- "errors": 0,
- "words": 15,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_098",
- "category": "question",
- "text": "What exactly are we waiting for at this point?",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_098.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "What exactly are we waiting for at this point?",
- "raw": {
- "normalized_reference": "what exactly are we waiting for at this point?",
- "normalized_hypothesis": "what exactly are we waiting for at this point?",
- "errors": 0,
- "words": 9,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "what exactly are we waiting for at this point",
- "normalized_hypothesis": "what exactly are we waiting for at this point",
- "errors": 0,
- "words": 9,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_099",
- "category": "short",
- "text": "I stand by what I said.",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_099.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "I stand by what I said.",
- "raw": {
- "normalized_reference": "i stand by what i said.",
- "normalized_hypothesis": "i stand by what i said.",
- "errors": 0,
- "words": 6,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "i stand by what i said",
- "normalized_hypothesis": "i stand by what i said",
- "errors": 0,
- "words": 6,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_100",
- "category": "technical",
- "text": "Double-check that value before you push anything to production.",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_100.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Double-check that value before you push anything to production.",
- "raw": {
- "normalized_reference": "double-check that value before you push anything to production.",
- "normalized_hypothesis": "double-check that value before you push anything to production.",
- "errors": 0,
- "words": 10,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "double check that value before you push anything to production",
- "normalized_hypothesis": "double check that value before you push anything to production",
- "errors": 0,
- "words": 10,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_101",
- "category": "conversational",
- "text": "I understand your point; I just do not agree with the conclusion.",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_101.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "I understand your point. I just do not agree with the conclusion.",
- "raw": {
- "normalized_reference": "i understand your point; i just do not agree with the conclusion.",
- "normalized_hypothesis": "i understand your point. i just do not agree with the conclusion.",
- "errors": 0,
- "words": 12,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "i understand your point i just do not agree with the conclusion",
- "normalized_hypothesis": "i understand your point i just do not agree with the conclusion",
- "errors": 0,
- "words": 12,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_102",
- "category": "descriptive",
- "text": "The sky that day was a strange shade of orange I'd never seen before.",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_102.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "The sky that day was a strange shade of orange I'd never seen before.",
- "raw": {
- "normalized_reference": "the sky that day was a strange shade of orange i'd never seen before.",
- "normalized_hypothesis": "the sky that day was a strange shade of orange i'd never seen before.",
- "errors": 0,
- "words": 14,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "the sky that day was a strange shade of orange i would never seen before",
- "normalized_hypothesis": "the sky that day was a strange shade of orange i would never seen before",
- "errors": 0,
- "words": 15,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_103",
- "category": "emotional",
- "text": "It hurt more than I expected, and I thought I was prepared.",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_103.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "It hurt more than I expected, and I thought I was prepared.",
- "raw": {
- "normalized_reference": "it hurt more than i expected, and i thought i was prepared.",
- "normalized_hypothesis": "it hurt more than i expected, and i thought i was prepared.",
- "errors": 0,
- "words": 12,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "it hurt more than i expected and i thought i was prepared",
- "normalized_hypothesis": "it hurt more than i expected and i thought i was prepared",
- "errors": 0,
- "words": 12,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_104",
- "category": "long",
- "text": "The hardest part wasn't making the decision, it was sitting with the uncertainty of not knowing whether it was the right one for months afterward.",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_104.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "The hardest part wasn't making the decision, it was sitting with the uncertainty of not knowing whether it was the right one for months afterward.",
- "raw": {
- "normalized_reference": "the hardest part wasn't making the decision, it was sitting with the uncertainty of not knowing whether it was the right one for months afterward.",
- "normalized_hypothesis": "the hardest part wasn't making the decision, it was sitting with the uncertainty of not knowing whether it was the right one for months afterward.",
- "errors": 0,
- "words": 25,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "the hardest part was not making the decision it was sitting with the uncertainty of not knowing whether it was the right one for months afterward",
- "normalized_hypothesis": "the hardest part was not making the decision it was sitting with the uncertainty of not knowing whether it was the right one for months afterward",
- "errors": 0,
- "words": 26,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_105",
- "category": "narrative",
- "text": "There was this one day last summer when everything just clicked into place.",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_105.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "There was this one day last summer when everything just clicked into place.",
- "raw": {
- "normalized_reference": "there was this one day last summer when everything just clicked into place.",
- "normalized_hypothesis": "there was this one day last summer when everything just clicked into place.",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "there was this one day last summer when everything just clicked into place",
- "normalized_hypothesis": "there was this one day last summer when everything just clicked into place",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_106",
- "category": "question",
- "text": "Have you ever thought about why we keep doing things the same way when they stop working?",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_106.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Have you ever thought about why we keep doing things the same way when they stop working?",
- "raw": {
- "normalized_reference": "have you ever thought about why we keep doing things the same way when they stop working?",
- "normalized_hypothesis": "have you ever thought about why we keep doing things the same way when they stop working?",
- "errors": 0,
- "words": 17,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "have you ever thought about why we keep doing things the same way when they stop working",
- "normalized_hypothesis": "have you ever thought about why we keep doing things the same way when they stop working",
- "errors": 0,
- "words": 17,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_107",
- "category": "short",
- "text": "I can't believe you just said that.",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_107.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "I can't believe you just said that.",
- "raw": {
- "normalized_reference": "i can't believe you just said that.",
- "normalized_hypothesis": "i can't believe you just said that.",
- "errors": 0,
- "words": 7,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "i can not believe you just said that",
- "normalized_hypothesis": "i can not believe you just said that",
- "errors": 0,
- "words": 8,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_108",
- "category": "technical",
- "text": "You'll want to isolate that variable before drawing any conclusions.",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_108.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "you'll want to isolate that variable before drawing any conclusions.",
- "raw": {
- "normalized_reference": "you'll want to isolate that variable before drawing any conclusions.",
- "normalized_hypothesis": "you'll want to isolate that variable before drawing any conclusions.",
- "errors": 0,
- "words": 10,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "you will want to isolate that variable before drawing any conclusions",
- "normalized_hypothesis": "you will want to isolate that variable before drawing any conclusions",
- "errors": 0,
- "words": 11,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_109",
- "category": "conversational",
- "text": "The meeting that could have been an email lasted two hours.",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_109.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "The meeting that could have been an email lasted two hours.",
- "raw": {
- "normalized_reference": "the meeting that could have been an email lasted two hours.",
- "normalized_hypothesis": "the meeting that could have been an email lasted two hours.",
- "errors": 0,
- "words": 11,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "the meeting that could have been an email lasted 2 hours",
- "normalized_hypothesis": "the meeting that could have been an email lasted 2 hours",
- "errors": 0,
- "words": 11,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_110",
- "category": "descriptive",
- "text": "There's a specific kind of tired that sleep doesn't fix.",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_110.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "There's a specific kind of tired that sleep doesn't fix.",
- "raw": {
- "normalized_reference": "there's a specific kind of tired that sleep doesn't fix.",
- "normalized_hypothesis": "there's a specific kind of tired that sleep doesn't fix.",
- "errors": 0,
- "words": 10,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "there is a specific kind of tired that sleep does not fix",
- "normalized_hypothesis": "there is a specific kind of tired that sleep does not fix",
- "errors": 0,
- "words": 12,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_111",
- "category": "emotional",
- "text": "Something about that moment broke something open in me that I hadn't realized was closed.",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_111.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Something about that moment broke something open in me that I hadn't realized was closed.",
- "raw": {
- "normalized_reference": "something about that moment broke something open in me that i hadn't realized was closed.",
- "normalized_hypothesis": "something about that moment broke something open in me that i hadn't realized was closed.",
- "errors": 0,
- "words": 15,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "something about that moment broke something open in me that i had not realized was closed",
- "normalized_hypothesis": "something about that moment broke something open in me that i had not realized was closed",
- "errors": 0,
- "words": 16,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_112",
- "category": "long",
- "text": "The thing about making something good is that you don't usually know it's good while you're making it. You just know you're not done yet.",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_112.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "The thing about making something good is that you don't usually know it's good while you're making it. You just know you're not done yet.",
- "raw": {
- "normalized_reference": "the thing about making something good is that you don't usually know it's good while you're making it. you just know you're not done yet.",
- "normalized_hypothesis": "the thing about making something good is that you don't usually know it's good while you're making it. you just know you're not done yet.",
- "errors": 0,
- "words": 25,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "the thing about making something good is that you do not usually know it is good while you are making it you just know you are not done yet",
- "normalized_hypothesis": "the thing about making something good is that you do not usually know it is good while you are making it you just know you are not done yet",
- "errors": 0,
- "words": 29,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_113",
- "category": "narrative",
- "text": "She opened the door slowly, as if she expected to find something she wasn't ready for.",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_113.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "She opened the door slowly, as if she expected to find something she wasn't ready for.",
- "raw": {
- "normalized_reference": "she opened the door slowly, as if she expected to find something she wasn't ready for.",
- "normalized_hypothesis": "she opened the door slowly, as if she expected to find something she wasn't ready for.",
- "errors": 0,
- "words": 16,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "she opened the door slowly as if she expected to find something she was not ready for",
- "normalized_hypothesis": "she opened the door slowly as if she expected to find something she was not ready for",
- "errors": 0,
- "words": 17,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_114",
- "category": "question",
- "text": "What would you do differently if you could go back?",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_114.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "What would you do differently if you could go back?",
- "raw": {
- "normalized_reference": "what would you do differently if you could go back?",
- "normalized_hypothesis": "what would you do differently if you could go back?",
- "errors": 0,
- "words": 10,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "what would you do differently if you could go back",
- "normalized_hypothesis": "what would you do differently if you could go back",
- "errors": 0,
- "words": 10,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_115",
- "category": "short",
- "text": "Okay. Okay okay okay.",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_115.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Okay, okay, okay, okay.",
- "raw": {
- "normalized_reference": "okay. okay okay okay.",
- "normalized_hypothesis": "okay, okay, okay, okay.",
- "errors": 0,
- "words": 4,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "okay okay okay okay",
- "normalized_hypothesis": "okay okay okay okay",
- "errors": 0,
- "words": 4,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_116",
- "category": "technical",
- "text": "The latency goes up significantly at scale, which is expected.",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_116.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "The latency goes up significantly at scale, which is expected.",
- "raw": {
- "normalized_reference": "the latency goes up significantly at scale, which is expected.",
- "normalized_hypothesis": "the latency goes up significantly at scale, which is expected.",
- "errors": 0,
- "words": 10,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "the latency goes up significantly at scale which is expected",
- "normalized_hypothesis": "the latency goes up significantly at scale which is expected",
- "errors": 0,
- "words": 10,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_117",
- "category": "conversational",
- "text": "We ended up staying until two in the morning talking about nothing.",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_117.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "We ended up staying until 2 in the morning talking about nothing.",
- "raw": {
- "normalized_reference": "we ended up staying until two in the morning talking about nothing.",
- "normalized_hypothesis": "we ended up staying until 2 in the morning talking about nothing.",
- "errors": 1,
- "words": 12,
- "wer": 0.08333333333333333
- },
- "semantic": {
- "normalized_reference": "we ended up staying until 2 in the morning talking about nothing",
- "normalized_hypothesis": "we ended up staying until 2 in the morning talking about nothing",
- "errors": 0,
- "words": 12,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_118",
- "category": "descriptive",
- "text": "The rain started slowly, and then all at once, the way it usually does.",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_118.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "The rain started slowly, and then all at once, the way it usually does.",
- "raw": {
- "normalized_reference": "the rain started slowly, and then all at once, the way it usually does.",
- "normalized_hypothesis": "the rain started slowly, and then all at once, the way it usually does.",
- "errors": 0,
- "words": 14,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "the rain started slowly and then all at once the way it usually does",
- "normalized_hypothesis": "the rain started slowly and then all at once the way it usually does",
- "errors": 0,
- "words": 14,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_119",
- "category": "emotional",
- "text": "I honestly did not think I was going to make it through that year.",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_119.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "I honestly did not think I was going to make it through that year.",
- "raw": {
- "normalized_reference": "i honestly did not think i was going to make it through that year.",
- "normalized_hypothesis": "i honestly did not think i was going to make it through that year.",
- "errors": 0,
- "words": 14,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "i honestly did not think i was going to make it through that year",
- "normalized_hypothesis": "i honestly did not think i was going to make it through that year",
- "errors": 0,
- "words": 14,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_120",
- "category": "long",
- "text": "There's an assumption that people who seem confident all the time actually feel confident, and in my experience that's almost never true.",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_120.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "There's an assumption that people who seem confident all the time actually feel confident, and in my experience that's almost never true.",
- "raw": {
- "normalized_reference": "there's an assumption that people who seem confident all the time actually feel confident, and in my experience that's almost never true.",
- "normalized_hypothesis": "there's an assumption that people who seem confident all the time actually feel confident, and in my experience that's almost never true.",
- "errors": 0,
- "words": 22,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "there is an assumption that people who seem confident all the time actually feel confident and in my experience that is almost never true",
- "normalized_hypothesis": "there is an assumption that people who seem confident all the time actually feel confident and in my experience that is almost never true",
- "errors": 0,
- "words": 24,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_121",
- "category": "narrative",
- "text": "The moment I opened the door I knew something was different, even if I couldn't say what.",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_121.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "The moment I opened the door I knew something was different, even if I couldn't say what.",
- "raw": {
- "normalized_reference": "the moment i opened the door i knew something was different, even if i couldn't say what.",
- "normalized_hypothesis": "the moment i opened the door i knew something was different, even if i couldn't say what.",
- "errors": 0,
- "words": 17,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "the moment i opened the door i knew something was different even if i could not say what",
- "normalized_hypothesis": "the moment i opened the door i knew something was different even if i could not say what",
- "errors": 0,
- "words": 18,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_122",
- "category": "question",
- "text": "How do you explain something that doesn't make sense even to you?",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_122.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "How do you explain something that doesn't make sense even to you?",
- "raw": {
- "normalized_reference": "how do you explain something that doesn't make sense even to you?",
- "normalized_hypothesis": "how do you explain something that doesn't make sense even to you?",
- "errors": 0,
- "words": 12,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "how do you explain something that does not make sense even to you",
- "normalized_hypothesis": "how do you explain something that does not make sense even to you",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_123",
- "category": "short",
- "text": "Hold on a second.",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_123.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Hold on a second.",
- "raw": {
- "normalized_reference": "hold on a second.",
- "normalized_hypothesis": "hold on a second.",
- "errors": 0,
- "words": 4,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "hold on a 2nd",
- "normalized_hypothesis": "hold on a 2nd",
- "errors": 0,
- "words": 4,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_124",
- "category": "technical",
- "text": "A deterministic split prevents duplicate normalized text from leaking into evaluation.",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_124.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "A deterministic split prevents duplicate normalized text from leaking into evaluation.",
- "raw": {
- "normalized_reference": "a deterministic split prevents duplicate normalized text from leaking into evaluation.",
- "normalized_hypothesis": "a deterministic split prevents duplicate normalized text from leaking into evaluation.",
- "errors": 0,
- "words": 11,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "a deterministic split prevents duplicate normalized text from leaking into evaluation",
- "normalized_hypothesis": "a deterministic split prevents duplicate normalized text from leaking into evaluation",
- "errors": 0,
- "words": 11,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_125",
- "category": "conversational",
- "text": "Honestly, that was much better than I expected, but the ending still felt rushed.",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_125.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Honestly, that was much better than I expected, but the ending still felt rushed.",
- "raw": {
- "normalized_reference": "honestly, that was much better than i expected, but the ending still felt rushed.",
- "normalized_hypothesis": "honestly, that was much better than i expected, but the ending still felt rushed.",
- "errors": 0,
- "words": 14,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "honestly that was much better than i expected but the ending still felt rushed",
- "normalized_hypothesis": "honestly that was much better than i expected but the ending still felt rushed",
- "errors": 0,
- "words": 14,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_126",
- "category": "descriptive",
- "text": "She had a habit of tilting her head slightly when thinking hard about something.",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_126.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "She had a habit of tilting her head slightly when thinking hard about something.",
- "raw": {
- "normalized_reference": "she had a habit of tilting her head slightly when thinking hard about something.",
- "normalized_hypothesis": "she had a habit of tilting her head slightly when thinking hard about something.",
- "errors": 0,
- "words": 14,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "she had a habit of tilting her head slightly when thinking hard about something",
- "normalized_hypothesis": "she had a habit of tilting her head slightly when thinking hard about something",
- "errors": 0,
- "words": 14,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_127",
- "category": "emotional",
- "text": "I cannot stress enough how much I did not want to be right about this.",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_127.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "I cannot stress enough how much I did not want to be right about this.",
- "raw": {
- "normalized_reference": "i cannot stress enough how much i did not want to be right about this.",
- "normalized_hypothesis": "i cannot stress enough how much i did not want to be right about this.",
- "errors": 0,
- "words": 15,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "i cannot stress enough how much i did not want to be right about this",
- "normalized_hypothesis": "i cannot stress enough how much i did not want to be right about this",
- "errors": 0,
- "words": 15,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_128",
- "category": "long",
- "text": "I used to think the goal was to get to a point where nothing was hard anymore, and now I think that point doesn't exist and I'm not sure I'd want it to.",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_128.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "I used to think the goal was to get to a point where nothing was hard anymore, and now I think that point doesn't exist and I'm not sure I'd want it to.",
- "raw": {
- "normalized_reference": "i used to think the goal was to get to a point where nothing was hard anymore, and now i think that point doesn't exist and i'm not sure i'd want it to.",
- "normalized_hypothesis": "i used to think the goal was to get to a point where nothing was hard anymore, and now i think that point doesn't exist and i'm not sure i'd want it to.",
- "errors": 0,
- "words": 33,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "i used to think the goal was to get to a point where nothing was hard anymore and now i think that point does not exist and i am not sure i would want it to",
- "normalized_hypothesis": "i used to think the goal was to get to a point where nothing was hard anymore and now i think that point does not exist and i am not sure i would want it to",
- "errors": 0,
- "words": 36,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_129",
- "category": "narrative",
- "text": "So this is the part of the story where everything goes completely sideways.",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_129.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "So this is the part of the story where everything goes completely sideways.",
- "raw": {
- "normalized_reference": "so this is the part of the story where everything goes completely sideways.",
- "normalized_hypothesis": "so this is the part of the story where everything goes completely sideways.",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "so this is the part of the story where everything goes completely sideways",
- "normalized_hypothesis": "so this is the part of the story where everything goes completely sideways",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_130",
- "category": "question",
- "text": "Is there a version of this where it actually works out okay?",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_130.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Is there a version of this where it actually works out okay?",
- "raw": {
- "normalized_reference": "is there a version of this where it actually works out okay?",
- "normalized_hypothesis": "is there a version of this where it actually works out okay?",
- "errors": 0,
- "words": 12,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "is there a version of this where it actually works out okay",
- "normalized_hypothesis": "is there a version of this where it actually works out okay",
- "errors": 0,
- "words": 12,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_131",
- "category": "short",
- "text": "Oh no. Oh no no no.",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_131.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Oh, no. Oh, no, no, no.",
- "raw": {
- "normalized_reference": "oh no. oh no no no.",
- "normalized_hypothesis": "oh, no. oh, no, no, no.",
- "errors": 0,
- "words": 6,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "0 no 0 no no no",
- "normalized_hypothesis": "0 no 0 no no no",
- "errors": 0,
- "words": 6,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_132",
- "category": "technical",
- "text": "Always test on a copy before modifying the original.",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_132.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Always test on a copy before modifying the original.",
- "raw": {
- "normalized_reference": "always test on a copy before modifying the original.",
- "normalized_hypothesis": "always test on a copy before modifying the original.",
- "errors": 0,
- "words": 9,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "always test on a copy before modifying the original",
- "normalized_hypothesis": "always test on a copy before modifying the original",
- "errors": 0,
- "words": 9,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_133",
- "category": "conversational",
- "text": "I wasn't even supposed to be there. That's the wild part.",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_133.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "I wasn't even supposed to be there. That's the wild part.",
- "raw": {
- "normalized_reference": "i wasn't even supposed to be there. that's the wild part.",
- "normalized_hypothesis": "i wasn't even supposed to be there. that's the wild part.",
- "errors": 0,
- "words": 11,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "i was not even supposed to be there that is the wild part",
- "normalized_hypothesis": "i was not even supposed to be there that is the wild part",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_134",
- "category": "descriptive",
- "text": "It's a small town. Not the kind you leave. The kind you leave and come back to.",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_134.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "It's a small town, not the kind you leave, the kind you leave and come back to.",
- "raw": {
- "normalized_reference": "it's a small town. not the kind you leave. the kind you leave and come back to.",
- "normalized_hypothesis": "it's a small town, not the kind you leave, the kind you leave and come back to.",
- "errors": 0,
- "words": 17,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "it is a small town not the kind you leave the kind you leave and come back to",
- "normalized_hypothesis": "it is a small town not the kind you leave the kind you leave and come back to",
- "errors": 0,
- "words": 18,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_135",
- "category": "emotional",
- "text": "She cried. I cried. We all just sort of stood there crying.",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_135.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "She cried. I cried. We all just sort of stood there crying.",
- "raw": {
- "normalized_reference": "she cried. i cried. we all just sort of stood there crying.",
- "normalized_hypothesis": "she cried. i cried. we all just sort of stood there crying.",
- "errors": 0,
- "words": 12,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "she cried i cried we all just sort of stood there crying",
- "normalized_hypothesis": "she cried i cried we all just sort of stood there crying",
- "errors": 0,
- "words": 12,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_136",
- "category": "long",
- "text": "There's a version of being busy that's actually just a very elaborate way of avoiding the one thing you know you should be doing.",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_136.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "There's a version of being busy that's actually just a very elaborate way of avoiding the one thing you know you should be doing.",
- "raw": {
- "normalized_reference": "there's a version of being busy that's actually just a very elaborate way of avoiding the one thing you know you should be doing.",
- "normalized_hypothesis": "there's a version of being busy that's actually just a very elaborate way of avoiding the one thing you know you should be doing.",
- "errors": 0,
- "words": 24,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "there is a version of being busy that is actually just a very elaborate way of avoiding the one thing you know you should be doing",
- "normalized_hypothesis": "there is a version of being busy that is actually just a very elaborate way of avoiding the one thing you know you should be doing",
- "errors": 0,
- "words": 26,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_137",
- "category": "narrative",
- "text": "The letter arrived on a Tuesday, which felt appropriate somehow.",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_137.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "The letter arrived on a Tuesday, which felt appropriate somehow.",
- "raw": {
- "normalized_reference": "the letter arrived on a tuesday, which felt appropriate somehow.",
- "normalized_hypothesis": "the letter arrived on a tuesday, which felt appropriate somehow.",
- "errors": 0,
- "words": 10,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "the letter arrived on a tuesday which felt appropriate somehow",
- "normalized_hypothesis": "the letter arrived on a tuesday which felt appropriate somehow",
- "errors": 0,
- "words": 10,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_138",
- "category": "question",
- "text": "What exactly did she say when you told her?",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_138.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "What exactly did she say when you told her?",
- "raw": {
- "normalized_reference": "what exactly did she say when you told her?",
- "normalized_hypothesis": "what exactly did she say when you told her?",
- "errors": 0,
- "words": 9,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "what exactly did she say when you told her",
- "normalized_hypothesis": "what exactly did she say when you told her",
- "errors": 0,
- "words": 9,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_139",
- "category": "short",
- "text": "No way. Absolutely not.",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_139.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "No way to absolutely not.",
- "raw": {
- "normalized_reference": "no way. absolutely not.",
- "normalized_hypothesis": "no way to absolutely not.",
- "errors": 1,
- "words": 4,
- "wer": 0.25
- },
- "semantic": {
- "normalized_reference": "no way absolutely not",
- "normalized_hypothesis": "no way to absolutely not",
- "errors": 1,
- "words": 4,
- "wer": 0.25
- }
- },
- {
- "id": "utmos200_140",
- "category": "technical",
- "text": "The first thing you want to do is make sure everything is backed up.",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_140.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "The first thing you want to do is make sure everything is backed up.",
- "raw": {
- "normalized_reference": "the first thing you want to do is make sure everything is backed up.",
- "normalized_hypothesis": "the first thing you want to do is make sure everything is backed up.",
- "errors": 0,
- "words": 14,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "the 1st thing you want to do is make sure everything is backed up",
- "normalized_hypothesis": "the 1st thing you want to do is make sure everything is backed up",
- "errors": 0,
- "words": 14,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_141",
- "category": "conversational",
- "text": "I have the most insane idea and you need to just hear me out.",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_141.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "I have the most insane idea and you need to just hear me out.",
- "raw": {
- "normalized_reference": "i have the most insane idea and you need to just hear me out.",
- "normalized_hypothesis": "i have the most insane idea and you need to just hear me out.",
- "errors": 0,
- "words": 14,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "i have the most insane idea and you need to just hear me out",
- "normalized_hypothesis": "i have the most insane idea and you need to just hear me out",
- "errors": 0,
- "words": 14,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_142",
- "category": "descriptive",
- "text": "There's a coffee shop on the corner that's been there longer than anyone can remember.",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_142.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "There's a coffee shop on the corner that's been there longer than anyone can remember.",
- "raw": {
- "normalized_reference": "there's a coffee shop on the corner that's been there longer than anyone can remember.",
- "normalized_hypothesis": "there's a coffee shop on the corner that's been there longer than anyone can remember.",
- "errors": 0,
- "words": 15,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "there is a coffee shop on the corner that has been there longer than anyone can remember",
- "normalized_hypothesis": "there is a coffee shop on the corner that has been there longer than anyone can remember",
- "errors": 0,
- "words": 17,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_143",
- "category": "emotional",
- "text": "I kept it together until I got to the car, and then I completely fell apart.",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_143.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "I kept it together until I got to the car, and then I completely fell apart.",
- "raw": {
- "normalized_reference": "i kept it together until i got to the car, and then i completely fell apart.",
- "normalized_hypothesis": "i kept it together until i got to the car, and then i completely fell apart.",
- "errors": 0,
- "words": 16,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "i kept it together until i got to the car and then i completely fell apart",
- "normalized_hypothesis": "i kept it together until i got to the car and then i completely fell apart",
- "errors": 0,
- "words": 16,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_144",
- "category": "long",
- "text": "The older I get, the more I think that most of what we call wisdom is just pattern recognition built from having been wrong enough times to start noticing.",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_144.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "The older I get, the more I think that most of what we call wisdom is just pattern recognition built from having been wrong enough times to start noticing.",
- "raw": {
- "normalized_reference": "the older i get, the more i think that most of what we call wisdom is just pattern recognition built from having been wrong enough times to start noticing.",
- "normalized_hypothesis": "the older i get, the more i think that most of what we call wisdom is just pattern recognition built from having been wrong enough times to start noticing.",
- "errors": 0,
- "words": 29,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "the older i get the more i think that most of what we call wisdom is just pattern recognition built from having been wrong enough times to start noticing",
- "normalized_hypothesis": "the older i get the more i think that most of what we call wisdom is just pattern recognition built from having been wrong enough times to start noticing",
- "errors": 0,
- "words": 29,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_145",
- "category": "narrative",
- "text": "I remember the exact moment I realized I had made a huge mistake.",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_145.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "I remember the exact moment I realized I had made a huge mistake.",
- "raw": {
- "normalized_reference": "i remember the exact moment i realized i had made a huge mistake.",
- "normalized_hypothesis": "i remember the exact moment i realized i had made a huge mistake.",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "i remember the exact moment i realized i had made a huge mistake",
- "normalized_hypothesis": "i remember the exact moment i realized i had made a huge mistake",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_146",
- "category": "question",
- "text": "Do you think it would have gone differently if we'd left earlier?",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_146.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Do you think it would have gone differently if we'd left earlier?",
- "raw": {
- "normalized_reference": "do you think it would have gone differently if we'd left earlier?",
- "normalized_hypothesis": "do you think it would have gone differently if we'd left earlier?",
- "errors": 0,
- "words": 12,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "do you think it would have gone differently if we would left earlier",
- "normalized_hypothesis": "do you think it would have gone differently if we would left earlier",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_147",
- "category": "short",
- "text": "You have to be kidding me.",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_147.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "You have to be kidding me.",
- "raw": {
- "normalized_reference": "you have to be kidding me.",
- "normalized_hypothesis": "you have to be kidding me.",
- "errors": 0,
- "words": 6,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "you have to be kidding me",
- "normalized_hypothesis": "you have to be kidding me",
- "errors": 0,
- "words": 6,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_148",
- "category": "technical",
- "text": "If the checksum fails, restart the worker process without deleting the cached tensors.",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_148.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "If the checksum fails, restart the worker process without deleting the cache tensors.",
- "raw": {
- "normalized_reference": "if the checksum fails, restart the worker process without deleting the cached tensors.",
- "normalized_hypothesis": "if the checksum fails, restart the worker process without deleting the cache tensors.",
- "errors": 1,
- "words": 13,
- "wer": 0.07692307692307693
- },
- "semantic": {
- "normalized_reference": "if the checksum fails restart the worker process without deleting the cached tensors",
- "normalized_hypothesis": "if the checksum fails restart the worker process without deleting the cache tensors",
- "errors": 1,
- "words": 13,
- "wer": 0.07692307692307693
- }
- },
- {
- "id": "utmos200_149",
- "category": "conversational",
- "text": "I need you to understand how close that was to not working.",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_149.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "I need you to understand how close that was to not working.",
- "raw": {
- "normalized_reference": "i need you to understand how close that was to not working.",
- "normalized_hypothesis": "i need you to understand how close that was to not working.",
- "errors": 0,
- "words": 12,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "i need you to understand how close that was to not working",
- "normalized_hypothesis": "i need you to understand how close that was to not working",
- "errors": 0,
- "words": 12,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_150",
- "category": "descriptive",
- "text": "It was the sort of morning where the whole world looks rinsed clean.",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_150.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "It was the sort of morning where the whole world looks rinsed clean.",
- "raw": {
- "normalized_reference": "it was the sort of morning where the whole world looks rinsed clean.",
- "normalized_hypothesis": "it was the sort of morning where the whole world looks rinsed clean.",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "it was the sort of morning where the whole world looks rinsed clean",
- "normalized_hypothesis": "it was the sort of morning where the whole world looks rinsed clean",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_151",
- "category": "emotional",
- "text": "She started laughing and I started laughing and we couldn't stop for a full minute.",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_151.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "She started laughing, and I started laughing, and we couldn't stop for a full minute.",
- "raw": {
- "normalized_reference": "she started laughing and i started laughing and we couldn't stop for a full minute.",
- "normalized_hypothesis": "she started laughing, and i started laughing, and we couldn't stop for a full minute.",
- "errors": 0,
- "words": 15,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "she started laughing and i started laughing and we could not stop for a full minute",
- "normalized_hypothesis": "she started laughing and i started laughing and we could not stop for a full minute",
- "errors": 0,
- "words": 16,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_152",
- "category": "long",
- "text": "We spent three years building something we were genuinely proud of, and then the market shifted and none of it mattered the way we thought it would.",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_152.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "We spent three years building something we were genuinely proud of, and then the market shifted and none of it mattered the way we thought it would.",
- "raw": {
- "normalized_reference": "we spent three years building something we were genuinely proud of, and then the market shifted and none of it mattered the way we thought it would.",
- "normalized_hypothesis": "we spent three years building something we were genuinely proud of, and then the market shifted and none of it mattered the way we thought it would.",
- "errors": 0,
- "words": 27,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "we spent 3 years building something we were genuinely proud of and then the market shifted and none of it mattered the way we thought it would",
- "normalized_hypothesis": "we spent 3 years building something we were genuinely proud of and then the market shifted and none of it mattered the way we thought it would",
- "errors": 0,
- "words": 27,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_153",
- "category": "narrative",
- "text": "And somehow, despite all of that, it actually turned out okay.",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_153.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "And somehow, despite all of that, it actually turned out okay.",
- "raw": {
- "normalized_reference": "and somehow, despite all of that, it actually turned out okay.",
- "normalized_hypothesis": "and somehow, despite all of that, it actually turned out okay.",
- "errors": 0,
- "words": 11,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "and somehow despite all of that it actually turned out okay",
- "normalized_hypothesis": "and somehow despite all of that it actually turned out okay",
- "errors": 0,
- "words": 11,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_154",
- "category": "question",
- "text": "Is that something you actually want, or something you think you should want?",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_154.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Is that something you actually want or something you think you should want?",
- "raw": {
- "normalized_reference": "is that something you actually want, or something you think you should want?",
- "normalized_hypothesis": "is that something you actually want or something you think you should want?",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "is that something you actually want or something you think you should want",
- "normalized_hypothesis": "is that something you actually want or something you think you should want",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_155",
- "category": "short",
- "text": "This is fine.",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_155.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "This is fine.",
- "raw": {
- "normalized_reference": "this is fine.",
- "normalized_hypothesis": "this is fine.",
- "errors": 0,
- "words": 3,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "this is fine",
- "normalized_hypothesis": "this is fine",
- "errors": 0,
- "words": 3,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_156",
- "category": "technical",
- "text": "The tricky part isn't the setup, it's making sure all dependencies line up.",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_156.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "The tricky part isn't the setup, it's making sure all dependencies line up.",
- "raw": {
- "normalized_reference": "the tricky part isn't the setup, it's making sure all dependencies line up.",
- "normalized_hypothesis": "the tricky part isn't the setup, it's making sure all dependencies line up.",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "the tricky part is not the setup it is making sure all dependencies line up",
- "normalized_hypothesis": "the tricky part is not the setup it is making sure all dependencies line up",
- "errors": 0,
- "words": 15,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_157",
- "category": "conversational",
- "text": "He called it a minor setback. I called it a disaster.",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_157.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "He called it a minor setback that I called it a disaster.",
- "raw": {
- "normalized_reference": "he called it a minor setback. i called it a disaster.",
- "normalized_hypothesis": "he called it a minor setback that i called it a disaster.",
- "errors": 1,
- "words": 11,
- "wer": 0.09090909090909091
- },
- "semantic": {
- "normalized_reference": "he called it a minor setback i called it a disaster",
- "normalized_hypothesis": "he called it a minor setback that i called it a disaster",
- "errors": 1,
- "words": 11,
- "wer": 0.09090909090909091
- }
- },
- {
- "id": "utmos200_158",
- "category": "descriptive",
- "text": "The whole place smelled like pine and old wood and cold air.",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_158.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "The whole place smelled like pine and old wood and cold air.",
- "raw": {
- "normalized_reference": "the whole place smelled like pine and old wood and cold air.",
- "normalized_hypothesis": "the whole place smelled like pine and old wood and cold air.",
- "errors": 0,
- "words": 12,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "the whole place smelled like pine and old wood and cold air",
- "normalized_hypothesis": "the whole place smelled like pine and old wood and cold air",
- "errors": 0,
- "words": 12,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_159",
- "category": "emotional",
- "text": "I've never been so relieved in my entire life.",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_159.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "I've never been so relieved in my entire life.",
- "raw": {
- "normalized_reference": "i've never been so relieved in my entire life.",
- "normalized_hypothesis": "i've never been so relieved in my entire life.",
- "errors": 0,
- "words": 9,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "i have never been so relieved in my entire life",
- "normalized_hypothesis": "i have never been so relieved in my entire life",
- "errors": 0,
- "words": 10,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_160",
- "category": "narrative",
- "text": "I wrote it all down that night so I wouldn't forget, and I still haven't looked at it.",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_160.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "I wrote it all down that night so I wouldn't forget, and I still haven't looked at it.",
- "raw": {
- "normalized_reference": "i wrote it all down that night so i wouldn't forget, and i still haven't looked at it.",
- "normalized_hypothesis": "i wrote it all down that night so i wouldn't forget, and i still haven't looked at it.",
- "errors": 0,
- "words": 18,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "i wrote it all down that night so i would not forget and i still have not looked at it",
- "normalized_hypothesis": "i wrote it all down that night so i would not forget and i still have not looked at it",
- "errors": 0,
- "words": 20,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_161",
- "category": "short",
- "text": "That's kind of wild actually.",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_161.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "That's kind of wild actually.",
- "raw": {
- "normalized_reference": "that's kind of wild actually.",
- "normalized_hypothesis": "that's kind of wild actually.",
- "errors": 0,
- "words": 5,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "that is kind of wild actually",
- "normalized_hypothesis": "that is kind of wild actually",
- "errors": 0,
- "words": 6,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_162",
- "category": "technical",
- "text": "Version pinning will save you a lot of pain later down the line.",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_162.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Version pinning will save you a lot of pain later down the line.",
- "raw": {
- "normalized_reference": "version pinning will save you a lot of pain later down the line.",
- "normalized_hypothesis": "version pinning will save you a lot of pain later down the line.",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "version pinning will save you a lot of pain later down the line",
- "normalized_hypothesis": "version pinning will save you a lot of pain later down the line",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_163",
- "category": "conversational",
- "text": "At some point you just have to commit and see what happens.",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_163.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "At some point you just have to commit and see what happens.",
- "raw": {
- "normalized_reference": "at some point you just have to commit and see what happens.",
- "normalized_hypothesis": "at some point you just have to commit and see what happens.",
- "errors": 0,
- "words": 12,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "at some point you just have to commit and see what happens",
- "normalized_hypothesis": "at some point you just have to commit and see what happens",
- "errors": 0,
- "words": 12,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_164",
- "category": "emotional",
- "text": "Nobody told me it was going to be this hard, and honestly, I'm grateful.",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_164.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Nobody told me it was going to be this hard, and honestly, I'm grateful.",
- "raw": {
- "normalized_reference": "nobody told me it was going to be this hard, and honestly, i'm grateful.",
- "normalized_hypothesis": "nobody told me it was going to be this hard, and honestly, i'm grateful.",
- "errors": 0,
- "words": 14,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "nobody told me it was going to be this hard and honestly i am grateful",
- "normalized_hypothesis": "nobody told me it was going to be this hard and honestly i am grateful",
- "errors": 0,
- "words": 15,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_165",
- "category": "narrative",
- "text": "Chapter seven. In which everything that could go wrong did, and a few things that couldn't.",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_165.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Chapter 7, in which everything that could go wrong did, and a few things that couldn't.",
- "raw": {
- "normalized_reference": "chapter seven. in which everything that could go wrong did, and a few things that couldn't.",
- "normalized_hypothesis": "chapter 7, in which everything that could go wrong did, and a few things that couldn't.",
- "errors": 1,
- "words": 16,
- "wer": 0.0625
- },
- "semantic": {
- "normalized_reference": "chapter 7 in which everything that could go wrong did and a few things that could not",
- "normalized_hypothesis": "chapter 7 in which everything that could go wrong did and a few things that could not",
- "errors": 0,
- "words": 17,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_166",
- "category": "short",
- "text": "Wait, what?",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_166.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Wait, what?",
- "raw": {
- "normalized_reference": "wait, what?",
- "normalized_hypothesis": "wait, what?",
- "errors": 0,
- "words": 2,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "wait what",
- "normalized_hypothesis": "wait what",
- "errors": 0,
- "words": 2,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_167",
- "category": "technical",
- "text": "The graphics processor remained idle because the data loader exhausted every central processing core.",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_167.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "The graphics processor remained idle because the data loader exhausted every central processing core.",
- "raw": {
- "normalized_reference": "the graphics processor remained idle because the data loader exhausted every central processing core.",
- "normalized_hypothesis": "the graphics processor remained idle because the data loader exhausted every central processing core.",
- "errors": 0,
- "words": 14,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "the graphics processor remained idle because the data loader exhausted every central processing core",
- "normalized_hypothesis": "the graphics processor remained idle because the data loader exhausted every central processing core",
- "errors": 0,
- "words": 14,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_168",
- "category": "conversational",
- "text": "Could you speak a little more slowly? I am trying to write every detail down.",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_168.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Could you speak a little more slowly? I'm trying to write every detail down.",
- "raw": {
- "normalized_reference": "could you speak a little more slowly? i am trying to write every detail down.",
- "normalized_hypothesis": "could you speak a little more slowly? i'm trying to write every detail down.",
- "errors": 2,
- "words": 15,
- "wer": 0.13333333333333333
- },
- "semantic": {
- "normalized_reference": "could you speak a little more slowly i am trying to write every detail down",
- "normalized_hypothesis": "could you speak a little more slowly i am trying to write every detail down",
- "errors": 0,
- "words": 15,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_169",
- "category": "emotional",
- "text": "I was so excited I could barely form a complete sentence.",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_169.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "I was so excited I could barely form a complete sentence.",
- "raw": {
- "normalized_reference": "i was so excited i could barely form a complete sentence.",
- "normalized_hypothesis": "i was so excited i could barely form a complete sentence.",
- "errors": 0,
- "words": 11,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "i was so excited i could barely form a complete sentence",
- "normalized_hypothesis": "i was so excited i could barely form a complete sentence",
- "errors": 0,
- "words": 11,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_170",
- "category": "narrative",
- "text": "He looked at me like I had said something in a language he didn't speak.",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_170.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "He looked at me like I had said something in a language he didn't speak.",
- "raw": {
- "normalized_reference": "he looked at me like i had said something in a language he didn't speak.",
- "normalized_hypothesis": "he looked at me like i had said something in a language he didn't speak.",
- "errors": 0,
- "words": 15,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "he looked at me like i had said something in a language he did not speak",
- "normalized_hypothesis": "he looked at me like i had said something in a language he did not speak",
- "errors": 0,
- "words": 16,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_171",
- "category": "short",
- "text": "That's actually hilarious.",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_171.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "That's actually hilarious.",
- "raw": {
- "normalized_reference": "that's actually hilarious.",
- "normalized_hypothesis": "that's actually hilarious.",
- "errors": 0,
- "words": 3,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "that is actually hilarious",
- "normalized_hypothesis": "that is actually hilarious",
- "errors": 0,
- "words": 4,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_172",
- "category": "technical",
- "text": "The encoder maps phonemes, stress, and phrase boundaries into a shared hidden representation.",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_172.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "The encoder maps phonemes, stress, and phrase boundaries into a shared hidden representation.",
- "raw": {
- "normalized_reference": "the encoder maps phonemes, stress, and phrase boundaries into a shared hidden representation.",
- "normalized_hypothesis": "the encoder maps phonemes, stress, and phrase boundaries into a shared hidden representation.",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "the encoder maps phonemes stress and phrase boundaries into a shared hidden representation",
- "normalized_hypothesis": "the encoder maps phonemes stress and phrase boundaries into a shared hidden representation",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_173",
- "category": "conversational",
- "text": "She showed up three hours late and acted like nothing had happened.",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_173.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "She showed up three hours late and acted like nothing had happened.",
- "raw": {
- "normalized_reference": "she showed up three hours late and acted like nothing had happened.",
- "normalized_hypothesis": "she showed up three hours late and acted like nothing had happened.",
- "errors": 0,
- "words": 12,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "she showed up 3 hours late and acted like nothing had happened",
- "normalized_hypothesis": "she showed up 3 hours late and acted like nothing had happened",
- "errors": 0,
- "words": 12,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_174",
- "category": "emotional",
- "text": "Oh my god. This is the best thing I've ever tasted.",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_174.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Oh my god, this is the best thing I've ever tasted.",
- "raw": {
- "normalized_reference": "oh my god. this is the best thing i've ever tasted.",
- "normalized_hypothesis": "oh my god, this is the best thing i've ever tasted.",
- "errors": 0,
- "words": 11,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "0 my god this is the best thing i have ever tasted",
- "normalized_hypothesis": "0 my god this is the best thing i have ever tasted",
- "errors": 0,
- "words": 12,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_175",
- "category": "narrative",
- "text": "The note was short, which was somehow more frightening than a long one would have been.",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_175.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "The note was short, which was somehow more frightening than a long one would have been.",
- "raw": {
- "normalized_reference": "the note was short, which was somehow more frightening than a long one would have been.",
- "normalized_hypothesis": "the note was short, which was somehow more frightening than a long one would have been.",
- "errors": 0,
- "words": 16,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "the note was short which was somehow more frightening than a long one would have been",
- "normalized_hypothesis": "the note was short which was somehow more frightening than a long one would have been",
- "errors": 0,
- "words": 16,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_176",
- "category": "short",
- "text": "Noted.",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_176.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Noted.",
- "raw": {
- "normalized_reference": "noted.",
- "normalized_hypothesis": "noted.",
- "errors": 0,
- "words": 1,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "noted",
- "normalized_hypothesis": "noted",
- "errors": 0,
- "words": 1,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_177",
- "category": "technical",
- "text": "It's a simple process once you understand the underlying structure.",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_177.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "It's a simple process once you understand the underlying structure.",
- "raw": {
- "normalized_reference": "it's a simple process once you understand the underlying structure.",
- "normalized_hypothesis": "it's a simple process once you understand the underlying structure.",
- "errors": 0,
- "words": 10,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "it is a simple process once you understand the underlying structure",
- "normalized_hypothesis": "it is a simple process once you understand the underlying structure",
- "errors": 0,
- "words": 11,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_178",
- "category": "conversational",
- "text": "You have to hear this story because it's almost too good to be true.",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_178.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "You have to hear this story because it's almost too good to be true.",
- "raw": {
- "normalized_reference": "you have to hear this story because it's almost too good to be true.",
- "normalized_hypothesis": "you have to hear this story because it's almost too good to be true.",
- "errors": 0,
- "words": 14,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "you have to hear this story because it is almost too good to be true",
- "normalized_hypothesis": "you have to hear this story because it is almost too good to be true",
- "errors": 0,
- "words": 15,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_179",
- "category": "emotional",
- "text": "I have never felt more seen in my entire life than in that moment.",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_179.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "I have never felt more seen in my entire life than in that moment.",
- "raw": {
- "normalized_reference": "i have never felt more seen in my entire life than in that moment.",
- "normalized_hypothesis": "i have never felt more seen in my entire life than in that moment.",
- "errors": 0,
- "words": 14,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "i have never felt more seen in my entire life than in that moment",
- "normalized_hypothesis": "i have never felt more seen in my entire life than in that moment",
- "errors": 0,
- "words": 14,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_180",
- "category": "narrative",
- "text": "She knew. She had always known. She just waited for the right moment.",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_180.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Ji knew but she had always known, she just waited for the right moment.",
- "raw": {
- "normalized_reference": "she knew. she had always known. she just waited for the right moment.",
- "normalized_hypothesis": "ji knew but she had always known, she just waited for the right moment.",
- "errors": 2,
- "words": 13,
- "wer": 0.15384615384615385
- },
- "semantic": {
- "normalized_reference": "she knew she had always known she just waited for the right moment",
- "normalized_hypothesis": "ji knew but she had always known she just waited for the right moment",
- "errors": 2,
- "words": 13,
- "wer": 0.15384615384615385
- }
- },
- {
- "id": "utmos200_181",
- "category": "short",
- "text": "Yeah, no. Hard pass.",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_181.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Yeah, no.Card pass.",
- "raw": {
- "normalized_reference": "yeah, no. hard pass.",
- "normalized_hypothesis": "yeah, no.card pass.",
- "errors": 1,
- "words": 4,
- "wer": 0.25
- },
- "semantic": {
- "normalized_reference": "yeah no hard pass",
- "normalized_hypothesis": "yeah no card pass",
- "errors": 1,
- "words": 4,
- "wer": 0.25
- }
- },
- {
- "id": "utmos200_182",
- "category": "technical",
- "text": "Please verify the sample rate, hop length, and mel filter configuration before inference.",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_182.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Please verify the sample rate, pop length, and MEL filter configuration before inference.",
- "raw": {
- "normalized_reference": "please verify the sample rate, hop length, and mel filter configuration before inference.",
- "normalized_hypothesis": "please verify the sample rate, pop length, and mel filter configuration before inference.",
- "errors": 1,
- "words": 13,
- "wer": 0.07692307692307693
- },
- "semantic": {
- "normalized_reference": "please verify the sample rate hop length and mel filter configuration before inference",
- "normalized_hypothesis": "please verify the sample rate pop length and mel filter configuration before inference",
- "errors": 1,
- "words": 13,
- "wer": 0.07692307692307693
- }
- },
- {
- "id": "utmos200_183",
- "category": "conversational",
- "text": "That's the worst idea I've ever heard and I'm completely in.",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_183.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "That's the worst idea I've ever heard and I'm completely in.",
- "raw": {
- "normalized_reference": "that's the worst idea i've ever heard and i'm completely in.",
- "normalized_hypothesis": "that's the worst idea i've ever heard and i'm completely in.",
- "errors": 0,
- "words": 11,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "that is the worst idea i have ever heard and i am completely in",
- "normalized_hypothesis": "that is the worst idea i have ever heard and i am completely in",
- "errors": 0,
- "words": 14,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_184",
- "category": "emotional",
- "text": "I remember thinking: this is it. This is the thing I've been waiting for.",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_184.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "I remember thinking, this is it. This is the thing I've been waiting for.",
- "raw": {
- "normalized_reference": "i remember thinking: this is it. this is the thing i've been waiting for.",
- "normalized_hypothesis": "i remember thinking, this is it. this is the thing i've been waiting for.",
- "errors": 0,
- "words": 14,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "i remember thinking this is it this is the thing i have been waiting for",
- "normalized_hypothesis": "i remember thinking this is it this is the thing i have been waiting for",
- "errors": 0,
- "words": 15,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_185",
- "category": "narrative",
- "text": "We never talked about it again, which I think was the right call.",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_185.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "We never talked about it again, which I think was the right call.",
- "raw": {
- "normalized_reference": "we never talked about it again, which i think was the right call.",
- "normalized_hypothesis": "we never talked about it again, which i think was the right call.",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "we never talked about it again which i think was the right call",
- "normalized_hypothesis": "we never talked about it again which i think was the right call",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_186",
- "category": "short",
- "text": "In hindsight, obvious.",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_186.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "In hindsight, obvious.",
- "raw": {
- "normalized_reference": "in hindsight, obvious.",
- "normalized_hypothesis": "in hindsight, obvious.",
- "errors": 0,
- "words": 3,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "in hindsight obvious",
- "normalized_hypothesis": "in hindsight obvious",
- "errors": 0,
- "words": 3,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_187",
- "category": "technical",
- "text": "Once it's configured correctly, it basically runs itself.",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_187.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Once it's configured correctly, it basically runs itself.",
- "raw": {
- "normalized_reference": "once it's configured correctly, it basically runs itself.",
- "normalized_hypothesis": "once it's configured correctly, it basically runs itself.",
- "errors": 0,
- "words": 8,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "once it is configured correctly it basically runs itself",
- "normalized_hypothesis": "once it is configured correctly it basically runs itself",
- "errors": 0,
- "words": 9,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_188",
- "category": "conversational",
- "text": "You got engaged? When? How? Tell me everything right now.",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_188.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "You got engaged? When? How? Tell me everything right now.",
- "raw": {
- "normalized_reference": "you got engaged? when? how? tell me everything right now.",
- "normalized_hypothesis": "you got engaged? when? how? tell me everything right now.",
- "errors": 0,
- "words": 10,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "you got engaged when how tell me everything right now",
- "normalized_hypothesis": "you got engaged when how tell me everything right now",
- "errors": 0,
- "words": 10,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_189",
- "category": "emotional",
- "text": "It's hard to explain how much that moment meant to me.",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_189.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "It's hard to explain how much that moment meant to me.",
- "raw": {
- "normalized_reference": "it's hard to explain how much that moment meant to me.",
- "normalized_hypothesis": "it's hard to explain how much that moment meant to me.",
- "errors": 0,
- "words": 11,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "it is hard to explain how much that moment meant to me",
- "normalized_hypothesis": "it is hard to explain how much that moment meant to me",
- "errors": 0,
- "words": 12,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_190",
- "category": "narrative",
- "text": "He told the story three different times that night and somehow it got better each time.",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_190.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "He told the story three different times that night, and somehow it got better each time.",
- "raw": {
- "normalized_reference": "he told the story three different times that night and somehow it got better each time.",
- "normalized_hypothesis": "he told the story three different times that night, and somehow it got better each time.",
- "errors": 0,
- "words": 16,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "he told the story 3 different times that night and somehow it got better each time",
- "normalized_hypothesis": "he told the story 3 different times that night and somehow it got better each time",
- "errors": 0,
- "words": 16,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_191",
- "category": "short",
- "text": "Right. Moving on.",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_191.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Right at moving on.",
- "raw": {
- "normalized_reference": "right. moving on.",
- "normalized_hypothesis": "right at moving on.",
- "errors": 1,
- "words": 3,
- "wer": 0.3333333333333333
- },
- "semantic": {
- "normalized_reference": "right moving on",
- "normalized_hypothesis": "right at moving on",
- "errors": 1,
- "words": 3,
- "wer": 0.3333333333333333
- }
- },
- {
- "id": "utmos200_192",
- "category": "conversational",
- "text": "No, I did not move your keys; check beside the small ceramic bowl.",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_192.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "No, I did not move your keys. Check beside the small ceramic bowl.",
- "raw": {
- "normalized_reference": "no, i did not move your keys; check beside the small ceramic bowl.",
- "normalized_hypothesis": "no, i did not move your keys. check beside the small ceramic bowl.",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "no i did not move your keys check beside the small ceramic bowl",
- "normalized_hypothesis": "no i did not move your keys check beside the small ceramic bowl",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_193",
- "category": "emotional",
- "text": "I was devastated. I'm not going to pretend otherwise.",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_193.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "I was devastated, I'm not going to pretend otherwise.",
- "raw": {
- "normalized_reference": "i was devastated. i'm not going to pretend otherwise.",
- "normalized_hypothesis": "i was devastated, i'm not going to pretend otherwise.",
- "errors": 0,
- "words": 9,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "i was devastated i am not going to pretend otherwise",
- "normalized_hypothesis": "i was devastated i am not going to pretend otherwise",
- "errors": 0,
- "words": 10,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_194",
- "category": "narrative",
- "text": "That was the last summer before everything changed, though we didn't know it yet.",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_194.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "That was the last summer before everything changed, though we didn't know it yet.",
- "raw": {
- "normalized_reference": "that was the last summer before everything changed, though we didn't know it yet.",
- "normalized_hypothesis": "that was the last summer before everything changed, though we didn't know it yet.",
- "errors": 0,
- "words": 14,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "that was the last summer before everything changed though we did not know it yet",
- "normalized_hypothesis": "that was the last summer before everything changed though we did not know it yet",
- "errors": 0,
- "words": 15,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_195",
- "category": "short",
- "text": "Bold of you to assume.",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_195.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Bold of you to assume.",
- "raw": {
- "normalized_reference": "bold of you to assume.",
- "normalized_hypothesis": "bold of you to assume.",
- "errors": 0,
- "words": 5,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "bold of you to assume",
- "normalized_hypothesis": "bold of you to assume",
- "errors": 0,
- "words": 5,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_196",
- "category": "conversational",
- "text": "It's one of those decisions that makes total sense in the moment.",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_196.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "It's one of those decisions that makes total sense in the moment.",
- "raw": {
- "normalized_reference": "it's one of those decisions that makes total sense in the moment.",
- "normalized_hypothesis": "it's one of those decisions that makes total sense in the moment.",
- "errors": 0,
- "words": 12,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "it is one of those decisions that makes total sense in the moment",
- "normalized_hypothesis": "it is one of those decisions that makes total sense in the moment",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_197",
- "category": "emotional",
- "text": "I was furious. Not the loud kind. The quiet kind that lasts for days.",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_197.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "I was furious, not the loud kind, the quiet kind that lasts for days.",
- "raw": {
- "normalized_reference": "i was furious. not the loud kind. the quiet kind that lasts for days.",
- "normalized_hypothesis": "i was furious, not the loud kind, the quiet kind that lasts for days.",
- "errors": 0,
- "words": 14,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "i was furious not the loud kind the quiet kind that lasts for days",
- "normalized_hypothesis": "i was furious not the loud kind the quiet kind that lasts for days",
- "errors": 0,
- "words": 14,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_198",
- "category": "narrative",
- "text": "That's the part of the story I usually leave out.",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_198.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "That's the part of the story I usually leave out.",
- "raw": {
- "normalized_reference": "that's the part of the story i usually leave out.",
- "normalized_hypothesis": "that's the part of the story i usually leave out.",
- "errors": 0,
- "words": 10,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "that is the part of the story i usually leave out",
- "normalized_hypothesis": "that is the part of the story i usually leave out",
- "errors": 0,
- "words": 11,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_199",
- "category": "short",
- "text": "I've said too much.",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_199.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "I've said too much.",
- "raw": {
- "normalized_reference": "i've said too much.",
- "normalized_hypothesis": "i've said too much.",
- "errors": 0,
- "words": 4,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "i have said too much",
- "normalized_hypothesis": "i have said too much",
- "errors": 0,
- "words": 5,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_200",
- "category": "conversational",
- "text": "I've been faking that I understood this the entire time.",
- "model": "inflect-nano-v2-finalist",
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_200.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "I've been faking that I understood this the entire time.",
- "raw": {
- "normalized_reference": "i've been faking that i understood this the entire time.",
- "normalized_hypothesis": "i've been faking that i understood this the entire time.",
- "errors": 0,
- "words": 10,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "i have been faking that i understood this the entire time",
- "normalized_hypothesis": "i have been faking that i understood this the entire time",
- "errors": 0,
- "words": 11,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_001",
- "category": "conversational",
- "text": "It wasn't until later that I realized what had actually happened.",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_001.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "It wasn't until later that I realized what had actually happened.",
- "raw": {
- "normalized_reference": "it wasn't until later that i realized what had actually happened.",
- "normalized_hypothesis": "it wasn't until later that i realized what had actually happened.",
- "errors": 0,
- "words": 11,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "it was not until later that i realized what had actually happened",
- "normalized_hypothesis": "it was not until later that i realized what had actually happened",
- "errors": 0,
- "words": 12,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_002",
- "category": "descriptive",
- "text": "The light at that time of afternoon is perfect for about twenty minutes.",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_002.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "The light at that time of afternoon is perfect for about 20 minutes.",
- "raw": {
- "normalized_reference": "the light at that time of afternoon is perfect for about twenty minutes.",
- "normalized_hypothesis": "the light at that time of afternoon is perfect for about 20 minutes.",
- "errors": 1,
- "words": 13,
- "wer": 0.07692307692307693
- },
- "semantic": {
- "normalized_reference": "the light at that time of afternoon is perfect for about 20 minutes",
- "normalized_hypothesis": "the light at that time of afternoon is perfect for about 20 minutes",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_003",
- "category": "emotional",
- "text": "That was the scariest thing I've ever done, and also maybe the best.",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_003.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "That was the scariest thing I've ever done, and also maybe the best.",
- "raw": {
- "normalized_reference": "that was the scariest thing i've ever done, and also maybe the best.",
- "normalized_hypothesis": "that was the scariest thing i've ever done, and also maybe the best.",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "that was the scariest thing i have ever done and also maybe the best",
- "normalized_hypothesis": "that was the scariest thing i have ever done and also maybe the best",
- "errors": 0,
- "words": 14,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_004",
- "category": "homographs",
- "text": "After I record the record, please present the present to the bass player.",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_004.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "After I record the record, please present the present to the bass player.",
- "raw": {
- "normalized_reference": "after i record the record, please present the present to the bass player.",
- "normalized_hypothesis": "after i record the record, please present the present to the bass player.",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "after i record the record please present the present to the bass player",
- "normalized_hypothesis": "after i record the record please present the present to the bass player",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_005",
- "category": "long",
- "text": "If I'm being completely honest, I had no idea what I was doing for most of it, and I think that might be the most normal thing about the whole experience.",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_005.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "If I am being completely honest, I had no idea what I was doing for most of it, and I think that might be the most normal thing about the whole experience.",
- "raw": {
- "normalized_reference": "if i'm being completely honest, i had no idea what i was doing for most of it, and i think that might be the most normal thing about the whole experience.",
- "normalized_hypothesis": "if i am being completely honest, i had no idea what i was doing for most of it, and i think that might be the most normal thing about the whole experience.",
- "errors": 2,
- "words": 31,
- "wer": 0.06451612903225806
- },
- "semantic": {
- "normalized_reference": "if i am being completely honest i had no idea what i was doing for most of it and i think that might be the most normal thing about the whole experience",
- "normalized_hypothesis": "if i am being completely honest i had no idea what i was doing for most of it and i think that might be the most normal thing about the whole experience",
- "errors": 0,
- "words": 32,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_006",
- "category": "long_range",
- "text": "The committee postponed its decision after discovering that three pages of the report had been printed in the wrong order.",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_006.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "The committee postponed its decision after discovering that three pages of the report had been printed in the wrong order.",
- "raw": {
- "normalized_reference": "the committee postponed its decision after discovering that three pages of the report had been printed in the wrong order.",
- "normalized_hypothesis": "the committee postponed its decision after discovering that three pages of the report had been printed in the wrong order.",
- "errors": 0,
- "words": 20,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "the committee postponed its decision after discovering that 3 pages of the report had been printed in the wrong order",
- "normalized_hypothesis": "the committee postponed its decision after discovering that 3 pages of the report had been printed in the wrong order",
- "errors": 0,
- "words": 20,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_007",
- "category": "names_places",
- "text": "Gwendolyn photographed the eucalyptus trees outside Ljubljana.",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_007.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Gwendolyn photographed the eucalyptus trees outside Ljubljana.",
- "raw": {
- "normalized_reference": "gwendolyn photographed the eucalyptus trees outside ljubljana.",
- "normalized_hypothesis": "gwendolyn photographed the eucalyptus trees outside ljubljana.",
- "errors": 0,
- "words": 7,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "gwendolyn photographed the eucalyptus trees outside ljubljana",
- "normalized_hypothesis": "gwendolyn photographed the eucalyptus trees outside ljubljana",
- "errors": 0,
- "words": 7,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_008",
- "category": "narrative",
- "text": "He had this theory, and the more time passes, the more I think he was right.",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_008.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "He had this theory, and the more time passes, the more I think he was right.",
- "raw": {
- "normalized_reference": "he had this theory, and the more time passes, the more i think he was right.",
- "normalized_hypothesis": "he had this theory, and the more time passes, the more i think he was right.",
- "errors": 0,
- "words": 16,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "he had this theory and the more time passes the more i think he was right",
- "normalized_hypothesis": "he had this theory and the more time passes the more i think he was right",
- "errors": 0,
- "words": 16,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_009",
- "category": "numbers",
- "text": "The package weighs twelve point six kilograms and arrived on July twenty-first.",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_009.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "The package weighs 12.6 kilograms and arrived on July 21st.",
- "raw": {
- "normalized_reference": "the package weighs twelve point six kilograms and arrived on july twenty-first.",
- "normalized_hypothesis": "the package weighs 12.6 kilograms and arrived on july 21st.",
- "errors": 5,
- "words": 13,
- "wer": 0.38461538461538464
- },
- "semantic": {
- "normalized_reference": "the package weighs 12.6 kilograms and arrived on july 21st",
- "normalized_hypothesis": "the package weighs 12.6 kilograms and arrived on july 21st",
- "errors": 0,
- "words": 11,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_010",
- "category": "punctuation",
- "text": "First, close the window; second, turn off the lamp; finally, lock the door.",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_010.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "First, close the window. Second, turn off the lamp. Finally, lock the door.",
- "raw": {
- "normalized_reference": "first, close the window; second, turn off the lamp; finally, lock the door.",
- "normalized_hypothesis": "first, close the window. second, turn off the lamp. finally, lock the door.",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "1st close the window 2nd turn off the lamp finally lock the door",
- "normalized_hypothesis": "1st close the window 2nd turn off the lamp finally lock the door",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_011",
- "category": "question",
- "text": "When did you first realize something was wrong?",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_011.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "When did you first realize something was wrong?",
- "raw": {
- "normalized_reference": "when did you first realize something was wrong?",
- "normalized_hypothesis": "when did you first realize something was wrong?",
- "errors": 0,
- "words": 8,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "when did you 1st realize something was wrong",
- "normalized_hypothesis": "when did you 1st realize something was wrong",
- "errors": 0,
- "words": 8,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_012",
- "category": "rare_words",
- "text": "The archaeologist catalogued a Byzantine reliquary and several anachronistic fragments.",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_012.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "the archaeologists cataloged a Byzantine reliquary and several anachronistic fragments.",
- "raw": {
- "normalized_reference": "the archaeologist catalogued a byzantine reliquary and several anachronistic fragments.",
- "normalized_hypothesis": "the archaeologists cataloged a byzantine reliquary and several anachronistic fragments.",
- "errors": 2,
- "words": 10,
- "wer": 0.2
- },
- "semantic": {
- "normalized_reference": "the archaeologist catalogued a byzantine reliquary and several anachronistic fragments",
- "normalized_hypothesis": "the archaeologists cataloged a byzantine reliquary and several anachronistic fragments",
- "errors": 2,
- "words": 10,
- "wer": 0.2
- }
- },
- {
- "id": "utmos200_013",
- "category": "short",
- "text": "I have no idea.",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_013.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "I have no idea.",
- "raw": {
- "normalized_reference": "i have no idea.",
- "normalized_hypothesis": "i have no idea.",
- "errors": 0,
- "words": 4,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "i have no idea",
- "normalized_hypothesis": "i have no idea",
- "errors": 0,
- "words": 4,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_014",
- "category": "technical",
- "text": "The system runs on three core components that all have to stay in sync.",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_014.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "The system runs on three core components that all have to stay in sync.",
- "raw": {
- "normalized_reference": "the system runs on three core components that all have to stay in sync.",
- "normalized_hypothesis": "the system runs on three core components that all have to stay in sync.",
- "errors": 0,
- "words": 14,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "the system runs on 3 core components that all have to stay in sync",
- "normalized_hypothesis": "the system runs on 3 core components that all have to stay in sync",
- "errors": 0,
- "words": 14,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_015",
- "category": "conversational",
- "text": "We improvised the entire second half and somehow nobody noticed.",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_015.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "We improvised the entire second half and somehow nobody noticed.",
- "raw": {
- "normalized_reference": "we improvised the entire second half and somehow nobody noticed.",
- "normalized_hypothesis": "we improvised the entire second half and somehow nobody noticed.",
- "errors": 0,
- "words": 10,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "we improvised the entire 2nd half and somehow nobody noticed",
- "normalized_hypothesis": "we improvised the entire 2nd half and somehow nobody noticed",
- "errors": 0,
- "words": 10,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_016",
- "category": "descriptive",
- "text": "There's a smell in old buildings that's impossible to describe but immediately familiar.",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_016.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "There's a smell in all buildings that's impossible to describe but immediately familiar.",
- "raw": {
- "normalized_reference": "there's a smell in old buildings that's impossible to describe but immediately familiar.",
- "normalized_hypothesis": "there's a smell in all buildings that's impossible to describe but immediately familiar.",
- "errors": 1,
- "words": 13,
- "wer": 0.07692307692307693
- },
- "semantic": {
- "normalized_reference": "there is a smell in old buildings that is impossible to describe but immediately familiar",
- "normalized_hypothesis": "there is a smell in all buildings that is impossible to describe but immediately familiar",
- "errors": 1,
- "words": 15,
- "wer": 0.06666666666666667
- }
- },
- {
- "id": "utmos200_017",
- "category": "emotional",
- "text": "I specifically asked for one thing. One thing. And it still didn't happen.",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_017.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "I specifically asked for one thing. One thing. And it still didn't happen.",
- "raw": {
- "normalized_reference": "i specifically asked for one thing. one thing. and it still didn't happen.",
- "normalized_hypothesis": "i specifically asked for one thing. one thing. and it still didn't happen.",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "i specifically asked for one thing one thing and it still did not happen",
- "normalized_hypothesis": "i specifically asked for one thing one thing and it still did not happen",
- "errors": 0,
- "words": 14,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_018",
- "category": "homographs",
- "text": "They refuse to place the refuse beside the produce aisle.",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_018.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "They refused to place the refuse beside the produce aisle.",
- "raw": {
- "normalized_reference": "they refuse to place the refuse beside the produce aisle.",
- "normalized_hypothesis": "they refused to place the refuse beside the produce aisle.",
- "errors": 1,
- "words": 10,
- "wer": 0.1
- },
- "semantic": {
- "normalized_reference": "they refuse to place the refuse beside the produce aisle",
- "normalized_hypothesis": "they refused to place the refuse beside the produce aisle",
- "errors": 1,
- "words": 10,
- "wer": 0.1
- }
- },
- {
- "id": "utmos200_019",
- "category": "long",
- "text": "There were days last year when I genuinely didn't know how I was going to get through it, and somehow I did, and I still don't fully understand how.",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_019.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "There were days last year when I genuinely didn't know how I was going to get through it, and somehow I did, and I still don't fully understand how.",
- "raw": {
- "normalized_reference": "there were days last year when i genuinely didn't know how i was going to get through it, and somehow i did, and i still don't fully understand how.",
- "normalized_hypothesis": "there were days last year when i genuinely didn't know how i was going to get through it, and somehow i did, and i still don't fully understand how.",
- "errors": 0,
- "words": 29,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "there were days last year when i genuinely did not know how i was going to get through it and somehow i did and i still do not fully understand how",
- "normalized_hypothesis": "there were days last year when i genuinely did not know how i was going to get through it and somehow i did and i still do not fully understand how",
- "errors": 0,
- "words": 31,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_020",
- "category": "long_range",
- "text": "Before announcing the winner, the judge thanked the volunteers who had worked through the storm to repair the damaged stage.",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_020.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Before announcing the winner, the judge thanked the volunteers who had worked through the storm to repair the damaged stage.",
- "raw": {
- "normalized_reference": "before announcing the winner, the judge thanked the volunteers who had worked through the storm to repair the damaged stage.",
- "normalized_hypothesis": "before announcing the winner, the judge thanked the volunteers who had worked through the storm to repair the damaged stage.",
- "errors": 0,
- "words": 20,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "before announcing the winner the judge thanked the volunteers who had worked through the storm to repair the damaged stage",
- "normalized_hypothesis": "before announcing the winner the judge thanked the volunteers who had worked through the storm to repair the damaged stage",
- "errors": 0,
- "words": 20,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_021",
- "category": "names_places",
- "text": "Professor Nguyen mailed the fluorescent poster to Reykjavik on Thursday.",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_021.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Professor Nguyen mailed the fluorescent poster to Reykjavik on Thursday.",
- "raw": {
- "normalized_reference": "professor nguyen mailed the fluorescent poster to reykjavik on thursday.",
- "normalized_hypothesis": "professor nguyen mailed the fluorescent poster to reykjavik on thursday.",
- "errors": 0,
- "words": 10,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "professor nguyen mailed the fluorescent poster to reykjavik on thursday",
- "normalized_hypothesis": "professor nguyen mailed the fluorescent poster to reykjavik on thursday",
- "errors": 0,
- "words": 10,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_022",
- "category": "narrative",
- "text": "The plan was simple. The execution was anything but.",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_022.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "The plan was simple. The execution was anything but.",
- "raw": {
- "normalized_reference": "the plan was simple. the execution was anything but.",
- "normalized_hypothesis": "the plan was simple. the execution was anything but.",
- "errors": 0,
- "words": 9,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "the plan was simple the execution was anything but",
- "normalized_hypothesis": "the plan was simple the execution was anything but",
- "errors": 0,
- "words": 9,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_023",
- "category": "numbers",
- "text": "Her score improved from eighty-two percent to ninety-seven percent in two weeks.",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_023.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Her score improved from 82% to 97% in two weeks.",
- "raw": {
- "normalized_reference": "her score improved from eighty-two percent to ninety-seven percent in two weeks.",
- "normalized_hypothesis": "her score improved from 82% to 97% in two weeks.",
- "errors": 6,
- "words": 14,
- "wer": 0.42857142857142855
- },
- "semantic": {
- "normalized_reference": "her score improved from 82% to 97% in 2 weeks",
- "normalized_hypothesis": "her score improved from 82% to 97% in 2 weeks",
- "errors": 0,
- "words": 10,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_024",
- "category": "punctuation",
- "text": "Why would anyone pack spoons, batteries, and a violin for one short trip?",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_024.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Why would anyone pack spoons, batteries, and a violin for one short trip?",
- "raw": {
- "normalized_reference": "why would anyone pack spoons, batteries, and a violin for one short trip?",
- "normalized_hypothesis": "why would anyone pack spoons, batteries, and a violin for one short trip?",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "why would anyone pack spoons batteries and a violin for one short trip",
- "normalized_hypothesis": "why would anyone pack spoons batteries and a violin for one short trip",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_025",
- "category": "question",
- "text": "How long have you been sitting on this?",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_025.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "How long have you been sitting on this?",
- "raw": {
- "normalized_reference": "how long have you been sitting on this?",
- "normalized_hypothesis": "how long have you been sitting on this?",
- "errors": 0,
- "words": 8,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "how long have you been sitting on this",
- "normalized_hypothesis": "how long have you been sitting on this",
- "errors": 0,
- "words": 8,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_026",
- "category": "rare_words",
- "text": "A mischievous chameleon rested beside the chrysanthemum and the bougainvillea.",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_026.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "A mischievous chameleon rested beside the chrysanthemum and the baugangelia.",
- "raw": {
- "normalized_reference": "a mischievous chameleon rested beside the chrysanthemum and the bougainvillea.",
- "normalized_hypothesis": "a mischievous chameleon rested beside the chrysanthemum and the baugangelia.",
- "errors": 1,
- "words": 10,
- "wer": 0.1
- },
- "semantic": {
- "normalized_reference": "a mischievous chameleon rested beside the chrysanthemum and the bougainvillea",
- "normalized_hypothesis": "a mischievous chameleon rested beside the chrysanthemum and the baugangelia",
- "errors": 1,
- "words": 10,
- "wer": 0.1
- }
- },
- {
- "id": "utmos200_027",
- "category": "short",
- "text": "We don't talk about that.",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_027.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "We don't talk about that.",
- "raw": {
- "normalized_reference": "we don't talk about that.",
- "normalized_hypothesis": "we don't talk about that.",
- "errors": 0,
- "words": 5,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "we do not talk about that",
- "normalized_hypothesis": "we do not talk about that",
- "errors": 0,
- "words": 6,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_028",
- "category": "technical",
- "text": "Hash the UTF eight payload with S H A two fifty-six before opening the checkpoint.",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_028.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Hash the UTF-8 payload with SHA-256 before opening the checkpoint.",
- "raw": {
- "normalized_reference": "hash the utf eight payload with s h a two fifty-six before opening the checkpoint.",
- "normalized_hypothesis": "hash the utf-8 payload with sha-256 before opening the checkpoint.",
- "errors": 7,
- "words": 16,
- "wer": 0.4375
- },
- "semantic": {
- "normalized_reference": "hash the utf 8 payload with s h a 256 before opening the checkpoint",
- "normalized_hypothesis": "hash the utf 8 payload with sha 256 before opening the checkpoint",
- "errors": 3,
- "words": 14,
- "wer": 0.21428571428571427
- }
- },
- {
- "id": "utmos200_029",
- "category": "conversational",
- "text": "I thought I had it all figured out, and then this happened.",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_029.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "I thought I had it all figured out, and then this happened.",
- "raw": {
- "normalized_reference": "i thought i had it all figured out, and then this happened.",
- "normalized_hypothesis": "i thought i had it all figured out, and then this happened.",
- "errors": 0,
- "words": 12,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "i thought i had it all figured out and then this happened",
- "normalized_hypothesis": "i thought i had it all figured out and then this happened",
- "errors": 0,
- "words": 12,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_030",
- "category": "descriptive",
- "text": "The kind of quiet you only get at three in the morning in a city.",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_030.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "The kind of quiet you only get at 3 in the morning in a city.",
- "raw": {
- "normalized_reference": "the kind of quiet you only get at three in the morning in a city.",
- "normalized_hypothesis": "the kind of quiet you only get at 3 in the morning in a city.",
- "errors": 1,
- "words": 15,
- "wer": 0.06666666666666667
- },
- "semantic": {
- "normalized_reference": "the kind of quiet you only get at 3 in the morning in a city",
- "normalized_hypothesis": "the kind of quiet you only get at 3 in the morning in a city",
- "errors": 0,
- "words": 15,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_031",
- "category": "emotional",
- "text": "The joy of it was almost overwhelming, like too much good all at once.",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_031.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "The joy of it was almost overwhelming, like too much good all at once.",
- "raw": {
- "normalized_reference": "the joy of it was almost overwhelming, like too much good all at once.",
- "normalized_hypothesis": "the joy of it was almost overwhelming, like too much good all at once.",
- "errors": 0,
- "words": 14,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "the joy of it was almost overwhelming like too much good all at once",
- "normalized_hypothesis": "the joy of it was almost overwhelming like too much good all at once",
- "errors": 0,
- "words": 14,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_032",
- "category": "homographs",
- "text": "Please close the door while I stand close to the microphone.",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_032.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Please close the door while I stand close to the microphone.",
- "raw": {
- "normalized_reference": "please close the door while i stand close to the microphone.",
- "normalized_hypothesis": "please close the door while i stand close to the microphone.",
- "errors": 0,
- "words": 11,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "please close the door while i stand close to the microphone",
- "normalized_hypothesis": "please close the door while i stand close to the microphone",
- "errors": 0,
- "words": 11,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_033",
- "category": "long",
- "text": "I think the most underrated skill in the world is knowing when to stop explaining yourself and just let people draw their own conclusions.",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_033.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "I think the most underrated skill in the world is knowing when to stop explaining yourself and just let people draw their own conclusions.",
- "raw": {
- "normalized_reference": "i think the most underrated skill in the world is knowing when to stop explaining yourself and just let people draw their own conclusions.",
- "normalized_hypothesis": "i think the most underrated skill in the world is knowing when to stop explaining yourself and just let people draw their own conclusions.",
- "errors": 0,
- "words": 24,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "i think the most underrated skill in the world is knowing when to stop explaining yourself and just let people draw their own conclusions",
- "normalized_hypothesis": "i think the most underrated skill in the world is knowing when to stop explaining yourself and just let people draw their own conclusions",
- "errors": 0,
- "words": 24,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_034",
- "category": "long_range",
- "text": "When the final bus disappeared around the corner, Nadia remembered that her umbrella, notebook, and lunch were still inside the quiet office.",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_034.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "When the final bus disappeared around the corner, Nadia remembered that her umbrella, notebook, and lunch were still inside the quiet office.",
- "raw": {
- "normalized_reference": "when the final bus disappeared around the corner, nadia remembered that her umbrella, notebook, and lunch were still inside the quiet office.",
- "normalized_hypothesis": "when the final bus disappeared around the corner, nadia remembered that her umbrella, notebook, and lunch were still inside the quiet office.",
- "errors": 0,
- "words": 22,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "when the final bus disappeared around the corner nadia remembered that her umbrella notebook and lunch were still inside the quiet office",
- "normalized_hypothesis": "when the final bus disappeared around the corner nadia remembered that her umbrella notebook and lunch were still inside the quiet office",
- "errors": 0,
- "words": 22,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_035",
- "category": "names_places",
- "text": "Aurelia and Joaquin crossed the Schuylkill River before noon.",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_035.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Aurelia and Joaquin crossed the Chiocul River before noon.",
- "raw": {
- "normalized_reference": "aurelia and joaquin crossed the schuylkill river before noon.",
- "normalized_hypothesis": "aurelia and joaquin crossed the chiocul river before noon.",
- "errors": 1,
- "words": 9,
- "wer": 0.1111111111111111
- },
- "semantic": {
- "normalized_reference": "aurelia and joaquin crossed the schuylkill river before noon",
- "normalized_hypothesis": "aurelia and joaquin crossed the chiocul river before noon",
- "errors": 1,
- "words": 9,
- "wer": 0.1111111111111111
- }
- },
- {
- "id": "utmos200_036",
- "category": "narrative",
- "text": "The village had a name, but nobody used it. Everyone just called it home.",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_036.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "The village had a name, but nobody used it. Everyone just called it home.",
- "raw": {
- "normalized_reference": "the village had a name, but nobody used it. everyone just called it home.",
- "normalized_hypothesis": "the village had a name, but nobody used it. everyone just called it home.",
- "errors": 0,
- "words": 14,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "the village had a name but nobody used it everyone just called it home",
- "normalized_hypothesis": "the village had a name but nobody used it everyone just called it home",
- "errors": 0,
- "words": 14,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_037",
- "category": "numbers",
- "text": "Increase the gain by three decibels, wait five seconds, then reduce it by one.",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_037.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Increase the gain by 3 decibels, wait 5 seconds, then reduce it by 1.",
- "raw": {
- "normalized_reference": "increase the gain by three decibels, wait five seconds, then reduce it by one.",
- "normalized_hypothesis": "increase the gain by 3 decibels, wait 5 seconds, then reduce it by 1.",
- "errors": 3,
- "words": 14,
- "wer": 0.21428571428571427
- },
- "semantic": {
- "normalized_reference": "increase the gain by 3 decibels wait 5 seconds then reduce it by one",
- "normalized_hypothesis": "increase the gain by 3 decibels wait 5 seconds then reduce it by one",
- "errors": 0,
- "words": 14,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_038",
- "category": "punctuation",
- "text": "The result was unexpected: clear, stable, and surprisingly natural.",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_038.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "The result was unexpected, clear, stable, and surprisingly natural.",
- "raw": {
- "normalized_reference": "the result was unexpected: clear, stable, and surprisingly natural.",
- "normalized_hypothesis": "the result was unexpected, clear, stable, and surprisingly natural.",
- "errors": 0,
- "words": 9,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "the result was unexpected clear stable and surprisingly natural",
- "normalized_hypothesis": "the result was unexpected clear stable and surprisingly natural",
- "errors": 0,
- "words": 9,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_039",
- "category": "question",
- "text": "Who told you that was the right way to do it?",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_039.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Who told you that was the right way to do it?",
- "raw": {
- "normalized_reference": "who told you that was the right way to do it?",
- "normalized_hypothesis": "who told you that was the right way to do it?",
- "errors": 0,
- "words": 11,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "who told you that was the right way to do it",
- "normalized_hypothesis": "who told you that was the right way to do it",
- "errors": 0,
- "words": 11,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_040",
- "category": "rare_words",
- "text": "The meticulous cartographer examined an iridescent archipelago through binoculars.",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_040.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "The meticulous cartographer examined an iridescent archipelago through binoculars.",
- "raw": {
- "normalized_reference": "the meticulous cartographer examined an iridescent archipelago through binoculars.",
- "normalized_hypothesis": "the meticulous cartographer examined an iridescent archipelago through binoculars.",
- "errors": 0,
- "words": 9,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "the meticulous cartographer examined an iridescent archipelago through binoculars",
- "normalized_hypothesis": "the meticulous cartographer examined an iridescent archipelago through binoculars",
- "errors": 0,
- "words": 9,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_041",
- "category": "short",
- "text": "I mean. Yeah. Kind of.",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_041.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "I mean, no, I know.",
- "raw": {
- "normalized_reference": "i mean. yeah. kind of.",
- "normalized_hypothesis": "i mean, no, i know.",
- "errors": 3,
- "words": 5,
- "wer": 0.6
- },
- "semantic": {
- "normalized_reference": "i mean yeah kind of",
- "normalized_hypothesis": "i mean no i know",
- "errors": 3,
- "words": 5,
- "wer": 0.6
- }
- },
- {
- "id": "utmos200_042",
- "category": "technical",
- "text": "The logs are your best friend when something behaves unexpectedly.",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_042.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "The logs are your best friend when something behaves unexpectedly.",
- "raw": {
- "normalized_reference": "the logs are your best friend when something behaves unexpectedly.",
- "normalized_hypothesis": "the logs are your best friend when something behaves unexpectedly.",
- "errors": 0,
- "words": 10,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "the logs are your best friend when something behaves unexpectedly",
- "normalized_hypothesis": "the logs are your best friend when something behaves unexpectedly",
- "errors": 0,
- "words": 10,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_043",
- "category": "conversational",
- "text": "He kept talking and I kept nodding and neither of us was listening.",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_043.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "He kept talking and I kept nodding and neither of us was listening.",
- "raw": {
- "normalized_reference": "he kept talking and i kept nodding and neither of us was listening.",
- "normalized_hypothesis": "he kept talking and i kept nodding and neither of us was listening.",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "he kept talking and i kept nodding and neither of us was listening",
- "normalized_hypothesis": "he kept talking and i kept nodding and neither of us was listening",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_044",
- "category": "descriptive",
- "text": "It's one of those places where everyone seems to know each other already.",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_044.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "It's one of those places where everyone seems to know each other already.",
- "raw": {
- "normalized_reference": "it's one of those places where everyone seems to know each other already.",
- "normalized_hypothesis": "it's one of those places where everyone seems to know each other already.",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "it is one of those places where everyone seems to know each other already",
- "normalized_hypothesis": "it is one of those places where everyone seems to know each other already",
- "errors": 0,
- "words": 14,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_045",
- "category": "emotional",
- "text": "He said he was proud of me and nobody had ever said that to me before.",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_045.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "He said he was proud of me and nobody had ever said that to me before.",
- "raw": {
- "normalized_reference": "he said he was proud of me and nobody had ever said that to me before.",
- "normalized_hypothesis": "he said he was proud of me and nobody had ever said that to me before.",
- "errors": 0,
- "words": 16,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "he said he was proud of me and nobody had ever said that to me before",
- "normalized_hypothesis": "he said he was proud of me and nobody had ever said that to me before",
- "errors": 0,
- "words": 16,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_046",
- "category": "homographs",
- "text": "The minute detail took a minute to explain clearly.",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_046.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "The minute detail took a minute to explain clearly.",
- "raw": {
- "normalized_reference": "the minute detail took a minute to explain clearly.",
- "normalized_hypothesis": "the minute detail took a minute to explain clearly.",
- "errors": 0,
- "words": 9,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "the minute detail took a minute to explain clearly",
- "normalized_hypothesis": "the minute detail took a minute to explain clearly",
- "errors": 0,
- "words": 9,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_047",
- "category": "long",
- "text": "The strange thing about really wanting something is that getting it never quite feels the way you imagined it would, and not getting it never quite feels as bad.",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_047.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "The strange thing about really wanting something is that getting it never quite feels the way you imagined it would, and not getting it never quite feels as bad.",
- "raw": {
- "normalized_reference": "the strange thing about really wanting something is that getting it never quite feels the way you imagined it would, and not getting it never quite feels as bad.",
- "normalized_hypothesis": "the strange thing about really wanting something is that getting it never quite feels the way you imagined it would, and not getting it never quite feels as bad.",
- "errors": 0,
- "words": 29,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "the strange thing about really wanting something is that getting it never quite feels the way you imagined it would and not getting it never quite feels as bad",
- "normalized_hypothesis": "the strange thing about really wanting something is that getting it never quite feels the way you imagined it would and not getting it never quite feels as bad",
- "errors": 0,
- "words": 29,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_048",
- "category": "long_range",
- "text": "Even though the recipe looked simple at first, preparing the sauce required patience, careful timing, and a much larger pan.",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_048.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Even though the recipe looked simple at first, preparing the sauce required patience, careful timing, and a much larger pan.",
- "raw": {
- "normalized_reference": "even though the recipe looked simple at first, preparing the sauce required patience, careful timing, and a much larger pan.",
- "normalized_hypothesis": "even though the recipe looked simple at first, preparing the sauce required patience, careful timing, and a much larger pan.",
- "errors": 0,
- "words": 20,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "even though the recipe looked simple at 1st preparing the sauce required patience careful timing and a much larger pan",
- "normalized_hypothesis": "even though the recipe looked simple at 1st preparing the sauce required patience careful timing and a much larger pan",
- "errors": 0,
- "words": 20,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_049",
- "category": "names_places",
- "text": "Xiomara met Siobhan near the old library in Worcestershire.",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_049.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "See Amaru Mechivan near the Old Library in Worcestershire.",
- "raw": {
- "normalized_reference": "xiomara met siobhan near the old library in worcestershire.",
- "normalized_hypothesis": "see amaru mechivan near the old library in worcestershire.",
- "errors": 3,
- "words": 9,
- "wer": 0.3333333333333333
- },
- "semantic": {
- "normalized_reference": "xiomara met siobhan near the old library in worcestershire",
- "normalized_hypothesis": "see amaru mechivan near the old library in worcestershire",
- "errors": 3,
- "words": 9,
- "wer": 0.3333333333333333
- }
- },
- {
- "id": "utmos200_050",
- "category": "narrative",
- "text": "She walked into the room and everyone went quiet, not because she was loud, but because she wasn't.",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_050.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "She walked into the room and everyone went quiet, not because she was loud, but because she wasn't.",
- "raw": {
- "normalized_reference": "she walked into the room and everyone went quiet, not because she was loud, but because she wasn't.",
- "normalized_hypothesis": "she walked into the room and everyone went quiet, not because she was loud, but because she wasn't.",
- "errors": 0,
- "words": 18,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "she walked into the room and everyone went quiet not because she was loud but because she was not",
- "normalized_hypothesis": "she walked into the room and everyone went quiet not because she was loud but because she was not",
- "errors": 0,
- "words": 19,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_051",
- "category": "numbers",
- "text": "At exactly seven forty-five, Morgan placed eighteen dollars and fifty cents beside the notebook.",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_051.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "At exactly 7.45, Morgan placed $18.50 beside the notebook.",
- "raw": {
- "normalized_reference": "at exactly seven forty-five, morgan placed eighteen dollars and fifty cents beside the notebook.",
- "normalized_hypothesis": "at exactly 7.45, morgan placed $18.50 beside the notebook.",
- "errors": 8,
- "words": 15,
- "wer": 0.5333333333333333
- },
- "semantic": {
- "normalized_reference": "at exactly 745 morgan placed $18.50 beside the notebook",
- "normalized_hypothesis": "at exactly 7.45 morgan placed $18.50 beside the notebook",
- "errors": 2,
- "words": 10,
- "wer": 0.2
- }
- },
- {
- "id": "utmos200_052",
- "category": "punctuation",
- "text": "Does punctuation work? Yes: commas, semicolons, and periods should create sensible boundaries.",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_052.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Does punctuation work? Yes. Commas, semicolons, and periods should create sensible boundaries.",
- "raw": {
- "normalized_reference": "does punctuation work? yes: commas, semicolons, and periods should create sensible boundaries.",
- "normalized_hypothesis": "does punctuation work? yes. commas, semicolons, and periods should create sensible boundaries.",
- "errors": 0,
- "words": 12,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "does punctuation work yes commas semicolons and periods should create sensible boundaries",
- "normalized_hypothesis": "does punctuation work yes commas semicolons and periods should create sensible boundaries",
- "errors": 0,
- "words": 12,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_053",
- "category": "question",
- "text": "Have you ever just stopped and thought about how strange all of this is?",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_053.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Have you ever just stopped and thought about how strange all of this is?",
- "raw": {
- "normalized_reference": "have you ever just stopped and thought about how strange all of this is?",
- "normalized_hypothesis": "have you ever just stopped and thought about how strange all of this is?",
- "errors": 0,
- "words": 14,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "have you ever just stopped and thought about how strange all of this is",
- "normalized_hypothesis": "have you ever just stopped and thought about how strange all of this is",
- "errors": 0,
- "words": 14,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_054",
- "category": "rare_words",
- "text": "Her idiosyncratic explanation sounded plausible but remained fundamentally ambiguous.",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_054.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Her idiosyncratic explanation sounded plausible but remained fundamentally ambiguous.",
- "raw": {
- "normalized_reference": "her idiosyncratic explanation sounded plausible but remained fundamentally ambiguous.",
- "normalized_hypothesis": "her idiosyncratic explanation sounded plausible but remained fundamentally ambiguous.",
- "errors": 0,
- "words": 9,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "her idiosyncratic explanation sounded plausible but remained fundamentally ambiguous",
- "normalized_hypothesis": "her idiosyncratic explanation sounded plausible but remained fundamentally ambiguous",
- "errors": 0,
- "words": 9,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_055",
- "category": "short",
- "text": "That tracks.",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_055.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "That tracks.",
- "raw": {
- "normalized_reference": "that tracks.",
- "normalized_hypothesis": "that tracks.",
- "errors": 0,
- "words": 2,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "that tracks",
- "normalized_hypothesis": "that tracks",
- "errors": 0,
- "words": 2,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_056",
- "category": "technical",
- "text": "The key thing to remember is that order matters here more than you'd think.",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_056.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "The key thing to remember is that order matters here more than you'd think.",
- "raw": {
- "normalized_reference": "the key thing to remember is that order matters here more than you'd think.",
- "normalized_hypothesis": "the key thing to remember is that order matters here more than you'd think.",
- "errors": 0,
- "words": 14,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "the key thing to remember is that order matters here more than you would think",
- "normalized_hypothesis": "the key thing to remember is that order matters here more than you would think",
- "errors": 0,
- "words": 15,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_057",
- "category": "conversational",
- "text": "I would not say that again in this room if I were you.",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_057.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "I would not say that again in this room if I were you.",
- "raw": {
- "normalized_reference": "i would not say that again in this room if i were you.",
- "normalized_hypothesis": "i would not say that again in this room if i were you.",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "i would not say that again in this room if i were you",
- "normalized_hypothesis": "i would not say that again in this room if i were you",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_058",
- "category": "descriptive",
- "text": "The room was filled with the quiet that means something is about to change.",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_058.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "The room was filled with a quiet that means something is about to change.",
- "raw": {
- "normalized_reference": "the room was filled with the quiet that means something is about to change.",
- "normalized_hypothesis": "the room was filled with a quiet that means something is about to change.",
- "errors": 1,
- "words": 14,
- "wer": 0.07142857142857142
- },
- "semantic": {
- "normalized_reference": "the room was filled with the quiet that means something is about to change",
- "normalized_hypothesis": "the room was filled with a quiet that means something is about to change",
- "errors": 1,
- "words": 14,
- "wer": 0.07142857142857142
- }
- },
- {
- "id": "utmos200_059",
- "category": "emotional",
- "text": "I didn't expect to feel that way about it, but I did.",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_059.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "I didn't expect to feel that way about it, but I did.",
- "raw": {
- "normalized_reference": "i didn't expect to feel that way about it, but i did.",
- "normalized_hypothesis": "i didn't expect to feel that way about it, but i did.",
- "errors": 0,
- "words": 12,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "i did not expect to feel that way about it but i did",
- "normalized_hypothesis": "i did not expect to feel that way about it but i did",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_060",
- "category": "homographs",
- "text": "I object to the strange object that was left near the entrance.",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_060.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "I object to the strange object that was left near the entrance.",
- "raw": {
- "normalized_reference": "i object to the strange object that was left near the entrance.",
- "normalized_hypothesis": "i object to the strange object that was left near the entrance.",
- "errors": 0,
- "words": 12,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "i object to the strange object that was left near the entrance",
- "normalized_hypothesis": "i object to the strange object that was left near the entrance",
- "errors": 0,
- "words": 12,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_061",
- "category": "long",
- "text": "I've learned more from the things that went wrong than from the things that worked, which is a frustrating way to have to learn things but apparently the only way that stuck.",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_061.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "I've learned more from the things that went wrong than from the things that worked, which is a frustrating way to have to learn things but apparently the only way that stuck.",
- "raw": {
- "normalized_reference": "i've learned more from the things that went wrong than from the things that worked, which is a frustrating way to have to learn things but apparently the only way that stuck.",
- "normalized_hypothesis": "i've learned more from the things that went wrong than from the things that worked, which is a frustrating way to have to learn things but apparently the only way that stuck.",
- "errors": 0,
- "words": 32,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "i have learned more from the things that went wrong than from the things that worked which is a frustrating way to have to learn things but apparently the only way that stuck",
- "normalized_hypothesis": "i have learned more from the things that went wrong than from the things that worked which is a frustrating way to have to learn things but apparently the only way that stuck",
- "errors": 0,
- "words": 33,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_062",
- "category": "long_range",
- "text": "Because every nearby hotel was full, the exhausted hikers followed the narrow road until they found a farmhouse with one empty room.",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_062.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Because every nearby hotel was full, the exhausted hikers followed the narrow road until they found a farmhouse with one empty room.",
- "raw": {
- "normalized_reference": "because every nearby hotel was full, the exhausted hikers followed the narrow road until they found a farmhouse with one empty room.",
- "normalized_hypothesis": "because every nearby hotel was full, the exhausted hikers followed the narrow road until they found a farmhouse with one empty room.",
- "errors": 0,
- "words": 22,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "because every nearby hotel was full the exhausted hikers followed the narrow road until they found a farmhouse with one empty room",
- "normalized_hypothesis": "because every nearby hotel was full the exhausted hikers followed the narrow road until they found a farmhouse with one empty room",
- "errors": 0,
- "words": 22,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_063",
- "category": "names_places",
- "text": "The itinerary sends Beatrice from Albuquerque to Des Moines and then to Boise.",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_063.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "The itinerary sends Beatrice from Albuquerque to Des Moines and then to Boise.",
- "raw": {
- "normalized_reference": "the itinerary sends beatrice from albuquerque to des moines and then to boise.",
- "normalized_hypothesis": "the itinerary sends beatrice from albuquerque to des moines and then to boise.",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "the itinerary sends beatrice from albuquerque to des moines and then to boise",
- "normalized_hypothesis": "the itinerary sends beatrice from albuquerque to des moines and then to boise",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_064",
- "category": "narrative",
- "text": "We had been driving for six hours when we finally saw the sign for the town.",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_064.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "We had been driving for six hours when we finally saw the sign for the town.",
- "raw": {
- "normalized_reference": "we had been driving for six hours when we finally saw the sign for the town.",
- "normalized_hypothesis": "we had been driving for six hours when we finally saw the sign for the town.",
- "errors": 0,
- "words": 16,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "we had been driving for 6 hours when we finally saw the sign for the town",
- "normalized_hypothesis": "we had been driving for 6 hours when we finally saw the sign for the town",
- "errors": 0,
- "words": 16,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_065",
- "category": "numbers",
- "text": "Call extension zero four nine after eight thirty in the morning.",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_065.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Call extension 049 after 8.30 in the morning.",
- "raw": {
- "normalized_reference": "call extension zero four nine after eight thirty in the morning.",
- "normalized_hypothesis": "call extension 049 after 8.30 in the morning.",
- "errors": 5,
- "words": 11,
- "wer": 0.45454545454545453
- },
- "semantic": {
- "normalized_reference": "call extension 049 after 830 in the morning",
- "normalized_hypothesis": "call extension 49 after 8.30 in the morning",
- "errors": 3,
- "words": 8,
- "wer": 0.375
- }
- },
- {
- "id": "utmos200_066",
- "category": "punctuation",
- "text": "Stop - listen carefully - then answer the question without guessing.",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_066.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Stop listen carefully then answer the question without guessing.",
- "raw": {
- "normalized_reference": "stop - listen carefully - then answer the question without guessing.",
- "normalized_hypothesis": "stop listen carefully then answer the question without guessing.",
- "errors": 0,
- "words": 9,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "stop listen carefully then answer the question without guessing",
- "normalized_hypothesis": "stop listen carefully then answer the question without guessing",
- "errors": 0,
- "words": 9,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_067",
- "category": "question",
- "text": "Does it matter at this point, honestly?",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_067.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Does it matter at this point, honestly?",
- "raw": {
- "normalized_reference": "does it matter at this point, honestly?",
- "normalized_hypothesis": "does it matter at this point, honestly?",
- "errors": 0,
- "words": 7,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "does it matter at this point honestly",
- "normalized_hypothesis": "does it matter at this point honestly",
- "errors": 0,
- "words": 7,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_068",
- "category": "rare_words",
- "text": "An entrepreneurial meteorologist described the unprecedented atmospheric phenomenon.",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_068.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "An entrepreneurial meteorologist described the unprecedented atmospheric phenomenon.",
- "raw": {
- "normalized_reference": "an entrepreneurial meteorologist described the unprecedented atmospheric phenomenon.",
- "normalized_hypothesis": "an entrepreneurial meteorologist described the unprecedented atmospheric phenomenon.",
- "errors": 0,
- "words": 8,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "an entrepreneurial meteorologist described the unprecedented atmospheric phenomenon",
- "normalized_hypothesis": "an entrepreneurial meteorologist described the unprecedented atmospheric phenomenon",
- "errors": 0,
- "words": 8,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_069",
- "category": "short",
- "text": "That's not what I expected.",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_069.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "That's not what I expected.",
- "raw": {
- "normalized_reference": "that's not what i expected.",
- "normalized_hypothesis": "that's not what i expected.",
- "errors": 0,
- "words": 5,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "that is not what i expected",
- "normalized_hypothesis": "that is not what i expected",
- "errors": 0,
- "words": 6,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_070",
- "category": "technical",
- "text": "There's a known issue with that version that causes failures under load.",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_070.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "There's a known issue with that version that causes failures under load.",
- "raw": {
- "normalized_reference": "there's a known issue with that version that causes failures under load.",
- "normalized_hypothesis": "there's a known issue with that version that causes failures under load.",
- "errors": 0,
- "words": 12,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "there is a known issue with that version that causes failures under load",
- "normalized_hypothesis": "there is a known issue with that version that causes failures under load",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_071",
- "category": "conversational",
- "text": "Don't look now but I think that's him over there.",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_071.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Don't look now, but I think that's him over there.",
- "raw": {
- "normalized_reference": "don't look now but i think that's him over there.",
- "normalized_hypothesis": "don't look now, but i think that's him over there.",
- "errors": 0,
- "words": 10,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "do not look now but i think that is him over there",
- "normalized_hypothesis": "do not look now but i think that is him over there",
- "errors": 0,
- "words": 12,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_072",
- "category": "descriptive",
- "text": "The city at night looks completely different from how it feels during the day.",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_072.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "The city at night looks completely different from how it feels during the day.",
- "raw": {
- "normalized_reference": "the city at night looks completely different from how it feels during the day.",
- "normalized_hypothesis": "the city at night looks completely different from how it feels during the day.",
- "errors": 0,
- "words": 14,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "the city at night looks completely different from how it feels during the day",
- "normalized_hypothesis": "the city at night looks completely different from how it feels during the day",
- "errors": 0,
- "words": 14,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_073",
- "category": "emotional",
- "text": "It's the kind of happy that makes you nervous because you know it can't last.",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_073.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "It's the kind of happy that makes you nervous because you know it can't last.",
- "raw": {
- "normalized_reference": "it's the kind of happy that makes you nervous because you know it can't last.",
- "normalized_hypothesis": "it's the kind of happy that makes you nervous because you know it can't last.",
- "errors": 0,
- "words": 15,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "it is the kind of happy that makes you nervous because you know it can not last",
- "normalized_hypothesis": "it is the kind of happy that makes you nervous because you know it can not last",
- "errors": 0,
- "words": 17,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_074",
- "category": "homographs",
- "text": "The dove dove beneath the branch while the wind began to wind the loose cable.",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_074.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "The dove dove beneath the branch while the wind began to wind the loose cable.",
- "raw": {
- "normalized_reference": "the dove dove beneath the branch while the wind began to wind the loose cable.",
- "normalized_hypothesis": "the dove dove beneath the branch while the wind began to wind the loose cable.",
- "errors": 0,
- "words": 15,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "the dove dove beneath the branch while the wind began to wind the loose cable",
- "normalized_hypothesis": "the dove dove beneath the branch while the wind began to wind the loose cable",
- "errors": 0,
- "words": 15,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_075",
- "category": "long",
- "text": "I've thought a lot about what I would do differently if I could go back, and the honest answer is probably not as much as I'd like to think.",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_075.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "I've thought a lot about what I would do differently if I could go back, and the honest answer is probably not as much as I'd like to think.",
- "raw": {
- "normalized_reference": "i've thought a lot about what i would do differently if i could go back, and the honest answer is probably not as much as i'd like to think.",
- "normalized_hypothesis": "i've thought a lot about what i would do differently if i could go back, and the honest answer is probably not as much as i'd like to think.",
- "errors": 0,
- "words": 29,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "i have thought a lot about what i would do differently if i could go back and the honest answer is probably not as much as i would like to think",
- "normalized_hypothesis": "i have thought a lot about what i would do differently if i could go back and the honest answer is probably not as much as i would like to think",
- "errors": 0,
- "words": 31,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_076",
- "category": "long_range",
- "text": "Although the weather changed without warning, the musicians finished packing before anyone opened the enormous glass door.",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_076.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Although the weather changed without warning, the musicians finished packing before anyone opened the enormous glass door.",
- "raw": {
- "normalized_reference": "although the weather changed without warning, the musicians finished packing before anyone opened the enormous glass door.",
- "normalized_hypothesis": "although the weather changed without warning, the musicians finished packing before anyone opened the enormous glass door.",
- "errors": 0,
- "words": 17,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "although the weather changed without warning the musicians finished packing before anyone opened the enormous glass door",
- "normalized_hypothesis": "although the weather changed without warning the musicians finished packing before anyone opened the enormous glass door",
- "errors": 0,
- "words": 17,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_077",
- "category": "names_places",
- "text": "Saskatchewan looked unusually bright beneath the pale February sunrise.",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_077.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Saskatchewan looked unusually bright beneath the pale February sunrise.",
- "raw": {
- "normalized_reference": "saskatchewan looked unusually bright beneath the pale february sunrise.",
- "normalized_hypothesis": "saskatchewan looked unusually bright beneath the pale february sunrise.",
- "errors": 0,
- "words": 9,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "saskatchewan looked unusually bright beneath the pale february sunrise",
- "normalized_hypothesis": "saskatchewan looked unusually bright beneath the pale february sunrise",
- "errors": 0,
- "words": 9,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_078",
- "category": "narrative",
- "text": "She handled it with a grace I genuinely did not expect.",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_078.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "She handled it with a grace I genuinely did not expect.",
- "raw": {
- "normalized_reference": "she handled it with a grace i genuinely did not expect.",
- "normalized_hypothesis": "she handled it with a grace i genuinely did not expect.",
- "errors": 0,
- "words": 11,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "she handled it with a grace i genuinely did not expect",
- "normalized_hypothesis": "she handled it with a grace i genuinely did not expect",
- "errors": 0,
- "words": 11,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_079",
- "category": "numbers",
- "text": "Apartment four hundred seven is across from room twenty-nine on the third floor.",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_079.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Apartment 407 is across from room 29 on the third floor.",
- "raw": {
- "normalized_reference": "apartment four hundred seven is across from room twenty-nine on the third floor.",
- "normalized_hypothesis": "apartment 407 is across from room 29 on the third floor.",
- "errors": 5,
- "words": 14,
- "wer": 0.35714285714285715
- },
- "semantic": {
- "normalized_reference": "apartment 407 is across from room 29 on the 3rd floor",
- "normalized_hypothesis": "apartment 407 is across from room 29 on the 3rd floor",
- "errors": 0,
- "words": 11,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_080",
- "category": "punctuation",
- "text": "\"Not yet,\" Rowan whispered. \"Give me another minute... please.\"",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_080.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Not yet, Rowan whispered. Give me another minute, please.",
- "raw": {
- "normalized_reference": "\"not yet,\" rowan whispered. \"give me another minute... please.\"",
- "normalized_hypothesis": "not yet, rowan whispered. give me another minute, please.",
- "errors": 0,
- "words": 9,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "not yet rowan whispered give me another minute please",
- "normalized_hypothesis": "not yet rowan whispered give me another minute please",
- "errors": 0,
- "words": 9,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_081",
- "category": "question",
- "text": "What are we actually trying to accomplish here?",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_081.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "What are we actually trying to accomplish here?",
- "raw": {
- "normalized_reference": "what are we actually trying to accomplish here?",
- "normalized_hypothesis": "what are we actually trying to accomplish here?",
- "errors": 0,
- "words": 8,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "what are we actually trying to accomplish here",
- "normalized_hypothesis": "what are we actually trying to accomplish here",
- "errors": 0,
- "words": 8,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_082",
- "category": "rare_words",
- "text": "The physician recommended a noninvasive procedure for the recurring neurological symptoms.",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_082.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "the physician recommended a non-invasive procedure for the recurring neurological symptoms.",
- "raw": {
- "normalized_reference": "the physician recommended a noninvasive procedure for the recurring neurological symptoms.",
- "normalized_hypothesis": "the physician recommended a non-invasive procedure for the recurring neurological symptoms.",
- "errors": 2,
- "words": 11,
- "wer": 0.18181818181818182
- },
- "semantic": {
- "normalized_reference": "the physician recommended a noninvasive procedure for the recurring neurological symptoms",
- "normalized_hypothesis": "the physician recommended a non invasive procedure for the recurring neurological symptoms",
- "errors": 2,
- "words": 11,
- "wer": 0.18181818181818182
- }
- },
- {
- "id": "utmos200_083",
- "category": "short",
- "text": "Are you serious right now?",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_083.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Are you serious right now?",
- "raw": {
- "normalized_reference": "are you serious right now?",
- "normalized_hypothesis": "are you serious right now?",
- "errors": 0,
- "words": 5,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "are you serious right now",
- "normalized_hypothesis": "are you serious right now",
- "errors": 0,
- "words": 5,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_084",
- "category": "technical",
- "text": "The performance improvement is real, but it comes with tradeoffs.",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_084.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "The performance improvement is real, but it comes with trade-offs.",
- "raw": {
- "normalized_reference": "the performance improvement is real, but it comes with tradeoffs.",
- "normalized_hypothesis": "the performance improvement is real, but it comes with trade-offs.",
- "errors": 2,
- "words": 10,
- "wer": 0.2
- },
- "semantic": {
- "normalized_reference": "the performance improvement is real but it comes with tradeoffs",
- "normalized_hypothesis": "the performance improvement is real but it comes with trade offs",
- "errors": 2,
- "words": 10,
- "wer": 0.2
- }
- },
- {
- "id": "utmos200_085",
- "category": "conversational",
- "text": "I don't know. It's just been one of those weeks, you know?",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_085.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "I don't know. It's just been one of those weeks, you know?",
- "raw": {
- "normalized_reference": "i don't know. it's just been one of those weeks, you know?",
- "normalized_hypothesis": "i don't know. it's just been one of those weeks, you know?",
- "errors": 0,
- "words": 12,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "i do not know it is just been one of those weeks you know",
- "normalized_hypothesis": "i do not know it is just been one of those weeks you know",
- "errors": 0,
- "words": 14,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_086",
- "category": "descriptive",
- "text": "The moment you stop expecting perfection, things get a lot more interesting.",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_086.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "The moment you stop expecting perfection, things get a lot more interesting.",
- "raw": {
- "normalized_reference": "the moment you stop expecting perfection, things get a lot more interesting.",
- "normalized_hypothesis": "the moment you stop expecting perfection, things get a lot more interesting.",
- "errors": 0,
- "words": 12,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "the moment you stop expecting perfection things get a lot more interesting",
- "normalized_hypothesis": "the moment you stop expecting perfection things get a lot more interesting",
- "errors": 0,
- "words": 12,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_087",
- "category": "emotional",
- "text": "It's the kind of thing that stays with you even years later.",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_087.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "It's the kind of thing that stays with you even years later.",
- "raw": {
- "normalized_reference": "it's the kind of thing that stays with you even years later.",
- "normalized_hypothesis": "it's the kind of thing that stays with you even years later.",
- "errors": 0,
- "words": 12,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "it is the kind of thing that stays with you even years later",
- "normalized_hypothesis": "it is the kind of thing that stays with you even years later",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_088",
- "category": "long",
- "text": "She told me once that the most important skill she'd developed was learning to disagree with someone without making them feel they were wrong.",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_088.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "She told me once that the most important skill she developed was learning to disagree with someone without making them feel they were wrong.",
- "raw": {
- "normalized_reference": "she told me once that the most important skill she'd developed was learning to disagree with someone without making them feel they were wrong.",
- "normalized_hypothesis": "she told me once that the most important skill she developed was learning to disagree with someone without making them feel they were wrong.",
- "errors": 1,
- "words": 24,
- "wer": 0.041666666666666664
- },
- "semantic": {
- "normalized_reference": "she told me once that the most important skill she would developed was learning to disagree with someone without making them feel they were wrong",
- "normalized_hypothesis": "she told me once that the most important skill she developed was learning to disagree with someone without making them feel they were wrong",
- "errors": 1,
- "words": 25,
- "wer": 0.04
- }
- },
- {
- "id": "utmos200_089",
- "category": "narrative",
- "text": "He stood at the edge of the cliff and wondered, not for the first time, how he'd gotten here.",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_089.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "He stood at the edge of the cliff and wondered, not for the first time, how he'd gotten here.",
- "raw": {
- "normalized_reference": "he stood at the edge of the cliff and wondered, not for the first time, how he'd gotten here.",
- "normalized_hypothesis": "he stood at the edge of the cliff and wondered, not for the first time, how he'd gotten here.",
- "errors": 0,
- "words": 19,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "he stood at the edge of the cliff and wondered not for the 1st time how he would gotten here",
- "normalized_hypothesis": "he stood at the edge of the cliff and wondered not for the 1st time how he would gotten here",
- "errors": 0,
- "words": 20,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_090",
- "category": "question",
- "text": "Why does asking for help feel so hard when you actually need it?",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_090.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Why does asking for help feel so hard when you actually need it?",
- "raw": {
- "normalized_reference": "why does asking for help feel so hard when you actually need it?",
- "normalized_hypothesis": "why does asking for help feel so hard when you actually need it?",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "why does asking for help feel so hard when you actually need it",
- "normalized_hypothesis": "why does asking for help feel so hard when you actually need it",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_091",
- "category": "short",
- "text": "That's fair.",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_091.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "That was fair.",
- "raw": {
- "normalized_reference": "that's fair.",
- "normalized_hypothesis": "that was fair.",
- "errors": 2,
- "words": 2,
- "wer": 1.0
- },
- "semantic": {
- "normalized_reference": "that is fair",
- "normalized_hypothesis": "that was fair",
- "errors": 1,
- "words": 3,
- "wer": 0.3333333333333333
- }
- },
- {
- "id": "utmos200_092",
- "category": "technical",
- "text": "If you run into that error, the fix is usually just restarting the service.",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_092.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "If you run into that error, the fix is usually just restarting the service.",
- "raw": {
- "normalized_reference": "if you run into that error, the fix is usually just restarting the service.",
- "normalized_hypothesis": "if you run into that error, the fix is usually just restarting the service.",
- "errors": 0,
- "words": 14,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "if you run into that error the fix is usually just restarting the service",
- "normalized_hypothesis": "if you run into that error the fix is usually just restarting the service",
- "errors": 0,
- "words": 14,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_093",
- "category": "conversational",
- "text": "It was a disaster in the best possible way.",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_093.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "It was a disaster in the best possible way.",
- "raw": {
- "normalized_reference": "it was a disaster in the best possible way.",
- "normalized_hypothesis": "it was a disaster in the best possible way.",
- "errors": 0,
- "words": 9,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "it was a disaster in the best possible way",
- "normalized_hypothesis": "it was a disaster in the best possible way",
- "errors": 0,
- "words": 9,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_094",
- "category": "descriptive",
- "text": "It had the look of a place that used to mean something to someone.",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_094.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "It had the look of a place that used to mean something to someone.",
- "raw": {
- "normalized_reference": "it had the look of a place that used to mean something to someone.",
- "normalized_hypothesis": "it had the look of a place that used to mean something to someone.",
- "errors": 0,
- "words": 14,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "it had the look of a place that used to mean something to someone",
- "normalized_hypothesis": "it had the look of a place that used to mean something to someone",
- "errors": 0,
- "words": 14,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_095",
- "category": "emotional",
- "text": "There's this feeling you get when something finally works after weeks of failure.",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_095.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "There's this feeling you get when something finally works after weeks of failure.",
- "raw": {
- "normalized_reference": "there's this feeling you get when something finally works after weeks of failure.",
- "normalized_hypothesis": "there's this feeling you get when something finally works after weeks of failure.",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "there is this feeling you get when something finally works after weeks of failure",
- "normalized_hypothesis": "there is this feeling you get when something finally works after weeks of failure",
- "errors": 0,
- "words": 14,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_096",
- "category": "long",
- "text": "He had a theory that most arguments aren't actually about what they seem to be about, and the longer I live the more I think he was onto something.",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_096.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "He had a theory that most arguments aren't actually about what they seem to be about, and the longer I live the more I think he was onto something.",
- "raw": {
- "normalized_reference": "he had a theory that most arguments aren't actually about what they seem to be about, and the longer i live the more i think he was onto something.",
- "normalized_hypothesis": "he had a theory that most arguments aren't actually about what they seem to be about, and the longer i live the more i think he was onto something.",
- "errors": 0,
- "words": 29,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "he had a theory that most arguments are not actually about what they seem to be about and the longer i live the more i think he was onto something",
- "normalized_hypothesis": "he had a theory that most arguments are not actually about what they seem to be about and the longer i live the more i think he was onto something",
- "errors": 0,
- "words": 30,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_097",
- "category": "narrative",
- "text": "The thing nobody tells you about starting over is how ordinary most of it feels.",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_097.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "The thing nobody tells you about starting over is how ordinary most of it feels.",
- "raw": {
- "normalized_reference": "the thing nobody tells you about starting over is how ordinary most of it feels.",
- "normalized_hypothesis": "the thing nobody tells you about starting over is how ordinary most of it feels.",
- "errors": 0,
- "words": 15,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "the thing nobody tells you about starting over is how ordinary most of it feels",
- "normalized_hypothesis": "the thing nobody tells you about starting over is how ordinary most of it feels",
- "errors": 0,
- "words": 15,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_098",
- "category": "question",
- "text": "What exactly are we waiting for at this point?",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_098.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "What exactly are we waiting for at this point?",
- "raw": {
- "normalized_reference": "what exactly are we waiting for at this point?",
- "normalized_hypothesis": "what exactly are we waiting for at this point?",
- "errors": 0,
- "words": 9,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "what exactly are we waiting for at this point",
- "normalized_hypothesis": "what exactly are we waiting for at this point",
- "errors": 0,
- "words": 9,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_099",
- "category": "short",
- "text": "I stand by what I said.",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_099.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "I stand by what I said.",
- "raw": {
- "normalized_reference": "i stand by what i said.",
- "normalized_hypothesis": "i stand by what i said.",
- "errors": 0,
- "words": 6,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "i stand by what i said",
- "normalized_hypothesis": "i stand by what i said",
- "errors": 0,
- "words": 6,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_100",
- "category": "technical",
- "text": "Double-check that value before you push anything to production.",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_100.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Double check that value before you push anything to production.",
- "raw": {
- "normalized_reference": "double-check that value before you push anything to production.",
- "normalized_hypothesis": "double check that value before you push anything to production.",
- "errors": 0,
- "words": 10,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "double check that value before you push anything to production",
- "normalized_hypothesis": "double check that value before you push anything to production",
- "errors": 0,
- "words": 10,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_101",
- "category": "conversational",
- "text": "I understand your point; I just do not agree with the conclusion.",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_101.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "I understand your point, I just do not agree with the conclusion.",
- "raw": {
- "normalized_reference": "i understand your point; i just do not agree with the conclusion.",
- "normalized_hypothesis": "i understand your point, i just do not agree with the conclusion.",
- "errors": 0,
- "words": 12,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "i understand your point i just do not agree with the conclusion",
- "normalized_hypothesis": "i understand your point i just do not agree with the conclusion",
- "errors": 0,
- "words": 12,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_102",
- "category": "descriptive",
- "text": "The sky that day was a strange shade of orange I'd never seen before.",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_102.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "The sky that day was a strange shade of orange I'd never seen before.",
- "raw": {
- "normalized_reference": "the sky that day was a strange shade of orange i'd never seen before.",
- "normalized_hypothesis": "the sky that day was a strange shade of orange i'd never seen before.",
- "errors": 0,
- "words": 14,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "the sky that day was a strange shade of orange i would never seen before",
- "normalized_hypothesis": "the sky that day was a strange shade of orange i would never seen before",
- "errors": 0,
- "words": 15,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_103",
- "category": "emotional",
- "text": "It hurt more than I expected, and I thought I was prepared.",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_103.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "It hurt more than I expected, and I thought I was prepared.",
- "raw": {
- "normalized_reference": "it hurt more than i expected, and i thought i was prepared.",
- "normalized_hypothesis": "it hurt more than i expected, and i thought i was prepared.",
- "errors": 0,
- "words": 12,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "it hurt more than i expected and i thought i was prepared",
- "normalized_hypothesis": "it hurt more than i expected and i thought i was prepared",
- "errors": 0,
- "words": 12,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_104",
- "category": "long",
- "text": "The hardest part wasn't making the decision, it was sitting with the uncertainty of not knowing whether it was the right one for months afterward.",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_104.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "The hardest part was in making the decision, it was sitting with the uncertainty of not knowing whether it was the right one for months afterward.",
- "raw": {
- "normalized_reference": "the hardest part wasn't making the decision, it was sitting with the uncertainty of not knowing whether it was the right one for months afterward.",
- "normalized_hypothesis": "the hardest part was in making the decision, it was sitting with the uncertainty of not knowing whether it was the right one for months afterward.",
- "errors": 2,
- "words": 25,
- "wer": 0.08
- },
- "semantic": {
- "normalized_reference": "the hardest part was not making the decision it was sitting with the uncertainty of not knowing whether it was the right one for months afterward",
- "normalized_hypothesis": "the hardest part was in making the decision it was sitting with the uncertainty of not knowing whether it was the right one for months afterward",
- "errors": 1,
- "words": 26,
- "wer": 0.038461538461538464
- }
- },
- {
- "id": "utmos200_105",
- "category": "narrative",
- "text": "There was this one day last summer when everything just clicked into place.",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_105.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "There was this one day last summer when everything just clicked into place.",
- "raw": {
- "normalized_reference": "there was this one day last summer when everything just clicked into place.",
- "normalized_hypothesis": "there was this one day last summer when everything just clicked into place.",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "there was this one day last summer when everything just clicked into place",
- "normalized_hypothesis": "there was this one day last summer when everything just clicked into place",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_106",
- "category": "question",
- "text": "Have you ever thought about why we keep doing things the same way when they stop working?",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_106.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Have you ever thought about why we keep doing things the same way when they stop working?",
- "raw": {
- "normalized_reference": "have you ever thought about why we keep doing things the same way when they stop working?",
- "normalized_hypothesis": "have you ever thought about why we keep doing things the same way when they stop working?",
- "errors": 0,
- "words": 17,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "have you ever thought about why we keep doing things the same way when they stop working",
- "normalized_hypothesis": "have you ever thought about why we keep doing things the same way when they stop working",
- "errors": 0,
- "words": 17,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_107",
- "category": "short",
- "text": "I can't believe you just said that.",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_107.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "I can't believe you just said that.",
- "raw": {
- "normalized_reference": "i can't believe you just said that.",
- "normalized_hypothesis": "i can't believe you just said that.",
- "errors": 0,
- "words": 7,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "i can not believe you just said that",
- "normalized_hypothesis": "i can not believe you just said that",
- "errors": 0,
- "words": 8,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_108",
- "category": "technical",
- "text": "You'll want to isolate that variable before drawing any conclusions.",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_108.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "You'll want to isolate that variable before drawing any conclusions.",
- "raw": {
- "normalized_reference": "you'll want to isolate that variable before drawing any conclusions.",
- "normalized_hypothesis": "you'll want to isolate that variable before drawing any conclusions.",
- "errors": 0,
- "words": 10,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "you will want to isolate that variable before drawing any conclusions",
- "normalized_hypothesis": "you will want to isolate that variable before drawing any conclusions",
- "errors": 0,
- "words": 11,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_109",
- "category": "conversational",
- "text": "The meeting that could have been an email lasted two hours.",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_109.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "The meeting that could have been an email lasted two hours.",
- "raw": {
- "normalized_reference": "the meeting that could have been an email lasted two hours.",
- "normalized_hypothesis": "the meeting that could have been an email lasted two hours.",
- "errors": 0,
- "words": 11,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "the meeting that could have been an email lasted 2 hours",
- "normalized_hypothesis": "the meeting that could have been an email lasted 2 hours",
- "errors": 0,
- "words": 11,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_110",
- "category": "descriptive",
- "text": "There's a specific kind of tired that sleep doesn't fix.",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_110.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "There's a specific kind of tired that sleep doesn't fix.",
- "raw": {
- "normalized_reference": "there's a specific kind of tired that sleep doesn't fix.",
- "normalized_hypothesis": "there's a specific kind of tired that sleep doesn't fix.",
- "errors": 0,
- "words": 10,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "there is a specific kind of tired that sleep does not fix",
- "normalized_hypothesis": "there is a specific kind of tired that sleep does not fix",
- "errors": 0,
- "words": 12,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_111",
- "category": "emotional",
- "text": "Something about that moment broke something open in me that I hadn't realized was closed.",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_111.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Something about that moment broke something open in me that I hadn't realized was closed.",
- "raw": {
- "normalized_reference": "something about that moment broke something open in me that i hadn't realized was closed.",
- "normalized_hypothesis": "something about that moment broke something open in me that i hadn't realized was closed.",
- "errors": 0,
- "words": 15,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "something about that moment broke something open in me that i had not realized was closed",
- "normalized_hypothesis": "something about that moment broke something open in me that i had not realized was closed",
- "errors": 0,
- "words": 16,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_112",
- "category": "long",
- "text": "The thing about making something good is that you don't usually know it's good while you're making it. You just know you're not done yet.",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_112.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "The thing about making something good is that you don't usually know it's good while you're making it. You just know you're not done yet.",
- "raw": {
- "normalized_reference": "the thing about making something good is that you don't usually know it's good while you're making it. you just know you're not done yet.",
- "normalized_hypothesis": "the thing about making something good is that you don't usually know it's good while you're making it. you just know you're not done yet.",
- "errors": 0,
- "words": 25,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "the thing about making something good is that you do not usually know it is good while you are making it you just know you are not done yet",
- "normalized_hypothesis": "the thing about making something good is that you do not usually know it is good while you are making it you just know you are not done yet",
- "errors": 0,
- "words": 29,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_113",
- "category": "narrative",
- "text": "She opened the door slowly, as if she expected to find something she wasn't ready for.",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_113.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "She opened the door slowly, as if she expected to find something she wasn't ready for.",
- "raw": {
- "normalized_reference": "she opened the door slowly, as if she expected to find something she wasn't ready for.",
- "normalized_hypothesis": "she opened the door slowly, as if she expected to find something she wasn't ready for.",
- "errors": 0,
- "words": 16,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "she opened the door slowly as if she expected to find something she was not ready for",
- "normalized_hypothesis": "she opened the door slowly as if she expected to find something she was not ready for",
- "errors": 0,
- "words": 17,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_114",
- "category": "question",
- "text": "What would you do differently if you could go back?",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_114.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "What would you do differently if you could go back?",
- "raw": {
- "normalized_reference": "what would you do differently if you could go back?",
- "normalized_hypothesis": "what would you do differently if you could go back?",
- "errors": 0,
- "words": 10,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "what would you do differently if you could go back",
- "normalized_hypothesis": "what would you do differently if you could go back",
- "errors": 0,
- "words": 10,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_115",
- "category": "short",
- "text": "Okay. Okay okay okay.",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_115.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Okay, okay, okay, okay.",
- "raw": {
- "normalized_reference": "okay. okay okay okay.",
- "normalized_hypothesis": "okay, okay, okay, okay.",
- "errors": 0,
- "words": 4,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "okay okay okay okay",
- "normalized_hypothesis": "okay okay okay okay",
- "errors": 0,
- "words": 4,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_116",
- "category": "technical",
- "text": "The latency goes up significantly at scale, which is expected.",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_116.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "The latency goes up significantly at scale, which is expected.",
- "raw": {
- "normalized_reference": "the latency goes up significantly at scale, which is expected.",
- "normalized_hypothesis": "the latency goes up significantly at scale, which is expected.",
- "errors": 0,
- "words": 10,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "the latency goes up significantly at scale which is expected",
- "normalized_hypothesis": "the latency goes up significantly at scale which is expected",
- "errors": 0,
- "words": 10,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_117",
- "category": "conversational",
- "text": "We ended up staying until two in the morning talking about nothing.",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_117.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "We ended up staying until 2 in the morning talking about nothing.",
- "raw": {
- "normalized_reference": "we ended up staying until two in the morning talking about nothing.",
- "normalized_hypothesis": "we ended up staying until 2 in the morning talking about nothing.",
- "errors": 1,
- "words": 12,
- "wer": 0.08333333333333333
- },
- "semantic": {
- "normalized_reference": "we ended up staying until 2 in the morning talking about nothing",
- "normalized_hypothesis": "we ended up staying until 2 in the morning talking about nothing",
- "errors": 0,
- "words": 12,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_118",
- "category": "descriptive",
- "text": "The rain started slowly, and then all at once, the way it usually does.",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_118.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "The rain started slowly and then all at once, the way it usually does.",
- "raw": {
- "normalized_reference": "the rain started slowly, and then all at once, the way it usually does.",
- "normalized_hypothesis": "the rain started slowly and then all at once, the way it usually does.",
- "errors": 0,
- "words": 14,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "the rain started slowly and then all at once the way it usually does",
- "normalized_hypothesis": "the rain started slowly and then all at once the way it usually does",
- "errors": 0,
- "words": 14,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_119",
- "category": "emotional",
- "text": "I honestly did not think I was going to make it through that year.",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_119.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "I honestly did not think I was going to make it through that year.",
- "raw": {
- "normalized_reference": "i honestly did not think i was going to make it through that year.",
- "normalized_hypothesis": "i honestly did not think i was going to make it through that year.",
- "errors": 0,
- "words": 14,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "i honestly did not think i was going to make it through that year",
- "normalized_hypothesis": "i honestly did not think i was going to make it through that year",
- "errors": 0,
- "words": 14,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_120",
- "category": "long",
- "text": "There's an assumption that people who seem confident all the time actually feel confident, and in my experience that's almost never true.",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_120.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "There's an assumption that people who seem confident all the time actually feel confident, and in my experience that's almost never true.",
- "raw": {
- "normalized_reference": "there's an assumption that people who seem confident all the time actually feel confident, and in my experience that's almost never true.",
- "normalized_hypothesis": "there's an assumption that people who seem confident all the time actually feel confident, and in my experience that's almost never true.",
- "errors": 0,
- "words": 22,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "there is an assumption that people who seem confident all the time actually feel confident and in my experience that is almost never true",
- "normalized_hypothesis": "there is an assumption that people who seem confident all the time actually feel confident and in my experience that is almost never true",
- "errors": 0,
- "words": 24,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_121",
- "category": "narrative",
- "text": "The moment I opened the door I knew something was different, even if I couldn't say what.",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_121.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "The moment I opened the door I knew something was different, even if I couldn't say what.",
- "raw": {
- "normalized_reference": "the moment i opened the door i knew something was different, even if i couldn't say what.",
- "normalized_hypothesis": "the moment i opened the door i knew something was different, even if i couldn't say what.",
- "errors": 0,
- "words": 17,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "the moment i opened the door i knew something was different even if i could not say what",
- "normalized_hypothesis": "the moment i opened the door i knew something was different even if i could not say what",
- "errors": 0,
- "words": 18,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_122",
- "category": "question",
- "text": "How do you explain something that doesn't make sense even to you?",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_122.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "How do you explain something that doesn't make sense even to you?",
- "raw": {
- "normalized_reference": "how do you explain something that doesn't make sense even to you?",
- "normalized_hypothesis": "how do you explain something that doesn't make sense even to you?",
- "errors": 0,
- "words": 12,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "how do you explain something that does not make sense even to you",
- "normalized_hypothesis": "how do you explain something that does not make sense even to you",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_123",
- "category": "short",
- "text": "Hold on a second.",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_123.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Hold on a second.",
- "raw": {
- "normalized_reference": "hold on a second.",
- "normalized_hypothesis": "hold on a second.",
- "errors": 0,
- "words": 4,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "hold on a 2nd",
- "normalized_hypothesis": "hold on a 2nd",
- "errors": 0,
- "words": 4,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_124",
- "category": "technical",
- "text": "A deterministic split prevents duplicate normalized text from leaking into evaluation.",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_124.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "A deterministic split prevents duplicate normalized hex from leaking into evaluation.",
- "raw": {
- "normalized_reference": "a deterministic split prevents duplicate normalized text from leaking into evaluation.",
- "normalized_hypothesis": "a deterministic split prevents duplicate normalized hex from leaking into evaluation.",
- "errors": 1,
- "words": 11,
- "wer": 0.09090909090909091
- },
- "semantic": {
- "normalized_reference": "a deterministic split prevents duplicate normalized text from leaking into evaluation",
- "normalized_hypothesis": "a deterministic split prevents duplicate normalized hex from leaking into evaluation",
- "errors": 1,
- "words": 11,
- "wer": 0.09090909090909091
- }
- },
- {
- "id": "utmos200_125",
- "category": "conversational",
- "text": "Honestly, that was much better than I expected, but the ending still felt rushed.",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_125.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Honestly, that was much better than I expected, but the ending still felt rushed.",
- "raw": {
- "normalized_reference": "honestly, that was much better than i expected, but the ending still felt rushed.",
- "normalized_hypothesis": "honestly, that was much better than i expected, but the ending still felt rushed.",
- "errors": 0,
- "words": 14,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "honestly that was much better than i expected but the ending still felt rushed",
- "normalized_hypothesis": "honestly that was much better than i expected but the ending still felt rushed",
- "errors": 0,
- "words": 14,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_126",
- "category": "descriptive",
- "text": "She had a habit of tilting her head slightly when thinking hard about something.",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_126.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "She had a habit of tilting her head slightly when thinking hard about something.",
- "raw": {
- "normalized_reference": "she had a habit of tilting her head slightly when thinking hard about something.",
- "normalized_hypothesis": "she had a habit of tilting her head slightly when thinking hard about something.",
- "errors": 0,
- "words": 14,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "she had a habit of tilting her head slightly when thinking hard about something",
- "normalized_hypothesis": "she had a habit of tilting her head slightly when thinking hard about something",
- "errors": 0,
- "words": 14,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_127",
- "category": "emotional",
- "text": "I cannot stress enough how much I did not want to be right about this.",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_127.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "I cannot stress enough how much I did not want to be right about this.",
- "raw": {
- "normalized_reference": "i cannot stress enough how much i did not want to be right about this.",
- "normalized_hypothesis": "i cannot stress enough how much i did not want to be right about this.",
- "errors": 0,
- "words": 15,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "i cannot stress enough how much i did not want to be right about this",
- "normalized_hypothesis": "i cannot stress enough how much i did not want to be right about this",
- "errors": 0,
- "words": 15,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_128",
- "category": "long",
- "text": "I used to think the goal was to get to a point where nothing was hard anymore, and now I think that point doesn't exist and I'm not sure I'd want it to.",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_128.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "I used to think the goal was to get to a point where nothing was hard anymore, and now I think that point doesn't exist and I'm not sure I'd want it to.",
- "raw": {
- "normalized_reference": "i used to think the goal was to get to a point where nothing was hard anymore, and now i think that point doesn't exist and i'm not sure i'd want it to.",
- "normalized_hypothesis": "i used to think the goal was to get to a point where nothing was hard anymore, and now i think that point doesn't exist and i'm not sure i'd want it to.",
- "errors": 0,
- "words": 33,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "i used to think the goal was to get to a point where nothing was hard anymore and now i think that point does not exist and i am not sure i would want it to",
- "normalized_hypothesis": "i used to think the goal was to get to a point where nothing was hard anymore and now i think that point does not exist and i am not sure i would want it to",
- "errors": 0,
- "words": 36,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_129",
- "category": "narrative",
- "text": "So this is the part of the story where everything goes completely sideways.",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_129.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "So this is the part of the story where everything goes completely sideways.",
- "raw": {
- "normalized_reference": "so this is the part of the story where everything goes completely sideways.",
- "normalized_hypothesis": "so this is the part of the story where everything goes completely sideways.",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "so this is the part of the story where everything goes completely sideways",
- "normalized_hypothesis": "so this is the part of the story where everything goes completely sideways",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_130",
- "category": "question",
- "text": "Is there a version of this where it actually works out okay?",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_130.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Is there a version of this where it actually works out okay?",
- "raw": {
- "normalized_reference": "is there a version of this where it actually works out okay?",
- "normalized_hypothesis": "is there a version of this where it actually works out okay?",
- "errors": 0,
- "words": 12,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "is there a version of this where it actually works out okay",
- "normalized_hypothesis": "is there a version of this where it actually works out okay",
- "errors": 0,
- "words": 12,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_131",
- "category": "short",
- "text": "Oh no. Oh no no no.",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_131.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Oh no. Oh no no no.",
- "raw": {
- "normalized_reference": "oh no. oh no no no.",
- "normalized_hypothesis": "oh no. oh no no no.",
- "errors": 0,
- "words": 6,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "0 no 0 no no no",
- "normalized_hypothesis": "0 no 0 no no no",
- "errors": 0,
- "words": 6,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_132",
- "category": "technical",
- "text": "Always test on a copy before modifying the original.",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_132.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Always test on a copy before modifying the original.",
- "raw": {
- "normalized_reference": "always test on a copy before modifying the original.",
- "normalized_hypothesis": "always test on a copy before modifying the original.",
- "errors": 0,
- "words": 9,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "always test on a copy before modifying the original",
- "normalized_hypothesis": "always test on a copy before modifying the original",
- "errors": 0,
- "words": 9,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_133",
- "category": "conversational",
- "text": "I wasn't even supposed to be there. That's the wild part.",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_133.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "I wasn't even supposed to be there. That's the wild part.",
- "raw": {
- "normalized_reference": "i wasn't even supposed to be there. that's the wild part.",
- "normalized_hypothesis": "i wasn't even supposed to be there. that's the wild part.",
- "errors": 0,
- "words": 11,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "i was not even supposed to be there that is the wild part",
- "normalized_hypothesis": "i was not even supposed to be there that is the wild part",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_134",
- "category": "descriptive",
- "text": "It's a small town. Not the kind you leave. The kind you leave and come back to.",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_134.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "It's a small town. Not the kind you leave. The kind you leave and come back to.",
- "raw": {
- "normalized_reference": "it's a small town. not the kind you leave. the kind you leave and come back to.",
- "normalized_hypothesis": "it's a small town. not the kind you leave. the kind you leave and come back to.",
- "errors": 0,
- "words": 17,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "it is a small town not the kind you leave the kind you leave and come back to",
- "normalized_hypothesis": "it is a small town not the kind you leave the kind you leave and come back to",
- "errors": 0,
- "words": 18,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_135",
- "category": "emotional",
- "text": "She cried. I cried. We all just sort of stood there crying.",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_135.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "She cried. I cried. We all just sort of stood there crying.",
- "raw": {
- "normalized_reference": "she cried. i cried. we all just sort of stood there crying.",
- "normalized_hypothesis": "she cried. i cried. we all just sort of stood there crying.",
- "errors": 0,
- "words": 12,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "she cried i cried we all just sort of stood there crying",
- "normalized_hypothesis": "she cried i cried we all just sort of stood there crying",
- "errors": 0,
- "words": 12,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_136",
- "category": "long",
- "text": "There's a version of being busy that's actually just a very elaborate way of avoiding the one thing you know you should be doing.",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_136.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "There's a version of being busy that's actually just a very elaborate way of avoiding the one thing you know you should be doing.",
- "raw": {
- "normalized_reference": "there's a version of being busy that's actually just a very elaborate way of avoiding the one thing you know you should be doing.",
- "normalized_hypothesis": "there's a version of being busy that's actually just a very elaborate way of avoiding the one thing you know you should be doing.",
- "errors": 0,
- "words": 24,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "there is a version of being busy that is actually just a very elaborate way of avoiding the one thing you know you should be doing",
- "normalized_hypothesis": "there is a version of being busy that is actually just a very elaborate way of avoiding the one thing you know you should be doing",
- "errors": 0,
- "words": 26,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_137",
- "category": "narrative",
- "text": "The letter arrived on a Tuesday, which felt appropriate somehow.",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_137.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "The letter arrived on a Tuesday, which felt appropriate somehow.",
- "raw": {
- "normalized_reference": "the letter arrived on a tuesday, which felt appropriate somehow.",
- "normalized_hypothesis": "the letter arrived on a tuesday, which felt appropriate somehow.",
- "errors": 0,
- "words": 10,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "the letter arrived on a tuesday which felt appropriate somehow",
- "normalized_hypothesis": "the letter arrived on a tuesday which felt appropriate somehow",
- "errors": 0,
- "words": 10,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_138",
- "category": "question",
- "text": "What exactly did she say when you told her?",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_138.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "What exactly did she say when you told her?",
- "raw": {
- "normalized_reference": "what exactly did she say when you told her?",
- "normalized_hypothesis": "what exactly did she say when you told her?",
- "errors": 0,
- "words": 9,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "what exactly did she say when you told her",
- "normalized_hypothesis": "what exactly did she say when you told her",
- "errors": 0,
- "words": 9,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_139",
- "category": "short",
- "text": "No way. Absolutely not.",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_139.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "No way. Absolutely not.",
- "raw": {
- "normalized_reference": "no way. absolutely not.",
- "normalized_hypothesis": "no way. absolutely not.",
- "errors": 0,
- "words": 4,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "no way absolutely not",
- "normalized_hypothesis": "no way absolutely not",
- "errors": 0,
- "words": 4,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_140",
- "category": "technical",
- "text": "The first thing you want to do is make sure everything is backed up.",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_140.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "The first thing you want to do is make sure everything is backed up.",
- "raw": {
- "normalized_reference": "the first thing you want to do is make sure everything is backed up.",
- "normalized_hypothesis": "the first thing you want to do is make sure everything is backed up.",
- "errors": 0,
- "words": 14,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "the 1st thing you want to do is make sure everything is backed up",
- "normalized_hypothesis": "the 1st thing you want to do is make sure everything is backed up",
- "errors": 0,
- "words": 14,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_141",
- "category": "conversational",
- "text": "I have the most insane idea and you need to just hear me out.",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_141.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "I have the most insane idea and you need to just hear me out.",
- "raw": {
- "normalized_reference": "i have the most insane idea and you need to just hear me out.",
- "normalized_hypothesis": "i have the most insane idea and you need to just hear me out.",
- "errors": 0,
- "words": 14,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "i have the most insane idea and you need to just hear me out",
- "normalized_hypothesis": "i have the most insane idea and you need to just hear me out",
- "errors": 0,
- "words": 14,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_142",
- "category": "descriptive",
- "text": "There's a coffee shop on the corner that's been there longer than anyone can remember.",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_142.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "There's a coffee shop on the corner that's been there longer than anyone can remember.",
- "raw": {
- "normalized_reference": "there's a coffee shop on the corner that's been there longer than anyone can remember.",
- "normalized_hypothesis": "there's a coffee shop on the corner that's been there longer than anyone can remember.",
- "errors": 0,
- "words": 15,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "there is a coffee shop on the corner that has been there longer than anyone can remember",
- "normalized_hypothesis": "there is a coffee shop on the corner that has been there longer than anyone can remember",
- "errors": 0,
- "words": 17,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_143",
- "category": "emotional",
- "text": "I kept it together until I got to the car, and then I completely fell apart.",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_143.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "I kept it together until I got to the car and then I completely fell apart.",
- "raw": {
- "normalized_reference": "i kept it together until i got to the car, and then i completely fell apart.",
- "normalized_hypothesis": "i kept it together until i got to the car and then i completely fell apart.",
- "errors": 0,
- "words": 16,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "i kept it together until i got to the car and then i completely fell apart",
- "normalized_hypothesis": "i kept it together until i got to the car and then i completely fell apart",
- "errors": 0,
- "words": 16,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_144",
- "category": "long",
- "text": "The older I get, the more I think that most of what we call wisdom is just pattern recognition built from having been wrong enough times to start noticing.",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_144.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "The older I get, the more I think that most of what we call wisdom is just pattern recognition built from having been wrong enough times to start noticing.",
- "raw": {
- "normalized_reference": "the older i get, the more i think that most of what we call wisdom is just pattern recognition built from having been wrong enough times to start noticing.",
- "normalized_hypothesis": "the older i get, the more i think that most of what we call wisdom is just pattern recognition built from having been wrong enough times to start noticing.",
- "errors": 0,
- "words": 29,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "the older i get the more i think that most of what we call wisdom is just pattern recognition built from having been wrong enough times to start noticing",
- "normalized_hypothesis": "the older i get the more i think that most of what we call wisdom is just pattern recognition built from having been wrong enough times to start noticing",
- "errors": 0,
- "words": 29,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_145",
- "category": "narrative",
- "text": "I remember the exact moment I realized I had made a huge mistake.",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_145.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "I remember the exact moment I realized I had made a huge mistake.",
- "raw": {
- "normalized_reference": "i remember the exact moment i realized i had made a huge mistake.",
- "normalized_hypothesis": "i remember the exact moment i realized i had made a huge mistake.",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "i remember the exact moment i realized i had made a huge mistake",
- "normalized_hypothesis": "i remember the exact moment i realized i had made a huge mistake",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_146",
- "category": "question",
- "text": "Do you think it would have gone differently if we'd left earlier?",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_146.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Do you think it would have gone differently if we'd left earlier?",
- "raw": {
- "normalized_reference": "do you think it would have gone differently if we'd left earlier?",
- "normalized_hypothesis": "do you think it would have gone differently if we'd left earlier?",
- "errors": 0,
- "words": 12,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "do you think it would have gone differently if we would left earlier",
- "normalized_hypothesis": "do you think it would have gone differently if we would left earlier",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_147",
- "category": "short",
- "text": "You have to be kidding me.",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_147.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "You have to be kidding me.",
- "raw": {
- "normalized_reference": "you have to be kidding me.",
- "normalized_hypothesis": "you have to be kidding me.",
- "errors": 0,
- "words": 6,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "you have to be kidding me",
- "normalized_hypothesis": "you have to be kidding me",
- "errors": 0,
- "words": 6,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_148",
- "category": "technical",
- "text": "If the checksum fails, restart the worker process without deleting the cached tensors.",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_148.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "If the checksum fails, restart the worker process without deleting the cache tensors.",
- "raw": {
- "normalized_reference": "if the checksum fails, restart the worker process without deleting the cached tensors.",
- "normalized_hypothesis": "if the checksum fails, restart the worker process without deleting the cache tensors.",
- "errors": 1,
- "words": 13,
- "wer": 0.07692307692307693
- },
- "semantic": {
- "normalized_reference": "if the checksum fails restart the worker process without deleting the cached tensors",
- "normalized_hypothesis": "if the checksum fails restart the worker process without deleting the cache tensors",
- "errors": 1,
- "words": 13,
- "wer": 0.07692307692307693
- }
- },
- {
- "id": "utmos200_149",
- "category": "conversational",
- "text": "I need you to understand how close that was to not working.",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_149.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "I need you to understand how close that was to not working.",
- "raw": {
- "normalized_reference": "i need you to understand how close that was to not working.",
- "normalized_hypothesis": "i need you to understand how close that was to not working.",
- "errors": 0,
- "words": 12,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "i need you to understand how close that was to not working",
- "normalized_hypothesis": "i need you to understand how close that was to not working",
- "errors": 0,
- "words": 12,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_150",
- "category": "descriptive",
- "text": "It was the sort of morning where the whole world looks rinsed clean.",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_150.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "It was the sort of morning where the whole world looks rinsed clean.",
- "raw": {
- "normalized_reference": "it was the sort of morning where the whole world looks rinsed clean.",
- "normalized_hypothesis": "it was the sort of morning where the whole world looks rinsed clean.",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "it was the sort of morning where the whole world looks rinsed clean",
- "normalized_hypothesis": "it was the sort of morning where the whole world looks rinsed clean",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_151",
- "category": "emotional",
- "text": "She started laughing and I started laughing and we couldn't stop for a full minute.",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_151.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "She started laughing and I started laughing and we couldn't stop for a full minute.",
- "raw": {
- "normalized_reference": "she started laughing and i started laughing and we couldn't stop for a full minute.",
- "normalized_hypothesis": "she started laughing and i started laughing and we couldn't stop for a full minute.",
- "errors": 0,
- "words": 15,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "she started laughing and i started laughing and we could not stop for a full minute",
- "normalized_hypothesis": "she started laughing and i started laughing and we could not stop for a full minute",
- "errors": 0,
- "words": 16,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_152",
- "category": "long",
- "text": "We spent three years building something we were genuinely proud of, and then the market shifted and none of it mattered the way we thought it would.",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_152.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "We spent three years building something we were genuinely proud of, and then the market shifted and none of it mattered the way we thought it would.",
- "raw": {
- "normalized_reference": "we spent three years building something we were genuinely proud of, and then the market shifted and none of it mattered the way we thought it would.",
- "normalized_hypothesis": "we spent three years building something we were genuinely proud of, and then the market shifted and none of it mattered the way we thought it would.",
- "errors": 0,
- "words": 27,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "we spent 3 years building something we were genuinely proud of and then the market shifted and none of it mattered the way we thought it would",
- "normalized_hypothesis": "we spent 3 years building something we were genuinely proud of and then the market shifted and none of it mattered the way we thought it would",
- "errors": 0,
- "words": 27,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_153",
- "category": "narrative",
- "text": "And somehow, despite all of that, it actually turned out okay.",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_153.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "And somehow, despite all of that, it actually turned out okay.",
- "raw": {
- "normalized_reference": "and somehow, despite all of that, it actually turned out okay.",
- "normalized_hypothesis": "and somehow, despite all of that, it actually turned out okay.",
- "errors": 0,
- "words": 11,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "and somehow despite all of that it actually turned out okay",
- "normalized_hypothesis": "and somehow despite all of that it actually turned out okay",
- "errors": 0,
- "words": 11,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_154",
- "category": "question",
- "text": "Is that something you actually want, or something you think you should want?",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_154.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Is that something you actually want or something you think you should want?",
- "raw": {
- "normalized_reference": "is that something you actually want, or something you think you should want?",
- "normalized_hypothesis": "is that something you actually want or something you think you should want?",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "is that something you actually want or something you think you should want",
- "normalized_hypothesis": "is that something you actually want or something you think you should want",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_155",
- "category": "short",
- "text": "This is fine.",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_155.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "This is fine.",
- "raw": {
- "normalized_reference": "this is fine.",
- "normalized_hypothesis": "this is fine.",
- "errors": 0,
- "words": 3,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "this is fine",
- "normalized_hypothesis": "this is fine",
- "errors": 0,
- "words": 3,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_156",
- "category": "technical",
- "text": "The tricky part isn't the setup, it's making sure all dependencies line up.",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_156.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "The tricky part isn't the setup, it's making sure all dependencies line up.",
- "raw": {
- "normalized_reference": "the tricky part isn't the setup, it's making sure all dependencies line up.",
- "normalized_hypothesis": "the tricky part isn't the setup, it's making sure all dependencies line up.",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "the tricky part is not the setup it is making sure all dependencies line up",
- "normalized_hypothesis": "the tricky part is not the setup it is making sure all dependencies line up",
- "errors": 0,
- "words": 15,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_157",
- "category": "conversational",
- "text": "He called it a minor setback. I called it a disaster.",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_157.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "He called it a minor setback. I called it a disaster.",
- "raw": {
- "normalized_reference": "he called it a minor setback. i called it a disaster.",
- "normalized_hypothesis": "he called it a minor setback. i called it a disaster.",
- "errors": 0,
- "words": 11,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "he called it a minor setback i called it a disaster",
- "normalized_hypothesis": "he called it a minor setback i called it a disaster",
- "errors": 0,
- "words": 11,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_158",
- "category": "descriptive",
- "text": "The whole place smelled like pine and old wood and cold air.",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_158.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "The whole place smelled like pine and old wood and cold air.",
- "raw": {
- "normalized_reference": "the whole place smelled like pine and old wood and cold air.",
- "normalized_hypothesis": "the whole place smelled like pine and old wood and cold air.",
- "errors": 0,
- "words": 12,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "the whole place smelled like pine and old wood and cold air",
- "normalized_hypothesis": "the whole place smelled like pine and old wood and cold air",
- "errors": 0,
- "words": 12,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_159",
- "category": "emotional",
- "text": "I've never been so relieved in my entire life.",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_159.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "I've never been so relieved in my entire life.",
- "raw": {
- "normalized_reference": "i've never been so relieved in my entire life.",
- "normalized_hypothesis": "i've never been so relieved in my entire life.",
- "errors": 0,
- "words": 9,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "i have never been so relieved in my entire life",
- "normalized_hypothesis": "i have never been so relieved in my entire life",
- "errors": 0,
- "words": 10,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_160",
- "category": "narrative",
- "text": "I wrote it all down that night so I wouldn't forget, and I still haven't looked at it.",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_160.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "I wrote it all down that night so I wouldn't forget, and I still haven't looked at it.",
- "raw": {
- "normalized_reference": "i wrote it all down that night so i wouldn't forget, and i still haven't looked at it.",
- "normalized_hypothesis": "i wrote it all down that night so i wouldn't forget, and i still haven't looked at it.",
- "errors": 0,
- "words": 18,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "i wrote it all down that night so i would not forget and i still have not looked at it",
- "normalized_hypothesis": "i wrote it all down that night so i would not forget and i still have not looked at it",
- "errors": 0,
- "words": 20,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_161",
- "category": "short",
- "text": "That's kind of wild actually.",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_161.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "That's kind of wild actually.",
- "raw": {
- "normalized_reference": "that's kind of wild actually.",
- "normalized_hypothesis": "that's kind of wild actually.",
- "errors": 0,
- "words": 5,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "that is kind of wild actually",
- "normalized_hypothesis": "that is kind of wild actually",
- "errors": 0,
- "words": 6,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_162",
- "category": "technical",
- "text": "Version pinning will save you a lot of pain later down the line.",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_162.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Version pinning will save you a lot of pain later down the line.",
- "raw": {
- "normalized_reference": "version pinning will save you a lot of pain later down the line.",
- "normalized_hypothesis": "version pinning will save you a lot of pain later down the line.",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "version pinning will save you a lot of pain later down the line",
- "normalized_hypothesis": "version pinning will save you a lot of pain later down the line",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_163",
- "category": "conversational",
- "text": "At some point you just have to commit and see what happens.",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_163.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "At some point you just have to commit and see what happens.",
- "raw": {
- "normalized_reference": "at some point you just have to commit and see what happens.",
- "normalized_hypothesis": "at some point you just have to commit and see what happens.",
- "errors": 0,
- "words": 12,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "at some point you just have to commit and see what happens",
- "normalized_hypothesis": "at some point you just have to commit and see what happens",
- "errors": 0,
- "words": 12,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_164",
- "category": "emotional",
- "text": "Nobody told me it was going to be this hard, and honestly, I'm grateful.",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_164.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Nobody told me it was going to be this hard and honestly, I'm grateful.",
- "raw": {
- "normalized_reference": "nobody told me it was going to be this hard, and honestly, i'm grateful.",
- "normalized_hypothesis": "nobody told me it was going to be this hard and honestly, i'm grateful.",
- "errors": 0,
- "words": 14,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "nobody told me it was going to be this hard and honestly i am grateful",
- "normalized_hypothesis": "nobody told me it was going to be this hard and honestly i am grateful",
- "errors": 0,
- "words": 15,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_165",
- "category": "narrative",
- "text": "Chapter seven. In which everything that could go wrong did, and a few things that couldn't.",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_165.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Chapter 7, in which everything that could go wrong did, and a few things that couldn't.",
- "raw": {
- "normalized_reference": "chapter seven. in which everything that could go wrong did, and a few things that couldn't.",
- "normalized_hypothesis": "chapter 7, in which everything that could go wrong did, and a few things that couldn't.",
- "errors": 1,
- "words": 16,
- "wer": 0.0625
- },
- "semantic": {
- "normalized_reference": "chapter 7 in which everything that could go wrong did and a few things that could not",
- "normalized_hypothesis": "chapter 7 in which everything that could go wrong did and a few things that could not",
- "errors": 0,
- "words": 17,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_166",
- "category": "short",
- "text": "Wait, what?",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_166.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Wait, what?",
- "raw": {
- "normalized_reference": "wait, what?",
- "normalized_hypothesis": "wait, what?",
- "errors": 0,
- "words": 2,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "wait what",
- "normalized_hypothesis": "wait what",
- "errors": 0,
- "words": 2,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_167",
- "category": "technical",
- "text": "The graphics processor remained idle because the data loader exhausted every central processing core.",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_167.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "The graphics processor remained idle because the data loader exhausted every central processing core.",
- "raw": {
- "normalized_reference": "the graphics processor remained idle because the data loader exhausted every central processing core.",
- "normalized_hypothesis": "the graphics processor remained idle because the data loader exhausted every central processing core.",
- "errors": 0,
- "words": 14,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "the graphics processor remained idle because the data loader exhausted every central processing core",
- "normalized_hypothesis": "the graphics processor remained idle because the data loader exhausted every central processing core",
- "errors": 0,
- "words": 14,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_168",
- "category": "conversational",
- "text": "Could you speak a little more slowly? I am trying to write every detail down.",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_168.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Could you speak a little more slowly? I am trying to write every detail down.",
- "raw": {
- "normalized_reference": "could you speak a little more slowly? i am trying to write every detail down.",
- "normalized_hypothesis": "could you speak a little more slowly? i am trying to write every detail down.",
- "errors": 0,
- "words": 15,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "could you speak a little more slowly i am trying to write every detail down",
- "normalized_hypothesis": "could you speak a little more slowly i am trying to write every detail down",
- "errors": 0,
- "words": 15,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_169",
- "category": "emotional",
- "text": "I was so excited I could barely form a complete sentence.",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_169.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "I was so excited I could barely form a complete sentence.",
- "raw": {
- "normalized_reference": "i was so excited i could barely form a complete sentence.",
- "normalized_hypothesis": "i was so excited i could barely form a complete sentence.",
- "errors": 0,
- "words": 11,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "i was so excited i could barely form a complete sentence",
- "normalized_hypothesis": "i was so excited i could barely form a complete sentence",
- "errors": 0,
- "words": 11,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_170",
- "category": "narrative",
- "text": "He looked at me like I had said something in a language he didn't speak.",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_170.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "He looked at me like I had said something in a language he didn't speak.",
- "raw": {
- "normalized_reference": "he looked at me like i had said something in a language he didn't speak.",
- "normalized_hypothesis": "he looked at me like i had said something in a language he didn't speak.",
- "errors": 0,
- "words": 15,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "he looked at me like i had said something in a language he did not speak",
- "normalized_hypothesis": "he looked at me like i had said something in a language he did not speak",
- "errors": 0,
- "words": 16,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_171",
- "category": "short",
- "text": "That's actually hilarious.",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_171.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "That's actually hilarious.",
- "raw": {
- "normalized_reference": "that's actually hilarious.",
- "normalized_hypothesis": "that's actually hilarious.",
- "errors": 0,
- "words": 3,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "that is actually hilarious",
- "normalized_hypothesis": "that is actually hilarious",
- "errors": 0,
- "words": 4,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_172",
- "category": "technical",
- "text": "The encoder maps phonemes, stress, and phrase boundaries into a shared hidden representation.",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_172.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "The encoder maps phonemes, stress, and phrase boundaries into a shared hidden representation.",
- "raw": {
- "normalized_reference": "the encoder maps phonemes, stress, and phrase boundaries into a shared hidden representation.",
- "normalized_hypothesis": "the encoder maps phonemes, stress, and phrase boundaries into a shared hidden representation.",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "the encoder maps phonemes stress and phrase boundaries into a shared hidden representation",
- "normalized_hypothesis": "the encoder maps phonemes stress and phrase boundaries into a shared hidden representation",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_173",
- "category": "conversational",
- "text": "She showed up three hours late and acted like nothing had happened.",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_173.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "She showed up three hours late and acted like nothing had happened.",
- "raw": {
- "normalized_reference": "she showed up three hours late and acted like nothing had happened.",
- "normalized_hypothesis": "she showed up three hours late and acted like nothing had happened.",
- "errors": 0,
- "words": 12,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "she showed up 3 hours late and acted like nothing had happened",
- "normalized_hypothesis": "she showed up 3 hours late and acted like nothing had happened",
- "errors": 0,
- "words": 12,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_174",
- "category": "emotional",
- "text": "Oh my god. This is the best thing I've ever tasted.",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_174.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Oh my god. This is the best thing I've ever tasted.",
- "raw": {
- "normalized_reference": "oh my god. this is the best thing i've ever tasted.",
- "normalized_hypothesis": "oh my god. this is the best thing i've ever tasted.",
- "errors": 0,
- "words": 11,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "0 my god this is the best thing i have ever tasted",
- "normalized_hypothesis": "0 my god this is the best thing i have ever tasted",
- "errors": 0,
- "words": 12,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_175",
- "category": "narrative",
- "text": "The note was short, which was somehow more frightening than a long one would have been.",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_175.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "The note was short, which was somehow more frightening than a long one would have been.",
- "raw": {
- "normalized_reference": "the note was short, which was somehow more frightening than a long one would have been.",
- "normalized_hypothesis": "the note was short, which was somehow more frightening than a long one would have been.",
- "errors": 0,
- "words": 16,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "the note was short which was somehow more frightening than a long one would have been",
- "normalized_hypothesis": "the note was short which was somehow more frightening than a long one would have been",
- "errors": 0,
- "words": 16,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_176",
- "category": "short",
- "text": "Noted.",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_176.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Noted.",
- "raw": {
- "normalized_reference": "noted.",
- "normalized_hypothesis": "noted.",
- "errors": 0,
- "words": 1,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "noted",
- "normalized_hypothesis": "noted",
- "errors": 0,
- "words": 1,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_177",
- "category": "technical",
- "text": "It's a simple process once you understand the underlying structure.",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_177.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "It's a simple process once you understand the underlying structure.",
- "raw": {
- "normalized_reference": "it's a simple process once you understand the underlying structure.",
- "normalized_hypothesis": "it's a simple process once you understand the underlying structure.",
- "errors": 0,
- "words": 10,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "it is a simple process once you understand the underlying structure",
- "normalized_hypothesis": "it is a simple process once you understand the underlying structure",
- "errors": 0,
- "words": 11,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_178",
- "category": "conversational",
- "text": "You have to hear this story because it's almost too good to be true.",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_178.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "You have to hear this story because it's almost too good to be true.",
- "raw": {
- "normalized_reference": "you have to hear this story because it's almost too good to be true.",
- "normalized_hypothesis": "you have to hear this story because it's almost too good to be true.",
- "errors": 0,
- "words": 14,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "you have to hear this story because it is almost too good to be true",
- "normalized_hypothesis": "you have to hear this story because it is almost too good to be true",
- "errors": 0,
- "words": 15,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_179",
- "category": "emotional",
- "text": "I have never felt more seen in my entire life than in that moment.",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_179.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "I have never felt more seen in my entire life than in that moment.",
- "raw": {
- "normalized_reference": "i have never felt more seen in my entire life than in that moment.",
- "normalized_hypothesis": "i have never felt more seen in my entire life than in that moment.",
- "errors": 0,
- "words": 14,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "i have never felt more seen in my entire life than in that moment",
- "normalized_hypothesis": "i have never felt more seen in my entire life than in that moment",
- "errors": 0,
- "words": 14,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_180",
- "category": "narrative",
- "text": "She knew. She had always known. She just waited for the right moment.",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_180.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "She knew. She had always known. She just waited for the right moment.",
- "raw": {
- "normalized_reference": "she knew. she had always known. she just waited for the right moment.",
- "normalized_hypothesis": "she knew. she had always known. she just waited for the right moment.",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "she knew she had always known she just waited for the right moment",
- "normalized_hypothesis": "she knew she had always known she just waited for the right moment",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_181",
- "category": "short",
- "text": "Yeah, no. Hard pass.",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_181.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Yeah, no. Hard pass.",
- "raw": {
- "normalized_reference": "yeah, no. hard pass.",
- "normalized_hypothesis": "yeah, no. hard pass.",
- "errors": 0,
- "words": 4,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "yeah no hard pass",
- "normalized_hypothesis": "yeah no hard pass",
- "errors": 0,
- "words": 4,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_182",
- "category": "technical",
- "text": "Please verify the sample rate, hop length, and mel filter configuration before inference.",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_182.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Please verify the sample rate, hop length, and MEL filter configuration before inference.",
- "raw": {
- "normalized_reference": "please verify the sample rate, hop length, and mel filter configuration before inference.",
- "normalized_hypothesis": "please verify the sample rate, hop length, and mel filter configuration before inference.",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "please verify the sample rate hop length and mel filter configuration before inference",
- "normalized_hypothesis": "please verify the sample rate hop length and mel filter configuration before inference",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_183",
- "category": "conversational",
- "text": "That's the worst idea I've ever heard and I'm completely in.",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_183.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "That's the worst idea I've ever heard and I'm completely in.",
- "raw": {
- "normalized_reference": "that's the worst idea i've ever heard and i'm completely in.",
- "normalized_hypothesis": "that's the worst idea i've ever heard and i'm completely in.",
- "errors": 0,
- "words": 11,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "that is the worst idea i have ever heard and i am completely in",
- "normalized_hypothesis": "that is the worst idea i have ever heard and i am completely in",
- "errors": 0,
- "words": 14,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_184",
- "category": "emotional",
- "text": "I remember thinking: this is it. This is the thing I've been waiting for.",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_184.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "I remember thinking, this is it. This is the thing I've been waiting for.",
- "raw": {
- "normalized_reference": "i remember thinking: this is it. this is the thing i've been waiting for.",
- "normalized_hypothesis": "i remember thinking, this is it. this is the thing i've been waiting for.",
- "errors": 0,
- "words": 14,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "i remember thinking this is it this is the thing i have been waiting for",
- "normalized_hypothesis": "i remember thinking this is it this is the thing i have been waiting for",
- "errors": 0,
- "words": 15,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_185",
- "category": "narrative",
- "text": "We never talked about it again, which I think was the right call.",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_185.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "We never talked about it again, which I think was the right call.",
- "raw": {
- "normalized_reference": "we never talked about it again, which i think was the right call.",
- "normalized_hypothesis": "we never talked about it again, which i think was the right call.",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "we never talked about it again which i think was the right call",
- "normalized_hypothesis": "we never talked about it again which i think was the right call",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_186",
- "category": "short",
- "text": "In hindsight, obvious.",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_186.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "In hindsight, obvious.",
- "raw": {
- "normalized_reference": "in hindsight, obvious.",
- "normalized_hypothesis": "in hindsight, obvious.",
- "errors": 0,
- "words": 3,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "in hindsight obvious",
- "normalized_hypothesis": "in hindsight obvious",
- "errors": 0,
- "words": 3,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_187",
- "category": "technical",
- "text": "Once it's configured correctly, it basically runs itself.",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_187.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Once it's configured correctly, it basically runs itself.",
- "raw": {
- "normalized_reference": "once it's configured correctly, it basically runs itself.",
- "normalized_hypothesis": "once it's configured correctly, it basically runs itself.",
- "errors": 0,
- "words": 8,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "once it is configured correctly it basically runs itself",
- "normalized_hypothesis": "once it is configured correctly it basically runs itself",
- "errors": 0,
- "words": 9,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_188",
- "category": "conversational",
- "text": "You got engaged? When? How? Tell me everything right now.",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_188.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "You got engaged. What? Oh. Tell me everything right now.",
- "raw": {
- "normalized_reference": "you got engaged? when? how? tell me everything right now.",
- "normalized_hypothesis": "you got engaged. what? oh. tell me everything right now.",
- "errors": 2,
- "words": 10,
- "wer": 0.2
- },
- "semantic": {
- "normalized_reference": "you got engaged when how tell me everything right now",
- "normalized_hypothesis": "you got engaged what 0 tell me everything right now",
- "errors": 2,
- "words": 10,
- "wer": 0.2
- }
- },
- {
- "id": "utmos200_189",
- "category": "emotional",
- "text": "It's hard to explain how much that moment meant to me.",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_189.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "It's hard to explain how much that moment meant to me.",
- "raw": {
- "normalized_reference": "it's hard to explain how much that moment meant to me.",
- "normalized_hypothesis": "it's hard to explain how much that moment meant to me.",
- "errors": 0,
- "words": 11,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "it is hard to explain how much that moment meant to me",
- "normalized_hypothesis": "it is hard to explain how much that moment meant to me",
- "errors": 0,
- "words": 12,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_190",
- "category": "narrative",
- "text": "He told the story three different times that night and somehow it got better each time.",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_190.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "He told the story three different times that night and somehow it got better each time.",
- "raw": {
- "normalized_reference": "he told the story three different times that night and somehow it got better each time.",
- "normalized_hypothesis": "he told the story three different times that night and somehow it got better each time.",
- "errors": 0,
- "words": 16,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "he told the story 3 different times that night and somehow it got better each time",
- "normalized_hypothesis": "he told the story 3 different times that night and somehow it got better each time",
- "errors": 0,
- "words": 16,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_191",
- "category": "short",
- "text": "Right. Moving on.",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_191.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Right, moving on.",
- "raw": {
- "normalized_reference": "right. moving on.",
- "normalized_hypothesis": "right, moving on.",
- "errors": 0,
- "words": 3,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "right moving on",
- "normalized_hypothesis": "right moving on",
- "errors": 0,
- "words": 3,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_192",
- "category": "conversational",
- "text": "No, I did not move your keys; check beside the small ceramic bowl.",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_192.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "No, I did not move your keys. Check beside the small ceramic bowl.",
- "raw": {
- "normalized_reference": "no, i did not move your keys; check beside the small ceramic bowl.",
- "normalized_hypothesis": "no, i did not move your keys. check beside the small ceramic bowl.",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "no i did not move your keys check beside the small ceramic bowl",
- "normalized_hypothesis": "no i did not move your keys check beside the small ceramic bowl",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_193",
- "category": "emotional",
- "text": "I was devastated. I'm not going to pretend otherwise.",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_193.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "I was devastated. I'm not going to pretend otherwise.",
- "raw": {
- "normalized_reference": "i was devastated. i'm not going to pretend otherwise.",
- "normalized_hypothesis": "i was devastated. i'm not going to pretend otherwise.",
- "errors": 0,
- "words": 9,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "i was devastated i am not going to pretend otherwise",
- "normalized_hypothesis": "i was devastated i am not going to pretend otherwise",
- "errors": 0,
- "words": 10,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_194",
- "category": "narrative",
- "text": "That was the last summer before everything changed, though we didn't know it yet.",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_194.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "That was the last summer before everything changed, though we didn't know it yet.",
- "raw": {
- "normalized_reference": "that was the last summer before everything changed, though we didn't know it yet.",
- "normalized_hypothesis": "that was the last summer before everything changed, though we didn't know it yet.",
- "errors": 0,
- "words": 14,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "that was the last summer before everything changed though we did not know it yet",
- "normalized_hypothesis": "that was the last summer before everything changed though we did not know it yet",
- "errors": 0,
- "words": 15,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_195",
- "category": "short",
- "text": "Bold of you to assume.",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_195.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Bold of you to assume.",
- "raw": {
- "normalized_reference": "bold of you to assume.",
- "normalized_hypothesis": "bold of you to assume.",
- "errors": 0,
- "words": 5,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "bold of you to assume",
- "normalized_hypothesis": "bold of you to assume",
- "errors": 0,
- "words": 5,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_196",
- "category": "conversational",
- "text": "It's one of those decisions that makes total sense in the moment.",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_196.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "It's one of those decisions that makes total sense in the moment.",
- "raw": {
- "normalized_reference": "it's one of those decisions that makes total sense in the moment.",
- "normalized_hypothesis": "it's one of those decisions that makes total sense in the moment.",
- "errors": 0,
- "words": 12,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "it is one of those decisions that makes total sense in the moment",
- "normalized_hypothesis": "it is one of those decisions that makes total sense in the moment",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_197",
- "category": "emotional",
- "text": "I was furious. Not the loud kind. The quiet kind that lasts for days.",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_197.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "I was furious. Not the loud kind. The quiet kind that lasts for days.",
- "raw": {
- "normalized_reference": "i was furious. not the loud kind. the quiet kind that lasts for days.",
- "normalized_hypothesis": "i was furious. not the loud kind. the quiet kind that lasts for days.",
- "errors": 0,
- "words": 14,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "i was furious not the loud kind the quiet kind that lasts for days",
- "normalized_hypothesis": "i was furious not the loud kind the quiet kind that lasts for days",
- "errors": 0,
- "words": 14,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_198",
- "category": "narrative",
- "text": "That's the part of the story I usually leave out.",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_198.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "That's the part of the story I usually leave out.",
- "raw": {
- "normalized_reference": "that's the part of the story i usually leave out.",
- "normalized_hypothesis": "that's the part of the story i usually leave out.",
- "errors": 0,
- "words": 10,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "that is the part of the story i usually leave out",
- "normalized_hypothesis": "that is the part of the story i usually leave out",
- "errors": 0,
- "words": 11,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_199",
- "category": "short",
- "text": "I've said too much.",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_199.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "I've said too much.",
- "raw": {
- "normalized_reference": "i've said too much.",
- "normalized_hypothesis": "i've said too much.",
- "errors": 0,
- "words": 4,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "i have said too much",
- "normalized_hypothesis": "i have said too much",
- "errors": 0,
- "words": 5,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_200",
- "category": "conversational",
- "text": "I've been faking that I understood this the entire time.",
- "model": "piper-ryan-low",
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_200.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "I have been faking that I understood this the entire time.",
- "raw": {
- "normalized_reference": "i've been faking that i understood this the entire time.",
- "normalized_hypothesis": "i have been faking that i understood this the entire time.",
- "errors": 2,
- "words": 10,
- "wer": 0.2
- },
- "semantic": {
- "normalized_reference": "i have been faking that i understood this the entire time",
- "normalized_hypothesis": "i have been faking that i understood this the entire time",
- "errors": 0,
- "words": 11,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_001",
- "category": "conversational",
- "text": "It wasn't until later that I realized what had actually happened.",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_001.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "It wasn't until later that I realized what had actually happened.",
- "raw": {
- "normalized_reference": "it wasn't until later that i realized what had actually happened.",
- "normalized_hypothesis": "it wasn't until later that i realized what had actually happened.",
- "errors": 0,
- "words": 11,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "it was not until later that i realized what had actually happened",
- "normalized_hypothesis": "it was not until later that i realized what had actually happened",
- "errors": 0,
- "words": 12,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_009",
- "category": "numbers",
- "text": "The package weighs twelve point six kilograms and arrived on July twenty-first.",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_009.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "The package weighs 12.6 kilograms and arrived on July 21.",
- "raw": {
- "normalized_reference": "the package weighs twelve point six kilograms and arrived on july twenty-first.",
- "normalized_hypothesis": "the package weighs 12.6 kilograms and arrived on july 21.",
- "errors": 5,
- "words": 13,
- "wer": 0.38461538461538464
- },
- "semantic": {
- "normalized_reference": "the package weighs 12.6 kilograms and arrived on july 21st",
- "normalized_hypothesis": "the package weighs 12.6 kilograms and arrived on july 21",
- "errors": 1,
- "words": 11,
- "wer": 0.09090909090909091
- }
- },
- {
- "id": "utmos200_017",
- "category": "emotional",
- "text": "I specifically asked for one thing. One thing. And it still didn't happen.",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_017.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "I specifically asked for one thing. One thing. And it still didn't happen.",
- "raw": {
- "normalized_reference": "i specifically asked for one thing. one thing. and it still didn't happen.",
- "normalized_hypothesis": "i specifically asked for one thing. one thing. and it still didn't happen.",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "i specifically asked for one thing one thing and it still did not happen",
- "normalized_hypothesis": "i specifically asked for one thing one thing and it still did not happen",
- "errors": 0,
- "words": 14,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_025",
- "category": "question",
- "text": "How long have you been sitting on this?",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_025.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "How long have you been sitting on this?",
- "raw": {
- "normalized_reference": "how long have you been sitting on this?",
- "normalized_hypothesis": "how long have you been sitting on this?",
- "errors": 0,
- "words": 8,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "how long have you been sitting on this",
- "normalized_hypothesis": "how long have you been sitting on this",
- "errors": 0,
- "words": 8,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_033",
- "category": "long",
- "text": "I think the most underrated skill in the world is knowing when to stop explaining yourself and just let people draw their own conclusions.",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_033.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "I think the most underrated skill in the world is knowing when to stop explaining yourself and just let people draw their own conclusions.",
- "raw": {
- "normalized_reference": "i think the most underrated skill in the world is knowing when to stop explaining yourself and just let people draw their own conclusions.",
- "normalized_hypothesis": "i think the most underrated skill in the world is knowing when to stop explaining yourself and just let people draw their own conclusions.",
- "errors": 0,
- "words": 24,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "i think the most underrated skill in the world is knowing when to stop explaining yourself and just let people draw their own conclusions",
- "normalized_hypothesis": "i think the most underrated skill in the world is knowing when to stop explaining yourself and just let people draw their own conclusions",
- "errors": 0,
- "words": 24,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_041",
- "category": "short",
- "text": "I mean. Yeah. Kind of.",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_041.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "I mean. Yes. Kind of.",
- "raw": {
- "normalized_reference": "i mean. yeah. kind of.",
- "normalized_hypothesis": "i mean. yes. kind of.",
- "errors": 1,
- "words": 5,
- "wer": 0.2
- },
- "semantic": {
- "normalized_reference": "i mean yeah kind of",
- "normalized_hypothesis": "i mean yes kind of",
- "errors": 1,
- "words": 5,
- "wer": 0.2
- }
- },
- {
- "id": "utmos200_049",
- "category": "names_places",
- "text": "Xiomara met Siobhan near the old library in Worcestershire.",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_049.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Xiomara met Siobhan near the old library in Worcestershire.",
- "raw": {
- "normalized_reference": "xiomara met siobhan near the old library in worcestershire.",
- "normalized_hypothesis": "xiomara met siobhan near the old library in worcestershire.",
- "errors": 0,
- "words": 9,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "xiomara met siobhan near the old library in worcestershire",
- "normalized_hypothesis": "xiomara met siobhan near the old library in worcestershire",
- "errors": 0,
- "words": 9,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_057",
- "category": "conversational",
- "text": "I would not say that again in this room if I were you.",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_057.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "I would not say that again in this room if I were you.",
- "raw": {
- "normalized_reference": "i would not say that again in this room if i were you.",
- "normalized_hypothesis": "i would not say that again in this room if i were you.",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "i would not say that again in this room if i were you",
- "normalized_hypothesis": "i would not say that again in this room if i were you",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_065",
- "category": "numbers",
- "text": "Call extension zero four nine after eight thirty in the morning.",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_065.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Call extension 049 after 8.30 in the morning.",
- "raw": {
- "normalized_reference": "call extension zero four nine after eight thirty in the morning.",
- "normalized_hypothesis": "call extension 049 after 8.30 in the morning.",
- "errors": 5,
- "words": 11,
- "wer": 0.45454545454545453
- },
- "semantic": {
- "normalized_reference": "call extension 049 after 830 in the morning",
- "normalized_hypothesis": "call extension 49 after 8.30 in the morning",
- "errors": 3,
- "words": 8,
- "wer": 0.375
- }
- },
- {
- "id": "utmos200_073",
- "category": "emotional",
- "text": "It's the kind of happy that makes you nervous because you know it can't last.",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_073.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "It's the kind of happy that makes you nervous because you know it can't last.",
- "raw": {
- "normalized_reference": "it's the kind of happy that makes you nervous because you know it can't last.",
- "normalized_hypothesis": "it's the kind of happy that makes you nervous because you know it can't last.",
- "errors": 0,
- "words": 15,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "it is the kind of happy that makes you nervous because you know it can not last",
- "normalized_hypothesis": "it is the kind of happy that makes you nervous because you know it can not last",
- "errors": 0,
- "words": 17,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_081",
- "category": "question",
- "text": "What are we actually trying to accomplish here?",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_081.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "What are we actually trying to accomplish here?",
- "raw": {
- "normalized_reference": "what are we actually trying to accomplish here?",
- "normalized_hypothesis": "what are we actually trying to accomplish here?",
- "errors": 0,
- "words": 8,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "what are we actually trying to accomplish here",
- "normalized_hypothesis": "what are we actually trying to accomplish here",
- "errors": 0,
- "words": 8,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_089",
- "category": "narrative",
- "text": "He stood at the edge of the cliff and wondered, not for the first time, how he'd gotten here.",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_089.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "He stood at the edge of the cliff and wondered, not for the first time, how he'd gotten here.",
- "raw": {
- "normalized_reference": "he stood at the edge of the cliff and wondered, not for the first time, how he'd gotten here.",
- "normalized_hypothesis": "he stood at the edge of the cliff and wondered, not for the first time, how he'd gotten here.",
- "errors": 0,
- "words": 19,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "he stood at the edge of the cliff and wondered not for the 1st time how he would gotten here",
- "normalized_hypothesis": "he stood at the edge of the cliff and wondered not for the 1st time how he would gotten here",
- "errors": 0,
- "words": 20,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_097",
- "category": "narrative",
- "text": "The thing nobody tells you about starting over is how ordinary most of it feels.",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_097.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "The thing nobody tells you about starting over is how ordinary most of it feels.",
- "raw": {
- "normalized_reference": "the thing nobody tells you about starting over is how ordinary most of it feels.",
- "normalized_hypothesis": "the thing nobody tells you about starting over is how ordinary most of it feels.",
- "errors": 0,
- "words": 15,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "the thing nobody tells you about starting over is how ordinary most of it feels",
- "normalized_hypothesis": "the thing nobody tells you about starting over is how ordinary most of it feels",
- "errors": 0,
- "words": 15,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_105",
- "category": "narrative",
- "text": "There was this one day last summer when everything just clicked into place.",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_105.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "There was this one day last summer when everything just clicked into place.",
- "raw": {
- "normalized_reference": "there was this one day last summer when everything just clicked into place.",
- "normalized_hypothesis": "there was this one day last summer when everything just clicked into place.",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "there was this one day last summer when everything just clicked into place",
- "normalized_hypothesis": "there was this one day last summer when everything just clicked into place",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_113",
- "category": "narrative",
- "text": "She opened the door slowly, as if she expected to find something she wasn't ready for.",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_113.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "She opened the door slowly, as if she expected to find something she was ready for.",
- "raw": {
- "normalized_reference": "she opened the door slowly, as if she expected to find something she wasn't ready for.",
- "normalized_hypothesis": "she opened the door slowly, as if she expected to find something she was ready for.",
- "errors": 1,
- "words": 16,
- "wer": 0.0625
- },
- "semantic": {
- "normalized_reference": "she opened the door slowly as if she expected to find something she was not ready for",
- "normalized_hypothesis": "she opened the door slowly as if she expected to find something she was ready for",
- "errors": 1,
- "words": 17,
- "wer": 0.058823529411764705
- }
- },
- {
- "id": "utmos200_121",
- "category": "narrative",
- "text": "The moment I opened the door I knew something was different, even if I couldn't say what.",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_121.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "The moment I opened the door I knew something was different, even if I couldn't say what.",
- "raw": {
- "normalized_reference": "the moment i opened the door i knew something was different, even if i couldn't say what.",
- "normalized_hypothesis": "the moment i opened the door i knew something was different, even if i couldn't say what.",
- "errors": 0,
- "words": 17,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "the moment i opened the door i knew something was different even if i could not say what",
- "normalized_hypothesis": "the moment i opened the door i knew something was different even if i could not say what",
- "errors": 0,
- "words": 18,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_129",
- "category": "narrative",
- "text": "So this is the part of the story where everything goes completely sideways.",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_129.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "So this is the part of the story where everything goes completely sideways.",
- "raw": {
- "normalized_reference": "so this is the part of the story where everything goes completely sideways.",
- "normalized_hypothesis": "so this is the part of the story where everything goes completely sideways.",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "so this is the part of the story where everything goes completely sideways",
- "normalized_hypothesis": "so this is the part of the story where everything goes completely sideways",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_137",
- "category": "narrative",
- "text": "The letter arrived on a Tuesday, which felt appropriate somehow.",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_137.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "The letter arrived on a Tuesday, which felt appropriate somehow.",
- "raw": {
- "normalized_reference": "the letter arrived on a tuesday, which felt appropriate somehow.",
- "normalized_hypothesis": "the letter arrived on a tuesday, which felt appropriate somehow.",
- "errors": 0,
- "words": 10,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "the letter arrived on a tuesday which felt appropriate somehow",
- "normalized_hypothesis": "the letter arrived on a tuesday which felt appropriate somehow",
- "errors": 0,
- "words": 10,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_145",
- "category": "narrative",
- "text": "I remember the exact moment I realized I had made a huge mistake.",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_145.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "I remember the exact moment I realized I had made a huge mistake.",
- "raw": {
- "normalized_reference": "i remember the exact moment i realized i had made a huge mistake.",
- "normalized_hypothesis": "i remember the exact moment i realized i had made a huge mistake.",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "i remember the exact moment i realized i had made a huge mistake",
- "normalized_hypothesis": "i remember the exact moment i realized i had made a huge mistake",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_153",
- "category": "narrative",
- "text": "And somehow, despite all of that, it actually turned out okay.",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_153.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "And somehow, despite all of that, it actually turned out okay.",
- "raw": {
- "normalized_reference": "and somehow, despite all of that, it actually turned out okay.",
- "normalized_hypothesis": "and somehow, despite all of that, it actually turned out okay.",
- "errors": 0,
- "words": 11,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "and somehow despite all of that it actually turned out okay",
- "normalized_hypothesis": "and somehow despite all of that it actually turned out okay",
- "errors": 0,
- "words": 11,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_161",
- "category": "short",
- "text": "That's kind of wild actually.",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_161.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "That's kind of wild actually.",
- "raw": {
- "normalized_reference": "that's kind of wild actually.",
- "normalized_hypothesis": "that's kind of wild actually.",
- "errors": 0,
- "words": 5,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "that is kind of wild actually",
- "normalized_hypothesis": "that is kind of wild actually",
- "errors": 0,
- "words": 6,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_169",
- "category": "emotional",
- "text": "I was so excited I could barely form a complete sentence.",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_169.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "I was so excited I could barely form a complete sentence.",
- "raw": {
- "normalized_reference": "i was so excited i could barely form a complete sentence.",
- "normalized_hypothesis": "i was so excited i could barely form a complete sentence.",
- "errors": 0,
- "words": 11,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "i was so excited i could barely form a complete sentence",
- "normalized_hypothesis": "i was so excited i could barely form a complete sentence",
- "errors": 0,
- "words": 11,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_177",
- "category": "technical",
- "text": "It's a simple process once you understand the underlying structure.",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_177.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "It's a simple process once you understand the underlying structure.",
- "raw": {
- "normalized_reference": "it's a simple process once you understand the underlying structure.",
- "normalized_hypothesis": "it's a simple process once you understand the underlying structure.",
- "errors": 0,
- "words": 10,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "it is a simple process once you understand the underlying structure",
- "normalized_hypothesis": "it is a simple process once you understand the underlying structure",
- "errors": 0,
- "words": 11,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_185",
- "category": "narrative",
- "text": "We never talked about it again, which I think was the right call.",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_185.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "We never talked about it again which I think was the right call.",
- "raw": {
- "normalized_reference": "we never talked about it again, which i think was the right call.",
- "normalized_hypothesis": "we never talked about it again which i think was the right call.",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "we never talked about it again which i think was the right call",
- "normalized_hypothesis": "we never talked about it again which i think was the right call",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_193",
- "category": "emotional",
- "text": "I was devastated. I'm not going to pretend otherwise.",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_193.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "I was devastated. I'm not going to pretend otherwise.",
- "raw": {
- "normalized_reference": "i was devastated. i'm not going to pretend otherwise.",
- "normalized_hypothesis": "i was devastated. i'm not going to pretend otherwise.",
- "errors": 0,
- "words": 9,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "i was devastated i am not going to pretend otherwise",
- "normalized_hypothesis": "i was devastated i am not going to pretend otherwise",
- "errors": 0,
- "words": 10,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_002",
- "category": "descriptive",
- "text": "The light at that time of afternoon is perfect for about twenty minutes.",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_002.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "The light at that time of afternoon is perfect for about 20 minutes.",
- "raw": {
- "normalized_reference": "the light at that time of afternoon is perfect for about twenty minutes.",
- "normalized_hypothesis": "the light at that time of afternoon is perfect for about 20 minutes.",
- "errors": 1,
- "words": 13,
- "wer": 0.07692307692307693
- },
- "semantic": {
- "normalized_reference": "the light at that time of afternoon is perfect for about 20 minutes",
- "normalized_hypothesis": "the light at that time of afternoon is perfect for about 20 minutes",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_010",
- "category": "punctuation",
- "text": "First, close the window; second, turn off the lamp; finally, lock the door.",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_010.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "First, close the window. Second, turn off the lamp. Finally, lock the door.",
- "raw": {
- "normalized_reference": "first, close the window; second, turn off the lamp; finally, lock the door.",
- "normalized_hypothesis": "first, close the window. second, turn off the lamp. finally, lock the door.",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "1st close the window 2nd turn off the lamp finally lock the door",
- "normalized_hypothesis": "1st close the window 2nd turn off the lamp finally lock the door",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_018",
- "category": "homographs",
- "text": "They refuse to place the refuse beside the produce aisle.",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_018.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "They refused to place the refuse beside the produce aisle.",
- "raw": {
- "normalized_reference": "they refuse to place the refuse beside the produce aisle.",
- "normalized_hypothesis": "they refused to place the refuse beside the produce aisle.",
- "errors": 1,
- "words": 10,
- "wer": 0.1
- },
- "semantic": {
- "normalized_reference": "they refuse to place the refuse beside the produce aisle",
- "normalized_hypothesis": "they refused to place the refuse beside the produce aisle",
- "errors": 1,
- "words": 10,
- "wer": 0.1
- }
- },
- {
- "id": "utmos200_026",
- "category": "rare_words",
- "text": "A mischievous chameleon rested beside the chrysanthemum and the bougainvillea.",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_026.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "A mischievous chameleon rested beside the chrysanthemum and the bougainvillea.",
- "raw": {
- "normalized_reference": "a mischievous chameleon rested beside the chrysanthemum and the bougainvillea.",
- "normalized_hypothesis": "a mischievous chameleon rested beside the chrysanthemum and the bougainvillea.",
- "errors": 0,
- "words": 10,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "a mischievous chameleon rested beside the chrysanthemum and the bougainvillea",
- "normalized_hypothesis": "a mischievous chameleon rested beside the chrysanthemum and the bougainvillea",
- "errors": 0,
- "words": 10,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_034",
- "category": "long_range",
- "text": "When the final bus disappeared around the corner, Nadia remembered that her umbrella, notebook, and lunch were still inside the quiet office.",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_034.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "When the final bus disappeared around the corner, Nadia remembered that her umbrella, notebook, and lunch were still inside the quiet office.",
- "raw": {
- "normalized_reference": "when the final bus disappeared around the corner, nadia remembered that her umbrella, notebook, and lunch were still inside the quiet office.",
- "normalized_hypothesis": "when the final bus disappeared around the corner, nadia remembered that her umbrella, notebook, and lunch were still inside the quiet office.",
- "errors": 0,
- "words": 22,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "when the final bus disappeared around the corner nadia remembered that her umbrella notebook and lunch were still inside the quiet office",
- "normalized_hypothesis": "when the final bus disappeared around the corner nadia remembered that her umbrella notebook and lunch were still inside the quiet office",
- "errors": 0,
- "words": 22,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_042",
- "category": "technical",
- "text": "The logs are your best friend when something behaves unexpectedly.",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_042.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "The logs are your best friend when something behaves unexpectedly.",
- "raw": {
- "normalized_reference": "the logs are your best friend when something behaves unexpectedly.",
- "normalized_hypothesis": "the logs are your best friend when something behaves unexpectedly.",
- "errors": 0,
- "words": 10,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "the logs are your best friend when something behaves unexpectedly",
- "normalized_hypothesis": "the logs are your best friend when something behaves unexpectedly",
- "errors": 0,
- "words": 10,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_050",
- "category": "narrative",
- "text": "She walked into the room and everyone went quiet, not because she was loud, but because she wasn't.",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_050.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "She walked into the room and everyone went quiet, not because she was loud, but because she was.",
- "raw": {
- "normalized_reference": "she walked into the room and everyone went quiet, not because she was loud, but because she wasn't.",
- "normalized_hypothesis": "she walked into the room and everyone went quiet, not because she was loud, but because she was.",
- "errors": 1,
- "words": 18,
- "wer": 0.05555555555555555
- },
- "semantic": {
- "normalized_reference": "she walked into the room and everyone went quiet not because she was loud but because she was not",
- "normalized_hypothesis": "she walked into the room and everyone went quiet not because she was loud but because she was",
- "errors": 1,
- "words": 19,
- "wer": 0.05263157894736842
- }
- },
- {
- "id": "utmos200_058",
- "category": "descriptive",
- "text": "The room was filled with the quiet that means something is about to change.",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_058.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "The room was filled with the quiet that means something is about to change.",
- "raw": {
- "normalized_reference": "the room was filled with the quiet that means something is about to change.",
- "normalized_hypothesis": "the room was filled with the quiet that means something is about to change.",
- "errors": 0,
- "words": 14,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "the room was filled with the quiet that means something is about to change",
- "normalized_hypothesis": "the room was filled with the quiet that means something is about to change",
- "errors": 0,
- "words": 14,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_066",
- "category": "punctuation",
- "text": "Stop - listen carefully - then answer the question without guessing.",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_066.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Stop listen carefully then answer the question without guessing.",
- "raw": {
- "normalized_reference": "stop - listen carefully - then answer the question without guessing.",
- "normalized_hypothesis": "stop listen carefully then answer the question without guessing.",
- "errors": 0,
- "words": 9,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "stop listen carefully then answer the question without guessing",
- "normalized_hypothesis": "stop listen carefully then answer the question without guessing",
- "errors": 0,
- "words": 9,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_074",
- "category": "homographs",
- "text": "The dove dove beneath the branch while the wind began to wind the loose cable.",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_074.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "The dove dove beneath the branch while the wind began to wind the loose cable.",
- "raw": {
- "normalized_reference": "the dove dove beneath the branch while the wind began to wind the loose cable.",
- "normalized_hypothesis": "the dove dove beneath the branch while the wind began to wind the loose cable.",
- "errors": 0,
- "words": 15,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "the dove dove beneath the branch while the wind began to wind the loose cable",
- "normalized_hypothesis": "the dove dove beneath the branch while the wind began to wind the loose cable",
- "errors": 0,
- "words": 15,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_082",
- "category": "rare_words",
- "text": "The physician recommended a noninvasive procedure for the recurring neurological symptoms.",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_082.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "The physician recommended a non-invasive procedure for the recurring neurological symptoms.",
- "raw": {
- "normalized_reference": "the physician recommended a noninvasive procedure for the recurring neurological symptoms.",
- "normalized_hypothesis": "the physician recommended a non-invasive procedure for the recurring neurological symptoms.",
- "errors": 2,
- "words": 11,
- "wer": 0.18181818181818182
- },
- "semantic": {
- "normalized_reference": "the physician recommended a noninvasive procedure for the recurring neurological symptoms",
- "normalized_hypothesis": "the physician recommended a non invasive procedure for the recurring neurological symptoms",
- "errors": 2,
- "words": 11,
- "wer": 0.18181818181818182
- }
- },
- {
- "id": "utmos200_090",
- "category": "question",
- "text": "Why does asking for help feel so hard when you actually need it?",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_090.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Why does asking for help feel so hard when you actually need it?",
- "raw": {
- "normalized_reference": "why does asking for help feel so hard when you actually need it?",
- "normalized_hypothesis": "why does asking for help feel so hard when you actually need it?",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "why does asking for help feel so hard when you actually need it",
- "normalized_hypothesis": "why does asking for help feel so hard when you actually need it",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_098",
- "category": "question",
- "text": "What exactly are we waiting for at this point?",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_098.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "What exactly are we waiting for at this point?",
- "raw": {
- "normalized_reference": "what exactly are we waiting for at this point?",
- "normalized_hypothesis": "what exactly are we waiting for at this point?",
- "errors": 0,
- "words": 9,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "what exactly are we waiting for at this point",
- "normalized_hypothesis": "what exactly are we waiting for at this point",
- "errors": 0,
- "words": 9,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_106",
- "category": "question",
- "text": "Have you ever thought about why we keep doing things the same way when they stop working?",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_106.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Have you ever thought about why we keep doing things the same way when they stop working?",
- "raw": {
- "normalized_reference": "have you ever thought about why we keep doing things the same way when they stop working?",
- "normalized_hypothesis": "have you ever thought about why we keep doing things the same way when they stop working?",
- "errors": 0,
- "words": 17,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "have you ever thought about why we keep doing things the same way when they stop working",
- "normalized_hypothesis": "have you ever thought about why we keep doing things the same way when they stop working",
- "errors": 0,
- "words": 17,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_114",
- "category": "question",
- "text": "What would you do differently if you could go back?",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_114.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "What would you do differently if you could go back?",
- "raw": {
- "normalized_reference": "what would you do differently if you could go back?",
- "normalized_hypothesis": "what would you do differently if you could go back?",
- "errors": 0,
- "words": 10,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "what would you do differently if you could go back",
- "normalized_hypothesis": "what would you do differently if you could go back",
- "errors": 0,
- "words": 10,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_122",
- "category": "question",
- "text": "How do you explain something that doesn't make sense even to you?",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_122.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "How do you explain something that doesn't make sense even to you?",
- "raw": {
- "normalized_reference": "how do you explain something that doesn't make sense even to you?",
- "normalized_hypothesis": "how do you explain something that doesn't make sense even to you?",
- "errors": 0,
- "words": 12,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "how do you explain something that does not make sense even to you",
- "normalized_hypothesis": "how do you explain something that does not make sense even to you",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_130",
- "category": "question",
- "text": "Is there a version of this where it actually works out okay?",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_130.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Is there a version of this where it actually works out okay?",
- "raw": {
- "normalized_reference": "is there a version of this where it actually works out okay?",
- "normalized_hypothesis": "is there a version of this where it actually works out okay?",
- "errors": 0,
- "words": 12,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "is there a version of this where it actually works out okay",
- "normalized_hypothesis": "is there a version of this where it actually works out okay",
- "errors": 0,
- "words": 12,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_138",
- "category": "question",
- "text": "What exactly did she say when you told her?",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_138.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "What exactly did she say when you told her?",
- "raw": {
- "normalized_reference": "what exactly did she say when you told her?",
- "normalized_hypothesis": "what exactly did she say when you told her?",
- "errors": 0,
- "words": 9,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "what exactly did she say when you told her",
- "normalized_hypothesis": "what exactly did she say when you told her",
- "errors": 0,
- "words": 9,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_146",
- "category": "question",
- "text": "Do you think it would have gone differently if we'd left earlier?",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_146.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Do you think it would have gone differently if we'd left earlier?",
- "raw": {
- "normalized_reference": "do you think it would have gone differently if we'd left earlier?",
- "normalized_hypothesis": "do you think it would have gone differently if we'd left earlier?",
- "errors": 0,
- "words": 12,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "do you think it would have gone differently if we would left earlier",
- "normalized_hypothesis": "do you think it would have gone differently if we would left earlier",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_154",
- "category": "question",
- "text": "Is that something you actually want, or something you think you should want?",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_154.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Is that something you actually want or something you think you should want?",
- "raw": {
- "normalized_reference": "is that something you actually want, or something you think you should want?",
- "normalized_hypothesis": "is that something you actually want or something you think you should want?",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "is that something you actually want or something you think you should want",
- "normalized_hypothesis": "is that something you actually want or something you think you should want",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_162",
- "category": "technical",
- "text": "Version pinning will save you a lot of pain later down the line.",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_162.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Version pinning will save you a lot of pain later down the line.",
- "raw": {
- "normalized_reference": "version pinning will save you a lot of pain later down the line.",
- "normalized_hypothesis": "version pinning will save you a lot of pain later down the line.",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "version pinning will save you a lot of pain later down the line",
- "normalized_hypothesis": "version pinning will save you a lot of pain later down the line",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_170",
- "category": "narrative",
- "text": "He looked at me like I had said something in a language he didn't speak.",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_170.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "He looked at me like I had said something in a language he didn't speak.",
- "raw": {
- "normalized_reference": "he looked at me like i had said something in a language he didn't speak.",
- "normalized_hypothesis": "he looked at me like i had said something in a language he didn't speak.",
- "errors": 0,
- "words": 15,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "he looked at me like i had said something in a language he did not speak",
- "normalized_hypothesis": "he looked at me like i had said something in a language he did not speak",
- "errors": 0,
- "words": 16,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_178",
- "category": "conversational",
- "text": "You have to hear this story because it's almost too good to be true.",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_178.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "You have to hear this story because it's almost too good to be true.",
- "raw": {
- "normalized_reference": "you have to hear this story because it's almost too good to be true.",
- "normalized_hypothesis": "you have to hear this story because it's almost too good to be true.",
- "errors": 0,
- "words": 14,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "you have to hear this story because it is almost too good to be true",
- "normalized_hypothesis": "you have to hear this story because it is almost too good to be true",
- "errors": 0,
- "words": 15,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_186",
- "category": "short",
- "text": "In hindsight, obvious.",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_186.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "In hindsight, obvious.",
- "raw": {
- "normalized_reference": "in hindsight, obvious.",
- "normalized_hypothesis": "in hindsight, obvious.",
- "errors": 0,
- "words": 3,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "in hindsight obvious",
- "normalized_hypothesis": "in hindsight obvious",
- "errors": 0,
- "words": 3,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_194",
- "category": "narrative",
- "text": "That was the last summer before everything changed, though we didn't know it yet.",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_194.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "That was the last summer before everything changed, though we didn't know it yet.",
- "raw": {
- "normalized_reference": "that was the last summer before everything changed, though we didn't know it yet.",
- "normalized_hypothesis": "that was the last summer before everything changed, though we didn't know it yet.",
- "errors": 0,
- "words": 14,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "that was the last summer before everything changed though we did not know it yet",
- "normalized_hypothesis": "that was the last summer before everything changed though we did not know it yet",
- "errors": 0,
- "words": 15,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_003",
- "category": "emotional",
- "text": "That was the scariest thing I've ever done, and also maybe the best.",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_003.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "That was the scariest thing I've ever done, and also maybe the best.",
- "raw": {
- "normalized_reference": "that was the scariest thing i've ever done, and also maybe the best.",
- "normalized_hypothesis": "that was the scariest thing i've ever done, and also maybe the best.",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "that was the scariest thing i have ever done and also maybe the best",
- "normalized_hypothesis": "that was the scariest thing i have ever done and also maybe the best",
- "errors": 0,
- "words": 14,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_011",
- "category": "question",
- "text": "When did you first realize something was wrong?",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_011.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "When did you first realize something was wrong?",
- "raw": {
- "normalized_reference": "when did you first realize something was wrong?",
- "normalized_hypothesis": "when did you first realize something was wrong?",
- "errors": 0,
- "words": 8,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "when did you 1st realize something was wrong",
- "normalized_hypothesis": "when did you 1st realize something was wrong",
- "errors": 0,
- "words": 8,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_019",
- "category": "long",
- "text": "There were days last year when I genuinely didn't know how I was going to get through it, and somehow I did, and I still don't fully understand how.",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_019.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "There were days last year when I genuinely didn't know how I was going to get through it, and somehow I did, and I still don't fully understand how.",
- "raw": {
- "normalized_reference": "there were days last year when i genuinely didn't know how i was going to get through it, and somehow i did, and i still don't fully understand how.",
- "normalized_hypothesis": "there were days last year when i genuinely didn't know how i was going to get through it, and somehow i did, and i still don't fully understand how.",
- "errors": 0,
- "words": 29,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "there were days last year when i genuinely did not know how i was going to get through it and somehow i did and i still do not fully understand how",
- "normalized_hypothesis": "there were days last year when i genuinely did not know how i was going to get through it and somehow i did and i still do not fully understand how",
- "errors": 0,
- "words": 31,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_027",
- "category": "short",
- "text": "We don't talk about that.",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_027.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "We don't talk about that.",
- "raw": {
- "normalized_reference": "we don't talk about that.",
- "normalized_hypothesis": "we don't talk about that.",
- "errors": 0,
- "words": 5,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "we do not talk about that",
- "normalized_hypothesis": "we do not talk about that",
- "errors": 0,
- "words": 6,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_035",
- "category": "names_places",
- "text": "Aurelia and Joaquin crossed the Schuylkill River before noon.",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_035.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Auralaya and Jokwan crossed the Shilkal river before noon.",
- "raw": {
- "normalized_reference": "aurelia and joaquin crossed the schuylkill river before noon.",
- "normalized_hypothesis": "auralaya and jokwan crossed the shilkal river before noon.",
- "errors": 3,
- "words": 9,
- "wer": 0.3333333333333333
- },
- "semantic": {
- "normalized_reference": "aurelia and joaquin crossed the schuylkill river before noon",
- "normalized_hypothesis": "auralaya and jokwan crossed the shilkal river before noon",
- "errors": 3,
- "words": 9,
- "wer": 0.3333333333333333
- }
- },
- {
- "id": "utmos200_043",
- "category": "conversational",
- "text": "He kept talking and I kept nodding and neither of us was listening.",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_043.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "He kept talking and I kept nodding and neither of us was listening.",
- "raw": {
- "normalized_reference": "he kept talking and i kept nodding and neither of us was listening.",
- "normalized_hypothesis": "he kept talking and i kept nodding and neither of us was listening.",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "he kept talking and i kept nodding and neither of us was listening",
- "normalized_hypothesis": "he kept talking and i kept nodding and neither of us was listening",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_051",
- "category": "numbers",
- "text": "At exactly seven forty-five, Morgan placed eighteen dollars and fifty cents beside the notebook.",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_051.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "At exactly 7.45, Morgan placed $18.50 beside the notebook.",
- "raw": {
- "normalized_reference": "at exactly seven forty-five, morgan placed eighteen dollars and fifty cents beside the notebook.",
- "normalized_hypothesis": "at exactly 7.45, morgan placed $18.50 beside the notebook.",
- "errors": 8,
- "words": 15,
- "wer": 0.5333333333333333
- },
- "semantic": {
- "normalized_reference": "at exactly 745 morgan placed $18.50 beside the notebook",
- "normalized_hypothesis": "at exactly 7.45 morgan placed $18.50 beside the notebook",
- "errors": 2,
- "words": 10,
- "wer": 0.2
- }
- },
- {
- "id": "utmos200_059",
- "category": "emotional",
- "text": "I didn't expect to feel that way about it, but I did.",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_059.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "I didn't expect to feel that way about it, but I did.",
- "raw": {
- "normalized_reference": "i didn't expect to feel that way about it, but i did.",
- "normalized_hypothesis": "i didn't expect to feel that way about it, but i did.",
- "errors": 0,
- "words": 12,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "i did not expect to feel that way about it but i did",
- "normalized_hypothesis": "i did not expect to feel that way about it but i did",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_067",
- "category": "question",
- "text": "Does it matter at this point, honestly?",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_067.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Does it matter at this point, honestly?",
- "raw": {
- "normalized_reference": "does it matter at this point, honestly?",
- "normalized_hypothesis": "does it matter at this point, honestly?",
- "errors": 0,
- "words": 7,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "does it matter at this point honestly",
- "normalized_hypothesis": "does it matter at this point honestly",
- "errors": 0,
- "words": 7,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_075",
- "category": "long",
- "text": "I've thought a lot about what I would do differently if I could go back, and the honest answer is probably not as much as I'd like to think.",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_075.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "I've thought a lot about what I would do differently if I could go back, and the honest answer is probably not as much as I'd like to think.",
- "raw": {
- "normalized_reference": "i've thought a lot about what i would do differently if i could go back, and the honest answer is probably not as much as i'd like to think.",
- "normalized_hypothesis": "i've thought a lot about what i would do differently if i could go back, and the honest answer is probably not as much as i'd like to think.",
- "errors": 0,
- "words": 29,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "i have thought a lot about what i would do differently if i could go back and the honest answer is probably not as much as i would like to think",
- "normalized_hypothesis": "i have thought a lot about what i would do differently if i could go back and the honest answer is probably not as much as i would like to think",
- "errors": 0,
- "words": 31,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_083",
- "category": "short",
- "text": "Are you serious right now?",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_083.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Are you serious right now?",
- "raw": {
- "normalized_reference": "are you serious right now?",
- "normalized_hypothesis": "are you serious right now?",
- "errors": 0,
- "words": 5,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "are you serious right now",
- "normalized_hypothesis": "are you serious right now",
- "errors": 0,
- "words": 5,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_091",
- "category": "short",
- "text": "That's fair.",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_091.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "That's fair.",
- "raw": {
- "normalized_reference": "that's fair.",
- "normalized_hypothesis": "that's fair.",
- "errors": 0,
- "words": 2,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "that is fair",
- "normalized_hypothesis": "that is fair",
- "errors": 0,
- "words": 3,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_099",
- "category": "short",
- "text": "I stand by what I said.",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_099.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "I stand by what I said.",
- "raw": {
- "normalized_reference": "i stand by what i said.",
- "normalized_hypothesis": "i stand by what i said.",
- "errors": 0,
- "words": 6,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "i stand by what i said",
- "normalized_hypothesis": "i stand by what i said",
- "errors": 0,
- "words": 6,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_107",
- "category": "short",
- "text": "I can't believe you just said that.",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_107.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "I can't believe you just said that.",
- "raw": {
- "normalized_reference": "i can't believe you just said that.",
- "normalized_hypothesis": "i can't believe you just said that.",
- "errors": 0,
- "words": 7,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "i can not believe you just said that",
- "normalized_hypothesis": "i can not believe you just said that",
- "errors": 0,
- "words": 8,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_115",
- "category": "short",
- "text": "Okay. Okay okay okay.",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_115.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Okay. Okay, okay, okay.",
- "raw": {
- "normalized_reference": "okay. okay okay okay.",
- "normalized_hypothesis": "okay. okay, okay, okay.",
- "errors": 0,
- "words": 4,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "okay okay okay okay",
- "normalized_hypothesis": "okay okay okay okay",
- "errors": 0,
- "words": 4,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_123",
- "category": "short",
- "text": "Hold on a second.",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_123.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Hold on a second.",
- "raw": {
- "normalized_reference": "hold on a second.",
- "normalized_hypothesis": "hold on a second.",
- "errors": 0,
- "words": 4,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "hold on a 2nd",
- "normalized_hypothesis": "hold on a 2nd",
- "errors": 0,
- "words": 4,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_131",
- "category": "short",
- "text": "Oh no. Oh no no no.",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_131.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Oh no. Oh no no no.",
- "raw": {
- "normalized_reference": "oh no. oh no no no.",
- "normalized_hypothesis": "oh no. oh no no no.",
- "errors": 0,
- "words": 6,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "0 no 0 no no no",
- "normalized_hypothesis": "0 no 0 no no no",
- "errors": 0,
- "words": 6,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_139",
- "category": "short",
- "text": "No way. Absolutely not.",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_139.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "No way. Absolutely not.",
- "raw": {
- "normalized_reference": "no way. absolutely not.",
- "normalized_hypothesis": "no way. absolutely not.",
- "errors": 0,
- "words": 4,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "no way absolutely not",
- "normalized_hypothesis": "no way absolutely not",
- "errors": 0,
- "words": 4,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_147",
- "category": "short",
- "text": "You have to be kidding me.",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_147.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "You have to be kidding me.",
- "raw": {
- "normalized_reference": "you have to be kidding me.",
- "normalized_hypothesis": "you have to be kidding me.",
- "errors": 0,
- "words": 6,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "you have to be kidding me",
- "normalized_hypothesis": "you have to be kidding me",
- "errors": 0,
- "words": 6,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_155",
- "category": "short",
- "text": "This is fine.",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_155.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "This is fine.",
- "raw": {
- "normalized_reference": "this is fine.",
- "normalized_hypothesis": "this is fine.",
- "errors": 0,
- "words": 3,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "this is fine",
- "normalized_hypothesis": "this is fine",
- "errors": 0,
- "words": 3,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_163",
- "category": "conversational",
- "text": "At some point you just have to commit and see what happens.",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_163.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "At some point you just have to commit and see what happens.",
- "raw": {
- "normalized_reference": "at some point you just have to commit and see what happens.",
- "normalized_hypothesis": "at some point you just have to commit and see what happens.",
- "errors": 0,
- "words": 12,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "at some point you just have to commit and see what happens",
- "normalized_hypothesis": "at some point you just have to commit and see what happens",
- "errors": 0,
- "words": 12,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_171",
- "category": "short",
- "text": "That's actually hilarious.",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_171.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "That's actually hilarious.",
- "raw": {
- "normalized_reference": "that's actually hilarious.",
- "normalized_hypothesis": "that's actually hilarious.",
- "errors": 0,
- "words": 3,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "that is actually hilarious",
- "normalized_hypothesis": "that is actually hilarious",
- "errors": 0,
- "words": 4,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_179",
- "category": "emotional",
- "text": "I have never felt more seen in my entire life than in that moment.",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_179.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "I have never felt more seen in my entire life than in that moment.",
- "raw": {
- "normalized_reference": "i have never felt more seen in my entire life than in that moment.",
- "normalized_hypothesis": "i have never felt more seen in my entire life than in that moment.",
- "errors": 0,
- "words": 14,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "i have never felt more seen in my entire life than in that moment",
- "normalized_hypothesis": "i have never felt more seen in my entire life than in that moment",
- "errors": 0,
- "words": 14,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_187",
- "category": "technical",
- "text": "Once it's configured correctly, it basically runs itself.",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_187.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Once it's configured correctly, it basically runs itself.",
- "raw": {
- "normalized_reference": "once it's configured correctly, it basically runs itself.",
- "normalized_hypothesis": "once it's configured correctly, it basically runs itself.",
- "errors": 0,
- "words": 8,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "once it is configured correctly it basically runs itself",
- "normalized_hypothesis": "once it is configured correctly it basically runs itself",
- "errors": 0,
- "words": 9,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_195",
- "category": "short",
- "text": "Bold of you to assume.",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_195.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Bold of you to assume.",
- "raw": {
- "normalized_reference": "bold of you to assume.",
- "normalized_hypothesis": "bold of you to assume.",
- "errors": 0,
- "words": 5,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "bold of you to assume",
- "normalized_hypothesis": "bold of you to assume",
- "errors": 0,
- "words": 5,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_004",
- "category": "homographs",
- "text": "After I record the record, please present the present to the bass player.",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_004.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "After I record the record, please present the present to the bass player.",
- "raw": {
- "normalized_reference": "after i record the record, please present the present to the bass player.",
- "normalized_hypothesis": "after i record the record, please present the present to the bass player.",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "after i record the record please present the present to the bass player",
- "normalized_hypothesis": "after i record the record please present the present to the bass player",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_012",
- "category": "rare_words",
- "text": "The archaeologist catalogued a Byzantine reliquary and several anachronistic fragments.",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_012.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "The archaeologist catalogued a Byzantine reliquary and several anachronistic fragments.",
- "raw": {
- "normalized_reference": "the archaeologist catalogued a byzantine reliquary and several anachronistic fragments.",
- "normalized_hypothesis": "the archaeologist catalogued a byzantine reliquary and several anachronistic fragments.",
- "errors": 0,
- "words": 10,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "the archaeologist catalogued a byzantine reliquary and several anachronistic fragments",
- "normalized_hypothesis": "the archaeologist catalogued a byzantine reliquary and several anachronistic fragments",
- "errors": 0,
- "words": 10,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_020",
- "category": "long_range",
- "text": "Before announcing the winner, the judge thanked the volunteers who had worked through the storm to repair the damaged stage.",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_020.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Before announcing the winner, the judge thanked the volunteers who had worked through the storm to repair the damaged stage.",
- "raw": {
- "normalized_reference": "before announcing the winner, the judge thanked the volunteers who had worked through the storm to repair the damaged stage.",
- "normalized_hypothesis": "before announcing the winner, the judge thanked the volunteers who had worked through the storm to repair the damaged stage.",
- "errors": 0,
- "words": 20,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "before announcing the winner the judge thanked the volunteers who had worked through the storm to repair the damaged stage",
- "normalized_hypothesis": "before announcing the winner the judge thanked the volunteers who had worked through the storm to repair the damaged stage",
- "errors": 0,
- "words": 20,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_028",
- "category": "technical",
- "text": "Hash the UTF eight payload with S H A two fifty-six before opening the checkpoint.",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_028.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Hash the UTF-8 payload with SHA-256 before opening the checkpoint.",
- "raw": {
- "normalized_reference": "hash the utf eight payload with s h a two fifty-six before opening the checkpoint.",
- "normalized_hypothesis": "hash the utf-8 payload with sha-256 before opening the checkpoint.",
- "errors": 7,
- "words": 16,
- "wer": 0.4375
- },
- "semantic": {
- "normalized_reference": "hash the utf 8 payload with s h a 256 before opening the checkpoint",
- "normalized_hypothesis": "hash the utf 8 payload with sha 256 before opening the checkpoint",
- "errors": 3,
- "words": 14,
- "wer": 0.21428571428571427
- }
- },
- {
- "id": "utmos200_036",
- "category": "narrative",
- "text": "The village had a name, but nobody used it. Everyone just called it home.",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_036.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "The village had a name, but nobody used it. Everyone just called it home.",
- "raw": {
- "normalized_reference": "the village had a name, but nobody used it. everyone just called it home.",
- "normalized_hypothesis": "the village had a name, but nobody used it. everyone just called it home.",
- "errors": 0,
- "words": 14,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "the village had a name but nobody used it everyone just called it home",
- "normalized_hypothesis": "the village had a name but nobody used it everyone just called it home",
- "errors": 0,
- "words": 14,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_044",
- "category": "descriptive",
- "text": "It's one of those places where everyone seems to know each other already.",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_044.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "It's one of those places where everyone seems to know each other already.",
- "raw": {
- "normalized_reference": "it's one of those places where everyone seems to know each other already.",
- "normalized_hypothesis": "it's one of those places where everyone seems to know each other already.",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "it is one of those places where everyone seems to know each other already",
- "normalized_hypothesis": "it is one of those places where everyone seems to know each other already",
- "errors": 0,
- "words": 14,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_052",
- "category": "punctuation",
- "text": "Does punctuation work? Yes: commas, semicolons, and periods should create sensible boundaries.",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_052.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Does punctuation work? Yes, commas, semicolons and periods should create sensible boundaries.",
- "raw": {
- "normalized_reference": "does punctuation work? yes: commas, semicolons, and periods should create sensible boundaries.",
- "normalized_hypothesis": "does punctuation work? yes, commas, semicolons and periods should create sensible boundaries.",
- "errors": 0,
- "words": 12,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "does punctuation work yes commas semicolons and periods should create sensible boundaries",
- "normalized_hypothesis": "does punctuation work yes commas semicolons and periods should create sensible boundaries",
- "errors": 0,
- "words": 12,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_060",
- "category": "homographs",
- "text": "I object to the strange object that was left near the entrance.",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_060.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "I object to the strange object that was left near the entrance.",
- "raw": {
- "normalized_reference": "i object to the strange object that was left near the entrance.",
- "normalized_hypothesis": "i object to the strange object that was left near the entrance.",
- "errors": 0,
- "words": 12,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "i object to the strange object that was left near the entrance",
- "normalized_hypothesis": "i object to the strange object that was left near the entrance",
- "errors": 0,
- "words": 12,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_068",
- "category": "rare_words",
- "text": "An entrepreneurial meteorologist described the unprecedented atmospheric phenomenon.",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_068.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "An entrepreneurial meteorologist described the unprecedented atmospheric phenomenon.",
- "raw": {
- "normalized_reference": "an entrepreneurial meteorologist described the unprecedented atmospheric phenomenon.",
- "normalized_hypothesis": "an entrepreneurial meteorologist described the unprecedented atmospheric phenomenon.",
- "errors": 0,
- "words": 8,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "an entrepreneurial meteorologist described the unprecedented atmospheric phenomenon",
- "normalized_hypothesis": "an entrepreneurial meteorologist described the unprecedented atmospheric phenomenon",
- "errors": 0,
- "words": 8,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_076",
- "category": "long_range",
- "text": "Although the weather changed without warning, the musicians finished packing before anyone opened the enormous glass door.",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_076.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Although the weather changed without warning, the musicians finished packing before anyone opened the enormous glass door.",
- "raw": {
- "normalized_reference": "although the weather changed without warning, the musicians finished packing before anyone opened the enormous glass door.",
- "normalized_hypothesis": "although the weather changed without warning, the musicians finished packing before anyone opened the enormous glass door.",
- "errors": 0,
- "words": 17,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "although the weather changed without warning the musicians finished packing before anyone opened the enormous glass door",
- "normalized_hypothesis": "although the weather changed without warning the musicians finished packing before anyone opened the enormous glass door",
- "errors": 0,
- "words": 17,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_084",
- "category": "technical",
- "text": "The performance improvement is real, but it comes with tradeoffs.",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_084.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "The performance improvement is real, but it comes with trade-offs.",
- "raw": {
- "normalized_reference": "the performance improvement is real, but it comes with tradeoffs.",
- "normalized_hypothesis": "the performance improvement is real, but it comes with trade-offs.",
- "errors": 2,
- "words": 10,
- "wer": 0.2
- },
- "semantic": {
- "normalized_reference": "the performance improvement is real but it comes with tradeoffs",
- "normalized_hypothesis": "the performance improvement is real but it comes with trade offs",
- "errors": 2,
- "words": 10,
- "wer": 0.2
- }
- },
- {
- "id": "utmos200_092",
- "category": "technical",
- "text": "If you run into that error, the fix is usually just restarting the service.",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_092.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "If you run into that error, the fix is usually just restarting the service.",
- "raw": {
- "normalized_reference": "if you run into that error, the fix is usually just restarting the service.",
- "normalized_hypothesis": "if you run into that error, the fix is usually just restarting the service.",
- "errors": 0,
- "words": 14,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "if you run into that error the fix is usually just restarting the service",
- "normalized_hypothesis": "if you run into that error the fix is usually just restarting the service",
- "errors": 0,
- "words": 14,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_100",
- "category": "technical",
- "text": "Double-check that value before you push anything to production.",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_100.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Double-check that value before you push anything to production.",
- "raw": {
- "normalized_reference": "double-check that value before you push anything to production.",
- "normalized_hypothesis": "double-check that value before you push anything to production.",
- "errors": 0,
- "words": 10,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "double check that value before you push anything to production",
- "normalized_hypothesis": "double check that value before you push anything to production",
- "errors": 0,
- "words": 10,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_108",
- "category": "technical",
- "text": "You'll want to isolate that variable before drawing any conclusions.",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_108.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "You'll want to isolate that variable before drawing any conclusions.",
- "raw": {
- "normalized_reference": "you'll want to isolate that variable before drawing any conclusions.",
- "normalized_hypothesis": "you'll want to isolate that variable before drawing any conclusions.",
- "errors": 0,
- "words": 10,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "you will want to isolate that variable before drawing any conclusions",
- "normalized_hypothesis": "you will want to isolate that variable before drawing any conclusions",
- "errors": 0,
- "words": 11,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_116",
- "category": "technical",
- "text": "The latency goes up significantly at scale, which is expected.",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_116.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "The latency goes up significantly at scale which is expected.",
- "raw": {
- "normalized_reference": "the latency goes up significantly at scale, which is expected.",
- "normalized_hypothesis": "the latency goes up significantly at scale which is expected.",
- "errors": 0,
- "words": 10,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "the latency goes up significantly at scale which is expected",
- "normalized_hypothesis": "the latency goes up significantly at scale which is expected",
- "errors": 0,
- "words": 10,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_124",
- "category": "technical",
- "text": "A deterministic split prevents duplicate normalized text from leaking into evaluation.",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_124.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "A deterministic split prevents duplicate normalized text from leaking into evaluation.",
- "raw": {
- "normalized_reference": "a deterministic split prevents duplicate normalized text from leaking into evaluation.",
- "normalized_hypothesis": "a deterministic split prevents duplicate normalized text from leaking into evaluation.",
- "errors": 0,
- "words": 11,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "a deterministic split prevents duplicate normalized text from leaking into evaluation",
- "normalized_hypothesis": "a deterministic split prevents duplicate normalized text from leaking into evaluation",
- "errors": 0,
- "words": 11,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_132",
- "category": "technical",
- "text": "Always test on a copy before modifying the original.",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_132.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Always test on a copy before modifying the original.",
- "raw": {
- "normalized_reference": "always test on a copy before modifying the original.",
- "normalized_hypothesis": "always test on a copy before modifying the original.",
- "errors": 0,
- "words": 9,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "always test on a copy before modifying the original",
- "normalized_hypothesis": "always test on a copy before modifying the original",
- "errors": 0,
- "words": 9,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_140",
- "category": "technical",
- "text": "The first thing you want to do is make sure everything is backed up.",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_140.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "The first thing you want to do is make sure everything is backed up.",
- "raw": {
- "normalized_reference": "the first thing you want to do is make sure everything is backed up.",
- "normalized_hypothesis": "the first thing you want to do is make sure everything is backed up.",
- "errors": 0,
- "words": 14,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "the 1st thing you want to do is make sure everything is backed up",
- "normalized_hypothesis": "the 1st thing you want to do is make sure everything is backed up",
- "errors": 0,
- "words": 14,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_148",
- "category": "technical",
- "text": "If the checksum fails, restart the worker process without deleting the cached tensors.",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_148.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "If the checksum fails, restart the worker process without deleting the cached tensors.",
- "raw": {
- "normalized_reference": "if the checksum fails, restart the worker process without deleting the cached tensors.",
- "normalized_hypothesis": "if the checksum fails, restart the worker process without deleting the cached tensors.",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "if the checksum fails restart the worker process without deleting the cached tensors",
- "normalized_hypothesis": "if the checksum fails restart the worker process without deleting the cached tensors",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_156",
- "category": "technical",
- "text": "The tricky part isn't the setup, it's making sure all dependencies line up.",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_156.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "The tricky part isn't the setup, it's making sure all dependencies line up.",
- "raw": {
- "normalized_reference": "the tricky part isn't the setup, it's making sure all dependencies line up.",
- "normalized_hypothesis": "the tricky part isn't the setup, it's making sure all dependencies line up.",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "the tricky part is not the setup it is making sure all dependencies line up",
- "normalized_hypothesis": "the tricky part is not the setup it is making sure all dependencies line up",
- "errors": 0,
- "words": 15,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_164",
- "category": "emotional",
- "text": "Nobody told me it was going to be this hard, and honestly, I'm grateful.",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_164.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Nobody told me it was going to be this hard and honestly, I'm grateful.",
- "raw": {
- "normalized_reference": "nobody told me it was going to be this hard, and honestly, i'm grateful.",
- "normalized_hypothesis": "nobody told me it was going to be this hard and honestly, i'm grateful.",
- "errors": 0,
- "words": 14,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "nobody told me it was going to be this hard and honestly i am grateful",
- "normalized_hypothesis": "nobody told me it was going to be this hard and honestly i am grateful",
- "errors": 0,
- "words": 15,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_172",
- "category": "technical",
- "text": "The encoder maps phonemes, stress, and phrase boundaries into a shared hidden representation.",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_172.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "The encoder maps phonemes, stress and phrase boundaries into a shared hidden representation.",
- "raw": {
- "normalized_reference": "the encoder maps phonemes, stress, and phrase boundaries into a shared hidden representation.",
- "normalized_hypothesis": "the encoder maps phonemes, stress and phrase boundaries into a shared hidden representation.",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "the encoder maps phonemes stress and phrase boundaries into a shared hidden representation",
- "normalized_hypothesis": "the encoder maps phonemes stress and phrase boundaries into a shared hidden representation",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_180",
- "category": "narrative",
- "text": "She knew. She had always known. She just waited for the right moment.",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_180.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "She knew. She had always known. She just waited for the right moment.",
- "raw": {
- "normalized_reference": "she knew. she had always known. she just waited for the right moment.",
- "normalized_hypothesis": "she knew. she had always known. she just waited for the right moment.",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "she knew she had always known she just waited for the right moment",
- "normalized_hypothesis": "she knew she had always known she just waited for the right moment",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_188",
- "category": "conversational",
- "text": "You got engaged? When? How? Tell me everything right now.",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_188.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "You got engaged. When? How? Tell me everything right now.",
- "raw": {
- "normalized_reference": "you got engaged? when? how? tell me everything right now.",
- "normalized_hypothesis": "you got engaged. when? how? tell me everything right now.",
- "errors": 0,
- "words": 10,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "you got engaged when how tell me everything right now",
- "normalized_hypothesis": "you got engaged when how tell me everything right now",
- "errors": 0,
- "words": 10,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_196",
- "category": "conversational",
- "text": "It's one of those decisions that makes total sense in the moment.",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_196.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "It's one of those decisions that makes total sense in the moment.",
- "raw": {
- "normalized_reference": "it's one of those decisions that makes total sense in the moment.",
- "normalized_hypothesis": "it's one of those decisions that makes total sense in the moment.",
- "errors": 0,
- "words": 12,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "it is one of those decisions that makes total sense in the moment",
- "normalized_hypothesis": "it is one of those decisions that makes total sense in the moment",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_005",
- "category": "long",
- "text": "If I'm being completely honest, I had no idea what I was doing for most of it, and I think that might be the most normal thing about the whole experience.",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_005.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "If I'm being completely honest, I had no idea what I was doing for most of it, and I think that might be the most normal thing about the whole experience.",
- "raw": {
- "normalized_reference": "if i'm being completely honest, i had no idea what i was doing for most of it, and i think that might be the most normal thing about the whole experience.",
- "normalized_hypothesis": "if i'm being completely honest, i had no idea what i was doing for most of it, and i think that might be the most normal thing about the whole experience.",
- "errors": 0,
- "words": 31,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "if i am being completely honest i had no idea what i was doing for most of it and i think that might be the most normal thing about the whole experience",
- "normalized_hypothesis": "if i am being completely honest i had no idea what i was doing for most of it and i think that might be the most normal thing about the whole experience",
- "errors": 0,
- "words": 32,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_013",
- "category": "short",
- "text": "I have no idea.",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_013.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "I have no idea.",
- "raw": {
- "normalized_reference": "i have no idea.",
- "normalized_hypothesis": "i have no idea.",
- "errors": 0,
- "words": 4,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "i have no idea",
- "normalized_hypothesis": "i have no idea",
- "errors": 0,
- "words": 4,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_021",
- "category": "names_places",
- "text": "Professor Nguyen mailed the fluorescent poster to Reykjavik on Thursday.",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_021.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Professor Nguyen mailed the fluorescent poster to Reykjavik on Thursday.",
- "raw": {
- "normalized_reference": "professor nguyen mailed the fluorescent poster to reykjavik on thursday.",
- "normalized_hypothesis": "professor nguyen mailed the fluorescent poster to reykjavik on thursday.",
- "errors": 0,
- "words": 10,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "professor nguyen mailed the fluorescent poster to reykjavik on thursday",
- "normalized_hypothesis": "professor nguyen mailed the fluorescent poster to reykjavik on thursday",
- "errors": 0,
- "words": 10,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_029",
- "category": "conversational",
- "text": "I thought I had it all figured out, and then this happened.",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_029.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "I thought I had it all figured out and then this happened.",
- "raw": {
- "normalized_reference": "i thought i had it all figured out, and then this happened.",
- "normalized_hypothesis": "i thought i had it all figured out and then this happened.",
- "errors": 0,
- "words": 12,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "i thought i had it all figured out and then this happened",
- "normalized_hypothesis": "i thought i had it all figured out and then this happened",
- "errors": 0,
- "words": 12,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_037",
- "category": "numbers",
- "text": "Increase the gain by three decibels, wait five seconds, then reduce it by one.",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_037.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Increase the gain by 3 decibels, wait 5 seconds, then reduce it by 1.",
- "raw": {
- "normalized_reference": "increase the gain by three decibels, wait five seconds, then reduce it by one.",
- "normalized_hypothesis": "increase the gain by 3 decibels, wait 5 seconds, then reduce it by 1.",
- "errors": 3,
- "words": 14,
- "wer": 0.21428571428571427
- },
- "semantic": {
- "normalized_reference": "increase the gain by 3 decibels wait 5 seconds then reduce it by one",
- "normalized_hypothesis": "increase the gain by 3 decibels wait 5 seconds then reduce it by one",
- "errors": 0,
- "words": 14,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_045",
- "category": "emotional",
- "text": "He said he was proud of me and nobody had ever said that to me before.",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_045.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "He said he was proud of me and nobody had ever said that to me before.",
- "raw": {
- "normalized_reference": "he said he was proud of me and nobody had ever said that to me before.",
- "normalized_hypothesis": "he said he was proud of me and nobody had ever said that to me before.",
- "errors": 0,
- "words": 16,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "he said he was proud of me and nobody had ever said that to me before",
- "normalized_hypothesis": "he said he was proud of me and nobody had ever said that to me before",
- "errors": 0,
- "words": 16,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_053",
- "category": "question",
- "text": "Have you ever just stopped and thought about how strange all of this is?",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_053.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Have you ever just stopped and thought about how strange all of this is?",
- "raw": {
- "normalized_reference": "have you ever just stopped and thought about how strange all of this is?",
- "normalized_hypothesis": "have you ever just stopped and thought about how strange all of this is?",
- "errors": 0,
- "words": 14,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "have you ever just stopped and thought about how strange all of this is",
- "normalized_hypothesis": "have you ever just stopped and thought about how strange all of this is",
- "errors": 0,
- "words": 14,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_061",
- "category": "long",
- "text": "I've learned more from the things that went wrong than from the things that worked, which is a frustrating way to have to learn things but apparently the only way that stuck.",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_061.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "I've learned more from the things that went wrong than from the things that worked, which is a frustrating way to have to learn things, but apparently the only way that stuck.",
- "raw": {
- "normalized_reference": "i've learned more from the things that went wrong than from the things that worked, which is a frustrating way to have to learn things but apparently the only way that stuck.",
- "normalized_hypothesis": "i've learned more from the things that went wrong than from the things that worked, which is a frustrating way to have to learn things, but apparently the only way that stuck.",
- "errors": 0,
- "words": 32,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "i have learned more from the things that went wrong than from the things that worked which is a frustrating way to have to learn things but apparently the only way that stuck",
- "normalized_hypothesis": "i have learned more from the things that went wrong than from the things that worked which is a frustrating way to have to learn things but apparently the only way that stuck",
- "errors": 0,
- "words": 33,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_069",
- "category": "short",
- "text": "That's not what I expected.",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_069.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "That's not what I expected.",
- "raw": {
- "normalized_reference": "that's not what i expected.",
- "normalized_hypothesis": "that's not what i expected.",
- "errors": 0,
- "words": 5,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "that is not what i expected",
- "normalized_hypothesis": "that is not what i expected",
- "errors": 0,
- "words": 6,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_077",
- "category": "names_places",
- "text": "Saskatchewan looked unusually bright beneath the pale February sunrise.",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_077.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Saskatchewan looked unusually bright beneath the pale February sunrise.",
- "raw": {
- "normalized_reference": "saskatchewan looked unusually bright beneath the pale february sunrise.",
- "normalized_hypothesis": "saskatchewan looked unusually bright beneath the pale february sunrise.",
- "errors": 0,
- "words": 9,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "saskatchewan looked unusually bright beneath the pale february sunrise",
- "normalized_hypothesis": "saskatchewan looked unusually bright beneath the pale february sunrise",
- "errors": 0,
- "words": 9,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_085",
- "category": "conversational",
- "text": "I don't know. It's just been one of those weeks, you know?",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_085.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "I don't know. It's just been one of those weeks, you know.",
- "raw": {
- "normalized_reference": "i don't know. it's just been one of those weeks, you know?",
- "normalized_hypothesis": "i don't know. it's just been one of those weeks, you know.",
- "errors": 0,
- "words": 12,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "i do not know it is just been one of those weeks you know",
- "normalized_hypothesis": "i do not know it is just been one of those weeks you know",
- "errors": 0,
- "words": 14,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_093",
- "category": "conversational",
- "text": "It was a disaster in the best possible way.",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_093.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "It was a disaster in the best possible way.",
- "raw": {
- "normalized_reference": "it was a disaster in the best possible way.",
- "normalized_hypothesis": "it was a disaster in the best possible way.",
- "errors": 0,
- "words": 9,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "it was a disaster in the best possible way",
- "normalized_hypothesis": "it was a disaster in the best possible way",
- "errors": 0,
- "words": 9,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_101",
- "category": "conversational",
- "text": "I understand your point; I just do not agree with the conclusion.",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_101.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "I understand your point, I just do not agree with the conclusion.",
- "raw": {
- "normalized_reference": "i understand your point; i just do not agree with the conclusion.",
- "normalized_hypothesis": "i understand your point, i just do not agree with the conclusion.",
- "errors": 0,
- "words": 12,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "i understand your point i just do not agree with the conclusion",
- "normalized_hypothesis": "i understand your point i just do not agree with the conclusion",
- "errors": 0,
- "words": 12,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_109",
- "category": "conversational",
- "text": "The meeting that could have been an email lasted two hours.",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_109.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "The meeting that could have been an email lasted two hours.",
- "raw": {
- "normalized_reference": "the meeting that could have been an email lasted two hours.",
- "normalized_hypothesis": "the meeting that could have been an email lasted two hours.",
- "errors": 0,
- "words": 11,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "the meeting that could have been an email lasted 2 hours",
- "normalized_hypothesis": "the meeting that could have been an email lasted 2 hours",
- "errors": 0,
- "words": 11,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_117",
- "category": "conversational",
- "text": "We ended up staying until two in the morning talking about nothing.",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_117.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "We ended up staying until two in the morning, talking about nothing.",
- "raw": {
- "normalized_reference": "we ended up staying until two in the morning talking about nothing.",
- "normalized_hypothesis": "we ended up staying until two in the morning, talking about nothing.",
- "errors": 0,
- "words": 12,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "we ended up staying until 2 in the morning talking about nothing",
- "normalized_hypothesis": "we ended up staying until 2 in the morning talking about nothing",
- "errors": 0,
- "words": 12,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_125",
- "category": "conversational",
- "text": "Honestly, that was much better than I expected, but the ending still felt rushed.",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_125.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Honestly, that was much better than expected, but the ending still felt rushed.",
- "raw": {
- "normalized_reference": "honestly, that was much better than i expected, but the ending still felt rushed.",
- "normalized_hypothesis": "honestly, that was much better than expected, but the ending still felt rushed.",
- "errors": 1,
- "words": 14,
- "wer": 0.07142857142857142
- },
- "semantic": {
- "normalized_reference": "honestly that was much better than i expected but the ending still felt rushed",
- "normalized_hypothesis": "honestly that was much better than expected but the ending still felt rushed",
- "errors": 1,
- "words": 14,
- "wer": 0.07142857142857142
- }
- },
- {
- "id": "utmos200_133",
- "category": "conversational",
- "text": "I wasn't even supposed to be there. That's the wild part.",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_133.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "I wasn't even supposed to be there. That's the wild part.",
- "raw": {
- "normalized_reference": "i wasn't even supposed to be there. that's the wild part.",
- "normalized_hypothesis": "i wasn't even supposed to be there. that's the wild part.",
- "errors": 0,
- "words": 11,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "i was not even supposed to be there that is the wild part",
- "normalized_hypothesis": "i was not even supposed to be there that is the wild part",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_141",
- "category": "conversational",
- "text": "I have the most insane idea and you need to just hear me out.",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_141.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "I have the most insane idea, and you need to just hear me out.",
- "raw": {
- "normalized_reference": "i have the most insane idea and you need to just hear me out.",
- "normalized_hypothesis": "i have the most insane idea, and you need to just hear me out.",
- "errors": 0,
- "words": 14,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "i have the most insane idea and you need to just hear me out",
- "normalized_hypothesis": "i have the most insane idea and you need to just hear me out",
- "errors": 0,
- "words": 14,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_149",
- "category": "conversational",
- "text": "I need you to understand how close that was to not working.",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_149.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "I need you to understand how close that was to not working.",
- "raw": {
- "normalized_reference": "i need you to understand how close that was to not working.",
- "normalized_hypothesis": "i need you to understand how close that was to not working.",
- "errors": 0,
- "words": 12,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "i need you to understand how close that was to not working",
- "normalized_hypothesis": "i need you to understand how close that was to not working",
- "errors": 0,
- "words": 12,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_157",
- "category": "conversational",
- "text": "He called it a minor setback. I called it a disaster.",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_157.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "He called it a minor setback. I called it a disaster.",
- "raw": {
- "normalized_reference": "he called it a minor setback. i called it a disaster.",
- "normalized_hypothesis": "he called it a minor setback. i called it a disaster.",
- "errors": 0,
- "words": 11,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "he called it a minor setback i called it a disaster",
- "normalized_hypothesis": "he called it a minor setback i called it a disaster",
- "errors": 0,
- "words": 11,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_165",
- "category": "narrative",
- "text": "Chapter seven. In which everything that could go wrong did, and a few things that couldn't.",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_165.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Chapter 7. In which everything that could go wrong did and a few things that couldn't.",
- "raw": {
- "normalized_reference": "chapter seven. in which everything that could go wrong did, and a few things that couldn't.",
- "normalized_hypothesis": "chapter 7. in which everything that could go wrong did and a few things that couldn't.",
- "errors": 1,
- "words": 16,
- "wer": 0.0625
- },
- "semantic": {
- "normalized_reference": "chapter 7 in which everything that could go wrong did and a few things that could not",
- "normalized_hypothesis": "chapter 7 in which everything that could go wrong did and a few things that could not",
- "errors": 0,
- "words": 17,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_173",
- "category": "conversational",
- "text": "She showed up three hours late and acted like nothing had happened.",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_173.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "She showed up three hours late and acted like nothing had happened.",
- "raw": {
- "normalized_reference": "she showed up three hours late and acted like nothing had happened.",
- "normalized_hypothesis": "she showed up three hours late and acted like nothing had happened.",
- "errors": 0,
- "words": 12,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "she showed up 3 hours late and acted like nothing had happened",
- "normalized_hypothesis": "she showed up 3 hours late and acted like nothing had happened",
- "errors": 0,
- "words": 12,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_181",
- "category": "short",
- "text": "Yeah, no. Hard pass.",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_181.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Yeth no. Hard pass.",
- "raw": {
- "normalized_reference": "yeah, no. hard pass.",
- "normalized_hypothesis": "yeth no. hard pass.",
- "errors": 1,
- "words": 4,
- "wer": 0.25
- },
- "semantic": {
- "normalized_reference": "yeah no hard pass",
- "normalized_hypothesis": "yeth no hard pass",
- "errors": 1,
- "words": 4,
- "wer": 0.25
- }
- },
- {
- "id": "utmos200_189",
- "category": "emotional",
- "text": "It's hard to explain how much that moment meant to me.",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_189.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "It's hard to explain how much that moment meant to me.",
- "raw": {
- "normalized_reference": "it's hard to explain how much that moment meant to me.",
- "normalized_hypothesis": "it's hard to explain how much that moment meant to me.",
- "errors": 0,
- "words": 11,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "it is hard to explain how much that moment meant to me",
- "normalized_hypothesis": "it is hard to explain how much that moment meant to me",
- "errors": 0,
- "words": 12,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_197",
- "category": "emotional",
- "text": "I was furious. Not the loud kind. The quiet kind that lasts for days.",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_197.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "I was furious. Not the loud kind. The quiet kind that lasts for days.",
- "raw": {
- "normalized_reference": "i was furious. not the loud kind. the quiet kind that lasts for days.",
- "normalized_hypothesis": "i was furious. not the loud kind. the quiet kind that lasts for days.",
- "errors": 0,
- "words": 14,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "i was furious not the loud kind the quiet kind that lasts for days",
- "normalized_hypothesis": "i was furious not the loud kind the quiet kind that lasts for days",
- "errors": 0,
- "words": 14,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_006",
- "category": "long_range",
- "text": "The committee postponed its decision after discovering that three pages of the report had been printed in the wrong order.",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_006.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "The committee postponed its decision after discovering that three pages of the report had been printed in the wrong order.",
- "raw": {
- "normalized_reference": "the committee postponed its decision after discovering that three pages of the report had been printed in the wrong order.",
- "normalized_hypothesis": "the committee postponed its decision after discovering that three pages of the report had been printed in the wrong order.",
- "errors": 0,
- "words": 20,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "the committee postponed its decision after discovering that 3 pages of the report had been printed in the wrong order",
- "normalized_hypothesis": "the committee postponed its decision after discovering that 3 pages of the report had been printed in the wrong order",
- "errors": 0,
- "words": 20,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_014",
- "category": "technical",
- "text": "The system runs on three core components that all have to stay in sync.",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_014.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "The system runs on three core components that all have to stay in sync.",
- "raw": {
- "normalized_reference": "the system runs on three core components that all have to stay in sync.",
- "normalized_hypothesis": "the system runs on three core components that all have to stay in sync.",
- "errors": 0,
- "words": 14,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "the system runs on 3 core components that all have to stay in sync",
- "normalized_hypothesis": "the system runs on 3 core components that all have to stay in sync",
- "errors": 0,
- "words": 14,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_022",
- "category": "narrative",
- "text": "The plan was simple. The execution was anything but.",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_022.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "The plan was simple. The execution was anything but.",
- "raw": {
- "normalized_reference": "the plan was simple. the execution was anything but.",
- "normalized_hypothesis": "the plan was simple. the execution was anything but.",
- "errors": 0,
- "words": 9,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "the plan was simple the execution was anything but",
- "normalized_hypothesis": "the plan was simple the execution was anything but",
- "errors": 0,
- "words": 9,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_030",
- "category": "descriptive",
- "text": "The kind of quiet you only get at three in the morning in a city.",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_030.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "The kind of quiet you only get at three in the morning in a city.",
- "raw": {
- "normalized_reference": "the kind of quiet you only get at three in the morning in a city.",
- "normalized_hypothesis": "the kind of quiet you only get at three in the morning in a city.",
- "errors": 0,
- "words": 15,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "the kind of quiet you only get at 3 in the morning in a city",
- "normalized_hypothesis": "the kind of quiet you only get at 3 in the morning in a city",
- "errors": 0,
- "words": 15,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_038",
- "category": "punctuation",
- "text": "The result was unexpected: clear, stable, and surprisingly natural.",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_038.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "The result was unexpected, clear, stable and surprisingly natural.",
- "raw": {
- "normalized_reference": "the result was unexpected: clear, stable, and surprisingly natural.",
- "normalized_hypothesis": "the result was unexpected, clear, stable and surprisingly natural.",
- "errors": 0,
- "words": 9,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "the result was unexpected clear stable and surprisingly natural",
- "normalized_hypothesis": "the result was unexpected clear stable and surprisingly natural",
- "errors": 0,
- "words": 9,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_046",
- "category": "homographs",
- "text": "The minute detail took a minute to explain clearly.",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_046.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "The minute detail took a minute to explain clearly.",
- "raw": {
- "normalized_reference": "the minute detail took a minute to explain clearly.",
- "normalized_hypothesis": "the minute detail took a minute to explain clearly.",
- "errors": 0,
- "words": 9,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "the minute detail took a minute to explain clearly",
- "normalized_hypothesis": "the minute detail took a minute to explain clearly",
- "errors": 0,
- "words": 9,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_054",
- "category": "rare_words",
- "text": "Her idiosyncratic explanation sounded plausible but remained fundamentally ambiguous.",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_054.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Her idiosyncratic explanation sounded plausible but remained fundamentally ambiguous.",
- "raw": {
- "normalized_reference": "her idiosyncratic explanation sounded plausible but remained fundamentally ambiguous.",
- "normalized_hypothesis": "her idiosyncratic explanation sounded plausible but remained fundamentally ambiguous.",
- "errors": 0,
- "words": 9,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "her idiosyncratic explanation sounded plausible but remained fundamentally ambiguous",
- "normalized_hypothesis": "her idiosyncratic explanation sounded plausible but remained fundamentally ambiguous",
- "errors": 0,
- "words": 9,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_062",
- "category": "long_range",
- "text": "Because every nearby hotel was full, the exhausted hikers followed the narrow road until they found a farmhouse with one empty room.",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_062.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Because every nearby hotel was full, the exhausted hikers followed the narrow road until they found a farmhouse with one empty room.",
- "raw": {
- "normalized_reference": "because every nearby hotel was full, the exhausted hikers followed the narrow road until they found a farmhouse with one empty room.",
- "normalized_hypothesis": "because every nearby hotel was full, the exhausted hikers followed the narrow road until they found a farmhouse with one empty room.",
- "errors": 0,
- "words": 22,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "because every nearby hotel was full the exhausted hikers followed the narrow road until they found a farmhouse with one empty room",
- "normalized_hypothesis": "because every nearby hotel was full the exhausted hikers followed the narrow road until they found a farmhouse with one empty room",
- "errors": 0,
- "words": 22,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_070",
- "category": "technical",
- "text": "There's a known issue with that version that causes failures under load.",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_070.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "There's a known issue with that version that causes failures under load.",
- "raw": {
- "normalized_reference": "there's a known issue with that version that causes failures under load.",
- "normalized_hypothesis": "there's a known issue with that version that causes failures under load.",
- "errors": 0,
- "words": 12,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "there is a known issue with that version that causes failures under load",
- "normalized_hypothesis": "there is a known issue with that version that causes failures under load",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_078",
- "category": "narrative",
- "text": "She handled it with a grace I genuinely did not expect.",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_078.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "She handled it with a grace I genuinely did not expect.",
- "raw": {
- "normalized_reference": "she handled it with a grace i genuinely did not expect.",
- "normalized_hypothesis": "she handled it with a grace i genuinely did not expect.",
- "errors": 0,
- "words": 11,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "she handled it with a grace i genuinely did not expect",
- "normalized_hypothesis": "she handled it with a grace i genuinely did not expect",
- "errors": 0,
- "words": 11,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_086",
- "category": "descriptive",
- "text": "The moment you stop expecting perfection, things get a lot more interesting.",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_086.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "The moment you stop expecting perfection, things get a lot more interesting.",
- "raw": {
- "normalized_reference": "the moment you stop expecting perfection, things get a lot more interesting.",
- "normalized_hypothesis": "the moment you stop expecting perfection, things get a lot more interesting.",
- "errors": 0,
- "words": 12,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "the moment you stop expecting perfection things get a lot more interesting",
- "normalized_hypothesis": "the moment you stop expecting perfection things get a lot more interesting",
- "errors": 0,
- "words": 12,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_094",
- "category": "descriptive",
- "text": "It had the look of a place that used to mean something to someone.",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_094.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "It had the look of a place that used to mean something to someone.",
- "raw": {
- "normalized_reference": "it had the look of a place that used to mean something to someone.",
- "normalized_hypothesis": "it had the look of a place that used to mean something to someone.",
- "errors": 0,
- "words": 14,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "it had the look of a place that used to mean something to someone",
- "normalized_hypothesis": "it had the look of a place that used to mean something to someone",
- "errors": 0,
- "words": 14,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_102",
- "category": "descriptive",
- "text": "The sky that day was a strange shade of orange I'd never seen before.",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_102.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "The sky that day was a strange shade of orange I'd never seen before.",
- "raw": {
- "normalized_reference": "the sky that day was a strange shade of orange i'd never seen before.",
- "normalized_hypothesis": "the sky that day was a strange shade of orange i'd never seen before.",
- "errors": 0,
- "words": 14,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "the sky that day was a strange shade of orange i would never seen before",
- "normalized_hypothesis": "the sky that day was a strange shade of orange i would never seen before",
- "errors": 0,
- "words": 15,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_110",
- "category": "descriptive",
- "text": "There's a specific kind of tired that sleep doesn't fix.",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_110.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "There's a specific kind of tired that sleep doesn't fix.",
- "raw": {
- "normalized_reference": "there's a specific kind of tired that sleep doesn't fix.",
- "normalized_hypothesis": "there's a specific kind of tired that sleep doesn't fix.",
- "errors": 0,
- "words": 10,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "there is a specific kind of tired that sleep does not fix",
- "normalized_hypothesis": "there is a specific kind of tired that sleep does not fix",
- "errors": 0,
- "words": 12,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_118",
- "category": "descriptive",
- "text": "The rain started slowly, and then all at once, the way it usually does.",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_118.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "The rain started slowly and then all at once, the way it usually does.",
- "raw": {
- "normalized_reference": "the rain started slowly, and then all at once, the way it usually does.",
- "normalized_hypothesis": "the rain started slowly and then all at once, the way it usually does.",
- "errors": 0,
- "words": 14,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "the rain started slowly and then all at once the way it usually does",
- "normalized_hypothesis": "the rain started slowly and then all at once the way it usually does",
- "errors": 0,
- "words": 14,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_126",
- "category": "descriptive",
- "text": "She had a habit of tilting her head slightly when thinking hard about something.",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_126.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "She had a habit of tilting her head slightly when thinking hard about something.",
- "raw": {
- "normalized_reference": "she had a habit of tilting her head slightly when thinking hard about something.",
- "normalized_hypothesis": "she had a habit of tilting her head slightly when thinking hard about something.",
- "errors": 0,
- "words": 14,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "she had a habit of tilting her head slightly when thinking hard about something",
- "normalized_hypothesis": "she had a habit of tilting her head slightly when thinking hard about something",
- "errors": 0,
- "words": 14,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_134",
- "category": "descriptive",
- "text": "It's a small town. Not the kind you leave. The kind you leave and come back to.",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_134.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "It's a small town. Not the kind you leave. The kind you leave and come back to.",
- "raw": {
- "normalized_reference": "it's a small town. not the kind you leave. the kind you leave and come back to.",
- "normalized_hypothesis": "it's a small town. not the kind you leave. the kind you leave and come back to.",
- "errors": 0,
- "words": 17,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "it is a small town not the kind you leave the kind you leave and come back to",
- "normalized_hypothesis": "it is a small town not the kind you leave the kind you leave and come back to",
- "errors": 0,
- "words": 18,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_142",
- "category": "descriptive",
- "text": "There's a coffee shop on the corner that's been there longer than anyone can remember.",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_142.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "There's a coffee shop on the corner that's been there longer than anyone can remember.",
- "raw": {
- "normalized_reference": "there's a coffee shop on the corner that's been there longer than anyone can remember.",
- "normalized_hypothesis": "there's a coffee shop on the corner that's been there longer than anyone can remember.",
- "errors": 0,
- "words": 15,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "there is a coffee shop on the corner that has been there longer than anyone can remember",
- "normalized_hypothesis": "there is a coffee shop on the corner that has been there longer than anyone can remember",
- "errors": 0,
- "words": 17,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_150",
- "category": "descriptive",
- "text": "It was the sort of morning where the whole world looks rinsed clean.",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_150.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "It was the sort of morning where the whole world looks rinsed clean.",
- "raw": {
- "normalized_reference": "it was the sort of morning where the whole world looks rinsed clean.",
- "normalized_hypothesis": "it was the sort of morning where the whole world looks rinsed clean.",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "it was the sort of morning where the whole world looks rinsed clean",
- "normalized_hypothesis": "it was the sort of morning where the whole world looks rinsed clean",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_158",
- "category": "descriptive",
- "text": "The whole place smelled like pine and old wood and cold air.",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_158.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "The whole place smelled like pine and old wood and cold air.",
- "raw": {
- "normalized_reference": "the whole place smelled like pine and old wood and cold air.",
- "normalized_hypothesis": "the whole place smelled like pine and old wood and cold air.",
- "errors": 0,
- "words": 12,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "the whole place smelled like pine and old wood and cold air",
- "normalized_hypothesis": "the whole place smelled like pine and old wood and cold air",
- "errors": 0,
- "words": 12,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_166",
- "category": "short",
- "text": "Wait, what?",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_166.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Wait, what?",
- "raw": {
- "normalized_reference": "wait, what?",
- "normalized_hypothesis": "wait, what?",
- "errors": 0,
- "words": 2,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "wait what",
- "normalized_hypothesis": "wait what",
- "errors": 0,
- "words": 2,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_174",
- "category": "emotional",
- "text": "Oh my god. This is the best thing I've ever tasted.",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_174.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Oh my god. This is the best thing I've ever tasted.",
- "raw": {
- "normalized_reference": "oh my god. this is the best thing i've ever tasted.",
- "normalized_hypothesis": "oh my god. this is the best thing i've ever tasted.",
- "errors": 0,
- "words": 11,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "0 my god this is the best thing i have ever tasted",
- "normalized_hypothesis": "0 my god this is the best thing i have ever tasted",
- "errors": 0,
- "words": 12,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_182",
- "category": "technical",
- "text": "Please verify the sample rate, hop length, and mel filter configuration before inference.",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_182.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Please verify the sample rate, hop length, and MEL filter configuration before inference.",
- "raw": {
- "normalized_reference": "please verify the sample rate, hop length, and mel filter configuration before inference.",
- "normalized_hypothesis": "please verify the sample rate, hop length, and mel filter configuration before inference.",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "please verify the sample rate hop length and mel filter configuration before inference",
- "normalized_hypothesis": "please verify the sample rate hop length and mel filter configuration before inference",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_190",
- "category": "narrative",
- "text": "He told the story three different times that night and somehow it got better each time.",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_190.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "He told the story three different times that night, and somehow it got better each time.",
- "raw": {
- "normalized_reference": "he told the story three different times that night and somehow it got better each time.",
- "normalized_hypothesis": "he told the story three different times that night, and somehow it got better each time.",
- "errors": 0,
- "words": 16,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "he told the story 3 different times that night and somehow it got better each time",
- "normalized_hypothesis": "he told the story 3 different times that night and somehow it got better each time",
- "errors": 0,
- "words": 16,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_198",
- "category": "narrative",
- "text": "That's the part of the story I usually leave out.",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_198.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "That's the part of the story I usually leave out.",
- "raw": {
- "normalized_reference": "that's the part of the story i usually leave out.",
- "normalized_hypothesis": "that's the part of the story i usually leave out.",
- "errors": 0,
- "words": 10,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "that is the part of the story i usually leave out",
- "normalized_hypothesis": "that is the part of the story i usually leave out",
- "errors": 0,
- "words": 11,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_007",
- "category": "names_places",
- "text": "Gwendolyn photographed the eucalyptus trees outside Ljubljana.",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_007.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Gwendolyn photographed the eucalyptus trees outside Ljubljana.",
- "raw": {
- "normalized_reference": "gwendolyn photographed the eucalyptus trees outside ljubljana.",
- "normalized_hypothesis": "gwendolyn photographed the eucalyptus trees outside ljubljana.",
- "errors": 0,
- "words": 7,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "gwendolyn photographed the eucalyptus trees outside ljubljana",
- "normalized_hypothesis": "gwendolyn photographed the eucalyptus trees outside ljubljana",
- "errors": 0,
- "words": 7,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_015",
- "category": "conversational",
- "text": "We improvised the entire second half and somehow nobody noticed.",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_015.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "We improvised the entire second half, and somehow nobody noticed.",
- "raw": {
- "normalized_reference": "we improvised the entire second half and somehow nobody noticed.",
- "normalized_hypothesis": "we improvised the entire second half, and somehow nobody noticed.",
- "errors": 0,
- "words": 10,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "we improvised the entire 2nd half and somehow nobody noticed",
- "normalized_hypothesis": "we improvised the entire 2nd half and somehow nobody noticed",
- "errors": 0,
- "words": 10,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_023",
- "category": "numbers",
- "text": "Her score improved from eighty-two percent to ninety-seven percent in two weeks.",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_023.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Her score improved from 82% to 97% in two weeks.",
- "raw": {
- "normalized_reference": "her score improved from eighty-two percent to ninety-seven percent in two weeks.",
- "normalized_hypothesis": "her score improved from 82% to 97% in two weeks.",
- "errors": 6,
- "words": 14,
- "wer": 0.42857142857142855
- },
- "semantic": {
- "normalized_reference": "her score improved from 82% to 97% in 2 weeks",
- "normalized_hypothesis": "her score improved from 82% to 97% in 2 weeks",
- "errors": 0,
- "words": 10,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_031",
- "category": "emotional",
- "text": "The joy of it was almost overwhelming, like too much good all at once.",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_031.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "The joy of it was almost overwhelming, like too much good all at once.",
- "raw": {
- "normalized_reference": "the joy of it was almost overwhelming, like too much good all at once.",
- "normalized_hypothesis": "the joy of it was almost overwhelming, like too much good all at once.",
- "errors": 0,
- "words": 14,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "the joy of it was almost overwhelming like too much good all at once",
- "normalized_hypothesis": "the joy of it was almost overwhelming like too much good all at once",
- "errors": 0,
- "words": 14,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_039",
- "category": "question",
- "text": "Who told you that was the right way to do it?",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_039.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Who told you that was the right way to do it?",
- "raw": {
- "normalized_reference": "who told you that was the right way to do it?",
- "normalized_hypothesis": "who told you that was the right way to do it?",
- "errors": 0,
- "words": 11,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "who told you that was the right way to do it",
- "normalized_hypothesis": "who told you that was the right way to do it",
- "errors": 0,
- "words": 11,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_047",
- "category": "long",
- "text": "The strange thing about really wanting something is that getting it never quite feels the way you imagined it would, and not getting it never quite feels as bad.",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_047.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "The strange thing about really wanting something is that getting it never quite feels the way you imagined it would, and not getting it never quite feels as bad.",
- "raw": {
- "normalized_reference": "the strange thing about really wanting something is that getting it never quite feels the way you imagined it would, and not getting it never quite feels as bad.",
- "normalized_hypothesis": "the strange thing about really wanting something is that getting it never quite feels the way you imagined it would, and not getting it never quite feels as bad.",
- "errors": 0,
- "words": 29,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "the strange thing about really wanting something is that getting it never quite feels the way you imagined it would and not getting it never quite feels as bad",
- "normalized_hypothesis": "the strange thing about really wanting something is that getting it never quite feels the way you imagined it would and not getting it never quite feels as bad",
- "errors": 0,
- "words": 29,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_055",
- "category": "short",
- "text": "That tracks.",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_055.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "That tracks.",
- "raw": {
- "normalized_reference": "that tracks.",
- "normalized_hypothesis": "that tracks.",
- "errors": 0,
- "words": 2,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "that tracks",
- "normalized_hypothesis": "that tracks",
- "errors": 0,
- "words": 2,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_063",
- "category": "names_places",
- "text": "The itinerary sends Beatrice from Albuquerque to Des Moines and then to Boise.",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_063.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "The itinerary sends Beatrice from Albuquerque to Des Moines and then to Boise.",
- "raw": {
- "normalized_reference": "the itinerary sends beatrice from albuquerque to des moines and then to boise.",
- "normalized_hypothesis": "the itinerary sends beatrice from albuquerque to des moines and then to boise.",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "the itinerary sends beatrice from albuquerque to des moines and then to boise",
- "normalized_hypothesis": "the itinerary sends beatrice from albuquerque to des moines and then to boise",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_071",
- "category": "conversational",
- "text": "Don't look now but I think that's him over there.",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_071.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Don't look now but I think that's him over there.",
- "raw": {
- "normalized_reference": "don't look now but i think that's him over there.",
- "normalized_hypothesis": "don't look now but i think that's him over there.",
- "errors": 0,
- "words": 10,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "do not look now but i think that is him over there",
- "normalized_hypothesis": "do not look now but i think that is him over there",
- "errors": 0,
- "words": 12,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_079",
- "category": "numbers",
- "text": "Apartment four hundred seven is across from room twenty-nine on the third floor.",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_079.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Apartment 407 is across from room 29 on the third floor.",
- "raw": {
- "normalized_reference": "apartment four hundred seven is across from room twenty-nine on the third floor.",
- "normalized_hypothesis": "apartment 407 is across from room 29 on the third floor.",
- "errors": 5,
- "words": 14,
- "wer": 0.35714285714285715
- },
- "semantic": {
- "normalized_reference": "apartment 407 is across from room 29 on the 3rd floor",
- "normalized_hypothesis": "apartment 407 is across from room 29 on the 3rd floor",
- "errors": 0,
- "words": 11,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_087",
- "category": "emotional",
- "text": "It's the kind of thing that stays with you even years later.",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_087.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "It's the kind of thing that stays with you even years later.",
- "raw": {
- "normalized_reference": "it's the kind of thing that stays with you even years later.",
- "normalized_hypothesis": "it's the kind of thing that stays with you even years later.",
- "errors": 0,
- "words": 12,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "it is the kind of thing that stays with you even years later",
- "normalized_hypothesis": "it is the kind of thing that stays with you even years later",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_095",
- "category": "emotional",
- "text": "There's this feeling you get when something finally works after weeks of failure.",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_095.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "There's this feeling you get when something finally works after weeks of failure.",
- "raw": {
- "normalized_reference": "there's this feeling you get when something finally works after weeks of failure.",
- "normalized_hypothesis": "there's this feeling you get when something finally works after weeks of failure.",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "there is this feeling you get when something finally works after weeks of failure",
- "normalized_hypothesis": "there is this feeling you get when something finally works after weeks of failure",
- "errors": 0,
- "words": 14,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_103",
- "category": "emotional",
- "text": "It hurt more than I expected, and I thought I was prepared.",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_103.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "It hurt more than I expected and I thought I was prepared.",
- "raw": {
- "normalized_reference": "it hurt more than i expected, and i thought i was prepared.",
- "normalized_hypothesis": "it hurt more than i expected and i thought i was prepared.",
- "errors": 0,
- "words": 12,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "it hurt more than i expected and i thought i was prepared",
- "normalized_hypothesis": "it hurt more than i expected and i thought i was prepared",
- "errors": 0,
- "words": 12,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_111",
- "category": "emotional",
- "text": "Something about that moment broke something open in me that I hadn't realized was closed.",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_111.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Something about that moment broke something open in me that I hadn't realized was closed.",
- "raw": {
- "normalized_reference": "something about that moment broke something open in me that i hadn't realized was closed.",
- "normalized_hypothesis": "something about that moment broke something open in me that i hadn't realized was closed.",
- "errors": 0,
- "words": 15,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "something about that moment broke something open in me that i had not realized was closed",
- "normalized_hypothesis": "something about that moment broke something open in me that i had not realized was closed",
- "errors": 0,
- "words": 16,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_119",
- "category": "emotional",
- "text": "I honestly did not think I was going to make it through that year.",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_119.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "I honestly did not think I was going to make it through that year.",
- "raw": {
- "normalized_reference": "i honestly did not think i was going to make it through that year.",
- "normalized_hypothesis": "i honestly did not think i was going to make it through that year.",
- "errors": 0,
- "words": 14,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "i honestly did not think i was going to make it through that year",
- "normalized_hypothesis": "i honestly did not think i was going to make it through that year",
- "errors": 0,
- "words": 14,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_127",
- "category": "emotional",
- "text": "I cannot stress enough how much I did not want to be right about this.",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_127.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "I cannot stress enough how much I did not want to be right about this.",
- "raw": {
- "normalized_reference": "i cannot stress enough how much i did not want to be right about this.",
- "normalized_hypothesis": "i cannot stress enough how much i did not want to be right about this.",
- "errors": 0,
- "words": 15,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "i cannot stress enough how much i did not want to be right about this",
- "normalized_hypothesis": "i cannot stress enough how much i did not want to be right about this",
- "errors": 0,
- "words": 15,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_135",
- "category": "emotional",
- "text": "She cried. I cried. We all just sort of stood there crying.",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_135.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "She cried. I cried. We all just sort of stood there crying.",
- "raw": {
- "normalized_reference": "she cried. i cried. we all just sort of stood there crying.",
- "normalized_hypothesis": "she cried. i cried. we all just sort of stood there crying.",
- "errors": 0,
- "words": 12,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "she cried i cried we all just sort of stood there crying",
- "normalized_hypothesis": "she cried i cried we all just sort of stood there crying",
- "errors": 0,
- "words": 12,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_143",
- "category": "emotional",
- "text": "I kept it together until I got to the car, and then I completely fell apart.",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_143.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "I kept it together until I got to the car and then I completely fell apart.",
- "raw": {
- "normalized_reference": "i kept it together until i got to the car, and then i completely fell apart.",
- "normalized_hypothesis": "i kept it together until i got to the car and then i completely fell apart.",
- "errors": 0,
- "words": 16,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "i kept it together until i got to the car and then i completely fell apart",
- "normalized_hypothesis": "i kept it together until i got to the car and then i completely fell apart",
- "errors": 0,
- "words": 16,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_151",
- "category": "emotional",
- "text": "She started laughing and I started laughing and we couldn't stop for a full minute.",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_151.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "She started laughing and I started laughing and we couldn't stop for a full minute.",
- "raw": {
- "normalized_reference": "she started laughing and i started laughing and we couldn't stop for a full minute.",
- "normalized_hypothesis": "she started laughing and i started laughing and we couldn't stop for a full minute.",
- "errors": 0,
- "words": 15,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "she started laughing and i started laughing and we could not stop for a full minute",
- "normalized_hypothesis": "she started laughing and i started laughing and we could not stop for a full minute",
- "errors": 0,
- "words": 16,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_159",
- "category": "emotional",
- "text": "I've never been so relieved in my entire life.",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_159.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "I've never been so relieved in my entire life.",
- "raw": {
- "normalized_reference": "i've never been so relieved in my entire life.",
- "normalized_hypothesis": "i've never been so relieved in my entire life.",
- "errors": 0,
- "words": 9,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "i have never been so relieved in my entire life",
- "normalized_hypothesis": "i have never been so relieved in my entire life",
- "errors": 0,
- "words": 10,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_167",
- "category": "technical",
- "text": "The graphics processor remained idle because the data loader exhausted every central processing core.",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_167.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "The graphics processor remained idle because the data loader exhausted every central processing core.",
- "raw": {
- "normalized_reference": "the graphics processor remained idle because the data loader exhausted every central processing core.",
- "normalized_hypothesis": "the graphics processor remained idle because the data loader exhausted every central processing core.",
- "errors": 0,
- "words": 14,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "the graphics processor remained idle because the data loader exhausted every central processing core",
- "normalized_hypothesis": "the graphics processor remained idle because the data loader exhausted every central processing core",
- "errors": 0,
- "words": 14,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_175",
- "category": "narrative",
- "text": "The note was short, which was somehow more frightening than a long one would have been.",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_175.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "The note was short, which was somehow more fright, ing than a long one would have been.",
- "raw": {
- "normalized_reference": "the note was short, which was somehow more frightening than a long one would have been.",
- "normalized_hypothesis": "the note was short, which was somehow more fright, ing than a long one would have been.",
- "errors": 2,
- "words": 16,
- "wer": 0.125
- },
- "semantic": {
- "normalized_reference": "the note was short which was somehow more frightening than a long one would have been",
- "normalized_hypothesis": "the note was short which was somehow more fright ing than a long one would have been",
- "errors": 2,
- "words": 16,
- "wer": 0.125
- }
- },
- {
- "id": "utmos200_183",
- "category": "conversational",
- "text": "That's the worst idea I've ever heard and I'm completely in.",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_183.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "That's the worst idea I've ever heard, and I'm completely in.",
- "raw": {
- "normalized_reference": "that's the worst idea i've ever heard and i'm completely in.",
- "normalized_hypothesis": "that's the worst idea i've ever heard, and i'm completely in.",
- "errors": 0,
- "words": 11,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "that is the worst idea i have ever heard and i am completely in",
- "normalized_hypothesis": "that is the worst idea i have ever heard and i am completely in",
- "errors": 0,
- "words": 14,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_191",
- "category": "short",
- "text": "Right. Moving on.",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_191.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Right. Moving on.",
- "raw": {
- "normalized_reference": "right. moving on.",
- "normalized_hypothesis": "right. moving on.",
- "errors": 0,
- "words": 3,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "right moving on",
- "normalized_hypothesis": "right moving on",
- "errors": 0,
- "words": 3,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_199",
- "category": "short",
- "text": "I've said too much.",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_199.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "I've said too much.",
- "raw": {
- "normalized_reference": "i've said too much.",
- "normalized_hypothesis": "i've said too much.",
- "errors": 0,
- "words": 4,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "i have said too much",
- "normalized_hypothesis": "i have said too much",
- "errors": 0,
- "words": 5,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_008",
- "category": "narrative",
- "text": "He had this theory, and the more time passes, the more I think he was right.",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_008.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "He had this theory and the more time passes the more I think he was right.",
- "raw": {
- "normalized_reference": "he had this theory, and the more time passes, the more i think he was right.",
- "normalized_hypothesis": "he had this theory and the more time passes the more i think he was right.",
- "errors": 0,
- "words": 16,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "he had this theory and the more time passes the more i think he was right",
- "normalized_hypothesis": "he had this theory and the more time passes the more i think he was right",
- "errors": 0,
- "words": 16,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_016",
- "category": "descriptive",
- "text": "There's a smell in old buildings that's impossible to describe but immediately familiar.",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_016.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "There's a smell in old buildings that's impossible to describe but immediately familiar.",
- "raw": {
- "normalized_reference": "there's a smell in old buildings that's impossible to describe but immediately familiar.",
- "normalized_hypothesis": "there's a smell in old buildings that's impossible to describe but immediately familiar.",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "there is a smell in old buildings that is impossible to describe but immediately familiar",
- "normalized_hypothesis": "there is a smell in old buildings that is impossible to describe but immediately familiar",
- "errors": 0,
- "words": 15,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_024",
- "category": "punctuation",
- "text": "Why would anyone pack spoons, batteries, and a violin for one short trip?",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_024.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Why would anyone pack spoons, batteries and a violin for one short trip?",
- "raw": {
- "normalized_reference": "why would anyone pack spoons, batteries, and a violin for one short trip?",
- "normalized_hypothesis": "why would anyone pack spoons, batteries and a violin for one short trip?",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "why would anyone pack spoons batteries and a violin for one short trip",
- "normalized_hypothesis": "why would anyone pack spoons batteries and a violin for one short trip",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_032",
- "category": "homographs",
- "text": "Please close the door while I stand close to the microphone.",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_032.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Please close the door while I stand close to the microphone.",
- "raw": {
- "normalized_reference": "please close the door while i stand close to the microphone.",
- "normalized_hypothesis": "please close the door while i stand close to the microphone.",
- "errors": 0,
- "words": 11,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "please close the door while i stand close to the microphone",
- "normalized_hypothesis": "please close the door while i stand close to the microphone",
- "errors": 0,
- "words": 11,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_040",
- "category": "rare_words",
- "text": "The meticulous cartographer examined an iridescent archipelago through binoculars.",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_040.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "The meticulous cartographer examined an iridescent archipelago through binoculars.",
- "raw": {
- "normalized_reference": "the meticulous cartographer examined an iridescent archipelago through binoculars.",
- "normalized_hypothesis": "the meticulous cartographer examined an iridescent archipelago through binoculars.",
- "errors": 0,
- "words": 9,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "the meticulous cartographer examined an iridescent archipelago through binoculars",
- "normalized_hypothesis": "the meticulous cartographer examined an iridescent archipelago through binoculars",
- "errors": 0,
- "words": 9,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_048",
- "category": "long_range",
- "text": "Even though the recipe looked simple at first, preparing the sauce required patience, careful timing, and a much larger pan.",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_048.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Even though the recipe looked simple at first, preparing the sauce required patience, careful timing, and a much larger pan.",
- "raw": {
- "normalized_reference": "even though the recipe looked simple at first, preparing the sauce required patience, careful timing, and a much larger pan.",
- "normalized_hypothesis": "even though the recipe looked simple at first, preparing the sauce required patience, careful timing, and a much larger pan.",
- "errors": 0,
- "words": 20,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "even though the recipe looked simple at 1st preparing the sauce required patience careful timing and a much larger pan",
- "normalized_hypothesis": "even though the recipe looked simple at 1st preparing the sauce required patience careful timing and a much larger pan",
- "errors": 0,
- "words": 20,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_056",
- "category": "technical",
- "text": "The key thing to remember is that order matters here more than you'd think.",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_056.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "The key thing to remember is that order matters here more than you'd think.",
- "raw": {
- "normalized_reference": "the key thing to remember is that order matters here more than you'd think.",
- "normalized_hypothesis": "the key thing to remember is that order matters here more than you'd think.",
- "errors": 0,
- "words": 14,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "the key thing to remember is that order matters here more than you would think",
- "normalized_hypothesis": "the key thing to remember is that order matters here more than you would think",
- "errors": 0,
- "words": 15,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_064",
- "category": "narrative",
- "text": "We had been driving for six hours when we finally saw the sign for the town.",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_064.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "We had been driving for six hours when we finally saw the sign for the town.",
- "raw": {
- "normalized_reference": "we had been driving for six hours when we finally saw the sign for the town.",
- "normalized_hypothesis": "we had been driving for six hours when we finally saw the sign for the town.",
- "errors": 0,
- "words": 16,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "we had been driving for 6 hours when we finally saw the sign for the town",
- "normalized_hypothesis": "we had been driving for 6 hours when we finally saw the sign for the town",
- "errors": 0,
- "words": 16,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_072",
- "category": "descriptive",
- "text": "The city at night looks completely different from how it feels during the day.",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_072.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "The city at night looks completely different from how it feels during the day.",
- "raw": {
- "normalized_reference": "the city at night looks completely different from how it feels during the day.",
- "normalized_hypothesis": "the city at night looks completely different from how it feels during the day.",
- "errors": 0,
- "words": 14,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "the city at night looks completely different from how it feels during the day",
- "normalized_hypothesis": "the city at night looks completely different from how it feels during the day",
- "errors": 0,
- "words": 14,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_080",
- "category": "punctuation",
- "text": "\"Not yet,\" Rowan whispered. \"Give me another minute... please.\"",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_080.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Not yet, Rowan whispered. Give me another minute. Please.",
- "raw": {
- "normalized_reference": "\"not yet,\" rowan whispered. \"give me another minute... please.\"",
- "normalized_hypothesis": "not yet, rowan whispered. give me another minute. please.",
- "errors": 0,
- "words": 9,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "not yet rowan whispered give me another minute please",
- "normalized_hypothesis": "not yet rowan whispered give me another minute please",
- "errors": 0,
- "words": 9,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_088",
- "category": "long",
- "text": "She told me once that the most important skill she'd developed was learning to disagree with someone without making them feel they were wrong.",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_088.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "She told me once that the most important skill she'd developed was learning to disagree with someone without making them feel they were wrong.",
- "raw": {
- "normalized_reference": "she told me once that the most important skill she'd developed was learning to disagree with someone without making them feel they were wrong.",
- "normalized_hypothesis": "she told me once that the most important skill she'd developed was learning to disagree with someone without making them feel they were wrong.",
- "errors": 0,
- "words": 24,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "she told me once that the most important skill she would developed was learning to disagree with someone without making them feel they were wrong",
- "normalized_hypothesis": "she told me once that the most important skill she would developed was learning to disagree with someone without making them feel they were wrong",
- "errors": 0,
- "words": 25,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_096",
- "category": "long",
- "text": "He had a theory that most arguments aren't actually about what they seem to be about, and the longer I live the more I think he was onto something.",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_096.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "He had a theory that most arguments aren't actually about what they seem to be about, and the longer I live the more I think he was onto something.",
- "raw": {
- "normalized_reference": "he had a theory that most arguments aren't actually about what they seem to be about, and the longer i live the more i think he was onto something.",
- "normalized_hypothesis": "he had a theory that most arguments aren't actually about what they seem to be about, and the longer i live the more i think he was onto something.",
- "errors": 0,
- "words": 29,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "he had a theory that most arguments are not actually about what they seem to be about and the longer i live the more i think he was onto something",
- "normalized_hypothesis": "he had a theory that most arguments are not actually about what they seem to be about and the longer i live the more i think he was onto something",
- "errors": 0,
- "words": 30,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_104",
- "category": "long",
- "text": "The hardest part wasn't making the decision, it was sitting with the uncertainty of not knowing whether it was the right one for months afterward.",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_104.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "The hardest part was, making the decision, it was sitting with the uncertainty of not knowing whether it was the right one for months afterward.",
- "raw": {
- "normalized_reference": "the hardest part wasn't making the decision, it was sitting with the uncertainty of not knowing whether it was the right one for months afterward.",
- "normalized_hypothesis": "the hardest part was, making the decision, it was sitting with the uncertainty of not knowing whether it was the right one for months afterward.",
- "errors": 1,
- "words": 25,
- "wer": 0.04
- },
- "semantic": {
- "normalized_reference": "the hardest part was not making the decision it was sitting with the uncertainty of not knowing whether it was the right one for months afterward",
- "normalized_hypothesis": "the hardest part was making the decision it was sitting with the uncertainty of not knowing whether it was the right one for months afterward",
- "errors": 1,
- "words": 26,
- "wer": 0.038461538461538464
- }
- },
- {
- "id": "utmos200_112",
- "category": "long",
- "text": "The thing about making something good is that you don't usually know it's good while you're making it. You just know you're not done yet.",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_112.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "The thing about making something good is that you don't usually know it's good while you're making it. You just know you're not done yet.",
- "raw": {
- "normalized_reference": "the thing about making something good is that you don't usually know it's good while you're making it. you just know you're not done yet.",
- "normalized_hypothesis": "the thing about making something good is that you don't usually know it's good while you're making it. you just know you're not done yet.",
- "errors": 0,
- "words": 25,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "the thing about making something good is that you do not usually know it is good while you are making it you just know you are not done yet",
- "normalized_hypothesis": "the thing about making something good is that you do not usually know it is good while you are making it you just know you are not done yet",
- "errors": 0,
- "words": 29,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_120",
- "category": "long",
- "text": "There's an assumption that people who seem confident all the time actually feel confident, and in my experience that's almost never true.",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_120.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "There's an assumption that people who seem confident all the time actually feel confident, and in my experience that's almost never true.",
- "raw": {
- "normalized_reference": "there's an assumption that people who seem confident all the time actually feel confident, and in my experience that's almost never true.",
- "normalized_hypothesis": "there's an assumption that people who seem confident all the time actually feel confident, and in my experience that's almost never true.",
- "errors": 0,
- "words": 22,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "there is an assumption that people who seem confident all the time actually feel confident and in my experience that is almost never true",
- "normalized_hypothesis": "there is an assumption that people who seem confident all the time actually feel confident and in my experience that is almost never true",
- "errors": 0,
- "words": 24,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_128",
- "category": "long",
- "text": "I used to think the goal was to get to a point where nothing was hard anymore, and now I think that point doesn't exist and I'm not sure I'd want it to.",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_128.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "I used to think the goal was to get to a point where nothing was hard anymore, and now I think that point doesn't exist, and I'm not sure I'd want it to.",
- "raw": {
- "normalized_reference": "i used to think the goal was to get to a point where nothing was hard anymore, and now i think that point doesn't exist and i'm not sure i'd want it to.",
- "normalized_hypothesis": "i used to think the goal was to get to a point where nothing was hard anymore, and now i think that point doesn't exist, and i'm not sure i'd want it to.",
- "errors": 0,
- "words": 33,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "i used to think the goal was to get to a point where nothing was hard anymore and now i think that point does not exist and i am not sure i would want it to",
- "normalized_hypothesis": "i used to think the goal was to get to a point where nothing was hard anymore and now i think that point does not exist and i am not sure i would want it to",
- "errors": 0,
- "words": 36,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_136",
- "category": "long",
- "text": "There's a version of being busy that's actually just a very elaborate way of avoiding the one thing you know you should be doing.",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_136.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "There's a version of being busy that's actually just a very elaborate way of avoiding the one thing you know you should be doing.",
- "raw": {
- "normalized_reference": "there's a version of being busy that's actually just a very elaborate way of avoiding the one thing you know you should be doing.",
- "normalized_hypothesis": "there's a version of being busy that's actually just a very elaborate way of avoiding the one thing you know you should be doing.",
- "errors": 0,
- "words": 24,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "there is a version of being busy that is actually just a very elaborate way of avoiding the one thing you know you should be doing",
- "normalized_hypothesis": "there is a version of being busy that is actually just a very elaborate way of avoiding the one thing you know you should be doing",
- "errors": 0,
- "words": 26,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_144",
- "category": "long",
- "text": "The older I get, the more I think that most of what we call wisdom is just pattern recognition built from having been wrong enough times to start noticing.",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_144.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "The older I get, the more I think that most of what we call wisdom is just pattern recognition built from having been wrong enough times to start noticing.",
- "raw": {
- "normalized_reference": "the older i get, the more i think that most of what we call wisdom is just pattern recognition built from having been wrong enough times to start noticing.",
- "normalized_hypothesis": "the older i get, the more i think that most of what we call wisdom is just pattern recognition built from having been wrong enough times to start noticing.",
- "errors": 0,
- "words": 29,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "the older i get the more i think that most of what we call wisdom is just pattern recognition built from having been wrong enough times to start noticing",
- "normalized_hypothesis": "the older i get the more i think that most of what we call wisdom is just pattern recognition built from having been wrong enough times to start noticing",
- "errors": 0,
- "words": 29,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_152",
- "category": "long",
- "text": "We spent three years building something we were genuinely proud of, and then the market shifted and none of it mattered the way we thought it would.",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_152.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "We spent three years building something we were genuinely proud of, and then the market shifted and none of it mattered the way we thought it would.",
- "raw": {
- "normalized_reference": "we spent three years building something we were genuinely proud of, and then the market shifted and none of it mattered the way we thought it would.",
- "normalized_hypothesis": "we spent three years building something we were genuinely proud of, and then the market shifted and none of it mattered the way we thought it would.",
- "errors": 0,
- "words": 27,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "we spent 3 years building something we were genuinely proud of and then the market shifted and none of it mattered the way we thought it would",
- "normalized_hypothesis": "we spent 3 years building something we were genuinely proud of and then the market shifted and none of it mattered the way we thought it would",
- "errors": 0,
- "words": 27,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_160",
- "category": "narrative",
- "text": "I wrote it all down that night so I wouldn't forget, and I still haven't looked at it.",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_160.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "I wrote it all down that night so I wouldn't forget, and I still haven't looked at it.",
- "raw": {
- "normalized_reference": "i wrote it all down that night so i wouldn't forget, and i still haven't looked at it.",
- "normalized_hypothesis": "i wrote it all down that night so i wouldn't forget, and i still haven't looked at it.",
- "errors": 0,
- "words": 18,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "i wrote it all down that night so i would not forget and i still have not looked at it",
- "normalized_hypothesis": "i wrote it all down that night so i would not forget and i still have not looked at it",
- "errors": 0,
- "words": 20,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_168",
- "category": "conversational",
- "text": "Could you speak a little more slowly? I am trying to write every detail down.",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_168.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Could you speak a little more slowly? I am trying to write every detail down.",
- "raw": {
- "normalized_reference": "could you speak a little more slowly? i am trying to write every detail down.",
- "normalized_hypothesis": "could you speak a little more slowly? i am trying to write every detail down.",
- "errors": 0,
- "words": 15,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "could you speak a little more slowly i am trying to write every detail down",
- "normalized_hypothesis": "could you speak a little more slowly i am trying to write every detail down",
- "errors": 0,
- "words": 15,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_176",
- "category": "short",
- "text": "Noted.",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_176.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "Noted.",
- "raw": {
- "normalized_reference": "noted.",
- "normalized_hypothesis": "noted.",
- "errors": 0,
- "words": 1,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "noted",
- "normalized_hypothesis": "noted",
- "errors": 0,
- "words": 1,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_184",
- "category": "emotional",
- "text": "I remember thinking: this is it. This is the thing I've been waiting for.",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_184.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "I remember thinking this is it. This is the thing I've been waiting for.",
- "raw": {
- "normalized_reference": "i remember thinking: this is it. this is the thing i've been waiting for.",
- "normalized_hypothesis": "i remember thinking this is it. this is the thing i've been waiting for.",
- "errors": 0,
- "words": 14,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "i remember thinking this is it this is the thing i have been waiting for",
- "normalized_hypothesis": "i remember thinking this is it this is the thing i have been waiting for",
- "errors": 0,
- "words": 15,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_192",
- "category": "conversational",
- "text": "No, I did not move your keys; check beside the small ceramic bowl.",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_192.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "No, I did not move your keys, check beside the small ceramic bowl.",
- "raw": {
- "normalized_reference": "no, i did not move your keys; check beside the small ceramic bowl.",
- "normalized_hypothesis": "no, i did not move your keys, check beside the small ceramic bowl.",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "no i did not move your keys check beside the small ceramic bowl",
- "normalized_hypothesis": "no i did not move your keys check beside the small ceramic bowl",
- "errors": 0,
- "words": 13,
- "wer": 0.0
- }
- },
- {
- "id": "utmos200_200",
- "category": "conversational",
- "text": "I've been faking that I understood this the entire time.",
- "model": "kitten-nano-bruno",
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_200.wav",
- "asr": "whisper-large-v3",
- "hypothesis": "I've been faking that I understood this the entire time.",
- "raw": {
- "normalized_reference": "i've been faking that i understood this the entire time.",
- "normalized_hypothesis": "i've been faking that i understood this the entire time.",
- "errors": 0,
- "words": 10,
- "wer": 0.0
- },
- "semantic": {
- "normalized_reference": "i have been faking that i understood this the entire time",
- "normalized_hypothesis": "i have been faking that i understood this the entire time",
- "errors": 0,
- "words": 11,
- "wer": 0.0
- }
- }
- ]
-}
diff --git a/evaluation/asr_failures.json b/evaluation/asr_failures.json
deleted file mode 100644
index e06d15ee9748e3591d0176395615b906b59bd3c2..0000000000000000000000000000000000000000
--- a/evaluation/asr_failures.json
+++ /dev/null
@@ -1,1961 +0,0 @@
-{
- "format": "inflect_tts_asr_failure_taxonomy_v1",
- "source_report": "release_polish_overnight_20260720/final_large200/asr/report.json",
- "asr": "whisper-large-v3",
- "systems": {
- "inflect-micro-v2-finalist": {
- "utterances": 200,
- "reference_words": 2608,
- "semantic_wer": 0.012269938650306749,
- "substitution_rate": 0.008819018404907975,
- "deletion_rate": 0.0011503067484662577,
- "insertion_rate": 0.0023006134969325155,
- "utterance_failure_rate": 0.105,
- "severe_failure_rate": 0.02,
- "empty_hypothesis_rate": 0.0,
- "leading_omission_rate": 0.0,
- "ending_omission_rate": 0.0,
- "insertion_burst_rate": 0.0,
- "adjacent_phrase_repetition_rate": 0.02
- },
- "inflect-nano-v2-finalist": {
- "utterances": 200,
- "reference_words": 2608,
- "semantic_wer": 0.015337423312883436,
- "substitution_rate": 0.009585889570552147,
- "deletion_rate": 0.0011503067484662577,
- "insertion_rate": 0.004601226993865031,
- "utterance_failure_rate": 0.115,
- "severe_failure_rate": 0.03,
- "empty_hypothesis_rate": 0.0,
- "leading_omission_rate": 0.005,
- "ending_omission_rate": 0.0,
- "insertion_burst_rate": 0.0,
- "adjacent_phrase_repetition_rate": 0.02
- },
- "kitten-nano-bruno": {
- "utterances": 200,
- "reference_words": 2608,
- "semantic_wer": 0.009585889570552147,
- "substitution_rate": 0.005368098159509203,
- "deletion_rate": 0.0023006134969325155,
- "insertion_rate": 0.0019171779141104294,
- "utterance_failure_rate": 0.075,
- "severe_failure_rate": 0.01,
- "empty_hypothesis_rate": 0.0,
- "leading_omission_rate": 0.0,
- "ending_omission_rate": 0.005,
- "insertion_burst_rate": 0.0,
- "adjacent_phrase_repetition_rate": 0.02
- },
- "piper-ryan-low": {
- "utterances": 200,
- "reference_words": 2608,
- "semantic_wer": 0.012269938650306749,
- "substitution_rate": 0.009585889570552147,
- "deletion_rate": 0.0011503067484662577,
- "insertion_rate": 0.0015337423312883436,
- "utterance_failure_rate": 0.095,
- "severe_failure_rate": 0.02,
- "empty_hypothesis_rate": 0.0,
- "leading_omission_rate": 0.0,
- "ending_omission_rate": 0.0,
- "insertion_burst_rate": 0.0,
- "adjacent_phrase_repetition_rate": 0.02
- }
- },
- "systems_by_category": {
- "inflect-micro-v2-finalist::conversational": {
- "utterances": 25,
- "reference_words": 309,
- "semantic_wer": 0.0,
- "substitution_rate": 0.0,
- "deletion_rate": 0.0,
- "insertion_rate": 0.0,
- "utterance_failure_rate": 0.0,
- "severe_failure_rate": 0.0,
- "empty_hypothesis_rate": 0.0,
- "leading_omission_rate": 0.0,
- "ending_omission_rate": 0.0,
- "insertion_burst_rate": 0.0,
- "adjacent_phrase_repetition_rate": 0.0
- },
- "inflect-micro-v2-finalist::descriptive": {
- "utterances": 16,
- "reference_words": 226,
- "semantic_wer": 0.0,
- "substitution_rate": 0.0,
- "deletion_rate": 0.0,
- "insertion_rate": 0.0,
- "utterance_failure_rate": 0.0,
- "severe_failure_rate": 0.0,
- "empty_hypothesis_rate": 0.0,
- "leading_omission_rate": 0.0,
- "ending_omission_rate": 0.0,
- "insertion_burst_rate": 0.0,
- "adjacent_phrase_repetition_rate": 0.0625
- },
- "inflect-micro-v2-finalist::emotional": {
- "utterances": 24,
- "reference_words": 329,
- "semantic_wer": 0.0,
- "substitution_rate": 0.0,
- "deletion_rate": 0.0,
- "insertion_rate": 0.0,
- "utterance_failure_rate": 0.0,
- "severe_failure_rate": 0.0,
- "empty_hypothesis_rate": 0.0,
- "leading_omission_rate": 0.0,
- "ending_omission_rate": 0.0,
- "insertion_burst_rate": 0.0,
- "adjacent_phrase_repetition_rate": 0.041666666666666664
- },
- "inflect-micro-v2-finalist::homographs": {
- "utterances": 6,
- "reference_words": 70,
- "semantic_wer": 0.014285714285714285,
- "substitution_rate": 0.014285714285714285,
- "deletion_rate": 0.0,
- "insertion_rate": 0.0,
- "utterance_failure_rate": 0.16666666666666666,
- "severe_failure_rate": 0.0,
- "empty_hypothesis_rate": 0.0,
- "leading_omission_rate": 0.0,
- "ending_omission_rate": 0.0,
- "insertion_burst_rate": 0.0,
- "adjacent_phrase_repetition_rate": 0.0
- },
- "inflect-micro-v2-finalist::long": {
- "utterances": 15,
- "reference_words": 432,
- "semantic_wer": 0.0023148148148148147,
- "substitution_rate": 0.0,
- "deletion_rate": 0.0023148148148148147,
- "insertion_rate": 0.0,
- "utterance_failure_rate": 0.06666666666666667,
- "severe_failure_rate": 0.0,
- "empty_hypothesis_rate": 0.0,
- "leading_omission_rate": 0.0,
- "ending_omission_rate": 0.0,
- "insertion_burst_rate": 0.0,
- "adjacent_phrase_repetition_rate": 0.0
- },
- "inflect-micro-v2-finalist::long_range": {
- "utterances": 6,
- "reference_words": 121,
- "semantic_wer": 0.0,
- "substitution_rate": 0.0,
- "deletion_rate": 0.0,
- "insertion_rate": 0.0,
- "utterance_failure_rate": 0.0,
- "severe_failure_rate": 0.0,
- "empty_hypothesis_rate": 0.0,
- "leading_omission_rate": 0.0,
- "ending_omission_rate": 0.0,
- "insertion_burst_rate": 0.0,
- "adjacent_phrase_repetition_rate": 0.0
- },
- "inflect-micro-v2-finalist::names_places": {
- "utterances": 6,
- "reference_words": 57,
- "semantic_wer": 0.07017543859649122,
- "substitution_rate": 0.07017543859649122,
- "deletion_rate": 0.0,
- "insertion_rate": 0.0,
- "utterance_failure_rate": 0.5,
- "severe_failure_rate": 0.0,
- "empty_hypothesis_rate": 0.0,
- "leading_omission_rate": 0.0,
- "ending_omission_rate": 0.0,
- "insertion_burst_rate": 0.0,
- "adjacent_phrase_repetition_rate": 0.0
- },
- "inflect-micro-v2-finalist::narrative": {
- "utterances": 24,
- "reference_words": 352,
- "semantic_wer": 0.002840909090909091,
- "substitution_rate": 0.002840909090909091,
- "deletion_rate": 0.0,
- "insertion_rate": 0.0,
- "utterance_failure_rate": 0.041666666666666664,
- "severe_failure_rate": 0.0,
- "empty_hypothesis_rate": 0.0,
- "leading_omission_rate": 0.0,
- "ending_omission_rate": 0.0,
- "insertion_burst_rate": 0.0,
- "adjacent_phrase_repetition_rate": 0.0
- },
- "inflect-micro-v2-finalist::numbers": {
- "utterances": 6,
- "reference_words": 64,
- "semantic_wer": 0.15625,
- "substitution_rate": 0.125,
- "deletion_rate": 0.0,
- "insertion_rate": 0.03125,
- "utterance_failure_rate": 0.8333333333333334,
- "severe_failure_rate": 0.3333333333333333,
- "empty_hypothesis_rate": 0.0,
- "leading_omission_rate": 0.0,
- "ending_omission_rate": 0.0,
- "insertion_burst_rate": 0.0,
- "adjacent_phrase_repetition_rate": 0.0
- },
- "inflect-micro-v2-finalist::punctuation": {
- "utterances": 6,
- "reference_words": 65,
- "semantic_wer": 0.0,
- "substitution_rate": 0.0,
- "deletion_rate": 0.0,
- "insertion_rate": 0.0,
- "utterance_failure_rate": 0.0,
- "severe_failure_rate": 0.0,
- "empty_hypothesis_rate": 0.0,
- "leading_omission_rate": 0.0,
- "ending_omission_rate": 0.0,
- "insertion_burst_rate": 0.0,
- "adjacent_phrase_repetition_rate": 0.0
- },
- "inflect-micro-v2-finalist::question": {
- "utterances": 15,
- "reference_words": 165,
- "semantic_wer": 0.0,
- "substitution_rate": 0.0,
- "deletion_rate": 0.0,
- "insertion_rate": 0.0,
- "utterance_failure_rate": 0.0,
- "severe_failure_rate": 0.0,
- "empty_hypothesis_rate": 0.0,
- "leading_omission_rate": 0.0,
- "ending_omission_rate": 0.0,
- "insertion_burst_rate": 0.0,
- "adjacent_phrase_repetition_rate": 0.0
- },
- "inflect-micro-v2-finalist::rare_words": {
- "utterances": 6,
- "reference_words": 57,
- "semantic_wer": 0.05263157894736842,
- "substitution_rate": 0.03508771929824561,
- "deletion_rate": 0.0,
- "insertion_rate": 0.017543859649122806,
- "utterance_failure_rate": 0.3333333333333333,
- "severe_failure_rate": 0.0,
- "empty_hypothesis_rate": 0.0,
- "leading_omission_rate": 0.0,
- "ending_omission_rate": 0.0,
- "insertion_burst_rate": 0.0,
- "adjacent_phrase_repetition_rate": 0.0
- },
- "inflect-micro-v2-finalist::short": {
- "utterances": 24,
- "reference_words": 105,
- "semantic_wer": 0.0380952380952381,
- "substitution_rate": 0.01904761904761905,
- "deletion_rate": 0.0,
- "insertion_rate": 0.01904761904761905,
- "utterance_failure_rate": 0.16666666666666666,
- "severe_failure_rate": 0.041666666666666664,
- "empty_hypothesis_rate": 0.0,
- "leading_omission_rate": 0.0,
- "ending_omission_rate": 0.0,
- "insertion_burst_rate": 0.0,
- "adjacent_phrase_repetition_rate": 0.08333333333333333
- },
- "inflect-micro-v2-finalist::technical": {
- "utterances": 21,
- "reference_words": 256,
- "semantic_wer": 0.03125,
- "substitution_rate": 0.01953125,
- "deletion_rate": 0.0078125,
- "insertion_rate": 0.00390625,
- "utterance_failure_rate": 0.19047619047619047,
- "severe_failure_rate": 0.047619047619047616,
- "empty_hypothesis_rate": 0.0,
- "leading_omission_rate": 0.0,
- "ending_omission_rate": 0.0,
- "insertion_burst_rate": 0.0,
- "adjacent_phrase_repetition_rate": 0.0
- },
- "inflect-nano-v2-finalist::conversational": {
- "utterances": 25,
- "reference_words": 309,
- "semantic_wer": 0.003236245954692557,
- "substitution_rate": 0.0,
- "deletion_rate": 0.0,
- "insertion_rate": 0.003236245954692557,
- "utterance_failure_rate": 0.04,
- "severe_failure_rate": 0.0,
- "empty_hypothesis_rate": 0.0,
- "leading_omission_rate": 0.0,
- "ending_omission_rate": 0.0,
- "insertion_burst_rate": 0.0,
- "adjacent_phrase_repetition_rate": 0.0
- },
- "inflect-nano-v2-finalist::descriptive": {
- "utterances": 16,
- "reference_words": 226,
- "semantic_wer": 0.0,
- "substitution_rate": 0.0,
- "deletion_rate": 0.0,
- "insertion_rate": 0.0,
- "utterance_failure_rate": 0.0,
- "severe_failure_rate": 0.0,
- "empty_hypothesis_rate": 0.0,
- "leading_omission_rate": 0.0,
- "ending_omission_rate": 0.0,
- "insertion_burst_rate": 0.0,
- "adjacent_phrase_repetition_rate": 0.0625
- },
- "inflect-nano-v2-finalist::emotional": {
- "utterances": 24,
- "reference_words": 329,
- "semantic_wer": 0.0,
- "substitution_rate": 0.0,
- "deletion_rate": 0.0,
- "insertion_rate": 0.0,
- "utterance_failure_rate": 0.0,
- "severe_failure_rate": 0.0,
- "empty_hypothesis_rate": 0.0,
- "leading_omission_rate": 0.0,
- "ending_omission_rate": 0.0,
- "insertion_burst_rate": 0.0,
- "adjacent_phrase_repetition_rate": 0.041666666666666664
- },
- "inflect-nano-v2-finalist::homographs": {
- "utterances": 6,
- "reference_words": 70,
- "semantic_wer": 0.05714285714285714,
- "substitution_rate": 0.05714285714285714,
- "deletion_rate": 0.0,
- "insertion_rate": 0.0,
- "utterance_failure_rate": 0.3333333333333333,
- "severe_failure_rate": 0.0,
- "empty_hypothesis_rate": 0.0,
- "leading_omission_rate": 0.0,
- "ending_omission_rate": 0.0,
- "insertion_burst_rate": 0.0,
- "adjacent_phrase_repetition_rate": 0.0
- },
- "inflect-nano-v2-finalist::long": {
- "utterances": 15,
- "reference_words": 432,
- "semantic_wer": 0.004629629629629629,
- "substitution_rate": 0.0023148148148148147,
- "deletion_rate": 0.0,
- "insertion_rate": 0.0023148148148148147,
- "utterance_failure_rate": 0.06666666666666667,
- "severe_failure_rate": 0.0,
- "empty_hypothesis_rate": 0.0,
- "leading_omission_rate": 0.0,
- "ending_omission_rate": 0.0,
- "insertion_burst_rate": 0.0,
- "adjacent_phrase_repetition_rate": 0.0
- },
- "inflect-nano-v2-finalist::long_range": {
- "utterances": 6,
- "reference_words": 121,
- "semantic_wer": 0.008264462809917356,
- "substitution_rate": 0.008264462809917356,
- "deletion_rate": 0.0,
- "insertion_rate": 0.0,
- "utterance_failure_rate": 0.16666666666666666,
- "severe_failure_rate": 0.0,
- "empty_hypothesis_rate": 0.0,
- "leading_omission_rate": 0.0,
- "ending_omission_rate": 0.0,
- "insertion_burst_rate": 0.0,
- "adjacent_phrase_repetition_rate": 0.0
- },
- "inflect-nano-v2-finalist::names_places": {
- "utterances": 6,
- "reference_words": 57,
- "semantic_wer": 0.08771929824561403,
- "substitution_rate": 0.08771929824561403,
- "deletion_rate": 0.0,
- "insertion_rate": 0.0,
- "utterance_failure_rate": 0.5,
- "severe_failure_rate": 0.16666666666666666,
- "empty_hypothesis_rate": 0.0,
- "leading_omission_rate": 0.0,
- "ending_omission_rate": 0.0,
- "insertion_burst_rate": 0.0,
- "adjacent_phrase_repetition_rate": 0.0
- },
- "inflect-nano-v2-finalist::narrative": {
- "utterances": 24,
- "reference_words": 352,
- "semantic_wer": 0.011363636363636364,
- "substitution_rate": 0.005681818181818182,
- "deletion_rate": 0.0,
- "insertion_rate": 0.005681818181818182,
- "utterance_failure_rate": 0.125,
- "severe_failure_rate": 0.0,
- "empty_hypothesis_rate": 0.0,
- "leading_omission_rate": 0.0,
- "ending_omission_rate": 0.0,
- "insertion_burst_rate": 0.0,
- "adjacent_phrase_repetition_rate": 0.0
- },
- "inflect-nano-v2-finalist::numbers": {
- "utterances": 6,
- "reference_words": 64,
- "semantic_wer": 0.125,
- "substitution_rate": 0.078125,
- "deletion_rate": 0.015625,
- "insertion_rate": 0.03125,
- "utterance_failure_rate": 0.5,
- "severe_failure_rate": 0.3333333333333333,
- "empty_hypothesis_rate": 0.0,
- "leading_omission_rate": 0.16666666666666666,
- "ending_omission_rate": 0.0,
- "insertion_burst_rate": 0.0,
- "adjacent_phrase_repetition_rate": 0.0
- },
- "inflect-nano-v2-finalist::punctuation": {
- "utterances": 6,
- "reference_words": 65,
- "semantic_wer": 0.0,
- "substitution_rate": 0.0,
- "deletion_rate": 0.0,
- "insertion_rate": 0.0,
- "utterance_failure_rate": 0.0,
- "severe_failure_rate": 0.0,
- "empty_hypothesis_rate": 0.0,
- "leading_omission_rate": 0.0,
- "ending_omission_rate": 0.0,
- "insertion_burst_rate": 0.0,
- "adjacent_phrase_repetition_rate": 0.0
- },
- "inflect-nano-v2-finalist::question": {
- "utterances": 15,
- "reference_words": 165,
- "semantic_wer": 0.0,
- "substitution_rate": 0.0,
- "deletion_rate": 0.0,
- "insertion_rate": 0.0,
- "utterance_failure_rate": 0.0,
- "severe_failure_rate": 0.0,
- "empty_hypothesis_rate": 0.0,
- "leading_omission_rate": 0.0,
- "ending_omission_rate": 0.0,
- "insertion_burst_rate": 0.0,
- "adjacent_phrase_repetition_rate": 0.0
- },
- "inflect-nano-v2-finalist::rare_words": {
- "utterances": 6,
- "reference_words": 57,
- "semantic_wer": 0.03508771929824561,
- "substitution_rate": 0.017543859649122806,
- "deletion_rate": 0.0,
- "insertion_rate": 0.017543859649122806,
- "utterance_failure_rate": 0.16666666666666666,
- "severe_failure_rate": 0.0,
- "empty_hypothesis_rate": 0.0,
- "leading_omission_rate": 0.0,
- "ending_omission_rate": 0.0,
- "insertion_burst_rate": 0.0,
- "adjacent_phrase_repetition_rate": 0.0
- },
- "inflect-nano-v2-finalist::short": {
- "utterances": 24,
- "reference_words": 105,
- "semantic_wer": 0.047619047619047616,
- "substitution_rate": 0.009523809523809525,
- "deletion_rate": 0.0,
- "insertion_rate": 0.0380952380952381,
- "utterance_failure_rate": 0.16666666666666666,
- "severe_failure_rate": 0.08333333333333333,
- "empty_hypothesis_rate": 0.0,
- "leading_omission_rate": 0.0,
- "ending_omission_rate": 0.0,
- "insertion_burst_rate": 0.0,
- "adjacent_phrase_repetition_rate": 0.08333333333333333
- },
- "inflect-nano-v2-finalist::technical": {
- "utterances": 21,
- "reference_words": 256,
- "semantic_wer": 0.03125,
- "substitution_rate": 0.01953125,
- "deletion_rate": 0.0078125,
- "insertion_rate": 0.00390625,
- "utterance_failure_rate": 0.19047619047619047,
- "severe_failure_rate": 0.047619047619047616,
- "empty_hypothesis_rate": 0.0,
- "leading_omission_rate": 0.0,
- "ending_omission_rate": 0.0,
- "insertion_burst_rate": 0.0,
- "adjacent_phrase_repetition_rate": 0.0
- },
- "kitten-nano-bruno::conversational": {
- "utterances": 25,
- "reference_words": 309,
- "semantic_wer": 0.003236245954692557,
- "substitution_rate": 0.0,
- "deletion_rate": 0.003236245954692557,
- "insertion_rate": 0.0,
- "utterance_failure_rate": 0.04,
- "severe_failure_rate": 0.0,
- "empty_hypothesis_rate": 0.0,
- "leading_omission_rate": 0.0,
- "ending_omission_rate": 0.0,
- "insertion_burst_rate": 0.0,
- "adjacent_phrase_repetition_rate": 0.0
- },
- "kitten-nano-bruno::descriptive": {
- "utterances": 16,
- "reference_words": 226,
- "semantic_wer": 0.0,
- "substitution_rate": 0.0,
- "deletion_rate": 0.0,
- "insertion_rate": 0.0,
- "utterance_failure_rate": 0.0,
- "severe_failure_rate": 0.0,
- "empty_hypothesis_rate": 0.0,
- "leading_omission_rate": 0.0,
- "ending_omission_rate": 0.0,
- "insertion_burst_rate": 0.0,
- "adjacent_phrase_repetition_rate": 0.0625
- },
- "kitten-nano-bruno::emotional": {
- "utterances": 24,
- "reference_words": 329,
- "semantic_wer": 0.0,
- "substitution_rate": 0.0,
- "deletion_rate": 0.0,
- "insertion_rate": 0.0,
- "utterance_failure_rate": 0.0,
- "severe_failure_rate": 0.0,
- "empty_hypothesis_rate": 0.0,
- "leading_omission_rate": 0.0,
- "ending_omission_rate": 0.0,
- "insertion_burst_rate": 0.0,
- "adjacent_phrase_repetition_rate": 0.041666666666666664
- },
- "kitten-nano-bruno::homographs": {
- "utterances": 6,
- "reference_words": 70,
- "semantic_wer": 0.014285714285714285,
- "substitution_rate": 0.014285714285714285,
- "deletion_rate": 0.0,
- "insertion_rate": 0.0,
- "utterance_failure_rate": 0.16666666666666666,
- "severe_failure_rate": 0.0,
- "empty_hypothesis_rate": 0.0,
- "leading_omission_rate": 0.0,
- "ending_omission_rate": 0.0,
- "insertion_burst_rate": 0.0,
- "adjacent_phrase_repetition_rate": 0.0
- },
- "kitten-nano-bruno::long": {
- "utterances": 15,
- "reference_words": 432,
- "semantic_wer": 0.0023148148148148147,
- "substitution_rate": 0.0,
- "deletion_rate": 0.0023148148148148147,
- "insertion_rate": 0.0,
- "utterance_failure_rate": 0.06666666666666667,
- "severe_failure_rate": 0.0,
- "empty_hypothesis_rate": 0.0,
- "leading_omission_rate": 0.0,
- "ending_omission_rate": 0.0,
- "insertion_burst_rate": 0.0,
- "adjacent_phrase_repetition_rate": 0.0
- },
- "kitten-nano-bruno::long_range": {
- "utterances": 6,
- "reference_words": 121,
- "semantic_wer": 0.0,
- "substitution_rate": 0.0,
- "deletion_rate": 0.0,
- "insertion_rate": 0.0,
- "utterance_failure_rate": 0.0,
- "severe_failure_rate": 0.0,
- "empty_hypothesis_rate": 0.0,
- "leading_omission_rate": 0.0,
- "ending_omission_rate": 0.0,
- "insertion_burst_rate": 0.0,
- "adjacent_phrase_repetition_rate": 0.0
- },
- "kitten-nano-bruno::names_places": {
- "utterances": 6,
- "reference_words": 57,
- "semantic_wer": 0.05263157894736842,
- "substitution_rate": 0.05263157894736842,
- "deletion_rate": 0.0,
- "insertion_rate": 0.0,
- "utterance_failure_rate": 0.16666666666666666,
- "severe_failure_rate": 0.16666666666666666,
- "empty_hypothesis_rate": 0.0,
- "leading_omission_rate": 0.0,
- "ending_omission_rate": 0.0,
- "insertion_burst_rate": 0.0,
- "adjacent_phrase_repetition_rate": 0.0
- },
- "kitten-nano-bruno::narrative": {
- "utterances": 24,
- "reference_words": 352,
- "semantic_wer": 0.011363636363636364,
- "substitution_rate": 0.002840909090909091,
- "deletion_rate": 0.005681818181818182,
- "insertion_rate": 0.002840909090909091,
- "utterance_failure_rate": 0.125,
- "severe_failure_rate": 0.0,
- "empty_hypothesis_rate": 0.0,
- "leading_omission_rate": 0.0,
- "ending_omission_rate": 0.041666666666666664,
- "insertion_burst_rate": 0.0,
- "adjacent_phrase_repetition_rate": 0.0
- },
- "kitten-nano-bruno::numbers": {
- "utterances": 6,
- "reference_words": 64,
- "semantic_wer": 0.09375,
- "substitution_rate": 0.0625,
- "deletion_rate": 0.0,
- "insertion_rate": 0.03125,
- "utterance_failure_rate": 0.5,
- "severe_failure_rate": 0.16666666666666666,
- "empty_hypothesis_rate": 0.0,
- "leading_omission_rate": 0.0,
- "ending_omission_rate": 0.0,
- "insertion_burst_rate": 0.0,
- "adjacent_phrase_repetition_rate": 0.0
- },
- "kitten-nano-bruno::punctuation": {
- "utterances": 6,
- "reference_words": 65,
- "semantic_wer": 0.0,
- "substitution_rate": 0.0,
- "deletion_rate": 0.0,
- "insertion_rate": 0.0,
- "utterance_failure_rate": 0.0,
- "severe_failure_rate": 0.0,
- "empty_hypothesis_rate": 0.0,
- "leading_omission_rate": 0.0,
- "ending_omission_rate": 0.0,
- "insertion_burst_rate": 0.0,
- "adjacent_phrase_repetition_rate": 0.0
- },
- "kitten-nano-bruno::question": {
- "utterances": 15,
- "reference_words": 165,
- "semantic_wer": 0.0,
- "substitution_rate": 0.0,
- "deletion_rate": 0.0,
- "insertion_rate": 0.0,
- "utterance_failure_rate": 0.0,
- "severe_failure_rate": 0.0,
- "empty_hypothesis_rate": 0.0,
- "leading_omission_rate": 0.0,
- "ending_omission_rate": 0.0,
- "insertion_burst_rate": 0.0,
- "adjacent_phrase_repetition_rate": 0.0
- },
- "kitten-nano-bruno::rare_words": {
- "utterances": 6,
- "reference_words": 57,
- "semantic_wer": 0.03508771929824561,
- "substitution_rate": 0.017543859649122806,
- "deletion_rate": 0.0,
- "insertion_rate": 0.017543859649122806,
- "utterance_failure_rate": 0.16666666666666666,
- "severe_failure_rate": 0.0,
- "empty_hypothesis_rate": 0.0,
- "leading_omission_rate": 0.0,
- "ending_omission_rate": 0.0,
- "insertion_burst_rate": 0.0,
- "adjacent_phrase_repetition_rate": 0.0
- },
- "kitten-nano-bruno::short": {
- "utterances": 24,
- "reference_words": 105,
- "semantic_wer": 0.01904761904761905,
- "substitution_rate": 0.01904761904761905,
- "deletion_rate": 0.0,
- "insertion_rate": 0.0,
- "utterance_failure_rate": 0.08333333333333333,
- "severe_failure_rate": 0.0,
- "empty_hypothesis_rate": 0.0,
- "leading_omission_rate": 0.0,
- "ending_omission_rate": 0.0,
- "insertion_burst_rate": 0.0,
- "adjacent_phrase_repetition_rate": 0.08333333333333333
- },
- "kitten-nano-bruno::technical": {
- "utterances": 21,
- "reference_words": 256,
- "semantic_wer": 0.01953125,
- "substitution_rate": 0.0078125,
- "deletion_rate": 0.0078125,
- "insertion_rate": 0.00390625,
- "utterance_failure_rate": 0.09523809523809523,
- "severe_failure_rate": 0.0,
- "empty_hypothesis_rate": 0.0,
- "leading_omission_rate": 0.0,
- "ending_omission_rate": 0.0,
- "insertion_burst_rate": 0.0,
- "adjacent_phrase_repetition_rate": 0.0
- },
- "piper-ryan-low::conversational": {
- "utterances": 25,
- "reference_words": 309,
- "semantic_wer": 0.006472491909385114,
- "substitution_rate": 0.006472491909385114,
- "deletion_rate": 0.0,
- "insertion_rate": 0.0,
- "utterance_failure_rate": 0.04,
- "severe_failure_rate": 0.0,
- "empty_hypothesis_rate": 0.0,
- "leading_omission_rate": 0.0,
- "ending_omission_rate": 0.0,
- "insertion_burst_rate": 0.0,
- "adjacent_phrase_repetition_rate": 0.0
- },
- "piper-ryan-low::descriptive": {
- "utterances": 16,
- "reference_words": 226,
- "semantic_wer": 0.008849557522123894,
- "substitution_rate": 0.008849557522123894,
- "deletion_rate": 0.0,
- "insertion_rate": 0.0,
- "utterance_failure_rate": 0.125,
- "severe_failure_rate": 0.0,
- "empty_hypothesis_rate": 0.0,
- "leading_omission_rate": 0.0,
- "ending_omission_rate": 0.0,
- "insertion_burst_rate": 0.0,
- "adjacent_phrase_repetition_rate": 0.0625
- },
- "piper-ryan-low::emotional": {
- "utterances": 24,
- "reference_words": 329,
- "semantic_wer": 0.0,
- "substitution_rate": 0.0,
- "deletion_rate": 0.0,
- "insertion_rate": 0.0,
- "utterance_failure_rate": 0.0,
- "severe_failure_rate": 0.0,
- "empty_hypothesis_rate": 0.0,
- "leading_omission_rate": 0.0,
- "ending_omission_rate": 0.0,
- "insertion_burst_rate": 0.0,
- "adjacent_phrase_repetition_rate": 0.041666666666666664
- },
- "piper-ryan-low::homographs": {
- "utterances": 6,
- "reference_words": 70,
- "semantic_wer": 0.014285714285714285,
- "substitution_rate": 0.014285714285714285,
- "deletion_rate": 0.0,
- "insertion_rate": 0.0,
- "utterance_failure_rate": 0.16666666666666666,
- "severe_failure_rate": 0.0,
- "empty_hypothesis_rate": 0.0,
- "leading_omission_rate": 0.0,
- "ending_omission_rate": 0.0,
- "insertion_burst_rate": 0.0,
- "adjacent_phrase_repetition_rate": 0.0
- },
- "piper-ryan-low::long": {
- "utterances": 15,
- "reference_words": 432,
- "semantic_wer": 0.004629629629629629,
- "substitution_rate": 0.0023148148148148147,
- "deletion_rate": 0.0023148148148148147,
- "insertion_rate": 0.0,
- "utterance_failure_rate": 0.13333333333333333,
- "severe_failure_rate": 0.0,
- "empty_hypothesis_rate": 0.0,
- "leading_omission_rate": 0.0,
- "ending_omission_rate": 0.0,
- "insertion_burst_rate": 0.0,
- "adjacent_phrase_repetition_rate": 0.0
- },
- "piper-ryan-low::long_range": {
- "utterances": 6,
- "reference_words": 121,
- "semantic_wer": 0.0,
- "substitution_rate": 0.0,
- "deletion_rate": 0.0,
- "insertion_rate": 0.0,
- "utterance_failure_rate": 0.0,
- "severe_failure_rate": 0.0,
- "empty_hypothesis_rate": 0.0,
- "leading_omission_rate": 0.0,
- "ending_omission_rate": 0.0,
- "insertion_burst_rate": 0.0,
- "adjacent_phrase_repetition_rate": 0.0
- },
- "piper-ryan-low::names_places": {
- "utterances": 6,
- "reference_words": 57,
- "semantic_wer": 0.07017543859649122,
- "substitution_rate": 0.07017543859649122,
- "deletion_rate": 0.0,
- "insertion_rate": 0.0,
- "utterance_failure_rate": 0.3333333333333333,
- "severe_failure_rate": 0.16666666666666666,
- "empty_hypothesis_rate": 0.0,
- "leading_omission_rate": 0.0,
- "ending_omission_rate": 0.0,
- "insertion_burst_rate": 0.0,
- "adjacent_phrase_repetition_rate": 0.0
- },
- "piper-ryan-low::narrative": {
- "utterances": 24,
- "reference_words": 352,
- "semantic_wer": 0.0,
- "substitution_rate": 0.0,
- "deletion_rate": 0.0,
- "insertion_rate": 0.0,
- "utterance_failure_rate": 0.0,
- "severe_failure_rate": 0.0,
- "empty_hypothesis_rate": 0.0,
- "leading_omission_rate": 0.0,
- "ending_omission_rate": 0.0,
- "insertion_burst_rate": 0.0,
- "adjacent_phrase_repetition_rate": 0.0
- },
- "piper-ryan-low::numbers": {
- "utterances": 6,
- "reference_words": 64,
- "semantic_wer": 0.078125,
- "substitution_rate": 0.046875,
- "deletion_rate": 0.0,
- "insertion_rate": 0.03125,
- "utterance_failure_rate": 0.3333333333333333,
- "severe_failure_rate": 0.16666666666666666,
- "empty_hypothesis_rate": 0.0,
- "leading_omission_rate": 0.0,
- "ending_omission_rate": 0.0,
- "insertion_burst_rate": 0.0,
- "adjacent_phrase_repetition_rate": 0.0
- },
- "piper-ryan-low::punctuation": {
- "utterances": 6,
- "reference_words": 65,
- "semantic_wer": 0.0,
- "substitution_rate": 0.0,
- "deletion_rate": 0.0,
- "insertion_rate": 0.0,
- "utterance_failure_rate": 0.0,
- "severe_failure_rate": 0.0,
- "empty_hypothesis_rate": 0.0,
- "leading_omission_rate": 0.0,
- "ending_omission_rate": 0.0,
- "insertion_burst_rate": 0.0,
- "adjacent_phrase_repetition_rate": 0.0
- },
- "piper-ryan-low::question": {
- "utterances": 15,
- "reference_words": 165,
- "semantic_wer": 0.0,
- "substitution_rate": 0.0,
- "deletion_rate": 0.0,
- "insertion_rate": 0.0,
- "utterance_failure_rate": 0.0,
- "severe_failure_rate": 0.0,
- "empty_hypothesis_rate": 0.0,
- "leading_omission_rate": 0.0,
- "ending_omission_rate": 0.0,
- "insertion_burst_rate": 0.0,
- "adjacent_phrase_repetition_rate": 0.0
- },
- "piper-ryan-low::rare_words": {
- "utterances": 6,
- "reference_words": 57,
- "semantic_wer": 0.08771929824561403,
- "substitution_rate": 0.07017543859649122,
- "deletion_rate": 0.0,
- "insertion_rate": 0.017543859649122806,
- "utterance_failure_rate": 0.5,
- "severe_failure_rate": 0.0,
- "empty_hypothesis_rate": 0.0,
- "leading_omission_rate": 0.0,
- "ending_omission_rate": 0.0,
- "insertion_burst_rate": 0.0,
- "adjacent_phrase_repetition_rate": 0.0
- },
- "piper-ryan-low::short": {
- "utterances": 24,
- "reference_words": 105,
- "semantic_wer": 0.0380952380952381,
- "substitution_rate": 0.0380952380952381,
- "deletion_rate": 0.0,
- "insertion_rate": 0.0,
- "utterance_failure_rate": 0.08333333333333333,
- "severe_failure_rate": 0.08333333333333333,
- "empty_hypothesis_rate": 0.0,
- "leading_omission_rate": 0.0,
- "ending_omission_rate": 0.0,
- "insertion_burst_rate": 0.0,
- "adjacent_phrase_repetition_rate": 0.08333333333333333
- },
- "piper-ryan-low::technical": {
- "utterances": 21,
- "reference_words": 256,
- "semantic_wer": 0.02734375,
- "substitution_rate": 0.015625,
- "deletion_rate": 0.0078125,
- "insertion_rate": 0.00390625,
- "utterance_failure_rate": 0.19047619047619047,
- "severe_failure_rate": 0.0,
- "empty_hypothesis_rate": 0.0,
- "leading_omission_rate": 0.0,
- "ending_omission_rate": 0.0,
- "insertion_burst_rate": 0.0,
- "adjacent_phrase_repetition_rate": 0.0
- }
- },
- "worst_utterances": [
- {
- "model": "inflect-micro-v2-finalist",
- "id": "utmos200_176",
- "category": "short",
- "reference": "Noted.",
- "hypothesis": "Noid.",
- "reference_words": 1,
- "hypothesis_words": 1,
- "substitutions": 1,
- "deletions": 0,
- "insertions": 0,
- "semantic_wer": 1.0,
- "utterance_failure": true,
- "severe_failure": true,
- "empty_hypothesis": false,
- "leading_omission": false,
- "ending_omission": false,
- "insertion_burst": false,
- "adjacent_phrase_repetition": false,
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_176.wav"
- },
- {
- "model": "piper-ryan-low",
- "id": "utmos200_041",
- "category": "short",
- "reference": "I mean. Yeah. Kind of.",
- "hypothesis": "I mean, no, I know.",
- "reference_words": 5,
- "hypothesis_words": 5,
- "substitutions": 3,
- "deletions": 0,
- "insertions": 0,
- "semantic_wer": 0.6,
- "utterance_failure": true,
- "severe_failure": true,
- "empty_hypothesis": false,
- "leading_omission": false,
- "ending_omission": false,
- "insertion_burst": false,
- "adjacent_phrase_repetition": false,
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_041.wav"
- },
- {
- "model": "inflect-nano-v2-finalist",
- "id": "utmos200_041",
- "category": "short",
- "reference": "I mean. Yeah. Kind of.",
- "hypothesis": "I mean that, yeah, that kind of,",
- "reference_words": 5,
- "hypothesis_words": 7,
- "substitutions": 0,
- "deletions": 0,
- "insertions": 2,
- "semantic_wer": 0.4,
- "utterance_failure": true,
- "severe_failure": true,
- "empty_hypothesis": false,
- "leading_omission": false,
- "ending_omission": false,
- "insertion_burst": false,
- "adjacent_phrase_repetition": false,
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_041.wav"
- },
- {
- "model": "inflect-micro-v2-finalist",
- "id": "utmos200_065",
- "category": "numbers",
- "reference": "Call extension zero four nine after eight thirty in the morning.",
- "hypothesis": "Call extension 049 after 8.30 in the morning.",
- "reference_words": 8,
- "hypothesis_words": 9,
- "substitutions": 2,
- "deletions": 0,
- "insertions": 1,
- "semantic_wer": 0.375,
- "utterance_failure": true,
- "severe_failure": true,
- "empty_hypothesis": false,
- "leading_omission": false,
- "ending_omission": false,
- "insertion_burst": false,
- "adjacent_phrase_repetition": false,
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_065.wav"
- },
- {
- "model": "inflect-nano-v2-finalist",
- "id": "utmos200_065",
- "category": "numbers",
- "reference": "Call extension zero four nine after eight thirty in the morning.",
- "hypothesis": "Call extension 049 after 8.30 in the morning.",
- "reference_words": 8,
- "hypothesis_words": 9,
- "substitutions": 2,
- "deletions": 0,
- "insertions": 1,
- "semantic_wer": 0.375,
- "utterance_failure": true,
- "severe_failure": true,
- "empty_hypothesis": false,
- "leading_omission": false,
- "ending_omission": false,
- "insertion_burst": false,
- "adjacent_phrase_repetition": false,
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_065.wav"
- },
- {
- "model": "piper-ryan-low",
- "id": "utmos200_065",
- "category": "numbers",
- "reference": "Call extension zero four nine after eight thirty in the morning.",
- "hypothesis": "Call extension 049 after 8.30 in the morning.",
- "reference_words": 8,
- "hypothesis_words": 9,
- "substitutions": 2,
- "deletions": 0,
- "insertions": 1,
- "semantic_wer": 0.375,
- "utterance_failure": true,
- "severe_failure": true,
- "empty_hypothesis": false,
- "leading_omission": false,
- "ending_omission": false,
- "insertion_burst": false,
- "adjacent_phrase_repetition": false,
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_065.wav"
- },
- {
- "model": "kitten-nano-bruno",
- "id": "utmos200_065",
- "category": "numbers",
- "reference": "Call extension zero four nine after eight thirty in the morning.",
- "hypothesis": "Call extension 049 after 8.30 in the morning.",
- "reference_words": 8,
- "hypothesis_words": 9,
- "substitutions": 2,
- "deletions": 0,
- "insertions": 1,
- "semantic_wer": 0.375,
- "utterance_failure": true,
- "severe_failure": true,
- "empty_hypothesis": false,
- "leading_omission": false,
- "ending_omission": false,
- "insertion_burst": false,
- "adjacent_phrase_repetition": false,
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_065.wav"
- },
- {
- "model": "inflect-nano-v2-finalist",
- "id": "utmos200_035",
- "category": "names_places",
- "reference": "Aurelia and Joaquin crossed the Schuylkill River before noon.",
- "hypothesis": "Uralaya and Jokwing crossed the Sialkal River before noon.",
- "reference_words": 9,
- "hypothesis_words": 9,
- "substitutions": 3,
- "deletions": 0,
- "insertions": 0,
- "semantic_wer": 0.3333333333333333,
- "utterance_failure": true,
- "severe_failure": true,
- "empty_hypothesis": false,
- "leading_omission": false,
- "ending_omission": false,
- "insertion_burst": false,
- "adjacent_phrase_repetition": false,
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_035.wav"
- },
- {
- "model": "inflect-nano-v2-finalist",
- "id": "utmos200_191",
- "category": "short",
- "reference": "Right. Moving on.",
- "hypothesis": "Right at moving on.",
- "reference_words": 3,
- "hypothesis_words": 4,
- "substitutions": 0,
- "deletions": 0,
- "insertions": 1,
- "semantic_wer": 0.3333333333333333,
- "utterance_failure": true,
- "severe_failure": true,
- "empty_hypothesis": false,
- "leading_omission": false,
- "ending_omission": false,
- "insertion_burst": false,
- "adjacent_phrase_repetition": false,
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_191.wav"
- },
- {
- "model": "piper-ryan-low",
- "id": "utmos200_049",
- "category": "names_places",
- "reference": "Xiomara met Siobhan near the old library in Worcestershire.",
- "hypothesis": "See Amaru Mechivan near the Old Library in Worcestershire.",
- "reference_words": 9,
- "hypothesis_words": 9,
- "substitutions": 3,
- "deletions": 0,
- "insertions": 0,
- "semantic_wer": 0.3333333333333333,
- "utterance_failure": true,
- "severe_failure": true,
- "empty_hypothesis": false,
- "leading_omission": false,
- "ending_omission": false,
- "insertion_burst": false,
- "adjacent_phrase_repetition": false,
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_049.wav"
- },
- {
- "model": "piper-ryan-low",
- "id": "utmos200_091",
- "category": "short",
- "reference": "That's fair.",
- "hypothesis": "That was fair.",
- "reference_words": 3,
- "hypothesis_words": 3,
- "substitutions": 1,
- "deletions": 0,
- "insertions": 0,
- "semantic_wer": 0.3333333333333333,
- "utterance_failure": true,
- "severe_failure": true,
- "empty_hypothesis": false,
- "leading_omission": false,
- "ending_omission": false,
- "insertion_burst": false,
- "adjacent_phrase_repetition": false,
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_091.wav"
- },
- {
- "model": "kitten-nano-bruno",
- "id": "utmos200_035",
- "category": "names_places",
- "reference": "Aurelia and Joaquin crossed the Schuylkill River before noon.",
- "hypothesis": "Auralaya and Jokwan crossed the Shilkal river before noon.",
- "reference_words": 9,
- "hypothesis_words": 9,
- "substitutions": 3,
- "deletions": 0,
- "insertions": 0,
- "semantic_wer": 0.3333333333333333,
- "utterance_failure": true,
- "severe_failure": true,
- "empty_hypothesis": false,
- "leading_omission": false,
- "ending_omission": false,
- "insertion_burst": false,
- "adjacent_phrase_repetition": false,
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_035.wav"
- },
- {
- "model": "inflect-micro-v2-finalist",
- "id": "utmos200_051",
- "category": "numbers",
- "reference": "At exactly seven forty-five, Morgan placed eighteen dollars and fifty cents beside the notebook.",
- "hypothesis": "At exactly $7.45, Morton placed $18.50 beside the notebook.",
- "reference_words": 10,
- "hypothesis_words": 11,
- "substitutions": 2,
- "deletions": 0,
- "insertions": 1,
- "semantic_wer": 0.3,
- "utterance_failure": true,
- "severe_failure": true,
- "empty_hypothesis": false,
- "leading_omission": false,
- "ending_omission": false,
- "insertion_burst": false,
- "adjacent_phrase_repetition": false,
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_051.wav"
- },
- {
- "model": "inflect-nano-v2-finalist",
- "id": "utmos200_051",
- "category": "numbers",
- "reference": "At exactly seven forty-five, Morgan placed eighteen dollars and fifty cents beside the notebook.",
- "hypothesis": "Exactly $7.45, Morgan placed $18.50 beside the notebook.",
- "reference_words": 10,
- "hypothesis_words": 10,
- "substitutions": 1,
- "deletions": 1,
- "insertions": 1,
- "semantic_wer": 0.3,
- "utterance_failure": true,
- "severe_failure": true,
- "empty_hypothesis": false,
- "leading_omission": true,
- "ending_omission": false,
- "insertion_burst": false,
- "adjacent_phrase_repetition": false,
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_051.wav"
- },
- {
- "model": "inflect-micro-v2-finalist",
- "id": "utmos200_028",
- "category": "technical",
- "reference": "Hash the UTF eight payload with S H A two fifty-six before opening the checkpoint.",
- "hypothesis": "Cache the UTF-8 payload with SH-256 before opening the checkpoint.",
- "reference_words": 14,
- "hypothesis_words": 12,
- "substitutions": 2,
- "deletions": 2,
- "insertions": 0,
- "semantic_wer": 0.2857142857142857,
- "utterance_failure": true,
- "severe_failure": true,
- "empty_hypothesis": false,
- "leading_omission": false,
- "ending_omission": false,
- "insertion_burst": false,
- "adjacent_phrase_repetition": false,
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_028.wav"
- },
- {
- "model": "inflect-nano-v2-finalist",
- "id": "utmos200_028",
- "category": "technical",
- "reference": "Hash the UTF eight payload with S H A two fifty-six before opening the checkpoint.",
- "hypothesis": "Ash the UTF-8 payload with SHA-256 before opening the checkpoint.",
- "reference_words": 14,
- "hypothesis_words": 12,
- "substitutions": 2,
- "deletions": 2,
- "insertions": 0,
- "semantic_wer": 0.2857142857142857,
- "utterance_failure": true,
- "severe_failure": true,
- "empty_hypothesis": false,
- "leading_omission": false,
- "ending_omission": false,
- "insertion_burst": false,
- "adjacent_phrase_repetition": false,
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_028.wav"
- },
- {
- "model": "inflect-micro-v2-finalist",
- "id": "utmos200_139",
- "category": "short",
- "reference": "No way. Absolutely not.",
- "hypothesis": "No wait, absolutely not.",
- "reference_words": 4,
- "hypothesis_words": 4,
- "substitutions": 1,
- "deletions": 0,
- "insertions": 0,
- "semantic_wer": 0.25,
- "utterance_failure": true,
- "severe_failure": false,
- "empty_hypothesis": false,
- "leading_omission": false,
- "ending_omission": false,
- "insertion_burst": false,
- "adjacent_phrase_repetition": false,
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_139.wav"
- },
- {
- "model": "inflect-micro-v2-finalist",
- "id": "utmos200_181",
- "category": "short",
- "reference": "Yeah, no. Hard pass.",
- "hypothesis": "Yeah, no but hard pass.",
- "reference_words": 4,
- "hypothesis_words": 5,
- "substitutions": 0,
- "deletions": 0,
- "insertions": 1,
- "semantic_wer": 0.25,
- "utterance_failure": true,
- "severe_failure": false,
- "empty_hypothesis": false,
- "leading_omission": false,
- "ending_omission": false,
- "insertion_burst": false,
- "adjacent_phrase_repetition": false,
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_181.wav"
- },
- {
- "model": "inflect-nano-v2-finalist",
- "id": "utmos200_139",
- "category": "short",
- "reference": "No way. Absolutely not.",
- "hypothesis": "No way to absolutely not.",
- "reference_words": 4,
- "hypothesis_words": 5,
- "substitutions": 0,
- "deletions": 0,
- "insertions": 1,
- "semantic_wer": 0.25,
- "utterance_failure": true,
- "severe_failure": false,
- "empty_hypothesis": false,
- "leading_omission": false,
- "ending_omission": false,
- "insertion_burst": false,
- "adjacent_phrase_repetition": false,
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_139.wav"
- },
- {
- "model": "inflect-nano-v2-finalist",
- "id": "utmos200_181",
- "category": "short",
- "reference": "Yeah, no. Hard pass.",
- "hypothesis": "Yeah, no.Card pass.",
- "reference_words": 4,
- "hypothesis_words": 4,
- "substitutions": 1,
- "deletions": 0,
- "insertions": 0,
- "semantic_wer": 0.25,
- "utterance_failure": true,
- "severe_failure": false,
- "empty_hypothesis": false,
- "leading_omission": false,
- "ending_omission": false,
- "insertion_burst": false,
- "adjacent_phrase_repetition": false,
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_181.wav"
- },
- {
- "model": "kitten-nano-bruno",
- "id": "utmos200_181",
- "category": "short",
- "reference": "Yeah, no. Hard pass.",
- "hypothesis": "Yeth no. Hard pass.",
- "reference_words": 4,
- "hypothesis_words": 4,
- "substitutions": 1,
- "deletions": 0,
- "insertions": 0,
- "semantic_wer": 0.25,
- "utterance_failure": true,
- "severe_failure": false,
- "empty_hypothesis": false,
- "leading_omission": false,
- "ending_omission": false,
- "insertion_burst": false,
- "adjacent_phrase_repetition": false,
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_181.wav"
- },
- {
- "model": "inflect-micro-v2-finalist",
- "id": "utmos200_035",
- "category": "names_places",
- "reference": "Aurelia and Joaquin crossed the Schuylkill River before noon.",
- "hypothesis": "Aurelia and Joachim crossed the Schalke river before noon.",
- "reference_words": 9,
- "hypothesis_words": 9,
- "substitutions": 2,
- "deletions": 0,
- "insertions": 0,
- "semantic_wer": 0.2222222222222222,
- "utterance_failure": true,
- "severe_failure": false,
- "empty_hypothesis": false,
- "leading_omission": false,
- "ending_omission": false,
- "insertion_burst": false,
- "adjacent_phrase_repetition": false,
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_035.wav"
- },
- {
- "model": "piper-ryan-low",
- "id": "utmos200_028",
- "category": "technical",
- "reference": "Hash the UTF eight payload with S H A two fifty-six before opening the checkpoint.",
- "hypothesis": "Hash the UTF-8 payload with SHA-256 before opening the checkpoint.",
- "reference_words": 14,
- "hypothesis_words": 12,
- "substitutions": 1,
- "deletions": 2,
- "insertions": 0,
- "semantic_wer": 0.21428571428571427,
- "utterance_failure": true,
- "severe_failure": false,
- "empty_hypothesis": false,
- "leading_omission": false,
- "ending_omission": false,
- "insertion_burst": false,
- "adjacent_phrase_repetition": false,
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_028.wav"
- },
- {
- "model": "kitten-nano-bruno",
- "id": "utmos200_028",
- "category": "technical",
- "reference": "Hash the UTF eight payload with S H A two fifty-six before opening the checkpoint.",
- "hypothesis": "Hash the UTF-8 payload with SHA-256 before opening the checkpoint.",
- "reference_words": 14,
- "hypothesis_words": 12,
- "substitutions": 1,
- "deletions": 2,
- "insertions": 0,
- "semantic_wer": 0.21428571428571427,
- "utterance_failure": true,
- "severe_failure": false,
- "empty_hypothesis": false,
- "leading_omission": false,
- "ending_omission": false,
- "insertion_burst": false,
- "adjacent_phrase_repetition": false,
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_028.wav"
- },
- {
- "model": "inflect-micro-v2-finalist",
- "id": "utmos200_041",
- "category": "short",
- "reference": "I mean. Yeah. Kind of.",
- "hypothesis": "I mean, yeah, I kind of.",
- "reference_words": 5,
- "hypothesis_words": 6,
- "substitutions": 0,
- "deletions": 0,
- "insertions": 1,
- "semantic_wer": 0.2,
- "utterance_failure": true,
- "severe_failure": false,
- "empty_hypothesis": false,
- "leading_omission": false,
- "ending_omission": false,
- "insertion_burst": false,
- "adjacent_phrase_repetition": false,
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_041.wav"
- },
- {
- "model": "inflect-micro-v2-finalist",
- "id": "utmos200_084",
- "category": "technical",
- "reference": "The performance improvement is real, but it comes with tradeoffs.",
- "hypothesis": "The performance improvement is real, but it comes with trade-offs.",
- "reference_words": 10,
- "hypothesis_words": 11,
- "substitutions": 1,
- "deletions": 0,
- "insertions": 1,
- "semantic_wer": 0.2,
- "utterance_failure": true,
- "severe_failure": false,
- "empty_hypothesis": false,
- "leading_omission": false,
- "ending_omission": false,
- "insertion_burst": false,
- "adjacent_phrase_repetition": false,
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_084.wav"
- },
- {
- "model": "inflect-nano-v2-finalist",
- "id": "utmos200_074",
- "category": "homographs",
- "reference": "The dove dove beneath the branch while the wind began to wind the loose cable.",
- "hypothesis": "The dub-dub beneath the branch wall the wind began to wind the loose cable.",
- "reference_words": 15,
- "hypothesis_words": 15,
- "substitutions": 3,
- "deletions": 0,
- "insertions": 0,
- "semantic_wer": 0.2,
- "utterance_failure": true,
- "severe_failure": false,
- "empty_hypothesis": false,
- "leading_omission": false,
- "ending_omission": false,
- "insertion_burst": false,
- "adjacent_phrase_repetition": false,
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_074.wav"
- },
- {
- "model": "inflect-nano-v2-finalist",
- "id": "utmos200_084",
- "category": "technical",
- "reference": "The performance improvement is real, but it comes with tradeoffs.",
- "hypothesis": "The performance improvement is real, but it comes with trade-offs.",
- "reference_words": 10,
- "hypothesis_words": 11,
- "substitutions": 1,
- "deletions": 0,
- "insertions": 1,
- "semantic_wer": 0.2,
- "utterance_failure": true,
- "severe_failure": false,
- "empty_hypothesis": false,
- "leading_omission": false,
- "ending_omission": false,
- "insertion_burst": false,
- "adjacent_phrase_repetition": false,
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_084.wav"
- },
- {
- "model": "piper-ryan-low",
- "id": "utmos200_012",
- "category": "rare_words",
- "reference": "The archaeologist catalogued a Byzantine reliquary and several anachronistic fragments.",
- "hypothesis": "the archaeologists cataloged a Byzantine reliquary and several anachronistic fragments.",
- "reference_words": 10,
- "hypothesis_words": 10,
- "substitutions": 2,
- "deletions": 0,
- "insertions": 0,
- "semantic_wer": 0.2,
- "utterance_failure": true,
- "severe_failure": false,
- "empty_hypothesis": false,
- "leading_omission": false,
- "ending_omission": false,
- "insertion_burst": false,
- "adjacent_phrase_repetition": false,
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_012.wav"
- },
- {
- "model": "piper-ryan-low",
- "id": "utmos200_051",
- "category": "numbers",
- "reference": "At exactly seven forty-five, Morgan placed eighteen dollars and fifty cents beside the notebook.",
- "hypothesis": "At exactly 7.45, Morgan placed $18.50 beside the notebook.",
- "reference_words": 10,
- "hypothesis_words": 11,
- "substitutions": 1,
- "deletions": 0,
- "insertions": 1,
- "semantic_wer": 0.2,
- "utterance_failure": true,
- "severe_failure": false,
- "empty_hypothesis": false,
- "leading_omission": false,
- "ending_omission": false,
- "insertion_burst": false,
- "adjacent_phrase_repetition": false,
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_051.wav"
- },
- {
- "model": "piper-ryan-low",
- "id": "utmos200_084",
- "category": "technical",
- "reference": "The performance improvement is real, but it comes with tradeoffs.",
- "hypothesis": "The performance improvement is real, but it comes with trade-offs.",
- "reference_words": 10,
- "hypothesis_words": 11,
- "substitutions": 1,
- "deletions": 0,
- "insertions": 1,
- "semantic_wer": 0.2,
- "utterance_failure": true,
- "severe_failure": false,
- "empty_hypothesis": false,
- "leading_omission": false,
- "ending_omission": false,
- "insertion_burst": false,
- "adjacent_phrase_repetition": false,
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_084.wav"
- },
- {
- "model": "piper-ryan-low",
- "id": "utmos200_188",
- "category": "conversational",
- "reference": "You got engaged? When? How? Tell me everything right now.",
- "hypothesis": "You got engaged. What? Oh. Tell me everything right now.",
- "reference_words": 10,
- "hypothesis_words": 10,
- "substitutions": 2,
- "deletions": 0,
- "insertions": 0,
- "semantic_wer": 0.2,
- "utterance_failure": true,
- "severe_failure": false,
- "empty_hypothesis": false,
- "leading_omission": false,
- "ending_omission": false,
- "insertion_burst": false,
- "adjacent_phrase_repetition": false,
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_188.wav"
- },
- {
- "model": "kitten-nano-bruno",
- "id": "utmos200_041",
- "category": "short",
- "reference": "I mean. Yeah. Kind of.",
- "hypothesis": "I mean. Yes. Kind of.",
- "reference_words": 5,
- "hypothesis_words": 5,
- "substitutions": 1,
- "deletions": 0,
- "insertions": 0,
- "semantic_wer": 0.2,
- "utterance_failure": true,
- "severe_failure": false,
- "empty_hypothesis": false,
- "leading_omission": false,
- "ending_omission": false,
- "insertion_burst": false,
- "adjacent_phrase_repetition": false,
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_041.wav"
- },
- {
- "model": "kitten-nano-bruno",
- "id": "utmos200_051",
- "category": "numbers",
- "reference": "At exactly seven forty-five, Morgan placed eighteen dollars and fifty cents beside the notebook.",
- "hypothesis": "At exactly 7.45, Morgan placed $18.50 beside the notebook.",
- "reference_words": 10,
- "hypothesis_words": 11,
- "substitutions": 1,
- "deletions": 0,
- "insertions": 1,
- "semantic_wer": 0.2,
- "utterance_failure": true,
- "severe_failure": false,
- "empty_hypothesis": false,
- "leading_omission": false,
- "ending_omission": false,
- "insertion_burst": false,
- "adjacent_phrase_repetition": false,
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_051.wav"
- },
- {
- "model": "kitten-nano-bruno",
- "id": "utmos200_084",
- "category": "technical",
- "reference": "The performance improvement is real, but it comes with tradeoffs.",
- "hypothesis": "The performance improvement is real, but it comes with trade-offs.",
- "reference_words": 10,
- "hypothesis_words": 11,
- "substitutions": 1,
- "deletions": 0,
- "insertions": 1,
- "semantic_wer": 0.2,
- "utterance_failure": true,
- "severe_failure": false,
- "empty_hypothesis": false,
- "leading_omission": false,
- "ending_omission": false,
- "insertion_burst": false,
- "adjacent_phrase_repetition": false,
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_084.wav"
- },
- {
- "model": "inflect-micro-v2-finalist",
- "id": "utmos200_009",
- "category": "numbers",
- "reference": "The package weighs twelve point six kilograms and arrived on July twenty-first.",
- "hypothesis": "The package weighs 12.6kg and arrived on July 21.",
- "reference_words": 11,
- "hypothesis_words": 11,
- "substitutions": 2,
- "deletions": 0,
- "insertions": 0,
- "semantic_wer": 0.18181818181818182,
- "utterance_failure": true,
- "severe_failure": false,
- "empty_hypothesis": false,
- "leading_omission": false,
- "ending_omission": false,
- "insertion_burst": false,
- "adjacent_phrase_repetition": false,
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_009.wav"
- },
- {
- "model": "inflect-micro-v2-finalist",
- "id": "utmos200_082",
- "category": "rare_words",
- "reference": "The physician recommended a noninvasive procedure for the recurring neurological symptoms.",
- "hypothesis": "The physician recommended a non-invasive procedure for the recurring neurological symptoms.",
- "reference_words": 11,
- "hypothesis_words": 12,
- "substitutions": 1,
- "deletions": 0,
- "insertions": 1,
- "semantic_wer": 0.18181818181818182,
- "utterance_failure": true,
- "severe_failure": false,
- "empty_hypothesis": false,
- "leading_omission": false,
- "ending_omission": false,
- "insertion_burst": false,
- "adjacent_phrase_repetition": false,
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_082.wav"
- },
- {
- "model": "inflect-nano-v2-finalist",
- "id": "utmos200_009",
- "category": "numbers",
- "reference": "The package weighs twelve point six kilograms and arrived on July twenty-first.",
- "hypothesis": "The package weighs 12.6kg and arrived on July 21.",
- "reference_words": 11,
- "hypothesis_words": 11,
- "substitutions": 2,
- "deletions": 0,
- "insertions": 0,
- "semantic_wer": 0.18181818181818182,
- "utterance_failure": true,
- "severe_failure": false,
- "empty_hypothesis": false,
- "leading_omission": false,
- "ending_omission": false,
- "insertion_burst": false,
- "adjacent_phrase_repetition": false,
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_009.wav"
- },
- {
- "model": "inflect-nano-v2-finalist",
- "id": "utmos200_082",
- "category": "rare_words",
- "reference": "The physician recommended a noninvasive procedure for the recurring neurological symptoms.",
- "hypothesis": "The physician recommended a non-invasive procedure for the recurring neurological symptoms.",
- "reference_words": 11,
- "hypothesis_words": 12,
- "substitutions": 1,
- "deletions": 0,
- "insertions": 1,
- "semantic_wer": 0.18181818181818182,
- "utterance_failure": true,
- "severe_failure": false,
- "empty_hypothesis": false,
- "leading_omission": false,
- "ending_omission": false,
- "insertion_burst": false,
- "adjacent_phrase_repetition": false,
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_082.wav"
- },
- {
- "model": "piper-ryan-low",
- "id": "utmos200_082",
- "category": "rare_words",
- "reference": "The physician recommended a noninvasive procedure for the recurring neurological symptoms.",
- "hypothesis": "the physician recommended a non-invasive procedure for the recurring neurological symptoms.",
- "reference_words": 11,
- "hypothesis_words": 12,
- "substitutions": 1,
- "deletions": 0,
- "insertions": 1,
- "semantic_wer": 0.18181818181818182,
- "utterance_failure": true,
- "severe_failure": false,
- "empty_hypothesis": false,
- "leading_omission": false,
- "ending_omission": false,
- "insertion_burst": false,
- "adjacent_phrase_repetition": false,
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_082.wav"
- },
- {
- "model": "kitten-nano-bruno",
- "id": "utmos200_082",
- "category": "rare_words",
- "reference": "The physician recommended a noninvasive procedure for the recurring neurological symptoms.",
- "hypothesis": "The physician recommended a non-invasive procedure for the recurring neurological symptoms.",
- "reference_words": 11,
- "hypothesis_words": 12,
- "substitutions": 1,
- "deletions": 0,
- "insertions": 1,
- "semantic_wer": 0.18181818181818182,
- "utterance_failure": true,
- "severe_failure": false,
- "empty_hypothesis": false,
- "leading_omission": false,
- "ending_omission": false,
- "insertion_burst": false,
- "adjacent_phrase_repetition": false,
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_082.wav"
- },
- {
- "model": "inflect-nano-v2-finalist",
- "id": "utmos200_180",
- "category": "narrative",
- "reference": "She knew. She had always known. She just waited for the right moment.",
- "hypothesis": "Ji knew but she had always known, she just waited for the right moment.",
- "reference_words": 13,
- "hypothesis_words": 14,
- "substitutions": 1,
- "deletions": 0,
- "insertions": 1,
- "semantic_wer": 0.15384615384615385,
- "utterance_failure": true,
- "severe_failure": false,
- "empty_hypothesis": false,
- "leading_omission": false,
- "ending_omission": false,
- "insertion_burst": false,
- "adjacent_phrase_repetition": false,
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_180.wav"
- },
- {
- "model": "kitten-nano-bruno",
- "id": "utmos200_175",
- "category": "narrative",
- "reference": "The note was short, which was somehow more frightening than a long one would have been.",
- "hypothesis": "The note was short, which was somehow more fright, ing than a long one would have been.",
- "reference_words": 16,
- "hypothesis_words": 17,
- "substitutions": 1,
- "deletions": 0,
- "insertions": 1,
- "semantic_wer": 0.125,
- "utterance_failure": true,
- "severe_failure": false,
- "empty_hypothesis": false,
- "leading_omission": false,
- "ending_omission": false,
- "insertion_burst": false,
- "adjacent_phrase_repetition": false,
- "audio": "utmos22_large200_v1/audio/kitten-nano-bruno/utmos200_175.wav"
- },
- {
- "model": "inflect-micro-v2-finalist",
- "id": "utmos200_049",
- "category": "names_places",
- "reference": "Xiomara met Siobhan near the old library in Worcestershire.",
- "hypothesis": "Zamara met Siobhan near the old library in Worcestershire.",
- "reference_words": 9,
- "hypothesis_words": 9,
- "substitutions": 1,
- "deletions": 0,
- "insertions": 0,
- "semantic_wer": 0.1111111111111111,
- "utterance_failure": true,
- "severe_failure": false,
- "empty_hypothesis": false,
- "leading_omission": false,
- "ending_omission": false,
- "insertion_burst": false,
- "adjacent_phrase_repetition": false,
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_049.wav"
- },
- {
- "model": "inflect-micro-v2-finalist",
- "id": "utmos200_077",
- "category": "names_places",
- "reference": "Saskatchewan looked unusually bright beneath the pale February sunrise.",
- "hypothesis": "Saskatchewan looked unusually bright beneath that pale February sunrise.",
- "reference_words": 9,
- "hypothesis_words": 9,
- "substitutions": 1,
- "deletions": 0,
- "insertions": 0,
- "semantic_wer": 0.1111111111111111,
- "utterance_failure": true,
- "severe_failure": false,
- "empty_hypothesis": false,
- "leading_omission": false,
- "ending_omission": false,
- "insertion_burst": false,
- "adjacent_phrase_repetition": false,
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_077.wav"
- },
- {
- "model": "inflect-nano-v2-finalist",
- "id": "utmos200_022",
- "category": "narrative",
- "reference": "The plan was simple. The execution was anything but.",
- "hypothesis": "The plan was simple. The execution was anything but perfect.",
- "reference_words": 9,
- "hypothesis_words": 10,
- "substitutions": 0,
- "deletions": 0,
- "insertions": 1,
- "semantic_wer": 0.1111111111111111,
- "utterance_failure": true,
- "severe_failure": false,
- "empty_hypothesis": false,
- "leading_omission": false,
- "ending_omission": false,
- "insertion_burst": false,
- "adjacent_phrase_repetition": false,
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_022.wav"
- },
- {
- "model": "inflect-nano-v2-finalist",
- "id": "utmos200_049",
- "category": "names_places",
- "reference": "Xiomara met Siobhan near the old library in Worcestershire.",
- "hypothesis": "Damara met Siobhan near the old library in Worcestershire.",
- "reference_words": 9,
- "hypothesis_words": 9,
- "substitutions": 1,
- "deletions": 0,
- "insertions": 0,
- "semantic_wer": 0.1111111111111111,
- "utterance_failure": true,
- "severe_failure": false,
- "empty_hypothesis": false,
- "leading_omission": false,
- "ending_omission": false,
- "insertion_burst": false,
- "adjacent_phrase_repetition": false,
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-nano-v2-finalist/utmos200_049.wav"
- },
- {
- "model": "piper-ryan-low",
- "id": "utmos200_035",
- "category": "names_places",
- "reference": "Aurelia and Joaquin crossed the Schuylkill River before noon.",
- "hypothesis": "Aurelia and Joaquin crossed the Chiocul River before noon.",
- "reference_words": 9,
- "hypothesis_words": 9,
- "substitutions": 1,
- "deletions": 0,
- "insertions": 0,
- "semantic_wer": 0.1111111111111111,
- "utterance_failure": true,
- "severe_failure": false,
- "empty_hypothesis": false,
- "leading_omission": false,
- "ending_omission": false,
- "insertion_burst": false,
- "adjacent_phrase_repetition": false,
- "audio": "utmos22_large200_v1/audio/piper-ryan-low/utmos200_035.wav"
- },
- {
- "model": "inflect-micro-v2-finalist",
- "id": "utmos200_018",
- "category": "homographs",
- "reference": "They refuse to place the refuse beside the produce aisle.",
- "hypothesis": "They refused to place the refuse beside the produce aisle.",
- "reference_words": 10,
- "hypothesis_words": 10,
- "substitutions": 1,
- "deletions": 0,
- "insertions": 0,
- "semantic_wer": 0.1,
- "utterance_failure": true,
- "severe_failure": false,
- "empty_hypothesis": false,
- "leading_omission": false,
- "ending_omission": false,
- "insertion_burst": false,
- "adjacent_phrase_repetition": false,
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_018.wav"
- },
- {
- "model": "inflect-micro-v2-finalist",
- "id": "utmos200_023",
- "category": "numbers",
- "reference": "Her score improved from eighty-two percent to ninety-seven percent in two weeks.",
- "hypothesis": "GERS score improved from 82% to 97% in two weeks.",
- "reference_words": 10,
- "hypothesis_words": 10,
- "substitutions": 1,
- "deletions": 0,
- "insertions": 0,
- "semantic_wer": 0.1,
- "utterance_failure": true,
- "severe_failure": false,
- "empty_hypothesis": false,
- "leading_omission": false,
- "ending_omission": false,
- "insertion_burst": false,
- "adjacent_phrase_repetition": false,
- "audio": "release_polish_overnight_20260720/final_large200/audio/inflect-micro-v2-finalist/utmos200_023.wav"
- }
- ]
-}
diff --git a/evaluation/asr_robustness.json b/evaluation/asr_robustness.json
deleted file mode 100644
index 48d3e263e3434cfc3ec295c99ebaa346e9b98a23..0000000000000000000000000000000000000000
--- a/evaluation/asr_robustness.json
+++ /dev/null
@@ -1,928 +0,0 @@
-{
- "format": "inflect_tts_asr_robustness_summary_v1",
- "source_format": "inflect_tts_multi_asr_wer_v1",
- "asr": "whisper-large-v3",
- "notes": [
- "Character error rate is calculated over semantic-normalized ASR transcripts.",
- "Sentence error rate counts any utterance with at least one semantic word edit.",
- "These metrics measure ASR recoverability and do not replace human listening."
- ],
- "systems": {
- "inflect-micro-v2-finalist": {
- "overall": {
- "utterances": 200,
- "semantic_errors": 32,
- "semantic_words": 2608,
- "semantic_wer": 0.012269938650306749,
- "failed_utterances": 21,
- "sentence_error_rate": 0.105,
- "exact_sentence_rate": 0.895,
- "character_errors": 62,
- "reference_characters": 13718,
- "character_error_rate": 0.004519609272488701,
- "mean_utterance_wer": 0.021512501387501385,
- "p95_utterance_wer": 0.18181818181818182,
- "worst_utterance_wer": 1.0
- },
- "categories": {
- "conversational": {
- "utterances": 25,
- "semantic_errors": 0,
- "semantic_words": 309,
- "semantic_wer": 0.0,
- "failed_utterances": 0,
- "sentence_error_rate": 0.0,
- "exact_sentence_rate": 1.0,
- "character_errors": 0,
- "reference_characters": 1501,
- "character_error_rate": 0.0,
- "mean_utterance_wer": 0.0,
- "p95_utterance_wer": 0.0,
- "worst_utterance_wer": 0.0
- },
- "descriptive": {
- "utterances": 16,
- "semantic_errors": 0,
- "semantic_words": 226,
- "semantic_wer": 0.0,
- "failed_utterances": 0,
- "sentence_error_rate": 0.0,
- "exact_sentence_rate": 1.0,
- "character_errors": 0,
- "reference_characters": 1146,
- "character_error_rate": 0.0,
- "mean_utterance_wer": 0.0,
- "p95_utterance_wer": 0.0,
- "worst_utterance_wer": 0.0
- },
- "emotional": {
- "utterances": 24,
- "semantic_errors": 0,
- "semantic_words": 329,
- "semantic_wer": 0.0,
- "failed_utterances": 0,
- "sentence_error_rate": 0.0,
- "exact_sentence_rate": 1.0,
- "character_errors": 0,
- "reference_characters": 1574,
- "character_error_rate": 0.0,
- "mean_utterance_wer": 0.0,
- "p95_utterance_wer": 0.0,
- "worst_utterance_wer": 0.0
- },
- "homographs": {
- "utterances": 6,
- "semantic_errors": 1,
- "semantic_words": 70,
- "semantic_wer": 0.014285714285714285,
- "failed_utterances": 1,
- "sentence_error_rate": 0.16666666666666666,
- "exact_sentence_rate": 0.8333333333333334,
- "character_errors": 1,
- "reference_characters": 375,
- "character_error_rate": 0.0026666666666666666,
- "mean_utterance_wer": 0.016666666666666666,
- "p95_utterance_wer": 0.07500000000000001,
- "worst_utterance_wer": 0.1
- },
- "long": {
- "utterances": 15,
- "semantic_errors": 1,
- "semantic_words": 432,
- "semantic_wer": 0.0023148148148148147,
- "failed_utterances": 1,
- "sentence_error_rate": 0.06666666666666667,
- "exact_sentence_rate": 0.9333333333333333,
- "character_errors": 6,
- "reference_characters": 2206,
- "character_error_rate": 0.0027198549410698096,
- "mean_utterance_wer": 0.0018518518518518517,
- "p95_utterance_wer": 0.008333333333333304,
- "worst_utterance_wer": 0.027777777777777776
- },
- "long_range": {
- "utterances": 6,
- "semantic_errors": 0,
- "semantic_words": 121,
- "semantic_wer": 0.0,
- "failed_utterances": 0,
- "sentence_error_rate": 0.0,
- "exact_sentence_rate": 1.0,
- "character_errors": 0,
- "reference_characters": 744,
- "character_error_rate": 0.0,
- "mean_utterance_wer": 0.0,
- "p95_utterance_wer": 0.0,
- "worst_utterance_wer": 0.0
- },
- "names_places": {
- "utterances": 6,
- "semantic_errors": 4,
- "semantic_words": 57,
- "semantic_wer": 0.07017543859649122,
- "failed_utterances": 3,
- "sentence_error_rate": 0.5,
- "exact_sentence_rate": 0.5,
- "character_errors": 13,
- "reference_characters": 397,
- "character_error_rate": 0.0327455919395466,
- "mean_utterance_wer": 0.07407407407407407,
- "p95_utterance_wer": 0.19444444444444442,
- "worst_utterance_wer": 0.2222222222222222
- },
- "narrative": {
- "utterances": 24,
- "semantic_errors": 1,
- "semantic_words": 352,
- "semantic_wer": 0.002840909090909091,
- "failed_utterances": 1,
- "sentence_error_rate": 0.041666666666666664,
- "exact_sentence_rate": 0.9583333333333334,
- "character_errors": 1,
- "reference_characters": 1762,
- "character_error_rate": 0.0005675368898978433,
- "mean_utterance_wer": 0.002777777777777778,
- "p95_utterance_wer": 0.0,
- "worst_utterance_wer": 0.06666666666666667
- },
- "numbers": {
- "utterances": 6,
- "semantic_errors": 10,
- "semantic_words": 64,
- "semantic_wer": 0.15625,
- "failed_utterances": 5,
- "sentence_error_rate": 0.8333333333333334,
- "exact_sentence_rate": 0.16666666666666663,
- "character_errors": 18,
- "reference_characters": 322,
- "character_error_rate": 0.055900621118012424,
- "mean_utterance_wer": 0.17462121212121215,
- "p95_utterance_wer": 0.35625,
- "worst_utterance_wer": 0.375
- },
- "punctuation": {
- "utterances": 6,
- "semantic_errors": 0,
- "semantic_words": 65,
- "semantic_wer": 0.0,
- "failed_utterances": 0,
- "sentence_error_rate": 0.0,
- "exact_sentence_rate": 1.0,
- "character_errors": 0,
- "reference_characters": 402,
- "character_error_rate": 0.0,
- "mean_utterance_wer": 0.0,
- "p95_utterance_wer": 0.0,
- "worst_utterance_wer": 0.0
- },
- "question": {
- "utterances": 15,
- "semantic_errors": 0,
- "semantic_words": 165,
- "semantic_wer": 0.0,
- "failed_utterances": 0,
- "sentence_error_rate": 0.0,
- "exact_sentence_rate": 1.0,
- "character_errors": 0,
- "reference_characters": 834,
- "character_error_rate": 0.0,
- "mean_utterance_wer": 0.0,
- "p95_utterance_wer": 0.0,
- "worst_utterance_wer": 0.0
- },
- "rare_words": {
- "utterances": 6,
- "semantic_errors": 3,
- "semantic_words": 57,
- "semantic_wer": 0.05263157894736842,
- "failed_utterances": 2,
- "sentence_error_rate": 0.3333333333333333,
- "exact_sentence_rate": 0.6666666666666667,
- "character_errors": 3,
- "reference_characters": 500,
- "character_error_rate": 0.006,
- "mean_utterance_wer": 0.046969696969696974,
- "p95_utterance_wer": 0.16136363636363635,
- "worst_utterance_wer": 0.18181818181818182
- },
- "short": {
- "utterances": 24,
- "semantic_errors": 4,
- "semantic_words": 105,
- "semantic_wer": 0.0380952380952381,
- "failed_utterances": 4,
- "sentence_error_rate": 0.16666666666666666,
- "exact_sentence_rate": 0.8333333333333334,
- "character_errors": 10,
- "reference_characters": 458,
- "character_error_rate": 0.021834061135371178,
- "mean_utterance_wer": 0.07083333333333333,
- "p95_utterance_wer": 0.25,
- "worst_utterance_wer": 1.0
- },
- "technical": {
- "utterances": 21,
- "semantic_errors": 8,
- "semantic_words": 256,
- "semantic_wer": 0.03125,
- "failed_utterances": 4,
- "sentence_error_rate": 0.19047619047619047,
- "exact_sentence_rate": 0.8095238095238095,
- "character_errors": 10,
- "reference_characters": 1497,
- "character_error_rate": 0.006680026720106881,
- "mean_utterance_wer": 0.030193615907901622,
- "p95_utterance_wer": 0.2,
- "worst_utterance_wer": 0.2857142857142857
- }
- }
- },
- "inflect-nano-v2-finalist": {
- "overall": {
- "utterances": 200,
- "semantic_errors": 40,
- "semantic_words": 2608,
- "semantic_wer": 0.015337423312883436,
- "failed_utterances": 23,
- "sentence_error_rate": 0.115,
- "exact_sentence_rate": 0.885,
- "character_errors": 102,
- "reference_characters": 13718,
- "character_error_rate": 0.007435486222481412,
- "mean_utterance_wer": 0.021362234987234986,
- "p95_utterance_wer": 0.18272727272727243,
- "worst_utterance_wer": 0.4
- },
- "categories": {
- "conversational": {
- "utterances": 25,
- "semantic_errors": 1,
- "semantic_words": 309,
- "semantic_wer": 0.003236245954692557,
- "failed_utterances": 1,
- "sentence_error_rate": 0.04,
- "exact_sentence_rate": 0.96,
- "character_errors": 5,
- "reference_characters": 1501,
- "character_error_rate": 0.0033311125916055963,
- "mean_utterance_wer": 0.0036363636363636364,
- "p95_utterance_wer": 0.0,
- "worst_utterance_wer": 0.09090909090909091
- },
- "descriptive": {
- "utterances": 16,
- "semantic_errors": 0,
- "semantic_words": 226,
- "semantic_wer": 0.0,
- "failed_utterances": 0,
- "sentence_error_rate": 0.0,
- "exact_sentence_rate": 1.0,
- "character_errors": 0,
- "reference_characters": 1146,
- "character_error_rate": 0.0,
- "mean_utterance_wer": 0.0,
- "p95_utterance_wer": 0.0,
- "worst_utterance_wer": 0.0
- },
- "emotional": {
- "utterances": 24,
- "semantic_errors": 0,
- "semantic_words": 329,
- "semantic_wer": 0.0,
- "failed_utterances": 0,
- "sentence_error_rate": 0.0,
- "exact_sentence_rate": 1.0,
- "character_errors": 0,
- "reference_characters": 1574,
- "character_error_rate": 0.0,
- "mean_utterance_wer": 0.0,
- "p95_utterance_wer": 0.0,
- "worst_utterance_wer": 0.0
- },
- "homographs": {
- "utterances": 6,
- "semantic_errors": 4,
- "semantic_words": 70,
- "semantic_wer": 0.05714285714285714,
- "failed_utterances": 2,
- "sentence_error_rate": 0.3333333333333333,
- "exact_sentence_rate": 0.6666666666666667,
- "character_errors": 10,
- "reference_characters": 375,
- "character_error_rate": 0.02666666666666667,
- "mean_utterance_wer": 0.05000000000000001,
- "p95_utterance_wer": 0.17500000000000002,
- "worst_utterance_wer": 0.2
- },
- "long": {
- "utterances": 15,
- "semantic_errors": 2,
- "semantic_words": 432,
- "semantic_wer": 0.004629629629629629,
- "failed_utterances": 1,
- "sentence_error_rate": 0.06666666666666667,
- "exact_sentence_rate": 0.9333333333333333,
- "character_errors": 5,
- "reference_characters": 2206,
- "character_error_rate": 0.0022665457842248413,
- "mean_utterance_wer": 0.00404040404040404,
- "p95_utterance_wer": 0.01818181818181812,
- "worst_utterance_wer": 0.06060606060606061
- },
- "long_range": {
- "utterances": 6,
- "semantic_errors": 1,
- "semantic_words": 121,
- "semantic_wer": 0.008264462809917356,
- "failed_utterances": 1,
- "sentence_error_rate": 0.16666666666666666,
- "exact_sentence_rate": 0.8333333333333334,
- "character_errors": 2,
- "reference_characters": 744,
- "character_error_rate": 0.002688172043010753,
- "mean_utterance_wer": 0.008333333333333333,
- "p95_utterance_wer": 0.037500000000000006,
- "worst_utterance_wer": 0.05
- },
- "names_places": {
- "utterances": 6,
- "semantic_errors": 5,
- "semantic_words": 57,
- "semantic_wer": 0.08771929824561403,
- "failed_utterances": 3,
- "sentence_error_rate": 0.5,
- "exact_sentence_rate": 0.5,
- "character_errors": 21,
- "reference_characters": 397,
- "character_error_rate": 0.05289672544080604,
- "mean_utterance_wer": 0.09074074074074073,
- "p95_utterance_wer": 0.2777777777777778,
- "worst_utterance_wer": 0.3333333333333333
- },
- "narrative": {
- "utterances": 24,
- "semantic_errors": 4,
- "semantic_words": 352,
- "semantic_wer": 0.011363636363636364,
- "failed_utterances": 3,
- "sentence_error_rate": 0.125,
- "exact_sentence_rate": 0.875,
- "character_errors": 19,
- "reference_characters": 1762,
- "character_error_rate": 0.010783200908059024,
- "mean_utterance_wer": 0.013123219373219372,
- "p95_utterance_wer": 0.10194444444444431,
- "worst_utterance_wer": 0.15384615384615385
- },
- "numbers": {
- "utterances": 6,
- "semantic_errors": 8,
- "semantic_words": 64,
- "semantic_wer": 0.125,
- "failed_utterances": 3,
- "sentence_error_rate": 0.5,
- "exact_sentence_rate": 0.5,
- "character_errors": 16,
- "reference_characters": 322,
- "character_error_rate": 0.049689440993788817,
- "mean_utterance_wer": 0.1428030303030303,
- "p95_utterance_wer": 0.35625,
- "worst_utterance_wer": 0.375
- },
- "punctuation": {
- "utterances": 6,
- "semantic_errors": 0,
- "semantic_words": 65,
- "semantic_wer": 0.0,
- "failed_utterances": 0,
- "sentence_error_rate": 0.0,
- "exact_sentence_rate": 1.0,
- "character_errors": 0,
- "reference_characters": 402,
- "character_error_rate": 0.0,
- "mean_utterance_wer": 0.0,
- "p95_utterance_wer": 0.0,
- "worst_utterance_wer": 0.0
- },
- "question": {
- "utterances": 15,
- "semantic_errors": 0,
- "semantic_words": 165,
- "semantic_wer": 0.0,
- "failed_utterances": 0,
- "sentence_error_rate": 0.0,
- "exact_sentence_rate": 1.0,
- "character_errors": 0,
- "reference_characters": 834,
- "character_error_rate": 0.0,
- "mean_utterance_wer": 0.0,
- "p95_utterance_wer": 0.0,
- "worst_utterance_wer": 0.0
- },
- "rare_words": {
- "utterances": 6,
- "semantic_errors": 2,
- "semantic_words": 57,
- "semantic_wer": 0.03508771929824561,
- "failed_utterances": 1,
- "sentence_error_rate": 0.16666666666666666,
- "exact_sentence_rate": 0.8333333333333334,
- "character_errors": 1,
- "reference_characters": 500,
- "character_error_rate": 0.002,
- "mean_utterance_wer": 0.030303030303030304,
- "p95_utterance_wer": 0.13636363636363635,
- "worst_utterance_wer": 0.18181818181818182
- },
- "short": {
- "utterances": 24,
- "semantic_errors": 5,
- "semantic_words": 105,
- "semantic_wer": 0.047619047619047616,
- "failed_utterances": 4,
- "sentence_error_rate": 0.16666666666666666,
- "exact_sentence_rate": 0.8333333333333334,
- "character_errors": 17,
- "reference_characters": 458,
- "character_error_rate": 0.03711790393013101,
- "mean_utterance_wer": 0.051388888888888894,
- "p95_utterance_wer": 0.32083333333333314,
- "worst_utterance_wer": 0.4
- },
- "technical": {
- "utterances": 21,
- "semantic_errors": 8,
- "semantic_words": 256,
- "semantic_wer": 0.03125,
- "failed_utterances": 4,
- "sentence_error_rate": 0.19047619047619047,
- "exact_sentence_rate": 0.8095238095238095,
- "character_errors": 6,
- "reference_characters": 1497,
- "character_error_rate": 0.004008016032064128,
- "mean_utterance_wer": 0.030455259026687595,
- "p95_utterance_wer": 0.2,
- "worst_utterance_wer": 0.2857142857142857
- }
- }
- },
- "kitten-nano-bruno": {
- "overall": {
- "utterances": 200,
- "semantic_errors": 25,
- "semantic_words": 2608,
- "semantic_wer": 0.009585889570552147,
- "failed_utterances": 15,
- "sentence_error_rate": 0.075,
- "exact_sentence_rate": 0.925,
- "character_errors": 41,
- "reference_characters": 13718,
- "character_error_rate": 0.002988773873742528,
- "mean_utterance_wer": 0.012458457692977818,
- "p95_utterance_wer": 0.09136363636363622,
- "worst_utterance_wer": 0.375
- },
- "categories": {
- "conversational": {
- "utterances": 25,
- "semantic_errors": 1,
- "semantic_words": 309,
- "semantic_wer": 0.003236245954692557,
- "failed_utterances": 1,
- "sentence_error_rate": 0.04,
- "exact_sentence_rate": 0.96,
- "character_errors": 2,
- "reference_characters": 1501,
- "character_error_rate": 0.0013324450366422385,
- "mean_utterance_wer": 0.002857142857142857,
- "p95_utterance_wer": 0.0,
- "worst_utterance_wer": 0.07142857142857142
- },
- "descriptive": {
- "utterances": 16,
- "semantic_errors": 0,
- "semantic_words": 226,
- "semantic_wer": 0.0,
- "failed_utterances": 0,
- "sentence_error_rate": 0.0,
- "exact_sentence_rate": 1.0,
- "character_errors": 0,
- "reference_characters": 1146,
- "character_error_rate": 0.0,
- "mean_utterance_wer": 0.0,
- "p95_utterance_wer": 0.0,
- "worst_utterance_wer": 0.0
- },
- "emotional": {
- "utterances": 24,
- "semantic_errors": 0,
- "semantic_words": 329,
- "semantic_wer": 0.0,
- "failed_utterances": 0,
- "sentence_error_rate": 0.0,
- "exact_sentence_rate": 1.0,
- "character_errors": 0,
- "reference_characters": 1574,
- "character_error_rate": 0.0,
- "mean_utterance_wer": 0.0,
- "p95_utterance_wer": 0.0,
- "worst_utterance_wer": 0.0
- },
- "homographs": {
- "utterances": 6,
- "semantic_errors": 1,
- "semantic_words": 70,
- "semantic_wer": 0.014285714285714285,
- "failed_utterances": 1,
- "sentence_error_rate": 0.16666666666666666,
- "exact_sentence_rate": 0.8333333333333334,
- "character_errors": 1,
- "reference_characters": 375,
- "character_error_rate": 0.0026666666666666666,
- "mean_utterance_wer": 0.016666666666666666,
- "p95_utterance_wer": 0.07500000000000001,
- "worst_utterance_wer": 0.1
- },
- "long": {
- "utterances": 15,
- "semantic_errors": 1,
- "semantic_words": 432,
- "semantic_wer": 0.0023148148148148147,
- "failed_utterances": 1,
- "sentence_error_rate": 0.06666666666666667,
- "exact_sentence_rate": 0.9333333333333333,
- "character_errors": 4,
- "reference_characters": 2206,
- "character_error_rate": 0.001813236627379873,
- "mean_utterance_wer": 0.002564102564102564,
- "p95_utterance_wer": 0.011538461538461497,
- "worst_utterance_wer": 0.038461538461538464
- },
- "long_range": {
- "utterances": 6,
- "semantic_errors": 0,
- "semantic_words": 121,
- "semantic_wer": 0.0,
- "failed_utterances": 0,
- "sentence_error_rate": 0.0,
- "exact_sentence_rate": 1.0,
- "character_errors": 0,
- "reference_characters": 744,
- "character_error_rate": 0.0,
- "mean_utterance_wer": 0.0,
- "p95_utterance_wer": 0.0,
- "worst_utterance_wer": 0.0
- },
- "names_places": {
- "utterances": 6,
- "semantic_errors": 3,
- "semantic_words": 57,
- "semantic_wer": 0.05263157894736842,
- "failed_utterances": 1,
- "sentence_error_rate": 0.16666666666666666,
- "exact_sentence_rate": 0.8333333333333334,
- "character_errors": 12,
- "reference_characters": 397,
- "character_error_rate": 0.030226700251889168,
- "mean_utterance_wer": 0.05555555555555555,
- "p95_utterance_wer": 0.25,
- "worst_utterance_wer": 0.3333333333333333
- },
- "narrative": {
- "utterances": 24,
- "semantic_errors": 4,
- "semantic_words": 352,
- "semantic_wer": 0.011363636363636364,
- "failed_utterances": 3,
- "sentence_error_rate": 0.125,
- "exact_sentence_rate": 0.875,
- "character_errors": 10,
- "reference_characters": 1762,
- "character_error_rate": 0.0056753688989784334,
- "mean_utterance_wer": 0.009852296181630547,
- "p95_utterance_wer": 0.05789473684210525,
- "worst_utterance_wer": 0.125
- },
- "numbers": {
- "utterances": 6,
- "semantic_errors": 6,
- "semantic_words": 64,
- "semantic_wer": 0.09375,
- "failed_utterances": 3,
- "sentence_error_rate": 0.5,
- "exact_sentence_rate": 0.5,
- "character_errors": 5,
- "reference_characters": 322,
- "character_error_rate": 0.015527950310559006,
- "mean_utterance_wer": 0.11098484848484848,
- "p95_utterance_wer": 0.33125,
- "worst_utterance_wer": 0.375
- },
- "punctuation": {
- "utterances": 6,
- "semantic_errors": 0,
- "semantic_words": 65,
- "semantic_wer": 0.0,
- "failed_utterances": 0,
- "sentence_error_rate": 0.0,
- "exact_sentence_rate": 1.0,
- "character_errors": 0,
- "reference_characters": 402,
- "character_error_rate": 0.0,
- "mean_utterance_wer": 0.0,
- "p95_utterance_wer": 0.0,
- "worst_utterance_wer": 0.0
- },
- "question": {
- "utterances": 15,
- "semantic_errors": 0,
- "semantic_words": 165,
- "semantic_wer": 0.0,
- "failed_utterances": 0,
- "sentence_error_rate": 0.0,
- "exact_sentence_rate": 1.0,
- "character_errors": 0,
- "reference_characters": 834,
- "character_error_rate": 0.0,
- "mean_utterance_wer": 0.0,
- "p95_utterance_wer": 0.0,
- "worst_utterance_wer": 0.0
- },
- "rare_words": {
- "utterances": 6,
- "semantic_errors": 2,
- "semantic_words": 57,
- "semantic_wer": 0.03508771929824561,
- "failed_utterances": 1,
- "sentence_error_rate": 0.16666666666666666,
- "exact_sentence_rate": 0.8333333333333334,
- "character_errors": 1,
- "reference_characters": 500,
- "character_error_rate": 0.002,
- "mean_utterance_wer": 0.030303030303030304,
- "p95_utterance_wer": 0.13636363636363635,
- "worst_utterance_wer": 0.18181818181818182
- },
- "short": {
- "utterances": 24,
- "semantic_errors": 2,
- "semantic_words": 105,
- "semantic_wer": 0.01904761904761905,
- "failed_utterances": 2,
- "sentence_error_rate": 0.08333333333333333,
- "exact_sentence_rate": 0.9166666666666666,
- "character_errors": 3,
- "reference_characters": 458,
- "character_error_rate": 0.006550218340611353,
- "mean_utterance_wer": 0.01875,
- "p95_utterance_wer": 0.1699999999999996,
- "worst_utterance_wer": 0.25
- },
- "technical": {
- "utterances": 21,
- "semantic_errors": 5,
- "semantic_words": 256,
- "semantic_wer": 0.01953125,
- "failed_utterances": 2,
- "sentence_error_rate": 0.09523809523809523,
- "exact_sentence_rate": 0.9047619047619048,
- "character_errors": 3,
- "reference_characters": 1497,
- "character_error_rate": 0.002004008016032064,
- "mean_utterance_wer": 0.019727891156462583,
- "p95_utterance_wer": 0.2,
- "worst_utterance_wer": 0.21428571428571427
- }
- }
- },
- "piper-ryan-low": {
- "overall": {
- "utterances": 200,
- "semantic_errors": 32,
- "semantic_words": 2608,
- "semantic_wer": 0.012269938650306749,
- "failed_utterances": 19,
- "sentence_error_rate": 0.095,
- "exact_sentence_rate": 0.905,
- "character_errors": 67,
- "reference_characters": 13718,
- "character_error_rate": 0.00488409389123779,
- "mean_utterance_wer": 0.01766635309135309,
- "p95_utterance_wer": 0.11464646464646343,
- "worst_utterance_wer": 0.6
- },
- "categories": {
- "conversational": {
- "utterances": 25,
- "semantic_errors": 2,
- "semantic_words": 309,
- "semantic_wer": 0.006472491909385114,
- "failed_utterances": 1,
- "sentence_error_rate": 0.04,
- "exact_sentence_rate": 0.96,
- "character_errors": 5,
- "reference_characters": 1501,
- "character_error_rate": 0.0033311125916055963,
- "mean_utterance_wer": 0.008,
- "p95_utterance_wer": 0.0,
- "worst_utterance_wer": 0.2
- },
- "descriptive": {
- "utterances": 16,
- "semantic_errors": 2,
- "semantic_words": 226,
- "semantic_wer": 0.008849557522123894,
- "failed_utterances": 2,
- "sentence_error_rate": 0.125,
- "exact_sentence_rate": 0.875,
- "character_errors": 5,
- "reference_characters": 1146,
- "character_error_rate": 0.004363001745200698,
- "mean_utterance_wer": 0.008630952380952382,
- "p95_utterance_wer": 0.06785714285714287,
- "worst_utterance_wer": 0.07142857142857142
- },
- "emotional": {
- "utterances": 24,
- "semantic_errors": 0,
- "semantic_words": 329,
- "semantic_wer": 0.0,
- "failed_utterances": 0,
- "sentence_error_rate": 0.0,
- "exact_sentence_rate": 1.0,
- "character_errors": 0,
- "reference_characters": 1574,
- "character_error_rate": 0.0,
- "mean_utterance_wer": 0.0,
- "p95_utterance_wer": 0.0,
- "worst_utterance_wer": 0.0
- },
- "homographs": {
- "utterances": 6,
- "semantic_errors": 1,
- "semantic_words": 70,
- "semantic_wer": 0.014285714285714285,
- "failed_utterances": 1,
- "sentence_error_rate": 0.16666666666666666,
- "exact_sentence_rate": 0.8333333333333334,
- "character_errors": 1,
- "reference_characters": 375,
- "character_error_rate": 0.0026666666666666666,
- "mean_utterance_wer": 0.016666666666666666,
- "p95_utterance_wer": 0.07500000000000001,
- "worst_utterance_wer": 0.1
- },
- "long": {
- "utterances": 15,
- "semantic_errors": 2,
- "semantic_words": 432,
- "semantic_wer": 0.004629629629629629,
- "failed_utterances": 2,
- "sentence_error_rate": 0.13333333333333333,
- "exact_sentence_rate": 0.8666666666666667,
- "character_errors": 9,
- "reference_characters": 2206,
- "character_error_rate": 0.004079782411604714,
- "mean_utterance_wer": 0.0052307692307692315,
- "p95_utterance_wer": 0.03892307692307692,
- "worst_utterance_wer": 0.04
- },
- "long_range": {
- "utterances": 6,
- "semantic_errors": 0,
- "semantic_words": 121,
- "semantic_wer": 0.0,
- "failed_utterances": 0,
- "sentence_error_rate": 0.0,
- "exact_sentence_rate": 1.0,
- "character_errors": 0,
- "reference_characters": 744,
- "character_error_rate": 0.0,
- "mean_utterance_wer": 0.0,
- "p95_utterance_wer": 0.0,
- "worst_utterance_wer": 0.0
- },
- "names_places": {
- "utterances": 6,
- "semantic_errors": 4,
- "semantic_words": 57,
- "semantic_wer": 0.07017543859649122,
- "failed_utterances": 2,
- "sentence_error_rate": 0.3333333333333333,
- "exact_sentence_rate": 0.6666666666666667,
- "character_errors": 19,
- "reference_characters": 397,
- "character_error_rate": 0.04785894206549118,
- "mean_utterance_wer": 0.07407407407407407,
- "p95_utterance_wer": 0.2777777777777778,
- "worst_utterance_wer": 0.3333333333333333
- },
- "narrative": {
- "utterances": 24,
- "semantic_errors": 0,
- "semantic_words": 352,
- "semantic_wer": 0.0,
- "failed_utterances": 0,
- "sentence_error_rate": 0.0,
- "exact_sentence_rate": 1.0,
- "character_errors": 0,
- "reference_characters": 1762,
- "character_error_rate": 0.0,
- "mean_utterance_wer": 0.0,
- "p95_utterance_wer": 0.0,
- "worst_utterance_wer": 0.0
- },
- "numbers": {
- "utterances": 6,
- "semantic_errors": 5,
- "semantic_words": 64,
- "semantic_wer": 0.078125,
- "failed_utterances": 2,
- "sentence_error_rate": 0.3333333333333333,
- "exact_sentence_rate": 0.6666666666666667,
- "character_errors": 3,
- "reference_characters": 322,
- "character_error_rate": 0.009316770186335404,
- "mean_utterance_wer": 0.09583333333333333,
- "p95_utterance_wer": 0.33125,
- "worst_utterance_wer": 0.375
- },
- "punctuation": {
- "utterances": 6,
- "semantic_errors": 0,
- "semantic_words": 65,
- "semantic_wer": 0.0,
- "failed_utterances": 0,
- "sentence_error_rate": 0.0,
- "exact_sentence_rate": 1.0,
- "character_errors": 0,
- "reference_characters": 402,
- "character_error_rate": 0.0,
- "mean_utterance_wer": 0.0,
- "p95_utterance_wer": 0.0,
- "worst_utterance_wer": 0.0
- },
- "question": {
- "utterances": 15,
- "semantic_errors": 0,
- "semantic_words": 165,
- "semantic_wer": 0.0,
- "failed_utterances": 0,
- "sentence_error_rate": 0.0,
- "exact_sentence_rate": 1.0,
- "character_errors": 0,
- "reference_characters": 834,
- "character_error_rate": 0.0,
- "mean_utterance_wer": 0.0,
- "p95_utterance_wer": 0.0,
- "worst_utterance_wer": 0.0
- },
- "rare_words": {
- "utterances": 6,
- "semantic_errors": 5,
- "semantic_words": 57,
- "semantic_wer": 0.08771929824561403,
- "failed_utterances": 3,
- "sentence_error_rate": 0.5,
- "exact_sentence_rate": 0.5,
- "character_errors": 9,
- "reference_characters": 500,
- "character_error_rate": 0.018,
- "mean_utterance_wer": 0.0803030303030303,
- "p95_utterance_wer": 0.1954545454545455,
- "worst_utterance_wer": 0.2
- },
- "short": {
- "utterances": 24,
- "semantic_errors": 4,
- "semantic_words": 105,
- "semantic_wer": 0.0380952380952381,
- "failed_utterances": 2,
- "sentence_error_rate": 0.08333333333333333,
- "exact_sentence_rate": 0.9166666666666666,
- "character_errors": 10,
- "reference_characters": 458,
- "character_error_rate": 0.021834061135371178,
- "mean_utterance_wer": 0.03888888888888889,
- "p95_utterance_wer": 0.2833333333333326,
- "worst_utterance_wer": 0.6
- },
- "technical": {
- "utterances": 21,
- "semantic_errors": 7,
- "semantic_words": 256,
- "semantic_wer": 0.02734375,
- "failed_utterances": 4,
- "sentence_error_rate": 0.19047619047619047,
- "exact_sentence_rate": 0.8095238095238095,
- "character_errors": 6,
- "reference_characters": 1497,
- "character_error_rate": 0.004008016032064128,
- "mean_utterance_wer": 0.027719899148470575,
- "p95_utterance_wer": 0.2,
- "worst_utterance_wer": 0.21428571428571427
- }
- }
- }
- }
-}
diff --git a/evaluation/final/evidence-summary.json b/evaluation/final/evidence-summary.json
new file mode 100644
index 0000000000000000000000000000000000000000..957e6d0d7d59cfe98a048d24929393cb15c7b356
--- /dev/null
+++ b/evaluation/final/evidence-summary.json
@@ -0,0 +1,1990 @@
+{
+ "format": "inflect_v2_release_evidence_v1",
+ "systems": {
+ "inflect-micro-v2": {
+ "label": "Inflect-Micro-v2",
+ "weight_mb": 37.529995,
+ "semantic_wer": {
+ "wer": 0.019293218720152817,
+ "ci": [
+ 0.014622733476311172,
+ 0.024179280518969924
+ ],
+ "categories": {
+ "conversational": {
+ "raw": {
+ "utterances": 25,
+ "errors": 0,
+ "words": 296,
+ "wer": 0.0
+ },
+ "semantic": {
+ "utterances": 25,
+ "errors": 0,
+ "words": 309,
+ "wer": 0.0
+ }
+ },
+ "descriptive": {
+ "raw": {
+ "utterances": 16,
+ "errors": 2,
+ "words": 217,
+ "wer": 0.009216589861751152
+ },
+ "semantic": {
+ "utterances": 16,
+ "errors": 1,
+ "words": 226,
+ "wer": 0.004424778761061947
+ }
+ },
+ "emotional": {
+ "raw": {
+ "utterances": 24,
+ "errors": 0,
+ "words": 314,
+ "wer": 0.0
+ },
+ "semantic": {
+ "utterances": 24,
+ "errors": 0,
+ "words": 329,
+ "wer": 0.0
+ }
+ },
+ "homographs": {
+ "raw": {
+ "utterances": 6,
+ "errors": 2,
+ "words": 70,
+ "wer": 0.02857142857142857
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 2,
+ "words": 70,
+ "wer": 0.02857142857142857
+ }
+ },
+ "long": {
+ "raw": {
+ "utterances": 15,
+ "errors": 3,
+ "words": 412,
+ "wer": 0.007281553398058253
+ },
+ "semantic": {
+ "utterances": 15,
+ "errors": 3,
+ "words": 432,
+ "wer": 0.006944444444444444
+ }
+ },
+ "long_range": {
+ "raw": {
+ "utterances": 6,
+ "errors": 0,
+ "words": 121,
+ "wer": 0.0
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 0,
+ "words": 121,
+ "wer": 0.0
+ }
+ },
+ "names_places": {
+ "raw": {
+ "utterances": 6,
+ "errors": 4,
+ "words": 57,
+ "wer": 0.07017543859649122
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 4,
+ "words": 57,
+ "wer": 0.07017543859649122
+ }
+ },
+ "narrative": {
+ "raw": {
+ "utterances": 24,
+ "errors": 2,
+ "words": 342,
+ "wer": 0.005847953216374269
+ },
+ "semantic": {
+ "utterances": 24,
+ "errors": 1,
+ "words": 352,
+ "wer": 0.002840909090909091
+ }
+ },
+ "numbers": {
+ "raw": {
+ "utterances": 6,
+ "errors": 33,
+ "words": 81,
+ "wer": 0.4074074074074074
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 8,
+ "words": 64,
+ "wer": 0.125
+ }
+ },
+ "phonetic_balance": {
+ "raw": {
+ "utterances": 300,
+ "errors": 82,
+ "words": 2605,
+ "wer": 0.03147792706333973
+ },
+ "semantic": {
+ "utterances": 300,
+ "errors": 68,
+ "words": 2627,
+ "wer": 0.02588503996954701
+ }
+ },
+ "punctuation": {
+ "raw": {
+ "utterances": 6,
+ "errors": 0,
+ "words": 65,
+ "wer": 0.0
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 0,
+ "words": 65,
+ "wer": 0.0
+ }
+ },
+ "question": {
+ "raw": {
+ "utterances": 15,
+ "errors": 0,
+ "words": 163,
+ "wer": 0.0
+ },
+ "semantic": {
+ "utterances": 15,
+ "errors": 0,
+ "words": 165,
+ "wer": 0.0
+ }
+ },
+ "rare_words": {
+ "raw": {
+ "utterances": 6,
+ "errors": 3,
+ "words": 57,
+ "wer": 0.05263157894736842
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 3,
+ "words": 57,
+ "wer": 0.05263157894736842
+ }
+ },
+ "short": {
+ "raw": {
+ "utterances": 24,
+ "errors": 3,
+ "words": 98,
+ "wer": 0.030612244897959183
+ },
+ "semantic": {
+ "utterances": 24,
+ "errors": 3,
+ "words": 105,
+ "wer": 0.02857142857142857
+ }
+ },
+ "technical": {
+ "raw": {
+ "utterances": 21,
+ "errors": 12,
+ "words": 251,
+ "wer": 0.04780876494023904
+ },
+ "semantic": {
+ "utterances": 21,
+ "errors": 8,
+ "words": 256,
+ "wer": 0.03125
+ }
+ }
+ }
+ },
+ "utmos22": {
+ "mean": 4.395012595653534,
+ "ci": [
+ 4.3812509111166,
+ 4.4079569326162344
+ ]
+ }
+ },
+ "inflect-nano-v2": {
+ "label": "Inflect-Nano-v2",
+ "weight_mb": 15.971083,
+ "semantic_wer": {
+ "wer": 0.024832855778414518,
+ "ci": [
+ 0.019525267993874426,
+ 0.030727866333781446
+ ],
+ "categories": {
+ "conversational": {
+ "raw": {
+ "utterances": 25,
+ "errors": 5,
+ "words": 296,
+ "wer": 0.016891891891891893
+ },
+ "semantic": {
+ "utterances": 25,
+ "errors": 3,
+ "words": 309,
+ "wer": 0.009708737864077669
+ }
+ },
+ "descriptive": {
+ "raw": {
+ "utterances": 16,
+ "errors": 2,
+ "words": 217,
+ "wer": 0.009216589861751152
+ },
+ "semantic": {
+ "utterances": 16,
+ "errors": 1,
+ "words": 226,
+ "wer": 0.004424778761061947
+ }
+ },
+ "emotional": {
+ "raw": {
+ "utterances": 24,
+ "errors": 1,
+ "words": 314,
+ "wer": 0.0031847133757961785
+ },
+ "semantic": {
+ "utterances": 24,
+ "errors": 1,
+ "words": 329,
+ "wer": 0.00303951367781155
+ }
+ },
+ "homographs": {
+ "raw": {
+ "utterances": 6,
+ "errors": 3,
+ "words": 70,
+ "wer": 0.04285714285714286
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 3,
+ "words": 70,
+ "wer": 0.04285714285714286
+ }
+ },
+ "long": {
+ "raw": {
+ "utterances": 15,
+ "errors": 3,
+ "words": 412,
+ "wer": 0.007281553398058253
+ },
+ "semantic": {
+ "utterances": 15,
+ "errors": 3,
+ "words": 432,
+ "wer": 0.006944444444444444
+ }
+ },
+ "long_range": {
+ "raw": {
+ "utterances": 6,
+ "errors": 1,
+ "words": 121,
+ "wer": 0.008264462809917356
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 1,
+ "words": 121,
+ "wer": 0.008264462809917356
+ }
+ },
+ "names_places": {
+ "raw": {
+ "utterances": 6,
+ "errors": 3,
+ "words": 57,
+ "wer": 0.05263157894736842
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 3,
+ "words": 57,
+ "wer": 0.05263157894736842
+ }
+ },
+ "narrative": {
+ "raw": {
+ "utterances": 24,
+ "errors": 5,
+ "words": 342,
+ "wer": 0.014619883040935672
+ },
+ "semantic": {
+ "utterances": 24,
+ "errors": 4,
+ "words": 352,
+ "wer": 0.011363636363636364
+ }
+ },
+ "numbers": {
+ "raw": {
+ "utterances": 6,
+ "errors": 33,
+ "words": 81,
+ "wer": 0.4074074074074074
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 6,
+ "words": 64,
+ "wer": 0.09375
+ }
+ },
+ "phonetic_balance": {
+ "raw": {
+ "utterances": 300,
+ "errors": 96,
+ "words": 2605,
+ "wer": 0.03685220729366603
+ },
+ "semantic": {
+ "utterances": 300,
+ "errors": 83,
+ "words": 2627,
+ "wer": 0.031594975256947086
+ }
+ },
+ "punctuation": {
+ "raw": {
+ "utterances": 6,
+ "errors": 0,
+ "words": 65,
+ "wer": 0.0
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 0,
+ "words": 65,
+ "wer": 0.0
+ }
+ },
+ "question": {
+ "raw": {
+ "utterances": 15,
+ "errors": 1,
+ "words": 163,
+ "wer": 0.006134969325153374
+ },
+ "semantic": {
+ "utterances": 15,
+ "errors": 1,
+ "words": 165,
+ "wer": 0.006060606060606061
+ }
+ },
+ "rare_words": {
+ "raw": {
+ "utterances": 6,
+ "errors": 5,
+ "words": 57,
+ "wer": 0.08771929824561403
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 5,
+ "words": 57,
+ "wer": 0.08771929824561403
+ }
+ },
+ "short": {
+ "raw": {
+ "utterances": 24,
+ "errors": 8,
+ "words": 98,
+ "wer": 0.08163265306122448
+ },
+ "semantic": {
+ "utterances": 24,
+ "errors": 8,
+ "words": 105,
+ "wer": 0.0761904761904762
+ }
+ },
+ "technical": {
+ "raw": {
+ "utterances": 21,
+ "errors": 12,
+ "words": 251,
+ "wer": 0.04780876494023904
+ },
+ "semantic": {
+ "utterances": 21,
+ "errors": 8,
+ "words": 256,
+ "wer": 0.03125
+ }
+ }
+ }
+ },
+ "utmos22": {
+ "mean": 4.3861008615493775,
+ "ci": [
+ 4.372184220826625,
+ 4.3987462292552
+ ]
+ }
+ },
+ "kitten-nano-bruno": {
+ "label": "KittenTTS Nano \u00b7 Bruno",
+ "weight_mb": 56.767095,
+ "semantic_wer": {
+ "wer": 0.009742120343839542,
+ "ci": [
+ 0.006622516556291391,
+ 0.013268998793727383
+ ],
+ "categories": {
+ "conversational": {
+ "raw": {
+ "utterances": 25,
+ "errors": 0,
+ "words": 296,
+ "wer": 0.0
+ },
+ "semantic": {
+ "utterances": 25,
+ "errors": 0,
+ "words": 309,
+ "wer": 0.0
+ }
+ },
+ "descriptive": {
+ "raw": {
+ "utterances": 16,
+ "errors": 1,
+ "words": 217,
+ "wer": 0.004608294930875576
+ },
+ "semantic": {
+ "utterances": 16,
+ "errors": 0,
+ "words": 226,
+ "wer": 0.0
+ }
+ },
+ "emotional": {
+ "raw": {
+ "utterances": 24,
+ "errors": 0,
+ "words": 314,
+ "wer": 0.0
+ },
+ "semantic": {
+ "utterances": 24,
+ "errors": 0,
+ "words": 329,
+ "wer": 0.0
+ }
+ },
+ "homographs": {
+ "raw": {
+ "utterances": 6,
+ "errors": 1,
+ "words": 70,
+ "wer": 0.014285714285714285
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 1,
+ "words": 70,
+ "wer": 0.014285714285714285
+ }
+ },
+ "long": {
+ "raw": {
+ "utterances": 15,
+ "errors": 2,
+ "words": 412,
+ "wer": 0.0048543689320388345
+ },
+ "semantic": {
+ "utterances": 15,
+ "errors": 2,
+ "words": 432,
+ "wer": 0.004629629629629629
+ }
+ },
+ "long_range": {
+ "raw": {
+ "utterances": 6,
+ "errors": 0,
+ "words": 121,
+ "wer": 0.0
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 0,
+ "words": 121,
+ "wer": 0.0
+ }
+ },
+ "names_places": {
+ "raw": {
+ "utterances": 6,
+ "errors": 2,
+ "words": 57,
+ "wer": 0.03508771929824561
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 2,
+ "words": 57,
+ "wer": 0.03508771929824561
+ }
+ },
+ "narrative": {
+ "raw": {
+ "utterances": 24,
+ "errors": 3,
+ "words": 342,
+ "wer": 0.008771929824561403
+ },
+ "semantic": {
+ "utterances": 24,
+ "errors": 2,
+ "words": 352,
+ "wer": 0.005681818181818182
+ }
+ },
+ "numbers": {
+ "raw": {
+ "utterances": 6,
+ "errors": 32,
+ "words": 81,
+ "wer": 0.3950617283950617
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 5,
+ "words": 64,
+ "wer": 0.078125
+ }
+ },
+ "phonetic_balance": {
+ "raw": {
+ "utterances": 300,
+ "errors": 39,
+ "words": 2605,
+ "wer": 0.014971209213051824
+ },
+ "semantic": {
+ "utterances": 300,
+ "errors": 30,
+ "words": 2627,
+ "wer": 0.011419870574800152
+ }
+ },
+ "punctuation": {
+ "raw": {
+ "utterances": 6,
+ "errors": 0,
+ "words": 65,
+ "wer": 0.0
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 0,
+ "words": 65,
+ "wer": 0.0
+ }
+ },
+ "question": {
+ "raw": {
+ "utterances": 15,
+ "errors": 0,
+ "words": 163,
+ "wer": 0.0
+ },
+ "semantic": {
+ "utterances": 15,
+ "errors": 0,
+ "words": 165,
+ "wer": 0.0
+ }
+ },
+ "rare_words": {
+ "raw": {
+ "utterances": 6,
+ "errors": 2,
+ "words": 57,
+ "wer": 0.03508771929824561
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 2,
+ "words": 57,
+ "wer": 0.03508771929824561
+ }
+ },
+ "short": {
+ "raw": {
+ "utterances": 24,
+ "errors": 2,
+ "words": 98,
+ "wer": 0.02040816326530612
+ },
+ "semantic": {
+ "utterances": 24,
+ "errors": 2,
+ "words": 105,
+ "wer": 0.01904761904761905
+ }
+ },
+ "technical": {
+ "raw": {
+ "utterances": 21,
+ "errors": 9,
+ "words": 251,
+ "wer": 0.035856573705179286
+ },
+ "semantic": {
+ "utterances": 21,
+ "errors": 5,
+ "words": 256,
+ "wer": 0.01953125
+ }
+ }
+ }
+ },
+ "utmos22": {
+ "mean": 4.21684304523468,
+ "ci": [
+ 4.193502563333512,
+ 4.239324071252346
+ ]
+ }
+ },
+ "kitten-nano-hugo": {
+ "label": "KittenTTS Nano \u00b7 Hugo",
+ "weight_mb": 56.767095,
+ "semantic_wer": {
+ "wer": 0.011461318051575931,
+ "ci": [
+ 0.007995319812792511,
+ 0.01523625843780135
+ ],
+ "categories": {
+ "conversational": {
+ "raw": {
+ "utterances": 25,
+ "errors": 1,
+ "words": 296,
+ "wer": 0.0033783783783783786
+ },
+ "semantic": {
+ "utterances": 25,
+ "errors": 1,
+ "words": 309,
+ "wer": 0.003236245954692557
+ }
+ },
+ "descriptive": {
+ "raw": {
+ "utterances": 16,
+ "errors": 1,
+ "words": 217,
+ "wer": 0.004608294930875576
+ },
+ "semantic": {
+ "utterances": 16,
+ "errors": 0,
+ "words": 226,
+ "wer": 0.0
+ }
+ },
+ "emotional": {
+ "raw": {
+ "utterances": 24,
+ "errors": 0,
+ "words": 314,
+ "wer": 0.0
+ },
+ "semantic": {
+ "utterances": 24,
+ "errors": 0,
+ "words": 329,
+ "wer": 0.0
+ }
+ },
+ "homographs": {
+ "raw": {
+ "utterances": 6,
+ "errors": 1,
+ "words": 70,
+ "wer": 0.014285714285714285
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 1,
+ "words": 70,
+ "wer": 0.014285714285714285
+ }
+ },
+ "long": {
+ "raw": {
+ "utterances": 15,
+ "errors": 3,
+ "words": 412,
+ "wer": 0.007281553398058253
+ },
+ "semantic": {
+ "utterances": 15,
+ "errors": 3,
+ "words": 432,
+ "wer": 0.006944444444444444
+ }
+ },
+ "long_range": {
+ "raw": {
+ "utterances": 6,
+ "errors": 0,
+ "words": 121,
+ "wer": 0.0
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 0,
+ "words": 121,
+ "wer": 0.0
+ }
+ },
+ "names_places": {
+ "raw": {
+ "utterances": 6,
+ "errors": 2,
+ "words": 57,
+ "wer": 0.03508771929824561
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 2,
+ "words": 57,
+ "wer": 0.03508771929824561
+ }
+ },
+ "narrative": {
+ "raw": {
+ "utterances": 24,
+ "errors": 6,
+ "words": 342,
+ "wer": 0.017543859649122806
+ },
+ "semantic": {
+ "utterances": 24,
+ "errors": 5,
+ "words": 352,
+ "wer": 0.014204545454545454
+ }
+ },
+ "numbers": {
+ "raw": {
+ "utterances": 6,
+ "errors": 32,
+ "words": 81,
+ "wer": 0.3950617283950617
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 5,
+ "words": 64,
+ "wer": 0.078125
+ }
+ },
+ "phonetic_balance": {
+ "raw": {
+ "utterances": 300,
+ "errors": 39,
+ "words": 2605,
+ "wer": 0.014971209213051824
+ },
+ "semantic": {
+ "utterances": 300,
+ "errors": 30,
+ "words": 2627,
+ "wer": 0.011419870574800152
+ }
+ },
+ "punctuation": {
+ "raw": {
+ "utterances": 6,
+ "errors": 0,
+ "words": 65,
+ "wer": 0.0
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 0,
+ "words": 65,
+ "wer": 0.0
+ }
+ },
+ "question": {
+ "raw": {
+ "utterances": 15,
+ "errors": 0,
+ "words": 163,
+ "wer": 0.0
+ },
+ "semantic": {
+ "utterances": 15,
+ "errors": 0,
+ "words": 165,
+ "wer": 0.0
+ }
+ },
+ "rare_words": {
+ "raw": {
+ "utterances": 6,
+ "errors": 3,
+ "words": 57,
+ "wer": 0.05263157894736842
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 3,
+ "words": 57,
+ "wer": 0.05263157894736842
+ }
+ },
+ "short": {
+ "raw": {
+ "utterances": 24,
+ "errors": 4,
+ "words": 98,
+ "wer": 0.04081632653061224
+ },
+ "semantic": {
+ "utterances": 24,
+ "errors": 4,
+ "words": 105,
+ "wer": 0.0380952380952381
+ }
+ },
+ "technical": {
+ "raw": {
+ "utterances": 21,
+ "errors": 12,
+ "words": 251,
+ "wer": 0.04780876494023904
+ },
+ "semantic": {
+ "utterances": 21,
+ "errors": 6,
+ "words": 256,
+ "wer": 0.0234375
+ }
+ }
+ }
+ },
+ "utmos22": {
+ "mean": 4.191078673839569,
+ "ci": [
+ 4.168552933418751,
+ 4.212353882300854
+ ]
+ }
+ },
+ "piper-ryan-low": {
+ "label": "Piper \u00b7 Ryan Low",
+ "weight_mb": 63.104526,
+ "semantic_wer": {
+ "wer": 0.023304680038204392,
+ "ci": [
+ 0.018028169014084508,
+ 0.02901655306718598
+ ],
+ "categories": {
+ "conversational": {
+ "raw": {
+ "utterances": 25,
+ "errors": 8,
+ "words": 296,
+ "wer": 0.02702702702702703
+ },
+ "semantic": {
+ "utterances": 25,
+ "errors": 5,
+ "words": 309,
+ "wer": 0.016181229773462782
+ }
+ },
+ "descriptive": {
+ "raw": {
+ "utterances": 16,
+ "errors": 3,
+ "words": 217,
+ "wer": 0.013824884792626729
+ },
+ "semantic": {
+ "utterances": 16,
+ "errors": 1,
+ "words": 226,
+ "wer": 0.004424778761061947
+ }
+ },
+ "emotional": {
+ "raw": {
+ "utterances": 24,
+ "errors": 1,
+ "words": 314,
+ "wer": 0.0031847133757961785
+ },
+ "semantic": {
+ "utterances": 24,
+ "errors": 1,
+ "words": 329,
+ "wer": 0.00303951367781155
+ }
+ },
+ "homographs": {
+ "raw": {
+ "utterances": 6,
+ "errors": 1,
+ "words": 70,
+ "wer": 0.014285714285714285
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 1,
+ "words": 70,
+ "wer": 0.014285714285714285
+ }
+ },
+ "long": {
+ "raw": {
+ "utterances": 15,
+ "errors": 3,
+ "words": 412,
+ "wer": 0.007281553398058253
+ },
+ "semantic": {
+ "utterances": 15,
+ "errors": 3,
+ "words": 432,
+ "wer": 0.006944444444444444
+ }
+ },
+ "long_range": {
+ "raw": {
+ "utterances": 6,
+ "errors": 1,
+ "words": 121,
+ "wer": 0.008264462809917356
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 1,
+ "words": 121,
+ "wer": 0.008264462809917356
+ }
+ },
+ "names_places": {
+ "raw": {
+ "utterances": 6,
+ "errors": 2,
+ "words": 57,
+ "wer": 0.03508771929824561
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 2,
+ "words": 57,
+ "wer": 0.03508771929824561
+ }
+ },
+ "narrative": {
+ "raw": {
+ "utterances": 24,
+ "errors": 1,
+ "words": 342,
+ "wer": 0.0029239766081871343
+ },
+ "semantic": {
+ "utterances": 24,
+ "errors": 0,
+ "words": 352,
+ "wer": 0.0
+ }
+ },
+ "numbers": {
+ "raw": {
+ "utterances": 6,
+ "errors": 32,
+ "words": 81,
+ "wer": 0.3950617283950617
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 6,
+ "words": 64,
+ "wer": 0.09375
+ }
+ },
+ "phonetic_balance": {
+ "raw": {
+ "utterances": 300,
+ "errors": 94,
+ "words": 2605,
+ "wer": 0.03608445297504798
+ },
+ "semantic": {
+ "utterances": 300,
+ "errors": 83,
+ "words": 2627,
+ "wer": 0.031594975256947086
+ }
+ },
+ "punctuation": {
+ "raw": {
+ "utterances": 6,
+ "errors": 1,
+ "words": 65,
+ "wer": 0.015384615384615385
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 1,
+ "words": 65,
+ "wer": 0.015384615384615385
+ }
+ },
+ "question": {
+ "raw": {
+ "utterances": 15,
+ "errors": 0,
+ "words": 163,
+ "wer": 0.0
+ },
+ "semantic": {
+ "utterances": 15,
+ "errors": 0,
+ "words": 165,
+ "wer": 0.0
+ }
+ },
+ "rare_words": {
+ "raw": {
+ "utterances": 6,
+ "errors": 4,
+ "words": 57,
+ "wer": 0.07017543859649122
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 4,
+ "words": 57,
+ "wer": 0.07017543859649122
+ }
+ },
+ "short": {
+ "raw": {
+ "utterances": 24,
+ "errors": 7,
+ "words": 98,
+ "wer": 0.07142857142857142
+ },
+ "semantic": {
+ "utterances": 24,
+ "errors": 8,
+ "words": 105,
+ "wer": 0.0761904761904762
+ }
+ },
+ "technical": {
+ "raw": {
+ "utterances": 21,
+ "errors": 10,
+ "words": 251,
+ "wer": 0.0398406374501992
+ },
+ "semantic": {
+ "utterances": 21,
+ "errors": 6,
+ "words": 256,
+ "wer": 0.0234375
+ }
+ }
+ }
+ },
+ "utmos22": {
+ "mean": 4.289378658294678,
+ "ci": [
+ 4.271461564397812,
+ 4.306235534155369
+ ]
+ }
+ },
+ "piper-danny-low": {
+ "label": "Piper \u00b7 Danny Low",
+ "weight_mb": 63.104526,
+ "semantic_wer": {
+ "wer": 0.018147086914995225,
+ "ci": [
+ 0.013549618320610687,
+ 0.022975077881619937
+ ],
+ "categories": {
+ "conversational": {
+ "raw": {
+ "utterances": 25,
+ "errors": 2,
+ "words": 296,
+ "wer": 0.006756756756756757
+ },
+ "semantic": {
+ "utterances": 25,
+ "errors": 2,
+ "words": 309,
+ "wer": 0.006472491909385114
+ }
+ },
+ "descriptive": {
+ "raw": {
+ "utterances": 16,
+ "errors": 2,
+ "words": 217,
+ "wer": 0.009216589861751152
+ },
+ "semantic": {
+ "utterances": 16,
+ "errors": 1,
+ "words": 226,
+ "wer": 0.004424778761061947
+ }
+ },
+ "emotional": {
+ "raw": {
+ "utterances": 24,
+ "errors": 1,
+ "words": 314,
+ "wer": 0.0031847133757961785
+ },
+ "semantic": {
+ "utterances": 24,
+ "errors": 1,
+ "words": 329,
+ "wer": 0.00303951367781155
+ }
+ },
+ "homographs": {
+ "raw": {
+ "utterances": 6,
+ "errors": 1,
+ "words": 70,
+ "wer": 0.014285714285714285
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 1,
+ "words": 70,
+ "wer": 0.014285714285714285
+ }
+ },
+ "long": {
+ "raw": {
+ "utterances": 15,
+ "errors": 7,
+ "words": 412,
+ "wer": 0.01699029126213592
+ },
+ "semantic": {
+ "utterances": 15,
+ "errors": 7,
+ "words": 432,
+ "wer": 0.016203703703703703
+ }
+ },
+ "long_range": {
+ "raw": {
+ "utterances": 6,
+ "errors": 0,
+ "words": 121,
+ "wer": 0.0
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 0,
+ "words": 121,
+ "wer": 0.0
+ }
+ },
+ "names_places": {
+ "raw": {
+ "utterances": 6,
+ "errors": 3,
+ "words": 57,
+ "wer": 0.05263157894736842
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 3,
+ "words": 57,
+ "wer": 0.05263157894736842
+ }
+ },
+ "narrative": {
+ "raw": {
+ "utterances": 24,
+ "errors": 1,
+ "words": 342,
+ "wer": 0.0029239766081871343
+ },
+ "semantic": {
+ "utterances": 24,
+ "errors": 0,
+ "words": 352,
+ "wer": 0.0
+ }
+ },
+ "numbers": {
+ "raw": {
+ "utterances": 6,
+ "errors": 33,
+ "words": 81,
+ "wer": 0.4074074074074074
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 7,
+ "words": 64,
+ "wer": 0.109375
+ }
+ },
+ "phonetic_balance": {
+ "raw": {
+ "utterances": 300,
+ "errors": 73,
+ "words": 2605,
+ "wer": 0.02802303262955854
+ },
+ "semantic": {
+ "utterances": 300,
+ "errors": 61,
+ "words": 2627,
+ "wer": 0.02322040350209364
+ }
+ },
+ "punctuation": {
+ "raw": {
+ "utterances": 6,
+ "errors": 0,
+ "words": 65,
+ "wer": 0.0
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 0,
+ "words": 65,
+ "wer": 0.0
+ }
+ },
+ "question": {
+ "raw": {
+ "utterances": 15,
+ "errors": 2,
+ "words": 163,
+ "wer": 0.012269938650306749
+ },
+ "semantic": {
+ "utterances": 15,
+ "errors": 2,
+ "words": 165,
+ "wer": 0.012121212121212121
+ }
+ },
+ "rare_words": {
+ "raw": {
+ "utterances": 6,
+ "errors": 2,
+ "words": 57,
+ "wer": 0.03508771929824561
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 2,
+ "words": 57,
+ "wer": 0.03508771929824561
+ }
+ },
+ "short": {
+ "raw": {
+ "utterances": 24,
+ "errors": 3,
+ "words": 98,
+ "wer": 0.030612244897959183
+ },
+ "semantic": {
+ "utterances": 24,
+ "errors": 3,
+ "words": 105,
+ "wer": 0.02857142857142857
+ }
+ },
+ "technical": {
+ "raw": {
+ "utterances": 21,
+ "errors": 9,
+ "words": 251,
+ "wer": 0.035856573705179286
+ },
+ "semantic": {
+ "utterances": 21,
+ "errors": 5,
+ "words": 256,
+ "wer": 0.01953125
+ }
+ }
+ }
+ },
+ "utmos22": {
+ "mean": 4.194989757061005,
+ "ci": [
+ 4.179823832964897,
+ 4.20924334000349
+ ]
+ }
+ },
+ "supertonic3-james-3step": {
+ "label": "Supertonic 3 \u00b7 James \u00b7 3-step",
+ "weight_mb": 398.075273,
+ "semantic_wer": {
+ "wer": 6.928175740210124,
+ "ci": [
+ 5.861435406698565,
+ 8.043254987161761
+ ],
+ "categories": {
+ "conversational": {
+ "raw": {
+ "utterances": 25,
+ "errors": 1295,
+ "words": 296,
+ "wer": 4.375
+ },
+ "semantic": {
+ "utterances": 25,
+ "errors": 1449,
+ "words": 309,
+ "wer": 4.689320388349515
+ }
+ },
+ "descriptive": {
+ "raw": {
+ "utterances": 16,
+ "errors": 902,
+ "words": 217,
+ "wer": 4.15668202764977
+ },
+ "semantic": {
+ "utterances": 16,
+ "errors": 998,
+ "words": 226,
+ "wer": 4.415929203539823
+ }
+ },
+ "emotional": {
+ "raw": {
+ "utterances": 24,
+ "errors": 2189,
+ "words": 314,
+ "wer": 6.971337579617835
+ },
+ "semantic": {
+ "utterances": 24,
+ "errors": 2306,
+ "words": 329,
+ "wer": 7.009118541033435
+ }
+ },
+ "homographs": {
+ "raw": {
+ "utterances": 6,
+ "errors": 70,
+ "words": 70,
+ "wer": 1.0
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 73,
+ "words": 70,
+ "wer": 1.042857142857143
+ }
+ },
+ "long": {
+ "raw": {
+ "utterances": 15,
+ "errors": 1113,
+ "words": 412,
+ "wer": 2.7014563106796117
+ },
+ "semantic": {
+ "utterances": 15,
+ "errors": 1190,
+ "words": 432,
+ "wer": 2.7546296296296298
+ }
+ },
+ "long_range": {
+ "raw": {
+ "utterances": 6,
+ "errors": 153,
+ "words": 121,
+ "wer": 1.2644628099173554
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 159,
+ "words": 121,
+ "wer": 1.3140495867768596
+ }
+ },
+ "names_places": {
+ "raw": {
+ "utterances": 6,
+ "errors": 666,
+ "words": 57,
+ "wer": 11.68421052631579
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 828,
+ "words": 57,
+ "wer": 14.526315789473685
+ }
+ },
+ "narrative": {
+ "raw": {
+ "utterances": 24,
+ "errors": 2415,
+ "words": 342,
+ "wer": 7.06140350877193
+ },
+ "semantic": {
+ "utterances": 24,
+ "errors": 2562,
+ "words": 352,
+ "wer": 7.278409090909091
+ }
+ },
+ "numbers": {
+ "raw": {
+ "utterances": 6,
+ "errors": 94,
+ "words": 81,
+ "wer": 1.1604938271604939
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 91,
+ "words": 64,
+ "wer": 1.421875
+ }
+ },
+ "phonetic_balance": {
+ "raw": {
+ "utterances": 300,
+ "errors": 19668,
+ "words": 2605,
+ "wer": 7.550095969289827
+ },
+ "semantic": {
+ "utterances": 300,
+ "errors": 21028,
+ "words": 2627,
+ "wer": 8.00456794822992
+ }
+ },
+ "punctuation": {
+ "raw": {
+ "utterances": 6,
+ "errors": 803,
+ "words": 65,
+ "wer": 12.353846153846154
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 807,
+ "words": 65,
+ "wer": 12.415384615384616
+ }
+ },
+ "question": {
+ "raw": {
+ "utterances": 15,
+ "errors": 1412,
+ "words": 163,
+ "wer": 8.662576687116564
+ },
+ "semantic": {
+ "utterances": 15,
+ "errors": 1529,
+ "words": 165,
+ "wer": 9.266666666666667
+ }
+ },
+ "rare_words": {
+ "raw": {
+ "utterances": 6,
+ "errors": 819,
+ "words": 57,
+ "wer": 14.368421052631579
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 820,
+ "words": 57,
+ "wer": 14.385964912280702
+ }
+ },
+ "short": {
+ "raw": {
+ "utterances": 24,
+ "errors": 436,
+ "words": 98,
+ "wer": 4.448979591836735
+ },
+ "semantic": {
+ "utterances": 24,
+ "errors": 447,
+ "words": 105,
+ "wer": 4.257142857142857
+ }
+ },
+ "technical": {
+ "raw": {
+ "utterances": 21,
+ "errors": 1976,
+ "words": 251,
+ "wer": 7.872509960159363
+ },
+ "semantic": {
+ "utterances": 21,
+ "errors": 1982,
+ "words": 256,
+ "wer": 7.7421875
+ }
+ }
+ }
+ },
+ "utmos22": {
+ "mean": 2.4712263617515564,
+ "ci": [
+ 2.412739716243744,
+ 2.5290928330957887
+ ]
+ }
+ },
+ "supertonic3-james-8step": {
+ "label": "Supertonic 3 \u00b7 James \u00b7 8-step",
+ "weight_mb": 398.075273,
+ "semantic_wer": {
+ "wer": 6.225214899713467,
+ "ci": [
+ 5.2127497621313035,
+ 7.2831739961759085
+ ],
+ "categories": {
+ "conversational": {
+ "raw": {
+ "utterances": 25,
+ "errors": 1746,
+ "words": 296,
+ "wer": 5.898648648648648
+ },
+ "semantic": {
+ "utterances": 25,
+ "errors": 1812,
+ "words": 309,
+ "wer": 5.864077669902913
+ }
+ },
+ "descriptive": {
+ "raw": {
+ "utterances": 16,
+ "errors": 1230,
+ "words": 217,
+ "wer": 5.668202764976958
+ },
+ "semantic": {
+ "utterances": 16,
+ "errors": 1234,
+ "words": 226,
+ "wer": 5.460176991150442
+ }
+ },
+ "emotional": {
+ "raw": {
+ "utterances": 24,
+ "errors": 1592,
+ "words": 314,
+ "wer": 5.070063694267516
+ },
+ "semantic": {
+ "utterances": 24,
+ "errors": 1670,
+ "words": 329,
+ "wer": 5.075987841945289
+ }
+ },
+ "homographs": {
+ "raw": {
+ "utterances": 6,
+ "errors": 326,
+ "words": 70,
+ "wer": 4.6571428571428575
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 413,
+ "words": 70,
+ "wer": 5.9
+ }
+ },
+ "long": {
+ "raw": {
+ "utterances": 15,
+ "errors": 1459,
+ "words": 412,
+ "wer": 3.54126213592233
+ },
+ "semantic": {
+ "utterances": 15,
+ "errors": 1468,
+ "words": 432,
+ "wer": 3.3981481481481484
+ }
+ },
+ "long_range": {
+ "raw": {
+ "utterances": 6,
+ "errors": 113,
+ "words": 121,
+ "wer": 0.9338842975206612
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 113,
+ "words": 121,
+ "wer": 0.9338842975206612
+ }
+ },
+ "names_places": {
+ "raw": {
+ "utterances": 6,
+ "errors": 381,
+ "words": 57,
+ "wer": 6.684210526315789
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 476,
+ "words": 57,
+ "wer": 8.350877192982455
+ }
+ },
+ "narrative": {
+ "raw": {
+ "utterances": 24,
+ "errors": 1338,
+ "words": 342,
+ "wer": 3.912280701754386
+ },
+ "semantic": {
+ "utterances": 24,
+ "errors": 1344,
+ "words": 352,
+ "wer": 3.8181818181818183
+ }
+ },
+ "numbers": {
+ "raw": {
+ "utterances": 6,
+ "errors": 90,
+ "words": 81,
+ "wer": 1.1111111111111112
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 87,
+ "words": 64,
+ "wer": 1.359375
+ }
+ },
+ "phonetic_balance": {
+ "raw": {
+ "utterances": 300,
+ "errors": 18274,
+ "words": 2605,
+ "wer": 7.014971209213051
+ },
+ "semantic": {
+ "utterances": 300,
+ "errors": 19044,
+ "words": 2627,
+ "wer": 7.2493338408831365
+ }
+ },
+ "punctuation": {
+ "raw": {
+ "utterances": 6,
+ "errors": 361,
+ "words": 65,
+ "wer": 5.553846153846154
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 438,
+ "words": 65,
+ "wer": 6.7384615384615385
+ }
+ },
+ "question": {
+ "raw": {
+ "utterances": 15,
+ "errors": 1153,
+ "words": 163,
+ "wer": 7.07361963190184
+ },
+ "semantic": {
+ "utterances": 15,
+ "errors": 1267,
+ "words": 165,
+ "wer": 7.678787878787879
+ }
+ },
+ "rare_words": {
+ "raw": {
+ "utterances": 6,
+ "errors": 438,
+ "words": 57,
+ "wer": 7.684210526315789
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 444,
+ "words": 57,
+ "wer": 7.7894736842105265
+ }
+ },
+ "short": {
+ "raw": {
+ "utterances": 24,
+ "errors": 1005,
+ "words": 98,
+ "wer": 10.255102040816327
+ },
+ "semantic": {
+ "utterances": 24,
+ "errors": 1057,
+ "words": 105,
+ "wer": 10.066666666666666
+ }
+ },
+ "technical": {
+ "raw": {
+ "utterances": 21,
+ "errors": 1585,
+ "words": 251,
+ "wer": 6.314741035856573
+ },
+ "semantic": {
+ "utterances": 21,
+ "errors": 1722,
+ "words": 256,
+ "wer": 6.7265625
+ }
+ }
+ }
+ },
+ "utmos22": {
+ "mean": 4.29470673418045,
+ "ci": [
+ 4.281224336981774,
+ 4.307890736854076
+ ]
+ }
+ }
+ },
+ "asr_robustness": {
+ "inflect-micro-v2": {
+ "whisper-large-v3": {
+ "wer": 0.019293218720152817,
+ "ci": [
+ 0.014622733476311172,
+ 0.024179280518969924
+ ]
+ },
+ "wav2vec2-large-lv60k-960h": {
+ "wer": 0.048137535816618914,
+ "ci": [
+ 0.03997704667176741,
+ 0.05712124077926991
+ ]
+ }
+ },
+ "inflect-nano-v2": {
+ "whisper-large-v3": {
+ "wer": 0.024832855778414518,
+ "ci": [
+ 0.019525267993874426,
+ 0.030727866333781446
+ ]
+ },
+ "wav2vec2-large-lv60k-960h": {
+ "wer": 0.05673352435530086,
+ "ci": [
+ 0.04819277108433735,
+ 0.06602129719264278
+ ]
+ }
+ },
+ "kitten-nano-bruno": {
+ "whisper-large-v3": {
+ "wer": 0.009742120343839542,
+ "ci": [
+ 0.006622516556291391,
+ 0.013268998793727383
+ ]
+ },
+ "wav2vec2-large-lv60k-960h": {
+ "wer": 0.03553008595988539,
+ "ci": [
+ 0.028724626579854463,
+ 0.04309006211180124
+ ]
+ }
+ },
+ "kitten-nano-hugo": {
+ "whisper-large-v3": {
+ "wer": 0.011461318051575931,
+ "ci": [
+ 0.007995319812792511,
+ 0.01523625843780135
+ ]
+ },
+ "wav2vec2-large-lv60k-960h": {
+ "wer": 0.03629417382999045,
+ "ci": [
+ 0.029451395572666024,
+ 0.043798449612403104
+ ]
+ }
+ },
+ "piper-ryan-low": {
+ "whisper-large-v3": {
+ "wer": 0.023304680038204392,
+ "ci": [
+ 0.018028169014084508,
+ 0.02901655306718598
+ ]
+ },
+ "wav2vec2-large-lv60k-960h": {
+ "wer": 0.0670487106017192,
+ "ci": [
+ 0.05838601710673113,
+ 0.07638619620007756
+ ]
+ }
+ },
+ "piper-danny-low": {
+ "whisper-large-v3": {
+ "wer": 0.018147086914995225,
+ "ci": [
+ 0.013549618320610687,
+ 0.022975077881619937
+ ]
+ },
+ "wav2vec2-large-lv60k-960h": {
+ "wer": 0.053104106972301814,
+ "ci": [
+ 0.04440632742519535,
+ 0.06227176628867961
+ ]
+ }
+ },
+ "supertonic3-james-3step": {
+ "whisper-large-v3": {
+ "wer": 6.928175740210124,
+ "ci": [
+ 5.861435406698565,
+ 8.043254987161761
+ ]
+ },
+ "wav2vec2-large-lv60k-960h": {
+ "wer": 0.1346704871060172,
+ "ci": [
+ 0.11719324026451139,
+ 0.15343099460292983
+ ]
+ }
+ },
+ "supertonic3-james-8step": {
+ "whisper-large-v3": {
+ "wer": 6.225214899713467,
+ "ci": [
+ 5.2127497621313035,
+ 7.2831739961759085
+ ]
+ },
+ "wav2vec2-large-lv60k-960h": {
+ "wer": 0.037631327602674305,
+ "ci": [
+ 0.030763314747164008,
+ 0.044854881266490766
+ ]
+ }
+ }
+ },
+ "family_averages": {
+ "kitten": {
+ "semantic_wer": 0.010601719197707736,
+ "utmos22": 4.2039608595371245
+ },
+ "piper": {
+ "semantic_wer": 0.02072588347659981,
+ "utmos22": 4.242184207677841
+ }
+ },
+ "notes": [
+ "Family averages are equal-weight macro averages over the two named voices.",
+ "Supertonic 3-step and 8-step remain separate.",
+ "UTMOS22 and WER are not combined into one score."
+ ]
+}
diff --git a/evaluation/final/raw/asr_combined/report.json b/evaluation/final/raw/asr_combined/report.json
new file mode 100644
index 0000000000000000000000000000000000000000..83c39ec2be3d67aa6cf7e3da07d98c33bf059828
--- /dev/null
+++ b/evaluation/final/raw/asr_combined/report.json
@@ -0,0 +1,198811 @@
+{
+ "format": "inflect_tts_multi_asr_merged_v1",
+ "sources": [
+ "benchmark/whisper_large_v3/report.json",
+ "benchmark/wav2vec2_large_lv60k_960h/report.json"
+ ],
+ "asr_models": [
+ "whisper-large-v3",
+ "wav2vec2-large-lv60k-960h"
+ ],
+ "normalization": {
+ "raw": "lowercase_punctuation_stripped",
+ "semantic": "transformers_whisper_EnglishTextNormalizer_then_lexical"
+ },
+ "systems": {
+ "inflect-micro-v2": {
+ "asr": {
+ "whisper-large-v3": {
+ "raw": {
+ "utterances": 500,
+ "errors": 146,
+ "words": 5149,
+ "wer": 0.028355020392309187
+ },
+ "raw_95ci": [
+ 0.02125193199381762,
+ 0.036153996482313856
+ ],
+ "semantic": {
+ "utterances": 500,
+ "errors": 101,
+ "words": 5235,
+ "wer": 0.019293218720152817
+ },
+ "semantic_95ci": [
+ 0.014622733476311172,
+ 0.024179280518969924
+ ],
+ "categories": {
+ "conversational": {
+ "raw": {
+ "utterances": 25,
+ "errors": 0,
+ "words": 296,
+ "wer": 0.0
+ },
+ "semantic": {
+ "utterances": 25,
+ "errors": 0,
+ "words": 309,
+ "wer": 0.0
+ }
+ },
+ "descriptive": {
+ "raw": {
+ "utterances": 16,
+ "errors": 2,
+ "words": 217,
+ "wer": 0.009216589861751152
+ },
+ "semantic": {
+ "utterances": 16,
+ "errors": 1,
+ "words": 226,
+ "wer": 0.004424778761061947
+ }
+ },
+ "emotional": {
+ "raw": {
+ "utterances": 24,
+ "errors": 0,
+ "words": 314,
+ "wer": 0.0
+ },
+ "semantic": {
+ "utterances": 24,
+ "errors": 0,
+ "words": 329,
+ "wer": 0.0
+ }
+ },
+ "homographs": {
+ "raw": {
+ "utterances": 6,
+ "errors": 2,
+ "words": 70,
+ "wer": 0.02857142857142857
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 2,
+ "words": 70,
+ "wer": 0.02857142857142857
+ }
+ },
+ "long": {
+ "raw": {
+ "utterances": 15,
+ "errors": 3,
+ "words": 412,
+ "wer": 0.007281553398058253
+ },
+ "semantic": {
+ "utterances": 15,
+ "errors": 3,
+ "words": 432,
+ "wer": 0.006944444444444444
+ }
+ },
+ "long_range": {
+ "raw": {
+ "utterances": 6,
+ "errors": 0,
+ "words": 121,
+ "wer": 0.0
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 0,
+ "words": 121,
+ "wer": 0.0
+ }
+ },
+ "names_places": {
+ "raw": {
+ "utterances": 6,
+ "errors": 4,
+ "words": 57,
+ "wer": 0.07017543859649122
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 4,
+ "words": 57,
+ "wer": 0.07017543859649122
+ }
+ },
+ "narrative": {
+ "raw": {
+ "utterances": 24,
+ "errors": 2,
+ "words": 342,
+ "wer": 0.005847953216374269
+ },
+ "semantic": {
+ "utterances": 24,
+ "errors": 1,
+ "words": 352,
+ "wer": 0.002840909090909091
+ }
+ },
+ "numbers": {
+ "raw": {
+ "utterances": 6,
+ "errors": 33,
+ "words": 81,
+ "wer": 0.4074074074074074
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 8,
+ "words": 64,
+ "wer": 0.125
+ }
+ },
+ "phonetic_balance": {
+ "raw": {
+ "utterances": 300,
+ "errors": 82,
+ "words": 2605,
+ "wer": 0.03147792706333973
+ },
+ "semantic": {
+ "utterances": 300,
+ "errors": 68,
+ "words": 2627,
+ "wer": 0.02588503996954701
+ }
+ },
+ "punctuation": {
+ "raw": {
+ "utterances": 6,
+ "errors": 0,
+ "words": 65,
+ "wer": 0.0
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 0,
+ "words": 65,
+ "wer": 0.0
+ }
+ },
+ "question": {
+ "raw": {
+ "utterances": 15,
+ "errors": 0,
+ "words": 163,
+ "wer": 0.0
+ },
+ "semantic": {
+ "utterances": 15,
+ "errors": 0,
+ "words": 165,
+ "wer": 0.0
+ }
+ },
+ "rare_words": {
+ "raw": {
+ "utterances": 6,
+ "errors": 3,
+ "words": 57,
+ "wer": 0.05263157894736842
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 3,
+ "words": 57,
+ "wer": 0.05263157894736842
+ }
+ },
+ "short": {
+ "raw": {
+ "utterances": 24,
+ "errors": 3,
+ "words": 98,
+ "wer": 0.030612244897959183
+ },
+ "semantic": {
+ "utterances": 24,
+ "errors": 3,
+ "words": 105,
+ "wer": 0.02857142857142857
+ }
+ },
+ "technical": {
+ "raw": {
+ "utterances": 21,
+ "errors": 12,
+ "words": 251,
+ "wer": 0.04780876494023904
+ },
+ "semantic": {
+ "utterances": 21,
+ "errors": 8,
+ "words": 256,
+ "wer": 0.03125
+ }
+ }
+ }
+ },
+ "wav2vec2-large-lv60k-960h": {
+ "raw": {
+ "utterances": 500,
+ "errors": 260,
+ "words": 5149,
+ "wer": 0.05049524179452321
+ },
+ "raw_95ci": [
+ 0.04179873217115689,
+ 0.059920859242509894
+ ],
+ "semantic": {
+ "utterances": 500,
+ "errors": 252,
+ "words": 5235,
+ "wer": 0.048137535816618914
+ },
+ "semantic_95ci": [
+ 0.03997704667176741,
+ 0.05712124077926991
+ ],
+ "categories": {
+ "conversational": {
+ "raw": {
+ "utterances": 25,
+ "errors": 4,
+ "words": 296,
+ "wer": 0.013513513513513514
+ },
+ "semantic": {
+ "utterances": 25,
+ "errors": 4,
+ "words": 309,
+ "wer": 0.012944983818770227
+ }
+ },
+ "descriptive": {
+ "raw": {
+ "utterances": 16,
+ "errors": 2,
+ "words": 217,
+ "wer": 0.009216589861751152
+ },
+ "semantic": {
+ "utterances": 16,
+ "errors": 2,
+ "words": 226,
+ "wer": 0.008849557522123894
+ }
+ },
+ "emotional": {
+ "raw": {
+ "utterances": 24,
+ "errors": 3,
+ "words": 314,
+ "wer": 0.009554140127388535
+ },
+ "semantic": {
+ "utterances": 24,
+ "errors": 2,
+ "words": 329,
+ "wer": 0.0060790273556231
+ }
+ },
+ "homographs": {
+ "raw": {
+ "utterances": 6,
+ "errors": 5,
+ "words": 70,
+ "wer": 0.07142857142857142
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 5,
+ "words": 70,
+ "wer": 0.07142857142857142
+ }
+ },
+ "long": {
+ "raw": {
+ "utterances": 15,
+ "errors": 9,
+ "words": 412,
+ "wer": 0.021844660194174758
+ },
+ "semantic": {
+ "utterances": 15,
+ "errors": 7,
+ "words": 432,
+ "wer": 0.016203703703703703
+ }
+ },
+ "long_range": {
+ "raw": {
+ "utterances": 6,
+ "errors": 5,
+ "words": 121,
+ "wer": 0.04132231404958678
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 6,
+ "words": 121,
+ "wer": 0.049586776859504134
+ }
+ },
+ "names_places": {
+ "raw": {
+ "utterances": 6,
+ "errors": 19,
+ "words": 57,
+ "wer": 0.3333333333333333
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 19,
+ "words": 57,
+ "wer": 0.3333333333333333
+ }
+ },
+ "narrative": {
+ "raw": {
+ "utterances": 24,
+ "errors": 4,
+ "words": 342,
+ "wer": 0.011695906432748537
+ },
+ "semantic": {
+ "utterances": 24,
+ "errors": 4,
+ "words": 352,
+ "wer": 0.011363636363636364
+ }
+ },
+ "numbers": {
+ "raw": {
+ "utterances": 6,
+ "errors": 10,
+ "words": 81,
+ "wer": 0.12345679012345678
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 6,
+ "words": 64,
+ "wer": 0.09375
+ }
+ },
+ "phonetic_balance": {
+ "raw": {
+ "utterances": 300,
+ "errors": 145,
+ "words": 2605,
+ "wer": 0.05566218809980806
+ },
+ "semantic": {
+ "utterances": 300,
+ "errors": 143,
+ "words": 2627,
+ "wer": 0.054434716406547394
+ }
+ },
+ "punctuation": {
+ "raw": {
+ "utterances": 6,
+ "errors": 2,
+ "words": 65,
+ "wer": 0.03076923076923077
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 2,
+ "words": 65,
+ "wer": 0.03076923076923077
+ }
+ },
+ "question": {
+ "raw": {
+ "utterances": 15,
+ "errors": 2,
+ "words": 163,
+ "wer": 0.012269938650306749
+ },
+ "semantic": {
+ "utterances": 15,
+ "errors": 2,
+ "words": 165,
+ "wer": 0.012121212121212121
+ }
+ },
+ "rare_words": {
+ "raw": {
+ "utterances": 6,
+ "errors": 11,
+ "words": 57,
+ "wer": 0.19298245614035087
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 11,
+ "words": 57,
+ "wer": 0.19298245614035087
+ }
+ },
+ "short": {
+ "raw": {
+ "utterances": 24,
+ "errors": 14,
+ "words": 98,
+ "wer": 0.14285714285714285
+ },
+ "semantic": {
+ "utterances": 24,
+ "errors": 14,
+ "words": 105,
+ "wer": 0.13333333333333333
+ }
+ },
+ "technical": {
+ "raw": {
+ "utterances": 21,
+ "errors": 25,
+ "words": 251,
+ "wer": 0.099601593625498
+ },
+ "semantic": {
+ "utterances": 21,
+ "errors": 25,
+ "words": 256,
+ "wer": 0.09765625
+ }
+ }
+ }
+ }
+ },
+ "cross_asr_mean_semantic_wer": 0.03371537726838587,
+ "cross_asr_worst_semantic_wer": 0.048137535816618914
+ },
+ "inflect-nano-v2": {
+ "asr": {
+ "whisper-large-v3": {
+ "raw": {
+ "utterances": 500,
+ "errors": 178,
+ "words": 5149,
+ "wer": 0.03456981938240435
+ },
+ "raw_95ci": [
+ 0.026621030614185207,
+ 0.043358297201419
+ ],
+ "semantic": {
+ "utterances": 500,
+ "errors": 130,
+ "words": 5235,
+ "wer": 0.024832855778414518
+ },
+ "semantic_95ci": [
+ 0.019525267993874426,
+ 0.030727866333781446
+ ],
+ "categories": {
+ "conversational": {
+ "raw": {
+ "utterances": 25,
+ "errors": 5,
+ "words": 296,
+ "wer": 0.016891891891891893
+ },
+ "semantic": {
+ "utterances": 25,
+ "errors": 3,
+ "words": 309,
+ "wer": 0.009708737864077669
+ }
+ },
+ "descriptive": {
+ "raw": {
+ "utterances": 16,
+ "errors": 2,
+ "words": 217,
+ "wer": 0.009216589861751152
+ },
+ "semantic": {
+ "utterances": 16,
+ "errors": 1,
+ "words": 226,
+ "wer": 0.004424778761061947
+ }
+ },
+ "emotional": {
+ "raw": {
+ "utterances": 24,
+ "errors": 1,
+ "words": 314,
+ "wer": 0.0031847133757961785
+ },
+ "semantic": {
+ "utterances": 24,
+ "errors": 1,
+ "words": 329,
+ "wer": 0.00303951367781155
+ }
+ },
+ "homographs": {
+ "raw": {
+ "utterances": 6,
+ "errors": 3,
+ "words": 70,
+ "wer": 0.04285714285714286
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 3,
+ "words": 70,
+ "wer": 0.04285714285714286
+ }
+ },
+ "long": {
+ "raw": {
+ "utterances": 15,
+ "errors": 3,
+ "words": 412,
+ "wer": 0.007281553398058253
+ },
+ "semantic": {
+ "utterances": 15,
+ "errors": 3,
+ "words": 432,
+ "wer": 0.006944444444444444
+ }
+ },
+ "long_range": {
+ "raw": {
+ "utterances": 6,
+ "errors": 1,
+ "words": 121,
+ "wer": 0.008264462809917356
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 1,
+ "words": 121,
+ "wer": 0.008264462809917356
+ }
+ },
+ "names_places": {
+ "raw": {
+ "utterances": 6,
+ "errors": 3,
+ "words": 57,
+ "wer": 0.05263157894736842
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 3,
+ "words": 57,
+ "wer": 0.05263157894736842
+ }
+ },
+ "narrative": {
+ "raw": {
+ "utterances": 24,
+ "errors": 5,
+ "words": 342,
+ "wer": 0.014619883040935672
+ },
+ "semantic": {
+ "utterances": 24,
+ "errors": 4,
+ "words": 352,
+ "wer": 0.011363636363636364
+ }
+ },
+ "numbers": {
+ "raw": {
+ "utterances": 6,
+ "errors": 33,
+ "words": 81,
+ "wer": 0.4074074074074074
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 6,
+ "words": 64,
+ "wer": 0.09375
+ }
+ },
+ "phonetic_balance": {
+ "raw": {
+ "utterances": 300,
+ "errors": 96,
+ "words": 2605,
+ "wer": 0.03685220729366603
+ },
+ "semantic": {
+ "utterances": 300,
+ "errors": 83,
+ "words": 2627,
+ "wer": 0.031594975256947086
+ }
+ },
+ "punctuation": {
+ "raw": {
+ "utterances": 6,
+ "errors": 0,
+ "words": 65,
+ "wer": 0.0
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 0,
+ "words": 65,
+ "wer": 0.0
+ }
+ },
+ "question": {
+ "raw": {
+ "utterances": 15,
+ "errors": 1,
+ "words": 163,
+ "wer": 0.006134969325153374
+ },
+ "semantic": {
+ "utterances": 15,
+ "errors": 1,
+ "words": 165,
+ "wer": 0.006060606060606061
+ }
+ },
+ "rare_words": {
+ "raw": {
+ "utterances": 6,
+ "errors": 5,
+ "words": 57,
+ "wer": 0.08771929824561403
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 5,
+ "words": 57,
+ "wer": 0.08771929824561403
+ }
+ },
+ "short": {
+ "raw": {
+ "utterances": 24,
+ "errors": 8,
+ "words": 98,
+ "wer": 0.08163265306122448
+ },
+ "semantic": {
+ "utterances": 24,
+ "errors": 8,
+ "words": 105,
+ "wer": 0.0761904761904762
+ }
+ },
+ "technical": {
+ "raw": {
+ "utterances": 21,
+ "errors": 12,
+ "words": 251,
+ "wer": 0.04780876494023904
+ },
+ "semantic": {
+ "utterances": 21,
+ "errors": 8,
+ "words": 256,
+ "wer": 0.03125
+ }
+ }
+ }
+ },
+ "wav2vec2-large-lv60k-960h": {
+ "raw": {
+ "utterances": 500,
+ "errors": 302,
+ "words": 5149,
+ "wer": 0.05865216546902311
+ },
+ "raw_95ci": [
+ 0.049577464788732394,
+ 0.06843965036265576
+ ],
+ "semantic": {
+ "utterances": 500,
+ "errors": 297,
+ "words": 5235,
+ "wer": 0.05673352435530086
+ },
+ "semantic_95ci": [
+ 0.04819277108433735,
+ 0.06602129719264278
+ ],
+ "categories": {
+ "conversational": {
+ "raw": {
+ "utterances": 25,
+ "errors": 12,
+ "words": 296,
+ "wer": 0.04054054054054054
+ },
+ "semantic": {
+ "utterances": 25,
+ "errors": 12,
+ "words": 309,
+ "wer": 0.038834951456310676
+ }
+ },
+ "descriptive": {
+ "raw": {
+ "utterances": 16,
+ "errors": 5,
+ "words": 217,
+ "wer": 0.02304147465437788
+ },
+ "semantic": {
+ "utterances": 16,
+ "errors": 5,
+ "words": 226,
+ "wer": 0.022123893805309734
+ }
+ },
+ "emotional": {
+ "raw": {
+ "utterances": 24,
+ "errors": 7,
+ "words": 314,
+ "wer": 0.022292993630573247
+ },
+ "semantic": {
+ "utterances": 24,
+ "errors": 7,
+ "words": 329,
+ "wer": 0.02127659574468085
+ }
+ },
+ "homographs": {
+ "raw": {
+ "utterances": 6,
+ "errors": 4,
+ "words": 70,
+ "wer": 0.05714285714285714
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 4,
+ "words": 70,
+ "wer": 0.05714285714285714
+ }
+ },
+ "long": {
+ "raw": {
+ "utterances": 15,
+ "errors": 8,
+ "words": 412,
+ "wer": 0.019417475728155338
+ },
+ "semantic": {
+ "utterances": 15,
+ "errors": 8,
+ "words": 432,
+ "wer": 0.018518518518518517
+ }
+ },
+ "long_range": {
+ "raw": {
+ "utterances": 6,
+ "errors": 6,
+ "words": 121,
+ "wer": 0.049586776859504134
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 7,
+ "words": 121,
+ "wer": 0.05785123966942149
+ }
+ },
+ "names_places": {
+ "raw": {
+ "utterances": 6,
+ "errors": 18,
+ "words": 57,
+ "wer": 0.3157894736842105
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 18,
+ "words": 57,
+ "wer": 0.3157894736842105
+ }
+ },
+ "narrative": {
+ "raw": {
+ "utterances": 24,
+ "errors": 7,
+ "words": 342,
+ "wer": 0.02046783625730994
+ },
+ "semantic": {
+ "utterances": 24,
+ "errors": 7,
+ "words": 352,
+ "wer": 0.019886363636363636
+ }
+ },
+ "numbers": {
+ "raw": {
+ "utterances": 6,
+ "errors": 10,
+ "words": 81,
+ "wer": 0.12345679012345678
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 6,
+ "words": 64,
+ "wer": 0.09375
+ }
+ },
+ "phonetic_balance": {
+ "raw": {
+ "utterances": 300,
+ "errors": 163,
+ "words": 2605,
+ "wer": 0.06257197696737044
+ },
+ "semantic": {
+ "utterances": 300,
+ "errors": 161,
+ "words": 2627,
+ "wer": 0.06128663875142749
+ }
+ },
+ "punctuation": {
+ "raw": {
+ "utterances": 6,
+ "errors": 3,
+ "words": 65,
+ "wer": 0.046153846153846156
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 3,
+ "words": 65,
+ "wer": 0.046153846153846156
+ }
+ },
+ "question": {
+ "raw": {
+ "utterances": 15,
+ "errors": 5,
+ "words": 163,
+ "wer": 0.03067484662576687
+ },
+ "semantic": {
+ "utterances": 15,
+ "errors": 5,
+ "words": 165,
+ "wer": 0.030303030303030304
+ }
+ },
+ "rare_words": {
+ "raw": {
+ "utterances": 6,
+ "errors": 9,
+ "words": 57,
+ "wer": 0.15789473684210525
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 9,
+ "words": 57,
+ "wer": 0.15789473684210525
+ }
+ },
+ "short": {
+ "raw": {
+ "utterances": 24,
+ "errors": 16,
+ "words": 98,
+ "wer": 0.16326530612244897
+ },
+ "semantic": {
+ "utterances": 24,
+ "errors": 16,
+ "words": 105,
+ "wer": 0.1523809523809524
+ }
+ },
+ "technical": {
+ "raw": {
+ "utterances": 21,
+ "errors": 29,
+ "words": 251,
+ "wer": 0.11553784860557768
+ },
+ "semantic": {
+ "utterances": 21,
+ "errors": 29,
+ "words": 256,
+ "wer": 0.11328125
+ }
+ }
+ }
+ }
+ },
+ "cross_asr_mean_semantic_wer": 0.04078319006685769,
+ "cross_asr_worst_semantic_wer": 0.05673352435530086
+ },
+ "kitten-nano-bruno": {
+ "asr": {
+ "whisper-large-v3": {
+ "raw": {
+ "utterances": 500,
+ "errors": 93,
+ "words": 5149,
+ "wer": 0.018061759564964072
+ },
+ "raw_95ci": [
+ 0.01197365795250449,
+ 0.024896265560165973
+ ],
+ "semantic": {
+ "utterances": 500,
+ "errors": 51,
+ "words": 5235,
+ "wer": 0.009742120343839542
+ },
+ "semantic_95ci": [
+ 0.006622516556291391,
+ 0.013268998793727383
+ ],
+ "categories": {
+ "conversational": {
+ "raw": {
+ "utterances": 25,
+ "errors": 0,
+ "words": 296,
+ "wer": 0.0
+ },
+ "semantic": {
+ "utterances": 25,
+ "errors": 0,
+ "words": 309,
+ "wer": 0.0
+ }
+ },
+ "descriptive": {
+ "raw": {
+ "utterances": 16,
+ "errors": 1,
+ "words": 217,
+ "wer": 0.004608294930875576
+ },
+ "semantic": {
+ "utterances": 16,
+ "errors": 0,
+ "words": 226,
+ "wer": 0.0
+ }
+ },
+ "emotional": {
+ "raw": {
+ "utterances": 24,
+ "errors": 0,
+ "words": 314,
+ "wer": 0.0
+ },
+ "semantic": {
+ "utterances": 24,
+ "errors": 0,
+ "words": 329,
+ "wer": 0.0
+ }
+ },
+ "homographs": {
+ "raw": {
+ "utterances": 6,
+ "errors": 1,
+ "words": 70,
+ "wer": 0.014285714285714285
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 1,
+ "words": 70,
+ "wer": 0.014285714285714285
+ }
+ },
+ "long": {
+ "raw": {
+ "utterances": 15,
+ "errors": 2,
+ "words": 412,
+ "wer": 0.0048543689320388345
+ },
+ "semantic": {
+ "utterances": 15,
+ "errors": 2,
+ "words": 432,
+ "wer": 0.004629629629629629
+ }
+ },
+ "long_range": {
+ "raw": {
+ "utterances": 6,
+ "errors": 0,
+ "words": 121,
+ "wer": 0.0
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 0,
+ "words": 121,
+ "wer": 0.0
+ }
+ },
+ "names_places": {
+ "raw": {
+ "utterances": 6,
+ "errors": 2,
+ "words": 57,
+ "wer": 0.03508771929824561
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 2,
+ "words": 57,
+ "wer": 0.03508771929824561
+ }
+ },
+ "narrative": {
+ "raw": {
+ "utterances": 24,
+ "errors": 3,
+ "words": 342,
+ "wer": 0.008771929824561403
+ },
+ "semantic": {
+ "utterances": 24,
+ "errors": 2,
+ "words": 352,
+ "wer": 0.005681818181818182
+ }
+ },
+ "numbers": {
+ "raw": {
+ "utterances": 6,
+ "errors": 32,
+ "words": 81,
+ "wer": 0.3950617283950617
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 5,
+ "words": 64,
+ "wer": 0.078125
+ }
+ },
+ "phonetic_balance": {
+ "raw": {
+ "utterances": 300,
+ "errors": 39,
+ "words": 2605,
+ "wer": 0.014971209213051824
+ },
+ "semantic": {
+ "utterances": 300,
+ "errors": 30,
+ "words": 2627,
+ "wer": 0.011419870574800152
+ }
+ },
+ "punctuation": {
+ "raw": {
+ "utterances": 6,
+ "errors": 0,
+ "words": 65,
+ "wer": 0.0
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 0,
+ "words": 65,
+ "wer": 0.0
+ }
+ },
+ "question": {
+ "raw": {
+ "utterances": 15,
+ "errors": 0,
+ "words": 163,
+ "wer": 0.0
+ },
+ "semantic": {
+ "utterances": 15,
+ "errors": 0,
+ "words": 165,
+ "wer": 0.0
+ }
+ },
+ "rare_words": {
+ "raw": {
+ "utterances": 6,
+ "errors": 2,
+ "words": 57,
+ "wer": 0.03508771929824561
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 2,
+ "words": 57,
+ "wer": 0.03508771929824561
+ }
+ },
+ "short": {
+ "raw": {
+ "utterances": 24,
+ "errors": 2,
+ "words": 98,
+ "wer": 0.02040816326530612
+ },
+ "semantic": {
+ "utterances": 24,
+ "errors": 2,
+ "words": 105,
+ "wer": 0.01904761904761905
+ }
+ },
+ "technical": {
+ "raw": {
+ "utterances": 21,
+ "errors": 9,
+ "words": 251,
+ "wer": 0.035856573705179286
+ },
+ "semantic": {
+ "utterances": 21,
+ "errors": 5,
+ "words": 256,
+ "wer": 0.01953125
+ }
+ }
+ }
+ },
+ "wav2vec2-large-lv60k-960h": {
+ "raw": {
+ "utterances": 500,
+ "errors": 196,
+ "words": 5149,
+ "wer": 0.03806564381433288
+ },
+ "raw_95ci": [
+ 0.03104042920099636,
+ 0.04564714946070878
+ ],
+ "semantic": {
+ "utterances": 500,
+ "errors": 186,
+ "words": 5235,
+ "wer": 0.03553008595988539
+ },
+ "semantic_95ci": [
+ 0.028724626579854463,
+ 0.04309006211180124
+ ],
+ "categories": {
+ "conversational": {
+ "raw": {
+ "utterances": 25,
+ "errors": 4,
+ "words": 296,
+ "wer": 0.013513513513513514
+ },
+ "semantic": {
+ "utterances": 25,
+ "errors": 4,
+ "words": 309,
+ "wer": 0.012944983818770227
+ }
+ },
+ "descriptive": {
+ "raw": {
+ "utterances": 16,
+ "errors": 2,
+ "words": 217,
+ "wer": 0.009216589861751152
+ },
+ "semantic": {
+ "utterances": 16,
+ "errors": 2,
+ "words": 226,
+ "wer": 0.008849557522123894
+ }
+ },
+ "emotional": {
+ "raw": {
+ "utterances": 24,
+ "errors": 6,
+ "words": 314,
+ "wer": 0.01910828025477707
+ },
+ "semantic": {
+ "utterances": 24,
+ "errors": 1,
+ "words": 329,
+ "wer": 0.00303951367781155
+ }
+ },
+ "homographs": {
+ "raw": {
+ "utterances": 6,
+ "errors": 4,
+ "words": 70,
+ "wer": 0.05714285714285714
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 4,
+ "words": 70,
+ "wer": 0.05714285714285714
+ }
+ },
+ "long": {
+ "raw": {
+ "utterances": 15,
+ "errors": 7,
+ "words": 412,
+ "wer": 0.01699029126213592
+ },
+ "semantic": {
+ "utterances": 15,
+ "errors": 7,
+ "words": 432,
+ "wer": 0.016203703703703703
+ }
+ },
+ "long_range": {
+ "raw": {
+ "utterances": 6,
+ "errors": 7,
+ "words": 121,
+ "wer": 0.05785123966942149
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 7,
+ "words": 121,
+ "wer": 0.05785123966942149
+ }
+ },
+ "names_places": {
+ "raw": {
+ "utterances": 6,
+ "errors": 15,
+ "words": 57,
+ "wer": 0.2631578947368421
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 15,
+ "words": 57,
+ "wer": 0.2631578947368421
+ }
+ },
+ "narrative": {
+ "raw": {
+ "utterances": 24,
+ "errors": 7,
+ "words": 342,
+ "wer": 0.02046783625730994
+ },
+ "semantic": {
+ "utterances": 24,
+ "errors": 7,
+ "words": 352,
+ "wer": 0.019886363636363636
+ }
+ },
+ "numbers": {
+ "raw": {
+ "utterances": 6,
+ "errors": 9,
+ "words": 81,
+ "wer": 0.1111111111111111
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 5,
+ "words": 64,
+ "wer": 0.078125
+ }
+ },
+ "phonetic_balance": {
+ "raw": {
+ "utterances": 300,
+ "errors": 88,
+ "words": 2605,
+ "wer": 0.03378119001919386
+ },
+ "semantic": {
+ "utterances": 300,
+ "errors": 89,
+ "words": 2627,
+ "wer": 0.03387894937190712
+ }
+ },
+ "punctuation": {
+ "raw": {
+ "utterances": 6,
+ "errors": 3,
+ "words": 65,
+ "wer": 0.046153846153846156
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 3,
+ "words": 65,
+ "wer": 0.046153846153846156
+ }
+ },
+ "question": {
+ "raw": {
+ "utterances": 15,
+ "errors": 2,
+ "words": 163,
+ "wer": 0.012269938650306749
+ },
+ "semantic": {
+ "utterances": 15,
+ "errors": 2,
+ "words": 165,
+ "wer": 0.012121212121212121
+ }
+ },
+ "rare_words": {
+ "raw": {
+ "utterances": 6,
+ "errors": 9,
+ "words": 57,
+ "wer": 0.15789473684210525
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 9,
+ "words": 57,
+ "wer": 0.15789473684210525
+ }
+ },
+ "short": {
+ "raw": {
+ "utterances": 24,
+ "errors": 9,
+ "words": 98,
+ "wer": 0.09183673469387756
+ },
+ "semantic": {
+ "utterances": 24,
+ "errors": 9,
+ "words": 105,
+ "wer": 0.08571428571428572
+ }
+ },
+ "technical": {
+ "raw": {
+ "utterances": 21,
+ "errors": 24,
+ "words": 251,
+ "wer": 0.09561752988047809
+ },
+ "semantic": {
+ "utterances": 21,
+ "errors": 22,
+ "words": 256,
+ "wer": 0.0859375
+ }
+ }
+ }
+ }
+ },
+ "cross_asr_mean_semantic_wer": 0.022636103151862465,
+ "cross_asr_worst_semantic_wer": 0.03553008595988539
+ },
+ "kitten-nano-hugo": {
+ "asr": {
+ "whisper-large-v3": {
+ "raw": {
+ "utterances": 500,
+ "errors": 104,
+ "words": 5149,
+ "wer": 0.020198096717809283
+ },
+ "raw_95ci": [
+ 0.01382394795690181,
+ 0.027267453103848385
+ ],
+ "semantic": {
+ "utterances": 500,
+ "errors": 60,
+ "words": 5235,
+ "wer": 0.011461318051575931
+ },
+ "semantic_95ci": [
+ 0.007995319812792511,
+ 0.01523625843780135
+ ],
+ "categories": {
+ "conversational": {
+ "raw": {
+ "utterances": 25,
+ "errors": 1,
+ "words": 296,
+ "wer": 0.0033783783783783786
+ },
+ "semantic": {
+ "utterances": 25,
+ "errors": 1,
+ "words": 309,
+ "wer": 0.003236245954692557
+ }
+ },
+ "descriptive": {
+ "raw": {
+ "utterances": 16,
+ "errors": 1,
+ "words": 217,
+ "wer": 0.004608294930875576
+ },
+ "semantic": {
+ "utterances": 16,
+ "errors": 0,
+ "words": 226,
+ "wer": 0.0
+ }
+ },
+ "emotional": {
+ "raw": {
+ "utterances": 24,
+ "errors": 0,
+ "words": 314,
+ "wer": 0.0
+ },
+ "semantic": {
+ "utterances": 24,
+ "errors": 0,
+ "words": 329,
+ "wer": 0.0
+ }
+ },
+ "homographs": {
+ "raw": {
+ "utterances": 6,
+ "errors": 1,
+ "words": 70,
+ "wer": 0.014285714285714285
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 1,
+ "words": 70,
+ "wer": 0.014285714285714285
+ }
+ },
+ "long": {
+ "raw": {
+ "utterances": 15,
+ "errors": 3,
+ "words": 412,
+ "wer": 0.007281553398058253
+ },
+ "semantic": {
+ "utterances": 15,
+ "errors": 3,
+ "words": 432,
+ "wer": 0.006944444444444444
+ }
+ },
+ "long_range": {
+ "raw": {
+ "utterances": 6,
+ "errors": 0,
+ "words": 121,
+ "wer": 0.0
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 0,
+ "words": 121,
+ "wer": 0.0
+ }
+ },
+ "names_places": {
+ "raw": {
+ "utterances": 6,
+ "errors": 2,
+ "words": 57,
+ "wer": 0.03508771929824561
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 2,
+ "words": 57,
+ "wer": 0.03508771929824561
+ }
+ },
+ "narrative": {
+ "raw": {
+ "utterances": 24,
+ "errors": 6,
+ "words": 342,
+ "wer": 0.017543859649122806
+ },
+ "semantic": {
+ "utterances": 24,
+ "errors": 5,
+ "words": 352,
+ "wer": 0.014204545454545454
+ }
+ },
+ "numbers": {
+ "raw": {
+ "utterances": 6,
+ "errors": 32,
+ "words": 81,
+ "wer": 0.3950617283950617
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 5,
+ "words": 64,
+ "wer": 0.078125
+ }
+ },
+ "phonetic_balance": {
+ "raw": {
+ "utterances": 300,
+ "errors": 39,
+ "words": 2605,
+ "wer": 0.014971209213051824
+ },
+ "semantic": {
+ "utterances": 300,
+ "errors": 30,
+ "words": 2627,
+ "wer": 0.011419870574800152
+ }
+ },
+ "punctuation": {
+ "raw": {
+ "utterances": 6,
+ "errors": 0,
+ "words": 65,
+ "wer": 0.0
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 0,
+ "words": 65,
+ "wer": 0.0
+ }
+ },
+ "question": {
+ "raw": {
+ "utterances": 15,
+ "errors": 0,
+ "words": 163,
+ "wer": 0.0
+ },
+ "semantic": {
+ "utterances": 15,
+ "errors": 0,
+ "words": 165,
+ "wer": 0.0
+ }
+ },
+ "rare_words": {
+ "raw": {
+ "utterances": 6,
+ "errors": 3,
+ "words": 57,
+ "wer": 0.05263157894736842
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 3,
+ "words": 57,
+ "wer": 0.05263157894736842
+ }
+ },
+ "short": {
+ "raw": {
+ "utterances": 24,
+ "errors": 4,
+ "words": 98,
+ "wer": 0.04081632653061224
+ },
+ "semantic": {
+ "utterances": 24,
+ "errors": 4,
+ "words": 105,
+ "wer": 0.0380952380952381
+ }
+ },
+ "technical": {
+ "raw": {
+ "utterances": 21,
+ "errors": 12,
+ "words": 251,
+ "wer": 0.04780876494023904
+ },
+ "semantic": {
+ "utterances": 21,
+ "errors": 6,
+ "words": 256,
+ "wer": 0.0234375
+ }
+ }
+ }
+ },
+ "wav2vec2-large-lv60k-960h": {
+ "raw": {
+ "utterances": 500,
+ "errors": 192,
+ "words": 5149,
+ "wer": 0.03728879394057098
+ },
+ "raw_95ci": [
+ 0.0302504816955684,
+ 0.04482286161675474
+ ],
+ "semantic": {
+ "utterances": 500,
+ "errors": 190,
+ "words": 5235,
+ "wer": 0.03629417382999045
+ },
+ "semantic_95ci": [
+ 0.029451395572666024,
+ 0.043798449612403104
+ ],
+ "categories": {
+ "conversational": {
+ "raw": {
+ "utterances": 25,
+ "errors": 4,
+ "words": 296,
+ "wer": 0.013513513513513514
+ },
+ "semantic": {
+ "utterances": 25,
+ "errors": 4,
+ "words": 309,
+ "wer": 0.012944983818770227
+ }
+ },
+ "descriptive": {
+ "raw": {
+ "utterances": 16,
+ "errors": 4,
+ "words": 217,
+ "wer": 0.018433179723502304
+ },
+ "semantic": {
+ "utterances": 16,
+ "errors": 4,
+ "words": 226,
+ "wer": 0.017699115044247787
+ }
+ },
+ "emotional": {
+ "raw": {
+ "utterances": 24,
+ "errors": 5,
+ "words": 314,
+ "wer": 0.01592356687898089
+ },
+ "semantic": {
+ "utterances": 24,
+ "errors": 4,
+ "words": 329,
+ "wer": 0.0121580547112462
+ }
+ },
+ "homographs": {
+ "raw": {
+ "utterances": 6,
+ "errors": 3,
+ "words": 70,
+ "wer": 0.04285714285714286
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 3,
+ "words": 70,
+ "wer": 0.04285714285714286
+ }
+ },
+ "long": {
+ "raw": {
+ "utterances": 15,
+ "errors": 8,
+ "words": 412,
+ "wer": 0.019417475728155338
+ },
+ "semantic": {
+ "utterances": 15,
+ "errors": 8,
+ "words": 432,
+ "wer": 0.018518518518518517
+ }
+ },
+ "long_range": {
+ "raw": {
+ "utterances": 6,
+ "errors": 7,
+ "words": 121,
+ "wer": 0.05785123966942149
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 7,
+ "words": 121,
+ "wer": 0.05785123966942149
+ }
+ },
+ "names_places": {
+ "raw": {
+ "utterances": 6,
+ "errors": 16,
+ "words": 57,
+ "wer": 0.2807017543859649
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 16,
+ "words": 57,
+ "wer": 0.2807017543859649
+ }
+ },
+ "narrative": {
+ "raw": {
+ "utterances": 24,
+ "errors": 8,
+ "words": 342,
+ "wer": 0.023391812865497075
+ },
+ "semantic": {
+ "utterances": 24,
+ "errors": 8,
+ "words": 352,
+ "wer": 0.022727272727272728
+ }
+ },
+ "numbers": {
+ "raw": {
+ "utterances": 6,
+ "errors": 13,
+ "words": 81,
+ "wer": 0.16049382716049382
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 14,
+ "words": 64,
+ "wer": 0.21875
+ }
+ },
+ "phonetic_balance": {
+ "raw": {
+ "utterances": 300,
+ "errors": 76,
+ "words": 2605,
+ "wer": 0.029174664107485603
+ },
+ "semantic": {
+ "utterances": 300,
+ "errors": 76,
+ "words": 2627,
+ "wer": 0.02893033878949372
+ }
+ },
+ "punctuation": {
+ "raw": {
+ "utterances": 6,
+ "errors": 3,
+ "words": 65,
+ "wer": 0.046153846153846156
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 3,
+ "words": 65,
+ "wer": 0.046153846153846156
+ }
+ },
+ "question": {
+ "raw": {
+ "utterances": 15,
+ "errors": 2,
+ "words": 163,
+ "wer": 0.012269938650306749
+ },
+ "semantic": {
+ "utterances": 15,
+ "errors": 2,
+ "words": 165,
+ "wer": 0.012121212121212121
+ }
+ },
+ "rare_words": {
+ "raw": {
+ "utterances": 6,
+ "errors": 10,
+ "words": 57,
+ "wer": 0.17543859649122806
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 10,
+ "words": 57,
+ "wer": 0.17543859649122806
+ }
+ },
+ "short": {
+ "raw": {
+ "utterances": 24,
+ "errors": 11,
+ "words": 98,
+ "wer": 0.11224489795918367
+ },
+ "semantic": {
+ "utterances": 24,
+ "errors": 11,
+ "words": 105,
+ "wer": 0.10476190476190476
+ }
+ },
+ "technical": {
+ "raw": {
+ "utterances": 21,
+ "errors": 22,
+ "words": 251,
+ "wer": 0.08764940239043825
+ },
+ "semantic": {
+ "utterances": 21,
+ "errors": 20,
+ "words": 256,
+ "wer": 0.078125
+ }
+ }
+ }
+ }
+ },
+ "cross_asr_mean_semantic_wer": 0.02387774594078319,
+ "cross_asr_worst_semantic_wer": 0.03629417382999045
+ },
+ "piper-danny-low": {
+ "asr": {
+ "whisper-large-v3": {
+ "raw": {
+ "utterances": 500,
+ "errors": 139,
+ "words": 5149,
+ "wer": 0.02699553311322587
+ },
+ "raw_95ci": [
+ 0.01997311311695794,
+ 0.034605980816249764
+ ],
+ "semantic": {
+ "utterances": 500,
+ "errors": 95,
+ "words": 5235,
+ "wer": 0.018147086914995225
+ },
+ "semantic_95ci": [
+ 0.013549618320610687,
+ 0.022975077881619937
+ ],
+ "categories": {
+ "conversational": {
+ "raw": {
+ "utterances": 25,
+ "errors": 2,
+ "words": 296,
+ "wer": 0.006756756756756757
+ },
+ "semantic": {
+ "utterances": 25,
+ "errors": 2,
+ "words": 309,
+ "wer": 0.006472491909385114
+ }
+ },
+ "descriptive": {
+ "raw": {
+ "utterances": 16,
+ "errors": 2,
+ "words": 217,
+ "wer": 0.009216589861751152
+ },
+ "semantic": {
+ "utterances": 16,
+ "errors": 1,
+ "words": 226,
+ "wer": 0.004424778761061947
+ }
+ },
+ "emotional": {
+ "raw": {
+ "utterances": 24,
+ "errors": 1,
+ "words": 314,
+ "wer": 0.0031847133757961785
+ },
+ "semantic": {
+ "utterances": 24,
+ "errors": 1,
+ "words": 329,
+ "wer": 0.00303951367781155
+ }
+ },
+ "homographs": {
+ "raw": {
+ "utterances": 6,
+ "errors": 1,
+ "words": 70,
+ "wer": 0.014285714285714285
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 1,
+ "words": 70,
+ "wer": 0.014285714285714285
+ }
+ },
+ "long": {
+ "raw": {
+ "utterances": 15,
+ "errors": 7,
+ "words": 412,
+ "wer": 0.01699029126213592
+ },
+ "semantic": {
+ "utterances": 15,
+ "errors": 7,
+ "words": 432,
+ "wer": 0.016203703703703703
+ }
+ },
+ "long_range": {
+ "raw": {
+ "utterances": 6,
+ "errors": 0,
+ "words": 121,
+ "wer": 0.0
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 0,
+ "words": 121,
+ "wer": 0.0
+ }
+ },
+ "names_places": {
+ "raw": {
+ "utterances": 6,
+ "errors": 3,
+ "words": 57,
+ "wer": 0.05263157894736842
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 3,
+ "words": 57,
+ "wer": 0.05263157894736842
+ }
+ },
+ "narrative": {
+ "raw": {
+ "utterances": 24,
+ "errors": 1,
+ "words": 342,
+ "wer": 0.0029239766081871343
+ },
+ "semantic": {
+ "utterances": 24,
+ "errors": 0,
+ "words": 352,
+ "wer": 0.0
+ }
+ },
+ "numbers": {
+ "raw": {
+ "utterances": 6,
+ "errors": 33,
+ "words": 81,
+ "wer": 0.4074074074074074
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 7,
+ "words": 64,
+ "wer": 0.109375
+ }
+ },
+ "phonetic_balance": {
+ "raw": {
+ "utterances": 300,
+ "errors": 73,
+ "words": 2605,
+ "wer": 0.02802303262955854
+ },
+ "semantic": {
+ "utterances": 300,
+ "errors": 61,
+ "words": 2627,
+ "wer": 0.02322040350209364
+ }
+ },
+ "punctuation": {
+ "raw": {
+ "utterances": 6,
+ "errors": 0,
+ "words": 65,
+ "wer": 0.0
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 0,
+ "words": 65,
+ "wer": 0.0
+ }
+ },
+ "question": {
+ "raw": {
+ "utterances": 15,
+ "errors": 2,
+ "words": 163,
+ "wer": 0.012269938650306749
+ },
+ "semantic": {
+ "utterances": 15,
+ "errors": 2,
+ "words": 165,
+ "wer": 0.012121212121212121
+ }
+ },
+ "rare_words": {
+ "raw": {
+ "utterances": 6,
+ "errors": 2,
+ "words": 57,
+ "wer": 0.03508771929824561
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 2,
+ "words": 57,
+ "wer": 0.03508771929824561
+ }
+ },
+ "short": {
+ "raw": {
+ "utterances": 24,
+ "errors": 3,
+ "words": 98,
+ "wer": 0.030612244897959183
+ },
+ "semantic": {
+ "utterances": 24,
+ "errors": 3,
+ "words": 105,
+ "wer": 0.02857142857142857
+ }
+ },
+ "technical": {
+ "raw": {
+ "utterances": 21,
+ "errors": 9,
+ "words": 251,
+ "wer": 0.035856573705179286
+ },
+ "semantic": {
+ "utterances": 21,
+ "errors": 5,
+ "words": 256,
+ "wer": 0.01953125
+ }
+ }
+ }
+ },
+ "wav2vec2-large-lv60k-960h": {
+ "raw": {
+ "utterances": 500,
+ "errors": 282,
+ "words": 5149,
+ "wer": 0.054767916100213636
+ },
+ "raw_95ci": [
+ 0.04591078066914498,
+ 0.0641751604379011
+ ],
+ "semantic": {
+ "utterances": 500,
+ "errors": 278,
+ "words": 5235,
+ "wer": 0.053104106972301814
+ },
+ "semantic_95ci": [
+ 0.04440632742519535,
+ 0.06227176628867961
+ ],
+ "categories": {
+ "conversational": {
+ "raw": {
+ "utterances": 25,
+ "errors": 7,
+ "words": 296,
+ "wer": 0.02364864864864865
+ },
+ "semantic": {
+ "utterances": 25,
+ "errors": 7,
+ "words": 309,
+ "wer": 0.022653721682847898
+ }
+ },
+ "descriptive": {
+ "raw": {
+ "utterances": 16,
+ "errors": 11,
+ "words": 217,
+ "wer": 0.05069124423963134
+ },
+ "semantic": {
+ "utterances": 16,
+ "errors": 11,
+ "words": 226,
+ "wer": 0.048672566371681415
+ }
+ },
+ "emotional": {
+ "raw": {
+ "utterances": 24,
+ "errors": 8,
+ "words": 314,
+ "wer": 0.025477707006369428
+ },
+ "semantic": {
+ "utterances": 24,
+ "errors": 6,
+ "words": 329,
+ "wer": 0.0182370820668693
+ }
+ },
+ "homographs": {
+ "raw": {
+ "utterances": 6,
+ "errors": 3,
+ "words": 70,
+ "wer": 0.04285714285714286
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 3,
+ "words": 70,
+ "wer": 0.04285714285714286
+ }
+ },
+ "long": {
+ "raw": {
+ "utterances": 15,
+ "errors": 9,
+ "words": 412,
+ "wer": 0.021844660194174758
+ },
+ "semantic": {
+ "utterances": 15,
+ "errors": 9,
+ "words": 432,
+ "wer": 0.020833333333333332
+ }
+ },
+ "long_range": {
+ "raw": {
+ "utterances": 6,
+ "errors": 8,
+ "words": 121,
+ "wer": 0.06611570247933884
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 8,
+ "words": 121,
+ "wer": 0.06611570247933884
+ }
+ },
+ "names_places": {
+ "raw": {
+ "utterances": 6,
+ "errors": 17,
+ "words": 57,
+ "wer": 0.2982456140350877
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 17,
+ "words": 57,
+ "wer": 0.2982456140350877
+ }
+ },
+ "narrative": {
+ "raw": {
+ "utterances": 24,
+ "errors": 6,
+ "words": 342,
+ "wer": 0.017543859649122806
+ },
+ "semantic": {
+ "utterances": 24,
+ "errors": 6,
+ "words": 352,
+ "wer": 0.017045454545454544
+ }
+ },
+ "numbers": {
+ "raw": {
+ "utterances": 6,
+ "errors": 9,
+ "words": 81,
+ "wer": 0.1111111111111111
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 5,
+ "words": 64,
+ "wer": 0.078125
+ }
+ },
+ "phonetic_balance": {
+ "raw": {
+ "utterances": 300,
+ "errors": 141,
+ "words": 2605,
+ "wer": 0.05412667946257198
+ },
+ "semantic": {
+ "utterances": 300,
+ "errors": 142,
+ "words": 2627,
+ "wer": 0.05405405405405406
+ }
+ },
+ "punctuation": {
+ "raw": {
+ "utterances": 6,
+ "errors": 5,
+ "words": 65,
+ "wer": 0.07692307692307693
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 5,
+ "words": 65,
+ "wer": 0.07692307692307693
+ }
+ },
+ "question": {
+ "raw": {
+ "utterances": 15,
+ "errors": 4,
+ "words": 163,
+ "wer": 0.024539877300613498
+ },
+ "semantic": {
+ "utterances": 15,
+ "errors": 5,
+ "words": 165,
+ "wer": 0.030303030303030304
+ }
+ },
+ "rare_words": {
+ "raw": {
+ "utterances": 6,
+ "errors": 13,
+ "words": 57,
+ "wer": 0.22807017543859648
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 13,
+ "words": 57,
+ "wer": 0.22807017543859648
+ }
+ },
+ "short": {
+ "raw": {
+ "utterances": 24,
+ "errors": 14,
+ "words": 98,
+ "wer": 0.14285714285714285
+ },
+ "semantic": {
+ "utterances": 24,
+ "errors": 14,
+ "words": 105,
+ "wer": 0.13333333333333333
+ }
+ },
+ "technical": {
+ "raw": {
+ "utterances": 21,
+ "errors": 27,
+ "words": 251,
+ "wer": 0.10756972111553785
+ },
+ "semantic": {
+ "utterances": 21,
+ "errors": 27,
+ "words": 256,
+ "wer": 0.10546875
+ }
+ }
+ }
+ }
+ },
+ "cross_asr_mean_semantic_wer": 0.035625596943648516,
+ "cross_asr_worst_semantic_wer": 0.053104106972301814
+ },
+ "piper-ryan-low": {
+ "asr": {
+ "whisper-large-v3": {
+ "raw": {
+ "utterances": 500,
+ "errors": 168,
+ "words": 5149,
+ "wer": 0.03262769469799961
+ },
+ "raw_95ci": [
+ 0.025009846396218983,
+ 0.04082039028275587
+ ],
+ "semantic": {
+ "utterances": 500,
+ "errors": 122,
+ "words": 5235,
+ "wer": 0.023304680038204392
+ },
+ "semantic_95ci": [
+ 0.018028169014084508,
+ 0.02901655306718598
+ ],
+ "categories": {
+ "conversational": {
+ "raw": {
+ "utterances": 25,
+ "errors": 8,
+ "words": 296,
+ "wer": 0.02702702702702703
+ },
+ "semantic": {
+ "utterances": 25,
+ "errors": 5,
+ "words": 309,
+ "wer": 0.016181229773462782
+ }
+ },
+ "descriptive": {
+ "raw": {
+ "utterances": 16,
+ "errors": 3,
+ "words": 217,
+ "wer": 0.013824884792626729
+ },
+ "semantic": {
+ "utterances": 16,
+ "errors": 1,
+ "words": 226,
+ "wer": 0.004424778761061947
+ }
+ },
+ "emotional": {
+ "raw": {
+ "utterances": 24,
+ "errors": 1,
+ "words": 314,
+ "wer": 0.0031847133757961785
+ },
+ "semantic": {
+ "utterances": 24,
+ "errors": 1,
+ "words": 329,
+ "wer": 0.00303951367781155
+ }
+ },
+ "homographs": {
+ "raw": {
+ "utterances": 6,
+ "errors": 1,
+ "words": 70,
+ "wer": 0.014285714285714285
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 1,
+ "words": 70,
+ "wer": 0.014285714285714285
+ }
+ },
+ "long": {
+ "raw": {
+ "utterances": 15,
+ "errors": 3,
+ "words": 412,
+ "wer": 0.007281553398058253
+ },
+ "semantic": {
+ "utterances": 15,
+ "errors": 3,
+ "words": 432,
+ "wer": 0.006944444444444444
+ }
+ },
+ "long_range": {
+ "raw": {
+ "utterances": 6,
+ "errors": 1,
+ "words": 121,
+ "wer": 0.008264462809917356
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 1,
+ "words": 121,
+ "wer": 0.008264462809917356
+ }
+ },
+ "names_places": {
+ "raw": {
+ "utterances": 6,
+ "errors": 2,
+ "words": 57,
+ "wer": 0.03508771929824561
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 2,
+ "words": 57,
+ "wer": 0.03508771929824561
+ }
+ },
+ "narrative": {
+ "raw": {
+ "utterances": 24,
+ "errors": 1,
+ "words": 342,
+ "wer": 0.0029239766081871343
+ },
+ "semantic": {
+ "utterances": 24,
+ "errors": 0,
+ "words": 352,
+ "wer": 0.0
+ }
+ },
+ "numbers": {
+ "raw": {
+ "utterances": 6,
+ "errors": 32,
+ "words": 81,
+ "wer": 0.3950617283950617
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 6,
+ "words": 64,
+ "wer": 0.09375
+ }
+ },
+ "phonetic_balance": {
+ "raw": {
+ "utterances": 300,
+ "errors": 94,
+ "words": 2605,
+ "wer": 0.03608445297504798
+ },
+ "semantic": {
+ "utterances": 300,
+ "errors": 83,
+ "words": 2627,
+ "wer": 0.031594975256947086
+ }
+ },
+ "punctuation": {
+ "raw": {
+ "utterances": 6,
+ "errors": 1,
+ "words": 65,
+ "wer": 0.015384615384615385
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 1,
+ "words": 65,
+ "wer": 0.015384615384615385
+ }
+ },
+ "question": {
+ "raw": {
+ "utterances": 15,
+ "errors": 0,
+ "words": 163,
+ "wer": 0.0
+ },
+ "semantic": {
+ "utterances": 15,
+ "errors": 0,
+ "words": 165,
+ "wer": 0.0
+ }
+ },
+ "rare_words": {
+ "raw": {
+ "utterances": 6,
+ "errors": 4,
+ "words": 57,
+ "wer": 0.07017543859649122
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 4,
+ "words": 57,
+ "wer": 0.07017543859649122
+ }
+ },
+ "short": {
+ "raw": {
+ "utterances": 24,
+ "errors": 7,
+ "words": 98,
+ "wer": 0.07142857142857142
+ },
+ "semantic": {
+ "utterances": 24,
+ "errors": 8,
+ "words": 105,
+ "wer": 0.0761904761904762
+ }
+ },
+ "technical": {
+ "raw": {
+ "utterances": 21,
+ "errors": 10,
+ "words": 251,
+ "wer": 0.0398406374501992
+ },
+ "semantic": {
+ "utterances": 21,
+ "errors": 6,
+ "words": 256,
+ "wer": 0.0234375
+ }
+ }
+ }
+ },
+ "wav2vec2-large-lv60k-960h": {
+ "raw": {
+ "utterances": 500,
+ "errors": 356,
+ "words": 5149,
+ "wer": 0.0691396387648087
+ },
+ "raw_95ci": [
+ 0.05994496855345912,
+ 0.07860090391039497
+ ],
+ "semantic": {
+ "utterances": 500,
+ "errors": 351,
+ "words": 5235,
+ "wer": 0.0670487106017192
+ },
+ "semantic_95ci": [
+ 0.05838601710673113,
+ 0.07638619620007756
+ ],
+ "categories": {
+ "conversational": {
+ "raw": {
+ "utterances": 25,
+ "errors": 12,
+ "words": 296,
+ "wer": 0.04054054054054054
+ },
+ "semantic": {
+ "utterances": 25,
+ "errors": 12,
+ "words": 309,
+ "wer": 0.038834951456310676
+ }
+ },
+ "descriptive": {
+ "raw": {
+ "utterances": 16,
+ "errors": 8,
+ "words": 217,
+ "wer": 0.03686635944700461
+ },
+ "semantic": {
+ "utterances": 16,
+ "errors": 8,
+ "words": 226,
+ "wer": 0.035398230088495575
+ }
+ },
+ "emotional": {
+ "raw": {
+ "utterances": 24,
+ "errors": 8,
+ "words": 314,
+ "wer": 0.025477707006369428
+ },
+ "semantic": {
+ "utterances": 24,
+ "errors": 8,
+ "words": 329,
+ "wer": 0.0243161094224924
+ }
+ },
+ "homographs": {
+ "raw": {
+ "utterances": 6,
+ "errors": 7,
+ "words": 70,
+ "wer": 0.1
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 7,
+ "words": 70,
+ "wer": 0.1
+ }
+ },
+ "long": {
+ "raw": {
+ "utterances": 15,
+ "errors": 15,
+ "words": 412,
+ "wer": 0.03640776699029126
+ },
+ "semantic": {
+ "utterances": 15,
+ "errors": 14,
+ "words": 432,
+ "wer": 0.032407407407407406
+ }
+ },
+ "long_range": {
+ "raw": {
+ "utterances": 6,
+ "errors": 10,
+ "words": 121,
+ "wer": 0.08264462809917356
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 10,
+ "words": 121,
+ "wer": 0.08264462809917356
+ }
+ },
+ "names_places": {
+ "raw": {
+ "utterances": 6,
+ "errors": 17,
+ "words": 57,
+ "wer": 0.2982456140350877
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 17,
+ "words": 57,
+ "wer": 0.2982456140350877
+ }
+ },
+ "narrative": {
+ "raw": {
+ "utterances": 24,
+ "errors": 8,
+ "words": 342,
+ "wer": 0.023391812865497075
+ },
+ "semantic": {
+ "utterances": 24,
+ "errors": 8,
+ "words": 352,
+ "wer": 0.022727272727272728
+ }
+ },
+ "numbers": {
+ "raw": {
+ "utterances": 6,
+ "errors": 9,
+ "words": 81,
+ "wer": 0.1111111111111111
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 8,
+ "words": 64,
+ "wer": 0.125
+ }
+ },
+ "phonetic_balance": {
+ "raw": {
+ "utterances": 300,
+ "errors": 175,
+ "words": 2605,
+ "wer": 0.0671785028790787
+ },
+ "semantic": {
+ "utterances": 300,
+ "errors": 169,
+ "words": 2627,
+ "wer": 0.06433193757137419
+ }
+ },
+ "punctuation": {
+ "raw": {
+ "utterances": 6,
+ "errors": 4,
+ "words": 65,
+ "wer": 0.06153846153846154
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 4,
+ "words": 65,
+ "wer": 0.06153846153846154
+ }
+ },
+ "question": {
+ "raw": {
+ "utterances": 15,
+ "errors": 6,
+ "words": 163,
+ "wer": 0.03680981595092025
+ },
+ "semantic": {
+ "utterances": 15,
+ "errors": 6,
+ "words": 165,
+ "wer": 0.03636363636363636
+ }
+ },
+ "rare_words": {
+ "raw": {
+ "utterances": 6,
+ "errors": 16,
+ "words": 57,
+ "wer": 0.2807017543859649
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 16,
+ "words": 57,
+ "wer": 0.2807017543859649
+ }
+ },
+ "short": {
+ "raw": {
+ "utterances": 24,
+ "errors": 22,
+ "words": 98,
+ "wer": 0.22448979591836735
+ },
+ "semantic": {
+ "utterances": 24,
+ "errors": 24,
+ "words": 105,
+ "wer": 0.22857142857142856
+ }
+ },
+ "technical": {
+ "raw": {
+ "utterances": 21,
+ "errors": 39,
+ "words": 251,
+ "wer": 0.1553784860557769
+ },
+ "semantic": {
+ "utterances": 21,
+ "errors": 40,
+ "words": 256,
+ "wer": 0.15625
+ }
+ }
+ }
+ }
+ },
+ "cross_asr_mean_semantic_wer": 0.045176695319961796,
+ "cross_asr_worst_semantic_wer": 0.0670487106017192
+ },
+ "supertonic3-james-3step": {
+ "asr": {
+ "whisper-large-v3": {
+ "raw": {
+ "utterances": 500,
+ "errors": 34011,
+ "words": 5149,
+ "wer": 6.605360264128957
+ },
+ "raw_95ci": [
+ 5.596401515151515,
+ 7.669077757685352
+ ],
+ "semantic": {
+ "utterances": 500,
+ "errors": 36269,
+ "words": 5235,
+ "wer": 6.928175740210124
+ },
+ "semantic_95ci": [
+ 5.861435406698565,
+ 8.043254987161761
+ ],
+ "categories": {
+ "conversational": {
+ "raw": {
+ "utterances": 25,
+ "errors": 1295,
+ "words": 296,
+ "wer": 4.375
+ },
+ "semantic": {
+ "utterances": 25,
+ "errors": 1449,
+ "words": 309,
+ "wer": 4.689320388349515
+ }
+ },
+ "descriptive": {
+ "raw": {
+ "utterances": 16,
+ "errors": 902,
+ "words": 217,
+ "wer": 4.15668202764977
+ },
+ "semantic": {
+ "utterances": 16,
+ "errors": 998,
+ "words": 226,
+ "wer": 4.415929203539823
+ }
+ },
+ "emotional": {
+ "raw": {
+ "utterances": 24,
+ "errors": 2189,
+ "words": 314,
+ "wer": 6.971337579617835
+ },
+ "semantic": {
+ "utterances": 24,
+ "errors": 2306,
+ "words": 329,
+ "wer": 7.009118541033435
+ }
+ },
+ "homographs": {
+ "raw": {
+ "utterances": 6,
+ "errors": 70,
+ "words": 70,
+ "wer": 1.0
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 73,
+ "words": 70,
+ "wer": 1.042857142857143
+ }
+ },
+ "long": {
+ "raw": {
+ "utterances": 15,
+ "errors": 1113,
+ "words": 412,
+ "wer": 2.7014563106796117
+ },
+ "semantic": {
+ "utterances": 15,
+ "errors": 1190,
+ "words": 432,
+ "wer": 2.7546296296296298
+ }
+ },
+ "long_range": {
+ "raw": {
+ "utterances": 6,
+ "errors": 153,
+ "words": 121,
+ "wer": 1.2644628099173554
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 159,
+ "words": 121,
+ "wer": 1.3140495867768596
+ }
+ },
+ "names_places": {
+ "raw": {
+ "utterances": 6,
+ "errors": 666,
+ "words": 57,
+ "wer": 11.68421052631579
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 828,
+ "words": 57,
+ "wer": 14.526315789473685
+ }
+ },
+ "narrative": {
+ "raw": {
+ "utterances": 24,
+ "errors": 2415,
+ "words": 342,
+ "wer": 7.06140350877193
+ },
+ "semantic": {
+ "utterances": 24,
+ "errors": 2562,
+ "words": 352,
+ "wer": 7.278409090909091
+ }
+ },
+ "numbers": {
+ "raw": {
+ "utterances": 6,
+ "errors": 94,
+ "words": 81,
+ "wer": 1.1604938271604939
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 91,
+ "words": 64,
+ "wer": 1.421875
+ }
+ },
+ "phonetic_balance": {
+ "raw": {
+ "utterances": 300,
+ "errors": 19668,
+ "words": 2605,
+ "wer": 7.550095969289827
+ },
+ "semantic": {
+ "utterances": 300,
+ "errors": 21028,
+ "words": 2627,
+ "wer": 8.00456794822992
+ }
+ },
+ "punctuation": {
+ "raw": {
+ "utterances": 6,
+ "errors": 803,
+ "words": 65,
+ "wer": 12.353846153846154
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 807,
+ "words": 65,
+ "wer": 12.415384615384616
+ }
+ },
+ "question": {
+ "raw": {
+ "utterances": 15,
+ "errors": 1412,
+ "words": 163,
+ "wer": 8.662576687116564
+ },
+ "semantic": {
+ "utterances": 15,
+ "errors": 1529,
+ "words": 165,
+ "wer": 9.266666666666667
+ }
+ },
+ "rare_words": {
+ "raw": {
+ "utterances": 6,
+ "errors": 819,
+ "words": 57,
+ "wer": 14.368421052631579
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 820,
+ "words": 57,
+ "wer": 14.385964912280702
+ }
+ },
+ "short": {
+ "raw": {
+ "utterances": 24,
+ "errors": 436,
+ "words": 98,
+ "wer": 4.448979591836735
+ },
+ "semantic": {
+ "utterances": 24,
+ "errors": 447,
+ "words": 105,
+ "wer": 4.257142857142857
+ }
+ },
+ "technical": {
+ "raw": {
+ "utterances": 21,
+ "errors": 1976,
+ "words": 251,
+ "wer": 7.872509960159363
+ },
+ "semantic": {
+ "utterances": 21,
+ "errors": 1982,
+ "words": 256,
+ "wer": 7.7421875
+ }
+ }
+ }
+ },
+ "wav2vec2-large-lv60k-960h": {
+ "raw": {
+ "utterances": 500,
+ "errors": 707,
+ "words": 5149,
+ "wer": 0.13730821518741504
+ },
+ "raw_95ci": [
+ 0.11963971020168396,
+ 0.15595238095238095
+ ],
+ "semantic": {
+ "utterances": 500,
+ "errors": 705,
+ "words": 5235,
+ "wer": 0.1346704871060172
+ },
+ "semantic_95ci": [
+ 0.11719324026451139,
+ 0.15343099460292983
+ ],
+ "categories": {
+ "conversational": {
+ "raw": {
+ "utterances": 25,
+ "errors": 16,
+ "words": 296,
+ "wer": 0.05405405405405406
+ },
+ "semantic": {
+ "utterances": 25,
+ "errors": 17,
+ "words": 309,
+ "wer": 0.05501618122977346
+ }
+ },
+ "descriptive": {
+ "raw": {
+ "utterances": 16,
+ "errors": 10,
+ "words": 217,
+ "wer": 0.04608294930875576
+ },
+ "semantic": {
+ "utterances": 16,
+ "errors": 11,
+ "words": 226,
+ "wer": 0.048672566371681415
+ }
+ },
+ "emotional": {
+ "raw": {
+ "utterances": 24,
+ "errors": 13,
+ "words": 314,
+ "wer": 0.041401273885350316
+ },
+ "semantic": {
+ "utterances": 24,
+ "errors": 13,
+ "words": 329,
+ "wer": 0.03951367781155015
+ }
+ },
+ "homographs": {
+ "raw": {
+ "utterances": 6,
+ "errors": 7,
+ "words": 70,
+ "wer": 0.1
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 7,
+ "words": 70,
+ "wer": 0.1
+ }
+ },
+ "long": {
+ "raw": {
+ "utterances": 15,
+ "errors": 17,
+ "words": 412,
+ "wer": 0.0412621359223301
+ },
+ "semantic": {
+ "utterances": 15,
+ "errors": 14,
+ "words": 432,
+ "wer": 0.032407407407407406
+ }
+ },
+ "long_range": {
+ "raw": {
+ "utterances": 6,
+ "errors": 5,
+ "words": 121,
+ "wer": 0.04132231404958678
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 5,
+ "words": 121,
+ "wer": 0.04132231404958678
+ }
+ },
+ "names_places": {
+ "raw": {
+ "utterances": 6,
+ "errors": 18,
+ "words": 57,
+ "wer": 0.3157894736842105
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 18,
+ "words": 57,
+ "wer": 0.3157894736842105
+ }
+ },
+ "narrative": {
+ "raw": {
+ "utterances": 24,
+ "errors": 8,
+ "words": 342,
+ "wer": 0.023391812865497075
+ },
+ "semantic": {
+ "utterances": 24,
+ "errors": 7,
+ "words": 352,
+ "wer": 0.019886363636363636
+ }
+ },
+ "numbers": {
+ "raw": {
+ "utterances": 6,
+ "errors": 6,
+ "words": 81,
+ "wer": 0.07407407407407407
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 2,
+ "words": 64,
+ "wer": 0.03125
+ }
+ },
+ "phonetic_balance": {
+ "raw": {
+ "utterances": 300,
+ "errors": 476,
+ "words": 2605,
+ "wer": 0.18272552783109405
+ },
+ "semantic": {
+ "utterances": 300,
+ "errors": 481,
+ "words": 2627,
+ "wer": 0.18309859154929578
+ }
+ },
+ "punctuation": {
+ "raw": {
+ "utterances": 6,
+ "errors": 10,
+ "words": 65,
+ "wer": 0.15384615384615385
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 10,
+ "words": 65,
+ "wer": 0.15384615384615385
+ }
+ },
+ "question": {
+ "raw": {
+ "utterances": 15,
+ "errors": 14,
+ "words": 163,
+ "wer": 0.08588957055214724
+ },
+ "semantic": {
+ "utterances": 15,
+ "errors": 14,
+ "words": 165,
+ "wer": 0.08484848484848485
+ }
+ },
+ "rare_words": {
+ "raw": {
+ "utterances": 6,
+ "errors": 12,
+ "words": 57,
+ "wer": 0.21052631578947367
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 12,
+ "words": 57,
+ "wer": 0.21052631578947367
+ }
+ },
+ "short": {
+ "raw": {
+ "utterances": 24,
+ "errors": 64,
+ "words": 98,
+ "wer": 0.6530612244897959
+ },
+ "semantic": {
+ "utterances": 24,
+ "errors": 63,
+ "words": 105,
+ "wer": 0.6
+ }
+ },
+ "technical": {
+ "raw": {
+ "utterances": 21,
+ "errors": 31,
+ "words": 251,
+ "wer": 0.12350597609561753
+ },
+ "semantic": {
+ "utterances": 21,
+ "errors": 31,
+ "words": 256,
+ "wer": 0.12109375
+ }
+ }
+ }
+ }
+ },
+ "cross_asr_mean_semantic_wer": 3.5314231136580707,
+ "cross_asr_worst_semantic_wer": 6.928175740210124
+ },
+ "supertonic3-james-8step": {
+ "asr": {
+ "whisper-large-v3": {
+ "raw": {
+ "utterances": 500,
+ "errors": 31091,
+ "words": 5149,
+ "wer": 6.038259856282774
+ },
+ "raw_95ci": [
+ 5.040967616074912,
+ 7.07244678773677
+ ],
+ "semantic": {
+ "utterances": 500,
+ "errors": 32589,
+ "words": 5235,
+ "wer": 6.225214899713467
+ },
+ "semantic_95ci": [
+ 5.2127497621313035,
+ 7.2831739961759085
+ ],
+ "categories": {
+ "conversational": {
+ "raw": {
+ "utterances": 25,
+ "errors": 1746,
+ "words": 296,
+ "wer": 5.898648648648648
+ },
+ "semantic": {
+ "utterances": 25,
+ "errors": 1812,
+ "words": 309,
+ "wer": 5.864077669902913
+ }
+ },
+ "descriptive": {
+ "raw": {
+ "utterances": 16,
+ "errors": 1230,
+ "words": 217,
+ "wer": 5.668202764976958
+ },
+ "semantic": {
+ "utterances": 16,
+ "errors": 1234,
+ "words": 226,
+ "wer": 5.460176991150442
+ }
+ },
+ "emotional": {
+ "raw": {
+ "utterances": 24,
+ "errors": 1592,
+ "words": 314,
+ "wer": 5.070063694267516
+ },
+ "semantic": {
+ "utterances": 24,
+ "errors": 1670,
+ "words": 329,
+ "wer": 5.075987841945289
+ }
+ },
+ "homographs": {
+ "raw": {
+ "utterances": 6,
+ "errors": 326,
+ "words": 70,
+ "wer": 4.6571428571428575
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 413,
+ "words": 70,
+ "wer": 5.9
+ }
+ },
+ "long": {
+ "raw": {
+ "utterances": 15,
+ "errors": 1459,
+ "words": 412,
+ "wer": 3.54126213592233
+ },
+ "semantic": {
+ "utterances": 15,
+ "errors": 1468,
+ "words": 432,
+ "wer": 3.3981481481481484
+ }
+ },
+ "long_range": {
+ "raw": {
+ "utterances": 6,
+ "errors": 113,
+ "words": 121,
+ "wer": 0.9338842975206612
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 113,
+ "words": 121,
+ "wer": 0.9338842975206612
+ }
+ },
+ "names_places": {
+ "raw": {
+ "utterances": 6,
+ "errors": 381,
+ "words": 57,
+ "wer": 6.684210526315789
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 476,
+ "words": 57,
+ "wer": 8.350877192982455
+ }
+ },
+ "narrative": {
+ "raw": {
+ "utterances": 24,
+ "errors": 1338,
+ "words": 342,
+ "wer": 3.912280701754386
+ },
+ "semantic": {
+ "utterances": 24,
+ "errors": 1344,
+ "words": 352,
+ "wer": 3.8181818181818183
+ }
+ },
+ "numbers": {
+ "raw": {
+ "utterances": 6,
+ "errors": 90,
+ "words": 81,
+ "wer": 1.1111111111111112
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 87,
+ "words": 64,
+ "wer": 1.359375
+ }
+ },
+ "phonetic_balance": {
+ "raw": {
+ "utterances": 300,
+ "errors": 18274,
+ "words": 2605,
+ "wer": 7.014971209213051
+ },
+ "semantic": {
+ "utterances": 300,
+ "errors": 19044,
+ "words": 2627,
+ "wer": 7.2493338408831365
+ }
+ },
+ "punctuation": {
+ "raw": {
+ "utterances": 6,
+ "errors": 361,
+ "words": 65,
+ "wer": 5.553846153846154
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 438,
+ "words": 65,
+ "wer": 6.7384615384615385
+ }
+ },
+ "question": {
+ "raw": {
+ "utterances": 15,
+ "errors": 1153,
+ "words": 163,
+ "wer": 7.07361963190184
+ },
+ "semantic": {
+ "utterances": 15,
+ "errors": 1267,
+ "words": 165,
+ "wer": 7.678787878787879
+ }
+ },
+ "rare_words": {
+ "raw": {
+ "utterances": 6,
+ "errors": 438,
+ "words": 57,
+ "wer": 7.684210526315789
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 444,
+ "words": 57,
+ "wer": 7.7894736842105265
+ }
+ },
+ "short": {
+ "raw": {
+ "utterances": 24,
+ "errors": 1005,
+ "words": 98,
+ "wer": 10.255102040816327
+ },
+ "semantic": {
+ "utterances": 24,
+ "errors": 1057,
+ "words": 105,
+ "wer": 10.066666666666666
+ }
+ },
+ "technical": {
+ "raw": {
+ "utterances": 21,
+ "errors": 1585,
+ "words": 251,
+ "wer": 6.314741035856573
+ },
+ "semantic": {
+ "utterances": 21,
+ "errors": 1722,
+ "words": 256,
+ "wer": 6.7265625
+ }
+ }
+ }
+ },
+ "wav2vec2-large-lv60k-960h": {
+ "raw": {
+ "utterances": 500,
+ "errors": 203,
+ "words": 5149,
+ "wer": 0.0394251310934162
+ },
+ "raw_95ci": [
+ 0.03225806451612903,
+ 0.04706565950029053
+ ],
+ "semantic": {
+ "utterances": 500,
+ "errors": 197,
+ "words": 5235,
+ "wer": 0.037631327602674305
+ },
+ "semantic_95ci": [
+ 0.030763314747164008,
+ 0.044854881266490766
+ ],
+ "categories": {
+ "conversational": {
+ "raw": {
+ "utterances": 25,
+ "errors": 2,
+ "words": 296,
+ "wer": 0.006756756756756757
+ },
+ "semantic": {
+ "utterances": 25,
+ "errors": 2,
+ "words": 309,
+ "wer": 0.006472491909385114
+ }
+ },
+ "descriptive": {
+ "raw": {
+ "utterances": 16,
+ "errors": 5,
+ "words": 217,
+ "wer": 0.02304147465437788
+ },
+ "semantic": {
+ "utterances": 16,
+ "errors": 5,
+ "words": 226,
+ "wer": 0.022123893805309734
+ }
+ },
+ "emotional": {
+ "raw": {
+ "utterances": 24,
+ "errors": 5,
+ "words": 314,
+ "wer": 0.01592356687898089
+ },
+ "semantic": {
+ "utterances": 24,
+ "errors": 1,
+ "words": 329,
+ "wer": 0.00303951367781155
+ }
+ },
+ "homographs": {
+ "raw": {
+ "utterances": 6,
+ "errors": 5,
+ "words": 70,
+ "wer": 0.07142857142857142
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 5,
+ "words": 70,
+ "wer": 0.07142857142857142
+ }
+ },
+ "long": {
+ "raw": {
+ "utterances": 15,
+ "errors": 5,
+ "words": 412,
+ "wer": 0.012135922330097087
+ },
+ "semantic": {
+ "utterances": 15,
+ "errors": 5,
+ "words": 432,
+ "wer": 0.011574074074074073
+ }
+ },
+ "long_range": {
+ "raw": {
+ "utterances": 6,
+ "errors": 5,
+ "words": 121,
+ "wer": 0.04132231404958678
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 5,
+ "words": 121,
+ "wer": 0.04132231404958678
+ }
+ },
+ "names_places": {
+ "raw": {
+ "utterances": 6,
+ "errors": 17,
+ "words": 57,
+ "wer": 0.2982456140350877
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 17,
+ "words": 57,
+ "wer": 0.2982456140350877
+ }
+ },
+ "narrative": {
+ "raw": {
+ "utterances": 24,
+ "errors": 2,
+ "words": 342,
+ "wer": 0.005847953216374269
+ },
+ "semantic": {
+ "utterances": 24,
+ "errors": 2,
+ "words": 352,
+ "wer": 0.005681818181818182
+ }
+ },
+ "numbers": {
+ "raw": {
+ "utterances": 6,
+ "errors": 7,
+ "words": 81,
+ "wer": 0.08641975308641975
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 3,
+ "words": 64,
+ "wer": 0.046875
+ }
+ },
+ "phonetic_balance": {
+ "raw": {
+ "utterances": 300,
+ "errors": 101,
+ "words": 2605,
+ "wer": 0.03877159309021113
+ },
+ "semantic": {
+ "utterances": 300,
+ "errors": 101,
+ "words": 2627,
+ "wer": 0.03844689760182718
+ }
+ },
+ "punctuation": {
+ "raw": {
+ "utterances": 6,
+ "errors": 1,
+ "words": 65,
+ "wer": 0.015384615384615385
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 1,
+ "words": 65,
+ "wer": 0.015384615384615385
+ }
+ },
+ "question": {
+ "raw": {
+ "utterances": 15,
+ "errors": 2,
+ "words": 163,
+ "wer": 0.012269938650306749
+ },
+ "semantic": {
+ "utterances": 15,
+ "errors": 2,
+ "words": 165,
+ "wer": 0.012121212121212121
+ }
+ },
+ "rare_words": {
+ "raw": {
+ "utterances": 6,
+ "errors": 12,
+ "words": 57,
+ "wer": 0.21052631578947367
+ },
+ "semantic": {
+ "utterances": 6,
+ "errors": 12,
+ "words": 57,
+ "wer": 0.21052631578947367
+ }
+ },
+ "short": {
+ "raw": {
+ "utterances": 24,
+ "errors": 8,
+ "words": 98,
+ "wer": 0.08163265306122448
+ },
+ "semantic": {
+ "utterances": 24,
+ "errors": 8,
+ "words": 105,
+ "wer": 0.0761904761904762
+ }
+ },
+ "technical": {
+ "raw": {
+ "utterances": 21,
+ "errors": 26,
+ "words": 251,
+ "wer": 0.10358565737051793
+ },
+ "semantic": {
+ "utterances": 21,
+ "errors": 28,
+ "words": 256,
+ "wer": 0.109375
+ }
+ }
+ }
+ }
+ },
+ "cross_asr_mean_semantic_wer": 3.1314231136580704,
+ "cross_asr_worst_semantic_wer": 6.225214899713467
+ }
+ },
+ "rows": [
+ {
+ "id": "release500_0001",
+ "category": "conversational",
+ "text": "It wasn't until later that I realized what had actually happened.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0001.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It wasn't until later that I realized what had actually happened.",
+ "raw": {
+ "normalized_reference": "it wasn't until later that i realized what had actually happened.",
+ "normalized_hypothesis": "it wasn't until later that i realized what had actually happened.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it was not until later that i realized what had actually happened",
+ "normalized_hypothesis": "it was not until later that i realized what had actually happened",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0002",
+ "category": "descriptive",
+ "text": "The light at that time of afternoon is perfect for about twenty minutes.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0002.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The light at that time of afternoon is perfect for about 20 minutes.",
+ "raw": {
+ "normalized_reference": "the light at that time of afternoon is perfect for about twenty minutes.",
+ "normalized_hypothesis": "the light at that time of afternoon is perfect for about 20 minutes.",
+ "errors": 1,
+ "words": 13,
+ "wer": 0.07692307692307693
+ },
+ "semantic": {
+ "normalized_reference": "the light at that time of afternoon is perfect for about 20 minutes",
+ "normalized_hypothesis": "the light at that time of afternoon is perfect for about 20 minutes",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0003",
+ "category": "emotional",
+ "text": "That was the scariest thing I've ever done, and also maybe the best.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0003.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "That was the scariest thing I've ever done, and also maybe the best.",
+ "raw": {
+ "normalized_reference": "that was the scariest thing i've ever done, and also maybe the best.",
+ "normalized_hypothesis": "that was the scariest thing i've ever done, and also maybe the best.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "that was the scariest thing i have ever done and also maybe the best",
+ "normalized_hypothesis": "that was the scariest thing i have ever done and also maybe the best",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0004",
+ "category": "homographs",
+ "text": "After I record the record, please present the present to the bass player.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0004.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "After I record the record, please present the present to the bass player.",
+ "raw": {
+ "normalized_reference": "after i record the record, please present the present to the bass player.",
+ "normalized_hypothesis": "after i record the record, please present the present to the bass player.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "after i record the record please present the present to the bass player",
+ "normalized_hypothesis": "after i record the record please present the present to the bass player",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0005",
+ "category": "long",
+ "text": "If I'm being completely honest, I had no idea what I was doing for most of it, and I think that might be the most normal thing about the whole experience.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0005.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "If I'm being completely honest, I had no idea what I was doing for most of it, and I think that might be the most normal thing about the whole experience.",
+ "raw": {
+ "normalized_reference": "if i'm being completely honest, i had no idea what i was doing for most of it, and i think that might be the most normal thing about the whole experience.",
+ "normalized_hypothesis": "if i'm being completely honest, i had no idea what i was doing for most of it, and i think that might be the most normal thing about the whole experience.",
+ "errors": 0,
+ "words": 31,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "if i am being completely honest i had no idea what i was doing for most of it and i think that might be the most normal thing about the whole experience",
+ "normalized_hypothesis": "if i am being completely honest i had no idea what i was doing for most of it and i think that might be the most normal thing about the whole experience",
+ "errors": 0,
+ "words": 32,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0006",
+ "category": "long_range",
+ "text": "The committee postponed its decision after discovering that three pages of the report had been printed in the wrong order.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0006.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The committee postponed its decision after discovering that three pages of the report had been printed in the wrong order.",
+ "raw": {
+ "normalized_reference": "the committee postponed its decision after discovering that three pages of the report had been printed in the wrong order.",
+ "normalized_hypothesis": "the committee postponed its decision after discovering that three pages of the report had been printed in the wrong order.",
+ "errors": 0,
+ "words": 20,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the committee postponed its decision after discovering that 3 pages of the report had been printed in the wrong order",
+ "normalized_hypothesis": "the committee postponed its decision after discovering that 3 pages of the report had been printed in the wrong order",
+ "errors": 0,
+ "words": 20,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0007",
+ "category": "names_places",
+ "text": "Gwendolyn photographed the eucalyptus trees outside Ljubljana.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0007.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Wendelin photographed the eucalyptus trees outside Ljubljana.",
+ "raw": {
+ "normalized_reference": "gwendolyn photographed the eucalyptus trees outside ljubljana.",
+ "normalized_hypothesis": "wendelin photographed the eucalyptus trees outside ljubljana.",
+ "errors": 1,
+ "words": 7,
+ "wer": 0.14285714285714285
+ },
+ "semantic": {
+ "normalized_reference": "gwendolyn photographed the eucalyptus trees outside ljubljana",
+ "normalized_hypothesis": "wendelin photographed the eucalyptus trees outside ljubljana",
+ "errors": 1,
+ "words": 7,
+ "wer": 0.14285714285714285
+ }
+ },
+ {
+ "id": "release500_0008",
+ "category": "narrative",
+ "text": "He had this theory, and the more time passes, the more I think he was right.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0008.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He had this theory, and the more time passes, the more I think he was right.",
+ "raw": {
+ "normalized_reference": "he had this theory, and the more time passes, the more i think he was right.",
+ "normalized_hypothesis": "he had this theory, and the more time passes, the more i think he was right.",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he had this theory and the more time passes the more i think he was right",
+ "normalized_hypothesis": "he had this theory and the more time passes the more i think he was right",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0009",
+ "category": "numbers",
+ "text": "The package weighs twelve point six kilograms and arrived on July twenty-first.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0009.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The package weighs 12.6 kilograms and arrived on July 21st.",
+ "raw": {
+ "normalized_reference": "the package weighs twelve point six kilograms and arrived on july twenty-first.",
+ "normalized_hypothesis": "the package weighs 12.6 kilograms and arrived on july 21st.",
+ "errors": 5,
+ "words": 13,
+ "wer": 0.38461538461538464
+ },
+ "semantic": {
+ "normalized_reference": "the package weighs 12.6 kilograms and arrived on july 21st",
+ "normalized_hypothesis": "the package weighs 12.6 kilograms and arrived on july 21st",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0010",
+ "category": "punctuation",
+ "text": "First, close the window; second, turn off the lamp; finally, lock the door.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0010.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "First, close the window. Second, turn off the lamp. Finally, lock the door.",
+ "raw": {
+ "normalized_reference": "first, close the window; second, turn off the lamp; finally, lock the door.",
+ "normalized_hypothesis": "first, close the window. second, turn off the lamp. finally, lock the door.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "1st close the window 2nd turn off the lamp finally lock the door",
+ "normalized_hypothesis": "1st close the window 2nd turn off the lamp finally lock the door",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0011",
+ "category": "question",
+ "text": "When did you first realize something was wrong?",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0011.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "When did you first realize something was wrong?",
+ "raw": {
+ "normalized_reference": "when did you first realize something was wrong?",
+ "normalized_hypothesis": "when did you first realize something was wrong?",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "when did you 1st realize something was wrong",
+ "normalized_hypothesis": "when did you 1st realize something was wrong",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0012",
+ "category": "rare_words",
+ "text": "The archaeologist catalogued a Byzantine reliquary and several anachronistic fragments.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0012.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The archaeologist catalogued the Byzantine reliquary and several anachronistic fragments.",
+ "raw": {
+ "normalized_reference": "the archaeologist catalogued a byzantine reliquary and several anachronistic fragments.",
+ "normalized_hypothesis": "the archaeologist catalogued the byzantine reliquary and several anachronistic fragments.",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ },
+ "semantic": {
+ "normalized_reference": "the archaeologist catalogued a byzantine reliquary and several anachronistic fragments",
+ "normalized_hypothesis": "the archaeologist catalogued the byzantine reliquary and several anachronistic fragments",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ }
+ },
+ {
+ "id": "release500_0013",
+ "category": "short",
+ "text": "I have no idea.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0013.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I have no idea.",
+ "raw": {
+ "normalized_reference": "i have no idea.",
+ "normalized_hypothesis": "i have no idea.",
+ "errors": 0,
+ "words": 4,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i have no idea",
+ "normalized_hypothesis": "i have no idea",
+ "errors": 0,
+ "words": 4,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0014",
+ "category": "technical",
+ "text": "The system runs on three core components that all have to stay in sync.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0014.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The system runs on three core components that all have to stay in sync.",
+ "raw": {
+ "normalized_reference": "the system runs on three core components that all have to stay in sync.",
+ "normalized_hypothesis": "the system runs on three core components that all have to stay in sync.",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the system runs on 3 core components that all have to stay in sync",
+ "normalized_hypothesis": "the system runs on 3 core components that all have to stay in sync",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0015",
+ "category": "conversational",
+ "text": "We improvised the entire second half and somehow nobody noticed.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0015.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "We improvised the entire second half and somehow nobody noticed.",
+ "raw": {
+ "normalized_reference": "we improvised the entire second half and somehow nobody noticed.",
+ "normalized_hypothesis": "we improvised the entire second half and somehow nobody noticed.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "we improvised the entire 2nd half and somehow nobody noticed",
+ "normalized_hypothesis": "we improvised the entire 2nd half and somehow nobody noticed",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0016",
+ "category": "descriptive",
+ "text": "There's a smell in old buildings that's impossible to describe but immediately familiar.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0016.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "There's a smell in old buildings that's impossible to describe but immediately familiar.",
+ "raw": {
+ "normalized_reference": "there's a smell in old buildings that's impossible to describe but immediately familiar.",
+ "normalized_hypothesis": "there's a smell in old buildings that's impossible to describe but immediately familiar.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "there is a smell in old buildings that is impossible to describe but immediately familiar",
+ "normalized_hypothesis": "there is a smell in old buildings that is impossible to describe but immediately familiar",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0017",
+ "category": "emotional",
+ "text": "I specifically asked for one thing. One thing. And it still didn't happen.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0017.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I specifically asked for one thing, one thing, and it still didn't happen.",
+ "raw": {
+ "normalized_reference": "i specifically asked for one thing. one thing. and it still didn't happen.",
+ "normalized_hypothesis": "i specifically asked for one thing, one thing, and it still didn't happen.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i specifically asked for one thing one thing and it still did not happen",
+ "normalized_hypothesis": "i specifically asked for one thing one thing and it still did not happen",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0018",
+ "category": "homographs",
+ "text": "They refuse to place the refuse beside the produce aisle.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0018.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "They refused to place the refuse beside the produce aisle.",
+ "raw": {
+ "normalized_reference": "they refuse to place the refuse beside the produce aisle.",
+ "normalized_hypothesis": "they refused to place the refuse beside the produce aisle.",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ },
+ "semantic": {
+ "normalized_reference": "they refuse to place the refuse beside the produce aisle",
+ "normalized_hypothesis": "they refused to place the refuse beside the produce aisle",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ }
+ },
+ {
+ "id": "release500_0019",
+ "category": "long",
+ "text": "There were days last year when I genuinely didn't know how I was going to get through it, and somehow I did, and I still don't fully understand how.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0019.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "There were days last year when I genuinely didn't know how I was going to get through it, and somehow I did, and I still don't fully understand how.",
+ "raw": {
+ "normalized_reference": "there were days last year when i genuinely didn't know how i was going to get through it, and somehow i did, and i still don't fully understand how.",
+ "normalized_hypothesis": "there were days last year when i genuinely didn't know how i was going to get through it, and somehow i did, and i still don't fully understand how.",
+ "errors": 0,
+ "words": 29,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "there were days last year when i genuinely did not know how i was going to get through it and somehow i did and i still do not fully understand how",
+ "normalized_hypothesis": "there were days last year when i genuinely did not know how i was going to get through it and somehow i did and i still do not fully understand how",
+ "errors": 0,
+ "words": 31,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0020",
+ "category": "long_range",
+ "text": "Before announcing the winner, the judge thanked the volunteers who had worked through the storm to repair the damaged stage.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0020.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Before announcing the winner, the judge thanked the volunteers who had worked through the storm to repair the damaged stage.",
+ "raw": {
+ "normalized_reference": "before announcing the winner, the judge thanked the volunteers who had worked through the storm to repair the damaged stage.",
+ "normalized_hypothesis": "before announcing the winner, the judge thanked the volunteers who had worked through the storm to repair the damaged stage.",
+ "errors": 0,
+ "words": 20,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "before announcing the winner the judge thanked the volunteers who had worked through the storm to repair the damaged stage",
+ "normalized_hypothesis": "before announcing the winner the judge thanked the volunteers who had worked through the storm to repair the damaged stage",
+ "errors": 0,
+ "words": 20,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0021",
+ "category": "names_places",
+ "text": "Professor Nguyen mailed the fluorescent poster to Reykjavik on Thursday.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0021.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Professor Nguyen mailed the fluorescent poster to Reykjavik on Thursday.",
+ "raw": {
+ "normalized_reference": "professor nguyen mailed the fluorescent poster to reykjavik on thursday.",
+ "normalized_hypothesis": "professor nguyen mailed the fluorescent poster to reykjavik on thursday.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "professor nguyen mailed the fluorescent poster to reykjavik on thursday",
+ "normalized_hypothesis": "professor nguyen mailed the fluorescent poster to reykjavik on thursday",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0022",
+ "category": "narrative",
+ "text": "The plan was simple. The execution was anything but.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0022.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The plan was simple. The execution was anything but.",
+ "raw": {
+ "normalized_reference": "the plan was simple. the execution was anything but.",
+ "normalized_hypothesis": "the plan was simple. the execution was anything but.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the plan was simple the execution was anything but",
+ "normalized_hypothesis": "the plan was simple the execution was anything but",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0023",
+ "category": "numbers",
+ "text": "Her score improved from eighty-two percent to ninety-seven percent in two weeks.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0023.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Her score improved from 82% to 97% in two weeks.",
+ "raw": {
+ "normalized_reference": "her score improved from eighty-two percent to ninety-seven percent in two weeks.",
+ "normalized_hypothesis": "her score improved from 82% to 97% in two weeks.",
+ "errors": 6,
+ "words": 14,
+ "wer": 0.42857142857142855
+ },
+ "semantic": {
+ "normalized_reference": "her score improved from 82% to 97% in 2 weeks",
+ "normalized_hypothesis": "her score improved from 82% to 97% in 2 weeks",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0024",
+ "category": "punctuation",
+ "text": "Why would anyone pack spoons, batteries, and a violin for one short trip?",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0024.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Why would anyone pack spoons, batteries, and a violin for one short trip?",
+ "raw": {
+ "normalized_reference": "why would anyone pack spoons, batteries, and a violin for one short trip?",
+ "normalized_hypothesis": "why would anyone pack spoons, batteries, and a violin for one short trip?",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "why would anyone pack spoons batteries and a violin for one short trip",
+ "normalized_hypothesis": "why would anyone pack spoons batteries and a violin for one short trip",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0025",
+ "category": "question",
+ "text": "How long have you been sitting on this?",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0025.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "How long have you been sitting on this?",
+ "raw": {
+ "normalized_reference": "how long have you been sitting on this?",
+ "normalized_hypothesis": "how long have you been sitting on this?",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "how long have you been sitting on this",
+ "normalized_hypothesis": "how long have you been sitting on this",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0026",
+ "category": "rare_words",
+ "text": "A mischievous chameleon rested beside the chrysanthemum and the bougainvillea.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0026.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "A mischievous chameleon rested beside the Chrysanthemum and the Bougainvillea.",
+ "raw": {
+ "normalized_reference": "a mischievous chameleon rested beside the chrysanthemum and the bougainvillea.",
+ "normalized_hypothesis": "a mischievous chameleon rested beside the chrysanthemum and the bougainvillea.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "a mischievous chameleon rested beside the chrysanthemum and the bougainvillea",
+ "normalized_hypothesis": "a mischievous chameleon rested beside the chrysanthemum and the bougainvillea",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0027",
+ "category": "short",
+ "text": "We don't talk about that.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0027.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "We don't talk about that.",
+ "raw": {
+ "normalized_reference": "we don't talk about that.",
+ "normalized_hypothesis": "we don't talk about that.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "we do not talk about that",
+ "normalized_hypothesis": "we do not talk about that",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0028",
+ "category": "technical",
+ "text": "Hash the UTF eight payload with S H A two fifty-six before opening the checkpoint.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0028.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Cache the UTF-8 payload with SH-256 before opening the checkpoint.",
+ "raw": {
+ "normalized_reference": "hash the utf eight payload with s h a two fifty-six before opening the checkpoint.",
+ "normalized_hypothesis": "cache the utf-8 payload with sh-256 before opening the checkpoint.",
+ "errors": 8,
+ "words": 16,
+ "wer": 0.5
+ },
+ "semantic": {
+ "normalized_reference": "hash the utf 8 payload with s h a 256 before opening the checkpoint",
+ "normalized_hypothesis": "cache the utf 8 payload with sh 256 before opening the checkpoint",
+ "errors": 4,
+ "words": 14,
+ "wer": 0.2857142857142857
+ }
+ },
+ {
+ "id": "release500_0029",
+ "category": "conversational",
+ "text": "I thought I had it all figured out, and then this happened.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0029.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I thought I had it all figured out, and then this happened.",
+ "raw": {
+ "normalized_reference": "i thought i had it all figured out, and then this happened.",
+ "normalized_hypothesis": "i thought i had it all figured out, and then this happened.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i thought i had it all figured out and then this happened",
+ "normalized_hypothesis": "i thought i had it all figured out and then this happened",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0030",
+ "category": "descriptive",
+ "text": "The kind of quiet you only get at three in the morning in a city.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0030.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The kind of quiet you only get at three in the morning in a city.",
+ "raw": {
+ "normalized_reference": "the kind of quiet you only get at three in the morning in a city.",
+ "normalized_hypothesis": "the kind of quiet you only get at three in the morning in a city.",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the kind of quiet you only get at 3 in the morning in a city",
+ "normalized_hypothesis": "the kind of quiet you only get at 3 in the morning in a city",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0031",
+ "category": "emotional",
+ "text": "The joy of it was almost overwhelming, like too much good all at once.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0031.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The joy of it was almost overwhelming, like too much good all at once.",
+ "raw": {
+ "normalized_reference": "the joy of it was almost overwhelming, like too much good all at once.",
+ "normalized_hypothesis": "the joy of it was almost overwhelming, like too much good all at once.",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the joy of it was almost overwhelming like too much good all at once",
+ "normalized_hypothesis": "the joy of it was almost overwhelming like too much good all at once",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0032",
+ "category": "homographs",
+ "text": "Please close the door while I stand close to the microphone.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0032.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Please close the door while I stand close to the microphone.",
+ "raw": {
+ "normalized_reference": "please close the door while i stand close to the microphone.",
+ "normalized_hypothesis": "please close the door while i stand close to the microphone.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "please close the door while i stand close to the microphone",
+ "normalized_hypothesis": "please close the door while i stand close to the microphone",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0033",
+ "category": "long",
+ "text": "I think the most underrated skill in the world is knowing when to stop explaining yourself and just let people draw their own conclusions.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0033.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I think the most underrated skill in the world is knowing when to stop explaining yourself and just let people draw their own conclusions.",
+ "raw": {
+ "normalized_reference": "i think the most underrated skill in the world is knowing when to stop explaining yourself and just let people draw their own conclusions.",
+ "normalized_hypothesis": "i think the most underrated skill in the world is knowing when to stop explaining yourself and just let people draw their own conclusions.",
+ "errors": 0,
+ "words": 24,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i think the most underrated skill in the world is knowing when to stop explaining yourself and just let people draw their own conclusions",
+ "normalized_hypothesis": "i think the most underrated skill in the world is knowing when to stop explaining yourself and just let people draw their own conclusions",
+ "errors": 0,
+ "words": 24,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0034",
+ "category": "long_range",
+ "text": "When the final bus disappeared around the corner, Nadia remembered that her umbrella, notebook, and lunch were still inside the quiet office.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0034.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "When the final bus disappeared around the corner, Nadia remembered that her umbrella, notebook, and lunch were still inside the quiet office.",
+ "raw": {
+ "normalized_reference": "when the final bus disappeared around the corner, nadia remembered that her umbrella, notebook, and lunch were still inside the quiet office.",
+ "normalized_hypothesis": "when the final bus disappeared around the corner, nadia remembered that her umbrella, notebook, and lunch were still inside the quiet office.",
+ "errors": 0,
+ "words": 22,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "when the final bus disappeared around the corner nadia remembered that her umbrella notebook and lunch were still inside the quiet office",
+ "normalized_hypothesis": "when the final bus disappeared around the corner nadia remembered that her umbrella notebook and lunch were still inside the quiet office",
+ "errors": 0,
+ "words": 22,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0035",
+ "category": "names_places",
+ "text": "Aurelia and Joaquin crossed the Schuylkill River before noon.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0035.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Aurelia and Joaquin crossed the Shilka River before noon.",
+ "raw": {
+ "normalized_reference": "aurelia and joaquin crossed the schuylkill river before noon.",
+ "normalized_hypothesis": "aurelia and joaquin crossed the shilka river before noon.",
+ "errors": 1,
+ "words": 9,
+ "wer": 0.1111111111111111
+ },
+ "semantic": {
+ "normalized_reference": "aurelia and joaquin crossed the schuylkill river before noon",
+ "normalized_hypothesis": "aurelia and joaquin crossed the shilka river before noon",
+ "errors": 1,
+ "words": 9,
+ "wer": 0.1111111111111111
+ }
+ },
+ {
+ "id": "release500_0036",
+ "category": "narrative",
+ "text": "The village had a name, but nobody used it. Everyone just called it home.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0036.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The village had a name, but nobody used it. Everyone just called it home.",
+ "raw": {
+ "normalized_reference": "the village had a name, but nobody used it. everyone just called it home.",
+ "normalized_hypothesis": "the village had a name, but nobody used it. everyone just called it home.",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the village had a name but nobody used it everyone just called it home",
+ "normalized_hypothesis": "the village had a name but nobody used it everyone just called it home",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0037",
+ "category": "numbers",
+ "text": "Increase the gain by three decibels, wait five seconds, then reduce it by one.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0037.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Increase the gain by 3 decibels, wait 5 seconds, then reduce it by 1.",
+ "raw": {
+ "normalized_reference": "increase the gain by three decibels, wait five seconds, then reduce it by one.",
+ "normalized_hypothesis": "increase the gain by 3 decibels, wait 5 seconds, then reduce it by 1.",
+ "errors": 3,
+ "words": 14,
+ "wer": 0.21428571428571427
+ },
+ "semantic": {
+ "normalized_reference": "increase the gain by 3 decibels wait 5 seconds then reduce it by one",
+ "normalized_hypothesis": "increase the gain by 3 decibels wait 5 seconds then reduce it by one",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0038",
+ "category": "punctuation",
+ "text": "The result was unexpected: clear, stable, and surprisingly natural.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0038.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The result was unexpected, clear, stable, and surprisingly natural.",
+ "raw": {
+ "normalized_reference": "the result was unexpected: clear, stable, and surprisingly natural.",
+ "normalized_hypothesis": "the result was unexpected, clear, stable, and surprisingly natural.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the result was unexpected clear stable and surprisingly natural",
+ "normalized_hypothesis": "the result was unexpected clear stable and surprisingly natural",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0039",
+ "category": "question",
+ "text": "Who told you that was the right way to do it?",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0039.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Who told you that was the right way to do it?",
+ "raw": {
+ "normalized_reference": "who told you that was the right way to do it?",
+ "normalized_hypothesis": "who told you that was the right way to do it?",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "who told you that was the right way to do it",
+ "normalized_hypothesis": "who told you that was the right way to do it",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0040",
+ "category": "rare_words",
+ "text": "The meticulous cartographer examined an iridescent archipelago through binoculars.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0040.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The meticulous cartographer examined an iridescent archipelago through binoculars.",
+ "raw": {
+ "normalized_reference": "the meticulous cartographer examined an iridescent archipelago through binoculars.",
+ "normalized_hypothesis": "the meticulous cartographer examined an iridescent archipelago through binoculars.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the meticulous cartographer examined an iridescent archipelago through binoculars",
+ "normalized_hypothesis": "the meticulous cartographer examined an iridescent archipelago through binoculars",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0041",
+ "category": "short",
+ "text": "I mean. Yeah. Kind of.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0041.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I mean, yeah, kind of.",
+ "raw": {
+ "normalized_reference": "i mean. yeah. kind of.",
+ "normalized_hypothesis": "i mean, yeah, kind of.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i mean yeah kind of",
+ "normalized_hypothesis": "i mean yeah kind of",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0042",
+ "category": "technical",
+ "text": "The logs are your best friend when something behaves unexpectedly.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0042.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The logs are your best friend when something behaves unexpectedly.",
+ "raw": {
+ "normalized_reference": "the logs are your best friend when something behaves unexpectedly.",
+ "normalized_hypothesis": "the logs are your best friend when something behaves unexpectedly.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the logs are your best friend when something behaves unexpectedly",
+ "normalized_hypothesis": "the logs are your best friend when something behaves unexpectedly",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0043",
+ "category": "conversational",
+ "text": "He kept talking and I kept nodding and neither of us was listening.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0043.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He kept talking, and I kept nodding, and neither of us was listening.",
+ "raw": {
+ "normalized_reference": "he kept talking and i kept nodding and neither of us was listening.",
+ "normalized_hypothesis": "he kept talking, and i kept nodding, and neither of us was listening.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he kept talking and i kept nodding and neither of us was listening",
+ "normalized_hypothesis": "he kept talking and i kept nodding and neither of us was listening",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0044",
+ "category": "descriptive",
+ "text": "It's one of those places where everyone seems to know each other already.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0044.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It's one of those places where everyone seems to know each other already.",
+ "raw": {
+ "normalized_reference": "it's one of those places where everyone seems to know each other already.",
+ "normalized_hypothesis": "it's one of those places where everyone seems to know each other already.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it is one of those places where everyone seems to know each other already",
+ "normalized_hypothesis": "it is one of those places where everyone seems to know each other already",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0045",
+ "category": "emotional",
+ "text": "He said he was proud of me and nobody had ever said that to me before.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0045.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He said he was proud of me and nobody had ever said that to me before.",
+ "raw": {
+ "normalized_reference": "he said he was proud of me and nobody had ever said that to me before.",
+ "normalized_hypothesis": "he said he was proud of me and nobody had ever said that to me before.",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he said he was proud of me and nobody had ever said that to me before",
+ "normalized_hypothesis": "he said he was proud of me and nobody had ever said that to me before",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0046",
+ "category": "homographs",
+ "text": "The minute detail took a minute to explain clearly.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0046.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The minute detail took a minute to explain clearly.",
+ "raw": {
+ "normalized_reference": "the minute detail took a minute to explain clearly.",
+ "normalized_hypothesis": "the minute detail took a minute to explain clearly.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the minute detail took a minute to explain clearly",
+ "normalized_hypothesis": "the minute detail took a minute to explain clearly",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0047",
+ "category": "long",
+ "text": "The strange thing about really wanting something is that getting it never quite feels the way you imagined it would, and not getting it never quite feels as bad.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0047.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The strange thing about really wanting something is that getting it never quite feels the way you imagined it would, and not getting it never quite feels as bad.",
+ "raw": {
+ "normalized_reference": "the strange thing about really wanting something is that getting it never quite feels the way you imagined it would, and not getting it never quite feels as bad.",
+ "normalized_hypothesis": "the strange thing about really wanting something is that getting it never quite feels the way you imagined it would, and not getting it never quite feels as bad.",
+ "errors": 0,
+ "words": 29,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the strange thing about really wanting something is that getting it never quite feels the way you imagined it would and not getting it never quite feels as bad",
+ "normalized_hypothesis": "the strange thing about really wanting something is that getting it never quite feels the way you imagined it would and not getting it never quite feels as bad",
+ "errors": 0,
+ "words": 29,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0048",
+ "category": "long_range",
+ "text": "Even though the recipe looked simple at first, preparing the sauce required patience, careful timing, and a much larger pan.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0048.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Even though the recipe looked simple at first, preparing the sauce required patience, careful timing, and a much larger pan.",
+ "raw": {
+ "normalized_reference": "even though the recipe looked simple at first, preparing the sauce required patience, careful timing, and a much larger pan.",
+ "normalized_hypothesis": "even though the recipe looked simple at first, preparing the sauce required patience, careful timing, and a much larger pan.",
+ "errors": 0,
+ "words": 20,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "even though the recipe looked simple at 1st preparing the sauce required patience careful timing and a much larger pan",
+ "normalized_hypothesis": "even though the recipe looked simple at 1st preparing the sauce required patience careful timing and a much larger pan",
+ "errors": 0,
+ "words": 20,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0049",
+ "category": "names_places",
+ "text": "Xiomara met Siobhan near the old library in Worcestershire.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0049.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Zamara met Siobhan near the old library in Worcestershire.",
+ "raw": {
+ "normalized_reference": "xiomara met siobhan near the old library in worcestershire.",
+ "normalized_hypothesis": "zamara met siobhan near the old library in worcestershire.",
+ "errors": 1,
+ "words": 9,
+ "wer": 0.1111111111111111
+ },
+ "semantic": {
+ "normalized_reference": "xiomara met siobhan near the old library in worcestershire",
+ "normalized_hypothesis": "zamara met siobhan near the old library in worcestershire",
+ "errors": 1,
+ "words": 9,
+ "wer": 0.1111111111111111
+ }
+ },
+ {
+ "id": "release500_0050",
+ "category": "narrative",
+ "text": "She walked into the room and everyone went quiet, not because she was loud, but because she wasn't.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0050.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "She walked into the room and everyone went quiet. Not because she was loud, but because she wasn't.",
+ "raw": {
+ "normalized_reference": "she walked into the room and everyone went quiet, not because she was loud, but because she wasn't.",
+ "normalized_hypothesis": "she walked into the room and everyone went quiet. not because she was loud, but because she wasn't.",
+ "errors": 0,
+ "words": 18,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "she walked into the room and everyone went quiet not because she was loud but because she was not",
+ "normalized_hypothesis": "she walked into the room and everyone went quiet not because she was loud but because she was not",
+ "errors": 0,
+ "words": 19,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0051",
+ "category": "numbers",
+ "text": "At exactly seven forty-five, Morgan placed eighteen dollars and fifty cents beside the notebook.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0051.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "At exactly $7.45, Morton placed $18.50 beside the notebook.",
+ "raw": {
+ "normalized_reference": "at exactly seven forty-five, morgan placed eighteen dollars and fifty cents beside the notebook.",
+ "normalized_hypothesis": "at exactly $7.45, morton placed $18.50 beside the notebook.",
+ "errors": 9,
+ "words": 15,
+ "wer": 0.6
+ },
+ "semantic": {
+ "normalized_reference": "at exactly 745 morgan placed $18.50 beside the notebook",
+ "normalized_hypothesis": "at exactly $7.45 morton placed $18.50 beside the notebook",
+ "errors": 3,
+ "words": 10,
+ "wer": 0.3
+ }
+ },
+ {
+ "id": "release500_0052",
+ "category": "punctuation",
+ "text": "Does punctuation work? Yes: commas, semicolons, and periods should create sensible boundaries.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0052.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Does punctuation work? Yes. Commas, semicolons, and periods should create sensible boundaries.",
+ "raw": {
+ "normalized_reference": "does punctuation work? yes: commas, semicolons, and periods should create sensible boundaries.",
+ "normalized_hypothesis": "does punctuation work? yes. commas, semicolons, and periods should create sensible boundaries.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "does punctuation work yes commas semicolons and periods should create sensible boundaries",
+ "normalized_hypothesis": "does punctuation work yes commas semicolons and periods should create sensible boundaries",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0053",
+ "category": "question",
+ "text": "Have you ever just stopped and thought about how strange all of this is?",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0053.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Have you ever just stopped and thought about how strange all of this is?",
+ "raw": {
+ "normalized_reference": "have you ever just stopped and thought about how strange all of this is?",
+ "normalized_hypothesis": "have you ever just stopped and thought about how strange all of this is?",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "have you ever just stopped and thought about how strange all of this is",
+ "normalized_hypothesis": "have you ever just stopped and thought about how strange all of this is",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0054",
+ "category": "rare_words",
+ "text": "Her idiosyncratic explanation sounded plausible but remained fundamentally ambiguous.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0054.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Her idiosyncratic explanation sounded plausible but remained fundamentally ambiguous.",
+ "raw": {
+ "normalized_reference": "her idiosyncratic explanation sounded plausible but remained fundamentally ambiguous.",
+ "normalized_hypothesis": "her idiosyncratic explanation sounded plausible but remained fundamentally ambiguous.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "her idiosyncratic explanation sounded plausible but remained fundamentally ambiguous",
+ "normalized_hypothesis": "her idiosyncratic explanation sounded plausible but remained fundamentally ambiguous",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0055",
+ "category": "short",
+ "text": "That tracks.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0055.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "That tracks.",
+ "raw": {
+ "normalized_reference": "that tracks.",
+ "normalized_hypothesis": "that tracks.",
+ "errors": 0,
+ "words": 2,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "that tracks",
+ "normalized_hypothesis": "that tracks",
+ "errors": 0,
+ "words": 2,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0056",
+ "category": "technical",
+ "text": "The key thing to remember is that order matters here more than you'd think.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0056.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The key thing to remember is that order matters here more than you'd think.",
+ "raw": {
+ "normalized_reference": "the key thing to remember is that order matters here more than you'd think.",
+ "normalized_hypothesis": "the key thing to remember is that order matters here more than you'd think.",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the key thing to remember is that order matters here more than you would think",
+ "normalized_hypothesis": "the key thing to remember is that order matters here more than you would think",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0057",
+ "category": "conversational",
+ "text": "I would not say that again in this room if I were you.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0057.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I would not say that again in this room if I were you.",
+ "raw": {
+ "normalized_reference": "i would not say that again in this room if i were you.",
+ "normalized_hypothesis": "i would not say that again in this room if i were you.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i would not say that again in this room if i were you",
+ "normalized_hypothesis": "i would not say that again in this room if i were you",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0058",
+ "category": "descriptive",
+ "text": "The room was filled with the quiet that means something is about to change.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0058.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The room was filled with the quiet that means something is about to change.",
+ "raw": {
+ "normalized_reference": "the room was filled with the quiet that means something is about to change.",
+ "normalized_hypothesis": "the room was filled with the quiet that means something is about to change.",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the room was filled with the quiet that means something is about to change",
+ "normalized_hypothesis": "the room was filled with the quiet that means something is about to change",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0059",
+ "category": "emotional",
+ "text": "I didn't expect to feel that way about it, but I did.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0059.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I didn't expect to feel that way about it, but I did.",
+ "raw": {
+ "normalized_reference": "i didn't expect to feel that way about it, but i did.",
+ "normalized_hypothesis": "i didn't expect to feel that way about it, but i did.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i did not expect to feel that way about it but i did",
+ "normalized_hypothesis": "i did not expect to feel that way about it but i did",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0060",
+ "category": "homographs",
+ "text": "I object to the strange object that was left near the entrance.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0060.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I object to the strange object that was left near the entrance.",
+ "raw": {
+ "normalized_reference": "i object to the strange object that was left near the entrance.",
+ "normalized_hypothesis": "i object to the strange object that was left near the entrance.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i object to the strange object that was left near the entrance",
+ "normalized_hypothesis": "i object to the strange object that was left near the entrance",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0061",
+ "category": "long",
+ "text": "I've learned more from the things that went wrong than from the things that worked, which is a frustrating way to have to learn things but apparently the only way that stuck.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0061.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I've learned more from the things that went wrong than from the things that worked, which is a frustrating way to have to learn things but apparently the only way that's stuck.",
+ "raw": {
+ "normalized_reference": "i've learned more from the things that went wrong than from the things that worked, which is a frustrating way to have to learn things but apparently the only way that stuck.",
+ "normalized_hypothesis": "i've learned more from the things that went wrong than from the things that worked, which is a frustrating way to have to learn things but apparently the only way that's stuck.",
+ "errors": 1,
+ "words": 32,
+ "wer": 0.03125
+ },
+ "semantic": {
+ "normalized_reference": "i have learned more from the things that went wrong than from the things that worked which is a frustrating way to have to learn things but apparently the only way that stuck",
+ "normalized_hypothesis": "i have learned more from the things that went wrong than from the things that worked which is a frustrating way to have to learn things but apparently the only way that is stuck",
+ "errors": 1,
+ "words": 33,
+ "wer": 0.030303030303030304
+ }
+ },
+ {
+ "id": "release500_0062",
+ "category": "long_range",
+ "text": "Because every nearby hotel was full, the exhausted hikers followed the narrow road until they found a farmhouse with one empty room.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0062.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Because every nearby hotel was full, the exhausted hikers followed the narrow road until they found a farmhouse with one empty room.",
+ "raw": {
+ "normalized_reference": "because every nearby hotel was full, the exhausted hikers followed the narrow road until they found a farmhouse with one empty room.",
+ "normalized_hypothesis": "because every nearby hotel was full, the exhausted hikers followed the narrow road until they found a farmhouse with one empty room.",
+ "errors": 0,
+ "words": 22,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "because every nearby hotel was full the exhausted hikers followed the narrow road until they found a farmhouse with one empty room",
+ "normalized_hypothesis": "because every nearby hotel was full the exhausted hikers followed the narrow road until they found a farmhouse with one empty room",
+ "errors": 0,
+ "words": 22,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0063",
+ "category": "names_places",
+ "text": "The itinerary sends Beatrice from Albuquerque to Des Moines and then to Boise.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0063.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The itinerary sends Beatrice from Albuquerque to Des Moines and then to Boise.",
+ "raw": {
+ "normalized_reference": "the itinerary sends beatrice from albuquerque to des moines and then to boise.",
+ "normalized_hypothesis": "the itinerary sends beatrice from albuquerque to des moines and then to boise.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the itinerary sends beatrice from albuquerque to des moines and then to boise",
+ "normalized_hypothesis": "the itinerary sends beatrice from albuquerque to des moines and then to boise",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0064",
+ "category": "narrative",
+ "text": "We had been driving for six hours when we finally saw the sign for the town.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0064.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "We had been driving for six hours when we finally saw the sign for the town.",
+ "raw": {
+ "normalized_reference": "we had been driving for six hours when we finally saw the sign for the town.",
+ "normalized_hypothesis": "we had been driving for six hours when we finally saw the sign for the town.",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "we had been driving for 6 hours when we finally saw the sign for the town",
+ "normalized_hypothesis": "we had been driving for 6 hours when we finally saw the sign for the town",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0065",
+ "category": "numbers",
+ "text": "Call extension zero four nine after eight thirty in the morning.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0065.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Call extension 049 after 8.30 in the morning.",
+ "raw": {
+ "normalized_reference": "call extension zero four nine after eight thirty in the morning.",
+ "normalized_hypothesis": "call extension 049 after 8.30 in the morning.",
+ "errors": 5,
+ "words": 11,
+ "wer": 0.45454545454545453
+ },
+ "semantic": {
+ "normalized_reference": "call extension 049 after 830 in the morning",
+ "normalized_hypothesis": "call extension 49 after 8.30 in the morning",
+ "errors": 3,
+ "words": 8,
+ "wer": 0.375
+ }
+ },
+ {
+ "id": "release500_0066",
+ "category": "punctuation",
+ "text": "Stop - listen carefully - then answer the question without guessing.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0066.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Stop, listen carefully, then answer the question without guessing.",
+ "raw": {
+ "normalized_reference": "stop - listen carefully - then answer the question without guessing.",
+ "normalized_hypothesis": "stop, listen carefully, then answer the question without guessing.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "stop listen carefully then answer the question without guessing",
+ "normalized_hypothesis": "stop listen carefully then answer the question without guessing",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0067",
+ "category": "question",
+ "text": "Does it matter at this point, honestly?",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0067.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Does it matter at this point, honestly?",
+ "raw": {
+ "normalized_reference": "does it matter at this point, honestly?",
+ "normalized_hypothesis": "does it matter at this point, honestly?",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "does it matter at this point honestly",
+ "normalized_hypothesis": "does it matter at this point honestly",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0068",
+ "category": "rare_words",
+ "text": "An entrepreneurial meteorologist described the unprecedented atmospheric phenomenon.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0068.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "An entrepreneurial meteorologist described the unprecedented atmospheric phenomenon.",
+ "raw": {
+ "normalized_reference": "an entrepreneurial meteorologist described the unprecedented atmospheric phenomenon.",
+ "normalized_hypothesis": "an entrepreneurial meteorologist described the unprecedented atmospheric phenomenon.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "an entrepreneurial meteorologist described the unprecedented atmospheric phenomenon",
+ "normalized_hypothesis": "an entrepreneurial meteorologist described the unprecedented atmospheric phenomenon",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0069",
+ "category": "short",
+ "text": "That's not what I expected.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0069.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "That's not what I expected.",
+ "raw": {
+ "normalized_reference": "that's not what i expected.",
+ "normalized_hypothesis": "that's not what i expected.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "that is not what i expected",
+ "normalized_hypothesis": "that is not what i expected",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0070",
+ "category": "technical",
+ "text": "There's a known issue with that version that causes failures under load.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0070.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "There's a known issue with that version that causes failures under load.",
+ "raw": {
+ "normalized_reference": "there's a known issue with that version that causes failures under load.",
+ "normalized_hypothesis": "there's a known issue with that version that causes failures under load.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "there is a known issue with that version that causes failures under load",
+ "normalized_hypothesis": "there is a known issue with that version that causes failures under load",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0071",
+ "category": "conversational",
+ "text": "Don't look now but I think that's him over there.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0071.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Don't look now, but I think that's him over there.",
+ "raw": {
+ "normalized_reference": "don't look now but i think that's him over there.",
+ "normalized_hypothesis": "don't look now, but i think that's him over there.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "do not look now but i think that is him over there",
+ "normalized_hypothesis": "do not look now but i think that is him over there",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0072",
+ "category": "descriptive",
+ "text": "The city at night looks completely different from how it feels during the day.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0072.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The city at night looks completely different from how it feels during the day.",
+ "raw": {
+ "normalized_reference": "the city at night looks completely different from how it feels during the day.",
+ "normalized_hypothesis": "the city at night looks completely different from how it feels during the day.",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the city at night looks completely different from how it feels during the day",
+ "normalized_hypothesis": "the city at night looks completely different from how it feels during the day",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0073",
+ "category": "emotional",
+ "text": "It's the kind of happy that makes you nervous because you know it can't last.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0073.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It's the kind of happy that makes you nervous because you know it can't last.",
+ "raw": {
+ "normalized_reference": "it's the kind of happy that makes you nervous because you know it can't last.",
+ "normalized_hypothesis": "it's the kind of happy that makes you nervous because you know it can't last.",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it is the kind of happy that makes you nervous because you know it can not last",
+ "normalized_hypothesis": "it is the kind of happy that makes you nervous because you know it can not last",
+ "errors": 0,
+ "words": 17,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0074",
+ "category": "homographs",
+ "text": "The dove dove beneath the branch while the wind began to wind the loose cable.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0074.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The dove dove beneath the branch wall the wind began to wind the loose cable.",
+ "raw": {
+ "normalized_reference": "the dove dove beneath the branch while the wind began to wind the loose cable.",
+ "normalized_hypothesis": "the dove dove beneath the branch wall the wind began to wind the loose cable.",
+ "errors": 1,
+ "words": 15,
+ "wer": 0.06666666666666667
+ },
+ "semantic": {
+ "normalized_reference": "the dove dove beneath the branch while the wind began to wind the loose cable",
+ "normalized_hypothesis": "the dove dove beneath the branch wall the wind began to wind the loose cable",
+ "errors": 1,
+ "words": 15,
+ "wer": 0.06666666666666667
+ }
+ },
+ {
+ "id": "release500_0075",
+ "category": "long",
+ "text": "I've thought a lot about what I would do differently if I could go back, and the honest answer is probably not as much as I'd like to think.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0075.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I've thought a lot about what I would do differently if I could go back, and the honest answer is probably not as much as I'd like to think.",
+ "raw": {
+ "normalized_reference": "i've thought a lot about what i would do differently if i could go back, and the honest answer is probably not as much as i'd like to think.",
+ "normalized_hypothesis": "i've thought a lot about what i would do differently if i could go back, and the honest answer is probably not as much as i'd like to think.",
+ "errors": 0,
+ "words": 29,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i have thought a lot about what i would do differently if i could go back and the honest answer is probably not as much as i would like to think",
+ "normalized_hypothesis": "i have thought a lot about what i would do differently if i could go back and the honest answer is probably not as much as i would like to think",
+ "errors": 0,
+ "words": 31,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0076",
+ "category": "long_range",
+ "text": "Although the weather changed without warning, the musicians finished packing before anyone opened the enormous glass door.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0076.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Although the weather changed without warning, the musicians finished packing before anyone opened the enormous glass door.",
+ "raw": {
+ "normalized_reference": "although the weather changed without warning, the musicians finished packing before anyone opened the enormous glass door.",
+ "normalized_hypothesis": "although the weather changed without warning, the musicians finished packing before anyone opened the enormous glass door.",
+ "errors": 0,
+ "words": 17,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "although the weather changed without warning the musicians finished packing before anyone opened the enormous glass door",
+ "normalized_hypothesis": "although the weather changed without warning the musicians finished packing before anyone opened the enormous glass door",
+ "errors": 0,
+ "words": 17,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0077",
+ "category": "names_places",
+ "text": "Saskatchewan looked unusually bright beneath the pale February sunrise.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0077.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Saskatchewan looked unusually bright beneath that pale February sunrise.",
+ "raw": {
+ "normalized_reference": "saskatchewan looked unusually bright beneath the pale february sunrise.",
+ "normalized_hypothesis": "saskatchewan looked unusually bright beneath that pale february sunrise.",
+ "errors": 1,
+ "words": 9,
+ "wer": 0.1111111111111111
+ },
+ "semantic": {
+ "normalized_reference": "saskatchewan looked unusually bright beneath the pale february sunrise",
+ "normalized_hypothesis": "saskatchewan looked unusually bright beneath that pale february sunrise",
+ "errors": 1,
+ "words": 9,
+ "wer": 0.1111111111111111
+ }
+ },
+ {
+ "id": "release500_0078",
+ "category": "narrative",
+ "text": "She handled it with a grace I genuinely did not expect.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0078.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "She handled it with a grace I genuinely did not expect.",
+ "raw": {
+ "normalized_reference": "she handled it with a grace i genuinely did not expect.",
+ "normalized_hypothesis": "she handled it with a grace i genuinely did not expect.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "she handled it with a grace i genuinely did not expect",
+ "normalized_hypothesis": "she handled it with a grace i genuinely did not expect",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0079",
+ "category": "numbers",
+ "text": "Apartment four hundred seven is across from room twenty-nine on the third floor.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0079.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Apartment 4-7 is across from room 29 on the third floor.",
+ "raw": {
+ "normalized_reference": "apartment four hundred seven is across from room twenty-nine on the third floor.",
+ "normalized_hypothesis": "apartment 4-7 is across from room 29 on the third floor.",
+ "errors": 5,
+ "words": 14,
+ "wer": 0.35714285714285715
+ },
+ "semantic": {
+ "normalized_reference": "apartment 407 is across from room 29 on the 3rd floor",
+ "normalized_hypothesis": "apartment 4 7 is across from room 29 on the 3rd floor",
+ "errors": 2,
+ "words": 11,
+ "wer": 0.18181818181818182
+ }
+ },
+ {
+ "id": "release500_0080",
+ "category": "punctuation",
+ "text": "\"Not yet,\" Rowan whispered. \"Give me another minute... please.\"",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0080.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Not yet, Rowan whispered. Give me another minute, please.",
+ "raw": {
+ "normalized_reference": "\"not yet,\" rowan whispered. \"give me another minute... please.\"",
+ "normalized_hypothesis": "not yet, rowan whispered. give me another minute, please.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "not yet rowan whispered give me another minute please",
+ "normalized_hypothesis": "not yet rowan whispered give me another minute please",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0081",
+ "category": "question",
+ "text": "What are we actually trying to accomplish here?",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0081.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "What are we actually trying to accomplish here?",
+ "raw": {
+ "normalized_reference": "what are we actually trying to accomplish here?",
+ "normalized_hypothesis": "what are we actually trying to accomplish here?",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "what are we actually trying to accomplish here",
+ "normalized_hypothesis": "what are we actually trying to accomplish here",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0082",
+ "category": "rare_words",
+ "text": "The physician recommended a noninvasive procedure for the recurring neurological symptoms.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0082.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The physician recommended a non-invasive procedure for the recurring neurological symptoms.",
+ "raw": {
+ "normalized_reference": "the physician recommended a noninvasive procedure for the recurring neurological symptoms.",
+ "normalized_hypothesis": "the physician recommended a non-invasive procedure for the recurring neurological symptoms.",
+ "errors": 2,
+ "words": 11,
+ "wer": 0.18181818181818182
+ },
+ "semantic": {
+ "normalized_reference": "the physician recommended a noninvasive procedure for the recurring neurological symptoms",
+ "normalized_hypothesis": "the physician recommended a non invasive procedure for the recurring neurological symptoms",
+ "errors": 2,
+ "words": 11,
+ "wer": 0.18181818181818182
+ }
+ },
+ {
+ "id": "release500_0083",
+ "category": "short",
+ "text": "Are you serious right now?",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0083.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Are you serious right now?",
+ "raw": {
+ "normalized_reference": "are you serious right now?",
+ "normalized_hypothesis": "are you serious right now?",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "are you serious right now",
+ "normalized_hypothesis": "are you serious right now",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0084",
+ "category": "technical",
+ "text": "The performance improvement is real, but it comes with tradeoffs.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0084.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The performance improvement is real, but it comes with trade-offs.",
+ "raw": {
+ "normalized_reference": "the performance improvement is real, but it comes with tradeoffs.",
+ "normalized_hypothesis": "the performance improvement is real, but it comes with trade-offs.",
+ "errors": 2,
+ "words": 10,
+ "wer": 0.2
+ },
+ "semantic": {
+ "normalized_reference": "the performance improvement is real but it comes with tradeoffs",
+ "normalized_hypothesis": "the performance improvement is real but it comes with trade offs",
+ "errors": 2,
+ "words": 10,
+ "wer": 0.2
+ }
+ },
+ {
+ "id": "release500_0085",
+ "category": "conversational",
+ "text": "I don't know. It's just been one of those weeks, you know?",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0085.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I don't know. It's just been one of those weeks, you know?",
+ "raw": {
+ "normalized_reference": "i don't know. it's just been one of those weeks, you know?",
+ "normalized_hypothesis": "i don't know. it's just been one of those weeks, you know?",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i do not know it is just been one of those weeks you know",
+ "normalized_hypothesis": "i do not know it is just been one of those weeks you know",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0086",
+ "category": "descriptive",
+ "text": "The moment you stop expecting perfection, things get a lot more interesting.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0086.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The moment you stop expecting perfection, things get a lot more interesting.",
+ "raw": {
+ "normalized_reference": "the moment you stop expecting perfection, things get a lot more interesting.",
+ "normalized_hypothesis": "the moment you stop expecting perfection, things get a lot more interesting.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the moment you stop expecting perfection things get a lot more interesting",
+ "normalized_hypothesis": "the moment you stop expecting perfection things get a lot more interesting",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0087",
+ "category": "emotional",
+ "text": "It's the kind of thing that stays with you even years later.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0087.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It's the kind of thing that stays with you even years later.",
+ "raw": {
+ "normalized_reference": "it's the kind of thing that stays with you even years later.",
+ "normalized_hypothesis": "it's the kind of thing that stays with you even years later.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it is the kind of thing that stays with you even years later",
+ "normalized_hypothesis": "it is the kind of thing that stays with you even years later",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0088",
+ "category": "long",
+ "text": "She told me once that the most important skill she'd developed was learning to disagree with someone without making them feel they were wrong.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0088.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "She told me once that the most important skill she'd developed was learning to disagree with someone without making them feel they were wrong.",
+ "raw": {
+ "normalized_reference": "she told me once that the most important skill she'd developed was learning to disagree with someone without making them feel they were wrong.",
+ "normalized_hypothesis": "she told me once that the most important skill she'd developed was learning to disagree with someone without making them feel they were wrong.",
+ "errors": 0,
+ "words": 24,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "she told me once that the most important skill she would developed was learning to disagree with someone without making them feel they were wrong",
+ "normalized_hypothesis": "she told me once that the most important skill she would developed was learning to disagree with someone without making them feel they were wrong",
+ "errors": 0,
+ "words": 25,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0089",
+ "category": "narrative",
+ "text": "He stood at the edge of the cliff and wondered, not for the first time, how he'd gotten here.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0089.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He stood at the edge of the cliff and wondered, not for the first time, how he'd gotten here.",
+ "raw": {
+ "normalized_reference": "he stood at the edge of the cliff and wondered, not for the first time, how he'd gotten here.",
+ "normalized_hypothesis": "he stood at the edge of the cliff and wondered, not for the first time, how he'd gotten here.",
+ "errors": 0,
+ "words": 19,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he stood at the edge of the cliff and wondered not for the 1st time how he would gotten here",
+ "normalized_hypothesis": "he stood at the edge of the cliff and wondered not for the 1st time how he would gotten here",
+ "errors": 0,
+ "words": 20,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0090",
+ "category": "question",
+ "text": "Why does asking for help feel so hard when you actually need it?",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0090.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Why does asking for help feel so hard when you actually need it?",
+ "raw": {
+ "normalized_reference": "why does asking for help feel so hard when you actually need it?",
+ "normalized_hypothesis": "why does asking for help feel so hard when you actually need it?",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "why does asking for help feel so hard when you actually need it",
+ "normalized_hypothesis": "why does asking for help feel so hard when you actually need it",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0091",
+ "category": "short",
+ "text": "That's fair.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0091.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "That's fair.",
+ "raw": {
+ "normalized_reference": "that's fair.",
+ "normalized_hypothesis": "that's fair.",
+ "errors": 0,
+ "words": 2,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "that is fair",
+ "normalized_hypothesis": "that is fair",
+ "errors": 0,
+ "words": 3,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0092",
+ "category": "technical",
+ "text": "If you run into that error, the fix is usually just restarting the service.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0092.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "If you run into that error, the fix is usually just restarting this service.",
+ "raw": {
+ "normalized_reference": "if you run into that error, the fix is usually just restarting the service.",
+ "normalized_hypothesis": "if you run into that error, the fix is usually just restarting this service.",
+ "errors": 1,
+ "words": 14,
+ "wer": 0.07142857142857142
+ },
+ "semantic": {
+ "normalized_reference": "if you run into that error the fix is usually just restarting the service",
+ "normalized_hypothesis": "if you run into that error the fix is usually just restarting this service",
+ "errors": 1,
+ "words": 14,
+ "wer": 0.07142857142857142
+ }
+ },
+ {
+ "id": "release500_0093",
+ "category": "conversational",
+ "text": "It was a disaster in the best possible way.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0093.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It was a disaster in the best possible way.",
+ "raw": {
+ "normalized_reference": "it was a disaster in the best possible way.",
+ "normalized_hypothesis": "it was a disaster in the best possible way.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it was a disaster in the best possible way",
+ "normalized_hypothesis": "it was a disaster in the best possible way",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0094",
+ "category": "descriptive",
+ "text": "It had the look of a place that used to mean something to someone.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0094.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It had the look of a place that used to mean something to someone.",
+ "raw": {
+ "normalized_reference": "it had the look of a place that used to mean something to someone.",
+ "normalized_hypothesis": "it had the look of a place that used to mean something to someone.",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it had the look of a place that used to mean something to someone",
+ "normalized_hypothesis": "it had the look of a place that used to mean something to someone",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0095",
+ "category": "emotional",
+ "text": "There's this feeling you get when something finally works after weeks of failure.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0095.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "There's this feeling you get when something finally works after weeks of failure.",
+ "raw": {
+ "normalized_reference": "there's this feeling you get when something finally works after weeks of failure.",
+ "normalized_hypothesis": "there's this feeling you get when something finally works after weeks of failure.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "there is this feeling you get when something finally works after weeks of failure",
+ "normalized_hypothesis": "there is this feeling you get when something finally works after weeks of failure",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0096",
+ "category": "long",
+ "text": "He had a theory that most arguments aren't actually about what they seem to be about, and the longer I live the more I think he was onto something.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0096.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He had a theory that most arguments aren't actually about what they seem to be about. And the longer I lived, the more I think he was onto something.",
+ "raw": {
+ "normalized_reference": "he had a theory that most arguments aren't actually about what they seem to be about, and the longer i live the more i think he was onto something.",
+ "normalized_hypothesis": "he had a theory that most arguments aren't actually about what they seem to be about. and the longer i lived, the more i think he was onto something.",
+ "errors": 1,
+ "words": 29,
+ "wer": 0.034482758620689655
+ },
+ "semantic": {
+ "normalized_reference": "he had a theory that most arguments are not actually about what they seem to be about and the longer i live the more i think he was onto something",
+ "normalized_hypothesis": "he had a theory that most arguments are not actually about what they seem to be about and the longer i lived the more i think he was onto something",
+ "errors": 1,
+ "words": 30,
+ "wer": 0.03333333333333333
+ }
+ },
+ {
+ "id": "release500_0097",
+ "category": "narrative",
+ "text": "The thing nobody tells you about starting over is how ordinary most of it feels.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0097.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The thing nobody tells you about starting overt is how ordinary most of it feels.",
+ "raw": {
+ "normalized_reference": "the thing nobody tells you about starting over is how ordinary most of it feels.",
+ "normalized_hypothesis": "the thing nobody tells you about starting overt is how ordinary most of it feels.",
+ "errors": 1,
+ "words": 15,
+ "wer": 0.06666666666666667
+ },
+ "semantic": {
+ "normalized_reference": "the thing nobody tells you about starting over is how ordinary most of it feels",
+ "normalized_hypothesis": "the thing nobody tells you about starting overt is how ordinary most of it feels",
+ "errors": 1,
+ "words": 15,
+ "wer": 0.06666666666666667
+ }
+ },
+ {
+ "id": "release500_0098",
+ "category": "question",
+ "text": "What exactly are we waiting for at this point?",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0098.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "What exactly are we waiting for at this point?",
+ "raw": {
+ "normalized_reference": "what exactly are we waiting for at this point?",
+ "normalized_hypothesis": "what exactly are we waiting for at this point?",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "what exactly are we waiting for at this point",
+ "normalized_hypothesis": "what exactly are we waiting for at this point",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0099",
+ "category": "short",
+ "text": "I stand by what I said.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0099.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I stand by what I said.",
+ "raw": {
+ "normalized_reference": "i stand by what i said.",
+ "normalized_hypothesis": "i stand by what i said.",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i stand by what i said",
+ "normalized_hypothesis": "i stand by what i said",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0100",
+ "category": "technical",
+ "text": "Double-check that value before you push anything to production.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0100.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Double check that value before you push anything to production.",
+ "raw": {
+ "normalized_reference": "double-check that value before you push anything to production.",
+ "normalized_hypothesis": "double check that value before you push anything to production.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "double check that value before you push anything to production",
+ "normalized_hypothesis": "double check that value before you push anything to production",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0101",
+ "category": "conversational",
+ "text": "I understand your point; I just do not agree with the conclusion.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0101.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I understand your point. I just do not agree with the conclusion.",
+ "raw": {
+ "normalized_reference": "i understand your point; i just do not agree with the conclusion.",
+ "normalized_hypothesis": "i understand your point. i just do not agree with the conclusion.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i understand your point i just do not agree with the conclusion",
+ "normalized_hypothesis": "i understand your point i just do not agree with the conclusion",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0102",
+ "category": "descriptive",
+ "text": "The sky that day was a strange shade of orange I'd never seen before.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0102.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The sky that day was a strange shade of orange I'd never seen before.",
+ "raw": {
+ "normalized_reference": "the sky that day was a strange shade of orange i'd never seen before.",
+ "normalized_hypothesis": "the sky that day was a strange shade of orange i'd never seen before.",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the sky that day was a strange shade of orange i would never seen before",
+ "normalized_hypothesis": "the sky that day was a strange shade of orange i would never seen before",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0103",
+ "category": "emotional",
+ "text": "It hurt more than I expected, and I thought I was prepared.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0103.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It hurt more than I expected, and I thought I was prepared.",
+ "raw": {
+ "normalized_reference": "it hurt more than i expected, and i thought i was prepared.",
+ "normalized_hypothesis": "it hurt more than i expected, and i thought i was prepared.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it hurt more than i expected and i thought i was prepared",
+ "normalized_hypothesis": "it hurt more than i expected and i thought i was prepared",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0104",
+ "category": "long",
+ "text": "The hardest part wasn't making the decision, it was sitting with the uncertainty of not knowing whether it was the right one for months afterward.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0104.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The hardest part wasn't making the decision. It was sitting with the uncertainty of not knowing whether it was the right one for months afterward.",
+ "raw": {
+ "normalized_reference": "the hardest part wasn't making the decision, it was sitting with the uncertainty of not knowing whether it was the right one for months afterward.",
+ "normalized_hypothesis": "the hardest part wasn't making the decision. it was sitting with the uncertainty of not knowing whether it was the right one for months afterward.",
+ "errors": 0,
+ "words": 25,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the hardest part was not making the decision it was sitting with the uncertainty of not knowing whether it was the right one for months afterward",
+ "normalized_hypothesis": "the hardest part was not making the decision it was sitting with the uncertainty of not knowing whether it was the right one for months afterward",
+ "errors": 0,
+ "words": 26,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0105",
+ "category": "narrative",
+ "text": "There was this one day last summer when everything just clicked into place.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0105.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "There was this one day last summer when everything just clicked into place.",
+ "raw": {
+ "normalized_reference": "there was this one day last summer when everything just clicked into place.",
+ "normalized_hypothesis": "there was this one day last summer when everything just clicked into place.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "there was this one day last summer when everything just clicked into place",
+ "normalized_hypothesis": "there was this one day last summer when everything just clicked into place",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0106",
+ "category": "question",
+ "text": "Have you ever thought about why we keep doing things the same way when they stop working?",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0106.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Have you ever thought about why we keep doing things the same way when they stop working?",
+ "raw": {
+ "normalized_reference": "have you ever thought about why we keep doing things the same way when they stop working?",
+ "normalized_hypothesis": "have you ever thought about why we keep doing things the same way when they stop working?",
+ "errors": 0,
+ "words": 17,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "have you ever thought about why we keep doing things the same way when they stop working",
+ "normalized_hypothesis": "have you ever thought about why we keep doing things the same way when they stop working",
+ "errors": 0,
+ "words": 17,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0107",
+ "category": "short",
+ "text": "I can't believe you just said that.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0107.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I can't believe you just said that.",
+ "raw": {
+ "normalized_reference": "i can't believe you just said that.",
+ "normalized_hypothesis": "i can't believe you just said that.",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i can not believe you just said that",
+ "normalized_hypothesis": "i can not believe you just said that",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0108",
+ "category": "technical",
+ "text": "You'll want to isolate that variable before drawing any conclusions.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0108.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "You'll want to isolate that variable before drawing any conclusions.",
+ "raw": {
+ "normalized_reference": "you'll want to isolate that variable before drawing any conclusions.",
+ "normalized_hypothesis": "you'll want to isolate that variable before drawing any conclusions.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "you will want to isolate that variable before drawing any conclusions",
+ "normalized_hypothesis": "you will want to isolate that variable before drawing any conclusions",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0109",
+ "category": "conversational",
+ "text": "The meeting that could have been an email lasted two hours.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0109.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The meeting that could have been an email lasted two hours.",
+ "raw": {
+ "normalized_reference": "the meeting that could have been an email lasted two hours.",
+ "normalized_hypothesis": "the meeting that could have been an email lasted two hours.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the meeting that could have been an email lasted 2 hours",
+ "normalized_hypothesis": "the meeting that could have been an email lasted 2 hours",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0110",
+ "category": "descriptive",
+ "text": "There's a specific kind of tired that sleep doesn't fix.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0110.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "There's a specific kind of tire that sleep doesn't fix.",
+ "raw": {
+ "normalized_reference": "there's a specific kind of tired that sleep doesn't fix.",
+ "normalized_hypothesis": "there's a specific kind of tire that sleep doesn't fix.",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ },
+ "semantic": {
+ "normalized_reference": "there is a specific kind of tired that sleep does not fix",
+ "normalized_hypothesis": "there is a specific kind of tire that sleep does not fix",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ }
+ },
+ {
+ "id": "release500_0111",
+ "category": "emotional",
+ "text": "Something about that moment broke something open in me that I hadn't realized was closed.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0111.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Something about that moment broke something open in me that I hadn't realized was closed.",
+ "raw": {
+ "normalized_reference": "something about that moment broke something open in me that i hadn't realized was closed.",
+ "normalized_hypothesis": "something about that moment broke something open in me that i hadn't realized was closed.",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "something about that moment broke something open in me that i had not realized was closed",
+ "normalized_hypothesis": "something about that moment broke something open in me that i had not realized was closed",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0112",
+ "category": "long",
+ "text": "The thing about making something good is that you don't usually know it's good while you're making it. You just know you're not done yet.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0112.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The thing about making something good is that you don't usually know it's good while you're making it. You just know you're not done yet.",
+ "raw": {
+ "normalized_reference": "the thing about making something good is that you don't usually know it's good while you're making it. you just know you're not done yet.",
+ "normalized_hypothesis": "the thing about making something good is that you don't usually know it's good while you're making it. you just know you're not done yet.",
+ "errors": 0,
+ "words": 25,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the thing about making something good is that you do not usually know it is good while you are making it you just know you are not done yet",
+ "normalized_hypothesis": "the thing about making something good is that you do not usually know it is good while you are making it you just know you are not done yet",
+ "errors": 0,
+ "words": 29,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0113",
+ "category": "narrative",
+ "text": "She opened the door slowly, as if she expected to find something she wasn't ready for.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0113.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "She opened the door slowly, as if she expected to find something she wasn't ready for.",
+ "raw": {
+ "normalized_reference": "she opened the door slowly, as if she expected to find something she wasn't ready for.",
+ "normalized_hypothesis": "she opened the door slowly, as if she expected to find something she wasn't ready for.",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "she opened the door slowly as if she expected to find something she was not ready for",
+ "normalized_hypothesis": "she opened the door slowly as if she expected to find something she was not ready for",
+ "errors": 0,
+ "words": 17,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0114",
+ "category": "question",
+ "text": "What would you do differently if you could go back?",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0114.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "What would you do differently if you could go back?",
+ "raw": {
+ "normalized_reference": "what would you do differently if you could go back?",
+ "normalized_hypothesis": "what would you do differently if you could go back?",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "what would you do differently if you could go back",
+ "normalized_hypothesis": "what would you do differently if you could go back",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0115",
+ "category": "short",
+ "text": "Okay. Okay okay okay.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0115.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Okay, okay, okay, okay.",
+ "raw": {
+ "normalized_reference": "okay. okay okay okay.",
+ "normalized_hypothesis": "okay, okay, okay, okay.",
+ "errors": 0,
+ "words": 4,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "okay okay okay okay",
+ "normalized_hypothesis": "okay okay okay okay",
+ "errors": 0,
+ "words": 4,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0116",
+ "category": "technical",
+ "text": "The latency goes up significantly at scale, which is expected.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0116.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The latency goes up significantly at scale, which is expected.",
+ "raw": {
+ "normalized_reference": "the latency goes up significantly at scale, which is expected.",
+ "normalized_hypothesis": "the latency goes up significantly at scale, which is expected.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the latency goes up significantly at scale which is expected",
+ "normalized_hypothesis": "the latency goes up significantly at scale which is expected",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0117",
+ "category": "conversational",
+ "text": "We ended up staying until two in the morning talking about nothing.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0117.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "We ended up staying until two in the morning talking about nothing.",
+ "raw": {
+ "normalized_reference": "we ended up staying until two in the morning talking about nothing.",
+ "normalized_hypothesis": "we ended up staying until two in the morning talking about nothing.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "we ended up staying until 2 in the morning talking about nothing",
+ "normalized_hypothesis": "we ended up staying until 2 in the morning talking about nothing",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0118",
+ "category": "descriptive",
+ "text": "The rain started slowly, and then all at once, the way it usually does.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0118.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The rain started slowly, and then all at once, the way it usually does.",
+ "raw": {
+ "normalized_reference": "the rain started slowly, and then all at once, the way it usually does.",
+ "normalized_hypothesis": "the rain started slowly, and then all at once, the way it usually does.",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the rain started slowly and then all at once the way it usually does",
+ "normalized_hypothesis": "the rain started slowly and then all at once the way it usually does",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0119",
+ "category": "emotional",
+ "text": "I honestly did not think I was going to make it through that year.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0119.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I honestly did not think I was going to make it through that year.",
+ "raw": {
+ "normalized_reference": "i honestly did not think i was going to make it through that year.",
+ "normalized_hypothesis": "i honestly did not think i was going to make it through that year.",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i honestly did not think i was going to make it through that year",
+ "normalized_hypothesis": "i honestly did not think i was going to make it through that year",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0120",
+ "category": "long",
+ "text": "There's an assumption that people who seem confident all the time actually feel confident, and in my experience that's almost never true.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0120.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "There's an assumption that people who seem confident all the time actually feel confident, and in my experience that's almost never true.",
+ "raw": {
+ "normalized_reference": "there's an assumption that people who seem confident all the time actually feel confident, and in my experience that's almost never true.",
+ "normalized_hypothesis": "there's an assumption that people who seem confident all the time actually feel confident, and in my experience that's almost never true.",
+ "errors": 0,
+ "words": 22,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "there is an assumption that people who seem confident all the time actually feel confident and in my experience that is almost never true",
+ "normalized_hypothesis": "there is an assumption that people who seem confident all the time actually feel confident and in my experience that is almost never true",
+ "errors": 0,
+ "words": 24,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0121",
+ "category": "narrative",
+ "text": "The moment I opened the door I knew something was different, even if I couldn't say what.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0121.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The moment I opened the door I knew something was different, even if I couldn't say what.",
+ "raw": {
+ "normalized_reference": "the moment i opened the door i knew something was different, even if i couldn't say what.",
+ "normalized_hypothesis": "the moment i opened the door i knew something was different, even if i couldn't say what.",
+ "errors": 0,
+ "words": 17,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the moment i opened the door i knew something was different even if i could not say what",
+ "normalized_hypothesis": "the moment i opened the door i knew something was different even if i could not say what",
+ "errors": 0,
+ "words": 18,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0122",
+ "category": "question",
+ "text": "How do you explain something that doesn't make sense even to you?",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0122.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "How do you explain something that doesn't make sense even to you?",
+ "raw": {
+ "normalized_reference": "how do you explain something that doesn't make sense even to you?",
+ "normalized_hypothesis": "how do you explain something that doesn't make sense even to you?",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "how do you explain something that does not make sense even to you",
+ "normalized_hypothesis": "how do you explain something that does not make sense even to you",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0123",
+ "category": "short",
+ "text": "Hold on a second.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0123.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Hold on a second.",
+ "raw": {
+ "normalized_reference": "hold on a second.",
+ "normalized_hypothesis": "hold on a second.",
+ "errors": 0,
+ "words": 4,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "hold on a 2nd",
+ "normalized_hypothesis": "hold on a 2nd",
+ "errors": 0,
+ "words": 4,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0124",
+ "category": "technical",
+ "text": "A deterministic split prevents duplicate normalized text from leaking into evaluation.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0124.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "A deterministic split prevents duplicate normalized text from leaking into evaluation.",
+ "raw": {
+ "normalized_reference": "a deterministic split prevents duplicate normalized text from leaking into evaluation.",
+ "normalized_hypothesis": "a deterministic split prevents duplicate normalized text from leaking into evaluation.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "a deterministic split prevents duplicate normalized text from leaking into evaluation",
+ "normalized_hypothesis": "a deterministic split prevents duplicate normalized text from leaking into evaluation",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0125",
+ "category": "conversational",
+ "text": "Honestly, that was much better than I expected, but the ending still felt rushed.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0125.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Honestly, that was much better than I expected, but the ending still felt rushed.",
+ "raw": {
+ "normalized_reference": "honestly, that was much better than i expected, but the ending still felt rushed.",
+ "normalized_hypothesis": "honestly, that was much better than i expected, but the ending still felt rushed.",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "honestly that was much better than i expected but the ending still felt rushed",
+ "normalized_hypothesis": "honestly that was much better than i expected but the ending still felt rushed",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0126",
+ "category": "descriptive",
+ "text": "She had a habit of tilting her head slightly when thinking hard about something.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0126.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "She had a habit of tilting her head slightly when thinking hard about something.",
+ "raw": {
+ "normalized_reference": "she had a habit of tilting her head slightly when thinking hard about something.",
+ "normalized_hypothesis": "she had a habit of tilting her head slightly when thinking hard about something.",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "she had a habit of tilting her head slightly when thinking hard about something",
+ "normalized_hypothesis": "she had a habit of tilting her head slightly when thinking hard about something",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0127",
+ "category": "emotional",
+ "text": "I cannot stress enough how much I did not want to be right about this.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0127.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I cannot stress enough how much I did not want to be right about this.",
+ "raw": {
+ "normalized_reference": "i cannot stress enough how much i did not want to be right about this.",
+ "normalized_hypothesis": "i cannot stress enough how much i did not want to be right about this.",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i cannot stress enough how much i did not want to be right about this",
+ "normalized_hypothesis": "i cannot stress enough how much i did not want to be right about this",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0128",
+ "category": "long",
+ "text": "I used to think the goal was to get to a point where nothing was hard anymore, and now I think that point doesn't exist and I'm not sure I'd want it to.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0128.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I used to think the goal was to get to a point where nothing was hard anymore, and now I think that point doesn't exist and I'm not sure I want it to.",
+ "raw": {
+ "normalized_reference": "i used to think the goal was to get to a point where nothing was hard anymore, and now i think that point doesn't exist and i'm not sure i'd want it to.",
+ "normalized_hypothesis": "i used to think the goal was to get to a point where nothing was hard anymore, and now i think that point doesn't exist and i'm not sure i want it to.",
+ "errors": 1,
+ "words": 33,
+ "wer": 0.030303030303030304
+ },
+ "semantic": {
+ "normalized_reference": "i used to think the goal was to get to a point where nothing was hard anymore and now i think that point does not exist and i am not sure i would want it to",
+ "normalized_hypothesis": "i used to think the goal was to get to a point where nothing was hard anymore and now i think that point does not exist and i am not sure i want it to",
+ "errors": 1,
+ "words": 36,
+ "wer": 0.027777777777777776
+ }
+ },
+ {
+ "id": "release500_0129",
+ "category": "narrative",
+ "text": "So this is the part of the story where everything goes completely sideways.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0129.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "So this is the part of the story where everything goes completely sideways.",
+ "raw": {
+ "normalized_reference": "so this is the part of the story where everything goes completely sideways.",
+ "normalized_hypothesis": "so this is the part of the story where everything goes completely sideways.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "so this is the part of the story where everything goes completely sideways",
+ "normalized_hypothesis": "so this is the part of the story where everything goes completely sideways",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0130",
+ "category": "question",
+ "text": "Is there a version of this where it actually works out okay?",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0130.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Is there a version of this where it actually works out okay?",
+ "raw": {
+ "normalized_reference": "is there a version of this where it actually works out okay?",
+ "normalized_hypothesis": "is there a version of this where it actually works out okay?",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "is there a version of this where it actually works out okay",
+ "normalized_hypothesis": "is there a version of this where it actually works out okay",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0131",
+ "category": "short",
+ "text": "Oh no. Oh no no no.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0131.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Oh, no. Oh, no, no, no.",
+ "raw": {
+ "normalized_reference": "oh no. oh no no no.",
+ "normalized_hypothesis": "oh, no. oh, no, no, no.",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "0 no 0 no no no",
+ "normalized_hypothesis": "0 no 0 no no no",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0132",
+ "category": "technical",
+ "text": "Always test on a copy before modifying the original.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0132.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Always test on a copy before modifying the original.",
+ "raw": {
+ "normalized_reference": "always test on a copy before modifying the original.",
+ "normalized_hypothesis": "always test on a copy before modifying the original.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "always test on a copy before modifying the original",
+ "normalized_hypothesis": "always test on a copy before modifying the original",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0133",
+ "category": "conversational",
+ "text": "I wasn't even supposed to be there. That's the wild part.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0133.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I wasn't even supposed to be there. That's the wild part.",
+ "raw": {
+ "normalized_reference": "i wasn't even supposed to be there. that's the wild part.",
+ "normalized_hypothesis": "i wasn't even supposed to be there. that's the wild part.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i was not even supposed to be there that is the wild part",
+ "normalized_hypothesis": "i was not even supposed to be there that is the wild part",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0134",
+ "category": "descriptive",
+ "text": "It's a small town. Not the kind you leave. The kind you leave and come back to.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0134.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It's a small town, not the kind you leave, the kind you leave and come back to.",
+ "raw": {
+ "normalized_reference": "it's a small town. not the kind you leave. the kind you leave and come back to.",
+ "normalized_hypothesis": "it's a small town, not the kind you leave, the kind you leave and come back to.",
+ "errors": 0,
+ "words": 17,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it is a small town not the kind you leave the kind you leave and come back to",
+ "normalized_hypothesis": "it is a small town not the kind you leave the kind you leave and come back to",
+ "errors": 0,
+ "words": 18,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0135",
+ "category": "emotional",
+ "text": "She cried. I cried. We all just sort of stood there crying.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0135.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "She cried. I cried. We all just sort of stood there crying.",
+ "raw": {
+ "normalized_reference": "she cried. i cried. we all just sort of stood there crying.",
+ "normalized_hypothesis": "she cried. i cried. we all just sort of stood there crying.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "she cried i cried we all just sort of stood there crying",
+ "normalized_hypothesis": "she cried i cried we all just sort of stood there crying",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0136",
+ "category": "long",
+ "text": "There's a version of being busy that's actually just a very elaborate way of avoiding the one thing you know you should be doing.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0136.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "There's a version of being busy that's actually just a very elaborate way of avoiding the one thing you know you should be doing.",
+ "raw": {
+ "normalized_reference": "there's a version of being busy that's actually just a very elaborate way of avoiding the one thing you know you should be doing.",
+ "normalized_hypothesis": "there's a version of being busy that's actually just a very elaborate way of avoiding the one thing you know you should be doing.",
+ "errors": 0,
+ "words": 24,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "there is a version of being busy that is actually just a very elaborate way of avoiding the one thing you know you should be doing",
+ "normalized_hypothesis": "there is a version of being busy that is actually just a very elaborate way of avoiding the one thing you know you should be doing",
+ "errors": 0,
+ "words": 26,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0137",
+ "category": "narrative",
+ "text": "The letter arrived on a Tuesday, which felt appropriate somehow.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0137.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The letter arrived on a Tuesday, which felt appropriate somehow.",
+ "raw": {
+ "normalized_reference": "the letter arrived on a tuesday, which felt appropriate somehow.",
+ "normalized_hypothesis": "the letter arrived on a tuesday, which felt appropriate somehow.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the letter arrived on a tuesday which felt appropriate somehow",
+ "normalized_hypothesis": "the letter arrived on a tuesday which felt appropriate somehow",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0138",
+ "category": "question",
+ "text": "What exactly did she say when you told her?",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0138.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "What exactly did she say when you told her?",
+ "raw": {
+ "normalized_reference": "what exactly did she say when you told her?",
+ "normalized_hypothesis": "what exactly did she say when you told her?",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "what exactly did she say when you told her",
+ "normalized_hypothesis": "what exactly did she say when you told her",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0139",
+ "category": "short",
+ "text": "No way. Absolutely not.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0139.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "No way, absolutely not.",
+ "raw": {
+ "normalized_reference": "no way. absolutely not.",
+ "normalized_hypothesis": "no way, absolutely not.",
+ "errors": 0,
+ "words": 4,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "no way absolutely not",
+ "normalized_hypothesis": "no way absolutely not",
+ "errors": 0,
+ "words": 4,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0140",
+ "category": "technical",
+ "text": "The first thing you want to do is make sure everything is backed up.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0140.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The first thing you want to do is make sure everything is backed up.",
+ "raw": {
+ "normalized_reference": "the first thing you want to do is make sure everything is backed up.",
+ "normalized_hypothesis": "the first thing you want to do is make sure everything is backed up.",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the 1st thing you want to do is make sure everything is backed up",
+ "normalized_hypothesis": "the 1st thing you want to do is make sure everything is backed up",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0141",
+ "category": "conversational",
+ "text": "I have the most insane idea and you need to just hear me out.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0141.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I have the most insane idea, and you need to just hear me out.",
+ "raw": {
+ "normalized_reference": "i have the most insane idea and you need to just hear me out.",
+ "normalized_hypothesis": "i have the most insane idea, and you need to just hear me out.",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i have the most insane idea and you need to just hear me out",
+ "normalized_hypothesis": "i have the most insane idea and you need to just hear me out",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0142",
+ "category": "descriptive",
+ "text": "There's a coffee shop on the corner that's been there longer than anyone can remember.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0142.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "There's a coffee shop on the corner that's been there longer than anyone can remember.",
+ "raw": {
+ "normalized_reference": "there's a coffee shop on the corner that's been there longer than anyone can remember.",
+ "normalized_hypothesis": "there's a coffee shop on the corner that's been there longer than anyone can remember.",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "there is a coffee shop on the corner that has been there longer than anyone can remember",
+ "normalized_hypothesis": "there is a coffee shop on the corner that has been there longer than anyone can remember",
+ "errors": 0,
+ "words": 17,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0143",
+ "category": "emotional",
+ "text": "I kept it together until I got to the car, and then I completely fell apart.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0143.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I kept it together until I got to the car, and then I completely fell apart.",
+ "raw": {
+ "normalized_reference": "i kept it together until i got to the car, and then i completely fell apart.",
+ "normalized_hypothesis": "i kept it together until i got to the car, and then i completely fell apart.",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i kept it together until i got to the car and then i completely fell apart",
+ "normalized_hypothesis": "i kept it together until i got to the car and then i completely fell apart",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0144",
+ "category": "long",
+ "text": "The older I get, the more I think that most of what we call wisdom is just pattern recognition built from having been wrong enough times to start noticing.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0144.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The older I get, the more I think that most of what we call wisdom is just pattern recognition built from having been wrong enough times to start noticing.",
+ "raw": {
+ "normalized_reference": "the older i get, the more i think that most of what we call wisdom is just pattern recognition built from having been wrong enough times to start noticing.",
+ "normalized_hypothesis": "the older i get, the more i think that most of what we call wisdom is just pattern recognition built from having been wrong enough times to start noticing.",
+ "errors": 0,
+ "words": 29,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the older i get the more i think that most of what we call wisdom is just pattern recognition built from having been wrong enough times to start noticing",
+ "normalized_hypothesis": "the older i get the more i think that most of what we call wisdom is just pattern recognition built from having been wrong enough times to start noticing",
+ "errors": 0,
+ "words": 29,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0145",
+ "category": "narrative",
+ "text": "I remember the exact moment I realized I had made a huge mistake.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0145.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I remember the exact moment I realized I had made a huge mistake.",
+ "raw": {
+ "normalized_reference": "i remember the exact moment i realized i had made a huge mistake.",
+ "normalized_hypothesis": "i remember the exact moment i realized i had made a huge mistake.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i remember the exact moment i realized i had made a huge mistake",
+ "normalized_hypothesis": "i remember the exact moment i realized i had made a huge mistake",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0146",
+ "category": "question",
+ "text": "Do you think it would have gone differently if we'd left earlier?",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0146.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Do you think it would have gone differently if we'd left earlier?",
+ "raw": {
+ "normalized_reference": "do you think it would have gone differently if we'd left earlier?",
+ "normalized_hypothesis": "do you think it would have gone differently if we'd left earlier?",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "do you think it would have gone differently if we would left earlier",
+ "normalized_hypothesis": "do you think it would have gone differently if we would left earlier",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0147",
+ "category": "short",
+ "text": "You have to be kidding me.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0147.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "You have to be kidding me.",
+ "raw": {
+ "normalized_reference": "you have to be kidding me.",
+ "normalized_hypothesis": "you have to be kidding me.",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "you have to be kidding me",
+ "normalized_hypothesis": "you have to be kidding me",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0148",
+ "category": "technical",
+ "text": "If the checksum fails, restart the worker process without deleting the cached tensors.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0148.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "If the checksum fails, restart the worker process without deleting the cache tensors.",
+ "raw": {
+ "normalized_reference": "if the checksum fails, restart the worker process without deleting the cached tensors.",
+ "normalized_hypothesis": "if the checksum fails, restart the worker process without deleting the cache tensors.",
+ "errors": 1,
+ "words": 13,
+ "wer": 0.07692307692307693
+ },
+ "semantic": {
+ "normalized_reference": "if the checksum fails restart the worker process without deleting the cached tensors",
+ "normalized_hypothesis": "if the checksum fails restart the worker process without deleting the cache tensors",
+ "errors": 1,
+ "words": 13,
+ "wer": 0.07692307692307693
+ }
+ },
+ {
+ "id": "release500_0149",
+ "category": "conversational",
+ "text": "I need you to understand how close that was to not working.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0149.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I need you to understand how close that was to not working.",
+ "raw": {
+ "normalized_reference": "i need you to understand how close that was to not working.",
+ "normalized_hypothesis": "i need you to understand how close that was to not working.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i need you to understand how close that was to not working",
+ "normalized_hypothesis": "i need you to understand how close that was to not working",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0150",
+ "category": "descriptive",
+ "text": "It was the sort of morning where the whole world looks rinsed clean.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0150.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It was the sort of morning where the whole world looks rinsed clean.",
+ "raw": {
+ "normalized_reference": "it was the sort of morning where the whole world looks rinsed clean.",
+ "normalized_hypothesis": "it was the sort of morning where the whole world looks rinsed clean.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it was the sort of morning where the whole world looks rinsed clean",
+ "normalized_hypothesis": "it was the sort of morning where the whole world looks rinsed clean",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0151",
+ "category": "emotional",
+ "text": "She started laughing and I started laughing and we couldn't stop for a full minute.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0151.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "She started laughing, and I started laughing, and we couldn't stop for a full minute.",
+ "raw": {
+ "normalized_reference": "she started laughing and i started laughing and we couldn't stop for a full minute.",
+ "normalized_hypothesis": "she started laughing, and i started laughing, and we couldn't stop for a full minute.",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "she started laughing and i started laughing and we could not stop for a full minute",
+ "normalized_hypothesis": "she started laughing and i started laughing and we could not stop for a full minute",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0152",
+ "category": "long",
+ "text": "We spent three years building something we were genuinely proud of, and then the market shifted and none of it mattered the way we thought it would.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0152.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "We spent three years building something we were genuinely proud of, and then the market shifted and none of it mattered the way we thought it would.",
+ "raw": {
+ "normalized_reference": "we spent three years building something we were genuinely proud of, and then the market shifted and none of it mattered the way we thought it would.",
+ "normalized_hypothesis": "we spent three years building something we were genuinely proud of, and then the market shifted and none of it mattered the way we thought it would.",
+ "errors": 0,
+ "words": 27,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "we spent 3 years building something we were genuinely proud of and then the market shifted and none of it mattered the way we thought it would",
+ "normalized_hypothesis": "we spent 3 years building something we were genuinely proud of and then the market shifted and none of it mattered the way we thought it would",
+ "errors": 0,
+ "words": 27,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0153",
+ "category": "narrative",
+ "text": "And somehow, despite all of that, it actually turned out okay.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0153.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "And somehow, despite all of that, it actually turned out okay.",
+ "raw": {
+ "normalized_reference": "and somehow, despite all of that, it actually turned out okay.",
+ "normalized_hypothesis": "and somehow, despite all of that, it actually turned out okay.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "and somehow despite all of that it actually turned out okay",
+ "normalized_hypothesis": "and somehow despite all of that it actually turned out okay",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0154",
+ "category": "question",
+ "text": "Is that something you actually want, or something you think you should want?",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0154.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Is that something you actually want or something you think you should want?",
+ "raw": {
+ "normalized_reference": "is that something you actually want, or something you think you should want?",
+ "normalized_hypothesis": "is that something you actually want or something you think you should want?",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "is that something you actually want or something you think you should want",
+ "normalized_hypothesis": "is that something you actually want or something you think you should want",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0155",
+ "category": "short",
+ "text": "This is fine.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0155.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "This is fine.",
+ "raw": {
+ "normalized_reference": "this is fine.",
+ "normalized_hypothesis": "this is fine.",
+ "errors": 0,
+ "words": 3,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "this is fine",
+ "normalized_hypothesis": "this is fine",
+ "errors": 0,
+ "words": 3,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0156",
+ "category": "technical",
+ "text": "The tricky part isn't the setup, it's making sure all dependencies line up.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0156.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The tricky part isn't the setup. It's making sure all dependencies line up.",
+ "raw": {
+ "normalized_reference": "the tricky part isn't the setup, it's making sure all dependencies line up.",
+ "normalized_hypothesis": "the tricky part isn't the setup. it's making sure all dependencies line up.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the tricky part is not the setup it is making sure all dependencies line up",
+ "normalized_hypothesis": "the tricky part is not the setup it is making sure all dependencies line up",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0157",
+ "category": "conversational",
+ "text": "He called it a minor setback. I called it a disaster.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0157.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He called it a minor setback. I called it a disaster.",
+ "raw": {
+ "normalized_reference": "he called it a minor setback. i called it a disaster.",
+ "normalized_hypothesis": "he called it a minor setback. i called it a disaster.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he called it a minor setback i called it a disaster",
+ "normalized_hypothesis": "he called it a minor setback i called it a disaster",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0158",
+ "category": "descriptive",
+ "text": "The whole place smelled like pine and old wood and cold air.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0158.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The whole place smelled like pine and old wood and cold air.",
+ "raw": {
+ "normalized_reference": "the whole place smelled like pine and old wood and cold air.",
+ "normalized_hypothesis": "the whole place smelled like pine and old wood and cold air.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the whole place smelled like pine and old wood and cold air",
+ "normalized_hypothesis": "the whole place smelled like pine and old wood and cold air",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0159",
+ "category": "emotional",
+ "text": "I've never been so relieved in my entire life.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0159.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I've never been so relieved in my entire life.",
+ "raw": {
+ "normalized_reference": "i've never been so relieved in my entire life.",
+ "normalized_hypothesis": "i've never been so relieved in my entire life.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i have never been so relieved in my entire life",
+ "normalized_hypothesis": "i have never been so relieved in my entire life",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0160",
+ "category": "narrative",
+ "text": "I wrote it all down that night so I wouldn't forget, and I still haven't looked at it.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0160.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I wrote it all down that night so I wouldn't forget, and I still haven't looked at it.",
+ "raw": {
+ "normalized_reference": "i wrote it all down that night so i wouldn't forget, and i still haven't looked at it.",
+ "normalized_hypothesis": "i wrote it all down that night so i wouldn't forget, and i still haven't looked at it.",
+ "errors": 0,
+ "words": 18,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i wrote it all down that night so i would not forget and i still have not looked at it",
+ "normalized_hypothesis": "i wrote it all down that night so i would not forget and i still have not looked at it",
+ "errors": 0,
+ "words": 20,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0161",
+ "category": "short",
+ "text": "That's kind of wild actually.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0161.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "That's kind of wild, actually.",
+ "raw": {
+ "normalized_reference": "that's kind of wild actually.",
+ "normalized_hypothesis": "that's kind of wild, actually.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "that is kind of wild actually",
+ "normalized_hypothesis": "that is kind of wild actually",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0162",
+ "category": "technical",
+ "text": "Version pinning will save you a lot of pain later down the line.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0162.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Version pinning will save you a lot of pain later down the line.",
+ "raw": {
+ "normalized_reference": "version pinning will save you a lot of pain later down the line.",
+ "normalized_hypothesis": "version pinning will save you a lot of pain later down the line.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "version pinning will save you a lot of pain later down the line",
+ "normalized_hypothesis": "version pinning will save you a lot of pain later down the line",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0163",
+ "category": "conversational",
+ "text": "At some point you just have to commit and see what happens.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0163.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "At some point you just have to commit and see what happens.",
+ "raw": {
+ "normalized_reference": "at some point you just have to commit and see what happens.",
+ "normalized_hypothesis": "at some point you just have to commit and see what happens.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "at some point you just have to commit and see what happens",
+ "normalized_hypothesis": "at some point you just have to commit and see what happens",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0164",
+ "category": "emotional",
+ "text": "Nobody told me it was going to be this hard, and honestly, I'm grateful.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0164.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Nobody told me it was going to be this hard, and honestly, I'm grateful.",
+ "raw": {
+ "normalized_reference": "nobody told me it was going to be this hard, and honestly, i'm grateful.",
+ "normalized_hypothesis": "nobody told me it was going to be this hard, and honestly, i'm grateful.",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "nobody told me it was going to be this hard and honestly i am grateful",
+ "normalized_hypothesis": "nobody told me it was going to be this hard and honestly i am grateful",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0165",
+ "category": "narrative",
+ "text": "Chapter seven. In which everything that could go wrong did, and a few things that couldn't.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0165.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Chapter 7, in which everything that could go wrong did, and a few things that couldn't.",
+ "raw": {
+ "normalized_reference": "chapter seven. in which everything that could go wrong did, and a few things that couldn't.",
+ "normalized_hypothesis": "chapter 7, in which everything that could go wrong did, and a few things that couldn't.",
+ "errors": 1,
+ "words": 16,
+ "wer": 0.0625
+ },
+ "semantic": {
+ "normalized_reference": "chapter 7 in which everything that could go wrong did and a few things that could not",
+ "normalized_hypothesis": "chapter 7 in which everything that could go wrong did and a few things that could not",
+ "errors": 0,
+ "words": 17,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0166",
+ "category": "short",
+ "text": "Wait, what?",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0166.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Wait, what?",
+ "raw": {
+ "normalized_reference": "wait, what?",
+ "normalized_hypothesis": "wait, what?",
+ "errors": 0,
+ "words": 2,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "wait what",
+ "normalized_hypothesis": "wait what",
+ "errors": 0,
+ "words": 2,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0167",
+ "category": "technical",
+ "text": "The graphics processor remained idle because the data loader exhausted every central processing core.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0167.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The graphics processor remained idle because the data loader exhausted every central processing core.",
+ "raw": {
+ "normalized_reference": "the graphics processor remained idle because the data loader exhausted every central processing core.",
+ "normalized_hypothesis": "the graphics processor remained idle because the data loader exhausted every central processing core.",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the graphics processor remained idle because the data loader exhausted every central processing core",
+ "normalized_hypothesis": "the graphics processor remained idle because the data loader exhausted every central processing core",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0168",
+ "category": "conversational",
+ "text": "Could you speak a little more slowly? I am trying to write every detail down.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0168.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Could you speak a little more slowly? I am trying to write every detail down.",
+ "raw": {
+ "normalized_reference": "could you speak a little more slowly? i am trying to write every detail down.",
+ "normalized_hypothesis": "could you speak a little more slowly? i am trying to write every detail down.",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "could you speak a little more slowly i am trying to write every detail down",
+ "normalized_hypothesis": "could you speak a little more slowly i am trying to write every detail down",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0169",
+ "category": "emotional",
+ "text": "I was so excited I could barely form a complete sentence.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0169.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I was so excited I could barely form a complete sentence.",
+ "raw": {
+ "normalized_reference": "i was so excited i could barely form a complete sentence.",
+ "normalized_hypothesis": "i was so excited i could barely form a complete sentence.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i was so excited i could barely form a complete sentence",
+ "normalized_hypothesis": "i was so excited i could barely form a complete sentence",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0170",
+ "category": "narrative",
+ "text": "He looked at me like I had said something in a language he didn't speak.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0170.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He looked at me like I had said something in a language he didn't speak.",
+ "raw": {
+ "normalized_reference": "he looked at me like i had said something in a language he didn't speak.",
+ "normalized_hypothesis": "he looked at me like i had said something in a language he didn't speak.",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he looked at me like i had said something in a language he did not speak",
+ "normalized_hypothesis": "he looked at me like i had said something in a language he did not speak",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0171",
+ "category": "short",
+ "text": "That's actually hilarious.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0171.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "That's actually hilarious.",
+ "raw": {
+ "normalized_reference": "that's actually hilarious.",
+ "normalized_hypothesis": "that's actually hilarious.",
+ "errors": 0,
+ "words": 3,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "that is actually hilarious",
+ "normalized_hypothesis": "that is actually hilarious",
+ "errors": 0,
+ "words": 4,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0172",
+ "category": "technical",
+ "text": "The encoder maps phonemes, stress, and phrase boundaries into a shared hidden representation.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0172.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The encoder maps phonemes, stress, and phrase boundaries into a shared hidden representation.",
+ "raw": {
+ "normalized_reference": "the encoder maps phonemes, stress, and phrase boundaries into a shared hidden representation.",
+ "normalized_hypothesis": "the encoder maps phonemes, stress, and phrase boundaries into a shared hidden representation.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the encoder maps phonemes stress and phrase boundaries into a shared hidden representation",
+ "normalized_hypothesis": "the encoder maps phonemes stress and phrase boundaries into a shared hidden representation",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0173",
+ "category": "conversational",
+ "text": "She showed up three hours late and acted like nothing had happened.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0173.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "She showed up three hours late and acted like nothing had happened.",
+ "raw": {
+ "normalized_reference": "she showed up three hours late and acted like nothing had happened.",
+ "normalized_hypothesis": "she showed up three hours late and acted like nothing had happened.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "she showed up 3 hours late and acted like nothing had happened",
+ "normalized_hypothesis": "she showed up 3 hours late and acted like nothing had happened",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0174",
+ "category": "emotional",
+ "text": "Oh my god. This is the best thing I've ever tasted.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0174.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Oh my god, this is the best thing I've ever tasted.",
+ "raw": {
+ "normalized_reference": "oh my god. this is the best thing i've ever tasted.",
+ "normalized_hypothesis": "oh my god, this is the best thing i've ever tasted.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "0 my god this is the best thing i have ever tasted",
+ "normalized_hypothesis": "0 my god this is the best thing i have ever tasted",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0175",
+ "category": "narrative",
+ "text": "The note was short, which was somehow more frightening than a long one would have been.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0175.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The note was short, which was somehow more frightening than a long one would have been.",
+ "raw": {
+ "normalized_reference": "the note was short, which was somehow more frightening than a long one would have been.",
+ "normalized_hypothesis": "the note was short, which was somehow more frightening than a long one would have been.",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the note was short which was somehow more frightening than a long one would have been",
+ "normalized_hypothesis": "the note was short which was somehow more frightening than a long one would have been",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0176",
+ "category": "short",
+ "text": "Noted.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0176.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Noid.",
+ "raw": {
+ "normalized_reference": "noted.",
+ "normalized_hypothesis": "noid.",
+ "errors": 1,
+ "words": 1,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "noted",
+ "normalized_hypothesis": "noid",
+ "errors": 1,
+ "words": 1,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0177",
+ "category": "technical",
+ "text": "It's a simple process once you understand the underlying structure.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0177.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It's a simple process once you understand the underlying structure.",
+ "raw": {
+ "normalized_reference": "it's a simple process once you understand the underlying structure.",
+ "normalized_hypothesis": "it's a simple process once you understand the underlying structure.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it is a simple process once you understand the underlying structure",
+ "normalized_hypothesis": "it is a simple process once you understand the underlying structure",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0178",
+ "category": "conversational",
+ "text": "You have to hear this story because it's almost too good to be true.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0178.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "You have to hear this story because it's almost too good to be true.",
+ "raw": {
+ "normalized_reference": "you have to hear this story because it's almost too good to be true.",
+ "normalized_hypothesis": "you have to hear this story because it's almost too good to be true.",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "you have to hear this story because it is almost too good to be true",
+ "normalized_hypothesis": "you have to hear this story because it is almost too good to be true",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0179",
+ "category": "emotional",
+ "text": "I have never felt more seen in my entire life than in that moment.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0179.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I have never felt more seen in my entire life than in that moment.",
+ "raw": {
+ "normalized_reference": "i have never felt more seen in my entire life than in that moment.",
+ "normalized_hypothesis": "i have never felt more seen in my entire life than in that moment.",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i have never felt more seen in my entire life than in that moment",
+ "normalized_hypothesis": "i have never felt more seen in my entire life than in that moment",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0180",
+ "category": "narrative",
+ "text": "She knew. She had always known. She just waited for the right moment.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0180.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "She knew. She had always known. She just waited for the right moment.",
+ "raw": {
+ "normalized_reference": "she knew. she had always known. she just waited for the right moment.",
+ "normalized_hypothesis": "she knew. she had always known. she just waited for the right moment.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "she knew she had always known she just waited for the right moment",
+ "normalized_hypothesis": "she knew she had always known she just waited for the right moment",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0181",
+ "category": "short",
+ "text": "Yeah, no. Hard pass.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0181.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Yeah, nailed a hard pass.",
+ "raw": {
+ "normalized_reference": "yeah, no. hard pass.",
+ "normalized_hypothesis": "yeah, nailed a hard pass.",
+ "errors": 2,
+ "words": 4,
+ "wer": 0.5
+ },
+ "semantic": {
+ "normalized_reference": "yeah no hard pass",
+ "normalized_hypothesis": "yeah nailed a hard pass",
+ "errors": 2,
+ "words": 4,
+ "wer": 0.5
+ }
+ },
+ {
+ "id": "release500_0182",
+ "category": "technical",
+ "text": "Please verify the sample rate, hop length, and mel filter configuration before inference.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0182.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Please verify the sample rate, hop length, and MEL filter configuration before inference.",
+ "raw": {
+ "normalized_reference": "please verify the sample rate, hop length, and mel filter configuration before inference.",
+ "normalized_hypothesis": "please verify the sample rate, hop length, and mel filter configuration before inference.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "please verify the sample rate hop length and mel filter configuration before inference",
+ "normalized_hypothesis": "please verify the sample rate hop length and mel filter configuration before inference",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0183",
+ "category": "conversational",
+ "text": "That's the worst idea I've ever heard and I'm completely in.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0183.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "That's the worst idea I've ever heard and I'm completely in.",
+ "raw": {
+ "normalized_reference": "that's the worst idea i've ever heard and i'm completely in.",
+ "normalized_hypothesis": "that's the worst idea i've ever heard and i'm completely in.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "that is the worst idea i have ever heard and i am completely in",
+ "normalized_hypothesis": "that is the worst idea i have ever heard and i am completely in",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0184",
+ "category": "emotional",
+ "text": "I remember thinking: this is it. This is the thing I've been waiting for.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0184.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I remember thinking, this is it. This is the thing I've been waiting for.",
+ "raw": {
+ "normalized_reference": "i remember thinking: this is it. this is the thing i've been waiting for.",
+ "normalized_hypothesis": "i remember thinking, this is it. this is the thing i've been waiting for.",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i remember thinking this is it this is the thing i have been waiting for",
+ "normalized_hypothesis": "i remember thinking this is it this is the thing i have been waiting for",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0185",
+ "category": "narrative",
+ "text": "We never talked about it again, which I think was the right call.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0185.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "We never talked about it again, which I think was the right call.",
+ "raw": {
+ "normalized_reference": "we never talked about it again, which i think was the right call.",
+ "normalized_hypothesis": "we never talked about it again, which i think was the right call.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "we never talked about it again which i think was the right call",
+ "normalized_hypothesis": "we never talked about it again which i think was the right call",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0186",
+ "category": "short",
+ "text": "In hindsight, obvious.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0186.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "In hindsight, obvious.",
+ "raw": {
+ "normalized_reference": "in hindsight, obvious.",
+ "normalized_hypothesis": "in hindsight, obvious.",
+ "errors": 0,
+ "words": 3,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "in hindsight obvious",
+ "normalized_hypothesis": "in hindsight obvious",
+ "errors": 0,
+ "words": 3,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0187",
+ "category": "technical",
+ "text": "Once it's configured correctly, it basically runs itself.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0187.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Once it's configured correctly, it basically runs itself.",
+ "raw": {
+ "normalized_reference": "once it's configured correctly, it basically runs itself.",
+ "normalized_hypothesis": "once it's configured correctly, it basically runs itself.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "once it is configured correctly it basically runs itself",
+ "normalized_hypothesis": "once it is configured correctly it basically runs itself",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0188",
+ "category": "conversational",
+ "text": "You got engaged? When? How? Tell me everything right now.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0188.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "You got engaged? When? How? Tell me everything right now.",
+ "raw": {
+ "normalized_reference": "you got engaged? when? how? tell me everything right now.",
+ "normalized_hypothesis": "you got engaged? when? how? tell me everything right now.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "you got engaged when how tell me everything right now",
+ "normalized_hypothesis": "you got engaged when how tell me everything right now",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0189",
+ "category": "emotional",
+ "text": "It's hard to explain how much that moment meant to me.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0189.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It's hard to explain how much that moment meant to me.",
+ "raw": {
+ "normalized_reference": "it's hard to explain how much that moment meant to me.",
+ "normalized_hypothesis": "it's hard to explain how much that moment meant to me.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it is hard to explain how much that moment meant to me",
+ "normalized_hypothesis": "it is hard to explain how much that moment meant to me",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0190",
+ "category": "narrative",
+ "text": "He told the story three different times that night and somehow it got better each time.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0190.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He told the story three different times that night, and somehow it got better each time.",
+ "raw": {
+ "normalized_reference": "he told the story three different times that night and somehow it got better each time.",
+ "normalized_hypothesis": "he told the story three different times that night, and somehow it got better each time.",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he told the story 3 different times that night and somehow it got better each time",
+ "normalized_hypothesis": "he told the story 3 different times that night and somehow it got better each time",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0191",
+ "category": "short",
+ "text": "Right. Moving on.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0191.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Right. Moving on.",
+ "raw": {
+ "normalized_reference": "right. moving on.",
+ "normalized_hypothesis": "right. moving on.",
+ "errors": 0,
+ "words": 3,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "right moving on",
+ "normalized_hypothesis": "right moving on",
+ "errors": 0,
+ "words": 3,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0192",
+ "category": "conversational",
+ "text": "No, I did not move your keys; check beside the small ceramic bowl.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0192.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "No, I did not move your keys. Check beside the small ceramic bowl.",
+ "raw": {
+ "normalized_reference": "no, i did not move your keys; check beside the small ceramic bowl.",
+ "normalized_hypothesis": "no, i did not move your keys. check beside the small ceramic bowl.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "no i did not move your keys check beside the small ceramic bowl",
+ "normalized_hypothesis": "no i did not move your keys check beside the small ceramic bowl",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0193",
+ "category": "emotional",
+ "text": "I was devastated. I'm not going to pretend otherwise.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0193.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I was devastated. I'm not going to pretend otherwise.",
+ "raw": {
+ "normalized_reference": "i was devastated. i'm not going to pretend otherwise.",
+ "normalized_hypothesis": "i was devastated. i'm not going to pretend otherwise.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i was devastated i am not going to pretend otherwise",
+ "normalized_hypothesis": "i was devastated i am not going to pretend otherwise",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0194",
+ "category": "narrative",
+ "text": "That was the last summer before everything changed, though we didn't know it yet.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0194.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "That was the last summer before everything changed, though we didn't know it yet.",
+ "raw": {
+ "normalized_reference": "that was the last summer before everything changed, though we didn't know it yet.",
+ "normalized_hypothesis": "that was the last summer before everything changed, though we didn't know it yet.",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "that was the last summer before everything changed though we did not know it yet",
+ "normalized_hypothesis": "that was the last summer before everything changed though we did not know it yet",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0195",
+ "category": "short",
+ "text": "Bold of you to assume.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0195.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Bold of you to assume.",
+ "raw": {
+ "normalized_reference": "bold of you to assume.",
+ "normalized_hypothesis": "bold of you to assume.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "bold of you to assume",
+ "normalized_hypothesis": "bold of you to assume",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0196",
+ "category": "conversational",
+ "text": "It's one of those decisions that makes total sense in the moment.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0196.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It's one of those decisions that makes total sense in the moment.",
+ "raw": {
+ "normalized_reference": "it's one of those decisions that makes total sense in the moment.",
+ "normalized_hypothesis": "it's one of those decisions that makes total sense in the moment.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it is one of those decisions that makes total sense in the moment",
+ "normalized_hypothesis": "it is one of those decisions that makes total sense in the moment",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0197",
+ "category": "emotional",
+ "text": "I was furious. Not the loud kind. The quiet kind that lasts for days.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0197.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I was furious. Not the loud kind. The quiet kind that lasts for days.",
+ "raw": {
+ "normalized_reference": "i was furious. not the loud kind. the quiet kind that lasts for days.",
+ "normalized_hypothesis": "i was furious. not the loud kind. the quiet kind that lasts for days.",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i was furious not the loud kind the quiet kind that lasts for days",
+ "normalized_hypothesis": "i was furious not the loud kind the quiet kind that lasts for days",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0198",
+ "category": "narrative",
+ "text": "That's the part of the story I usually leave out.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0198.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "That's the part of the story I usually leave out.",
+ "raw": {
+ "normalized_reference": "that's the part of the story i usually leave out.",
+ "normalized_hypothesis": "that's the part of the story i usually leave out.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "that is the part of the story i usually leave out",
+ "normalized_hypothesis": "that is the part of the story i usually leave out",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0199",
+ "category": "short",
+ "text": "I've said too much.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0199.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I've said too much.",
+ "raw": {
+ "normalized_reference": "i've said too much.",
+ "normalized_hypothesis": "i've said too much.",
+ "errors": 0,
+ "words": 4,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i have said too much",
+ "normalized_hypothesis": "i have said too much",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0200",
+ "category": "conversational",
+ "text": "I've been faking that I understood this the entire time.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0200.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I've been faking that I understood this the entire time.",
+ "raw": {
+ "normalized_reference": "i've been faking that i understood this the entire time.",
+ "normalized_hypothesis": "i've been faking that i understood this the entire time.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i have been faking that i understood this the entire time",
+ "normalized_hypothesis": "i have been faking that i understood this the entire time",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0201",
+ "category": "phonetic_balance",
+ "text": "Peace be unto you and grace before the Lord.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0201.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Peace be unto you and grace before the Lord.",
+ "raw": {
+ "normalized_reference": "peace be unto you and grace before the lord.",
+ "normalized_hypothesis": "peace be unto you and grace before the lord.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "peace be unto you and grace before the lord",
+ "normalized_hypothesis": "peace be unto you and grace before the lord",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0202",
+ "category": "phonetic_balance",
+ "text": "From now on we're pals.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0202.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "From now on, Rare Pals.",
+ "raw": {
+ "normalized_reference": "from now on we're pals.",
+ "normalized_hypothesis": "from now on, rare pals.",
+ "errors": 1,
+ "words": 5,
+ "wer": 0.2
+ },
+ "semantic": {
+ "normalized_reference": "from now on we are pals",
+ "normalized_hypothesis": "from now on rare pals",
+ "errors": 2,
+ "words": 6,
+ "wer": 0.3333333333333333
+ }
+ },
+ {
+ "id": "release500_0203",
+ "category": "phonetic_balance",
+ "text": "Here we allow our solicitors to look after our legal work.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0203.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Here we allow our solicitors to look after our legal work.",
+ "raw": {
+ "normalized_reference": "here we allow our solicitors to look after our legal work.",
+ "normalized_hypothesis": "here we allow our solicitors to look after our legal work.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "here we allow our solicitors to look after our legal work",
+ "normalized_hypothesis": "here we allow our solicitors to look after our legal work",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0204",
+ "category": "phonetic_balance",
+ "text": "Her words sent a strange chill through Philip.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0204.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Her words send a strange chill through Philip.",
+ "raw": {
+ "normalized_reference": "her words sent a strange chill through philip.",
+ "normalized_hypothesis": "her words send a strange chill through philip.",
+ "errors": 1,
+ "words": 8,
+ "wer": 0.125
+ },
+ "semantic": {
+ "normalized_reference": "her words sent a strange chill through philip",
+ "normalized_hypothesis": "her words send a strange chill through philip",
+ "errors": 1,
+ "words": 8,
+ "wer": 0.125
+ }
+ },
+ {
+ "id": "release500_0205",
+ "category": "phonetic_balance",
+ "text": "It was introduced by Representative Dick of Ohio.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0205.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It was introduced by Representative Dick of Ohio.",
+ "raw": {
+ "normalized_reference": "it was introduced by representative dick of ohio.",
+ "normalized_hypothesis": "it was introduced by representative dick of ohio.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it was introduced by representative dick of ohio",
+ "normalized_hypothesis": "it was introduced by representative dick of ohio",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0206",
+ "category": "phonetic_balance",
+ "text": "It's the aurora borealis.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0206.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It's the Aurora Borealis.",
+ "raw": {
+ "normalized_reference": "it's the aurora borealis.",
+ "normalized_hypothesis": "it's the aurora borealis.",
+ "errors": 0,
+ "words": 4,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it is the aurora borealis",
+ "normalized_hypothesis": "it is the aurora borealis",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0207",
+ "category": "phonetic_balance",
+ "text": "The boy grew and prospered.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0207.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The boy grew and prospered.",
+ "raw": {
+ "normalized_reference": "the boy grew and prospered.",
+ "normalized_hypothesis": "the boy grew and prospered.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the boy grew and prospered",
+ "normalized_hypothesis": "the boy grew and prospered",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0208",
+ "category": "phonetic_balance",
+ "text": "His voice was passionately rebellious.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0208.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "His voice was passionately rebellious.",
+ "raw": {
+ "normalized_reference": "his voice was passionately rebellious.",
+ "normalized_hypothesis": "his voice was passionately rebellious.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "his voice was passionately rebellious",
+ "normalized_hypothesis": "his voice was passionately rebellious",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0209",
+ "category": "phonetic_balance",
+ "text": "That's where they cut off the Scottish Chiefs and killed all hands.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0209.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "That's where they cut off the Scottish chiefs and killed all hands.",
+ "raw": {
+ "normalized_reference": "that's where they cut off the scottish chiefs and killed all hands.",
+ "normalized_hypothesis": "that's where they cut off the scottish chiefs and killed all hands.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "that is where they cut off the scottish chiefs and killed all hands",
+ "normalized_hypothesis": "that is where they cut off the scottish chiefs and killed all hands",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0210",
+ "category": "phonetic_balance",
+ "text": "And Tom King patiently endured.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0210.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "And Tom King patiently endured.",
+ "raw": {
+ "normalized_reference": "and tom king patiently endured.",
+ "normalized_hypothesis": "and tom king patiently endured.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "and tom king patiently endured",
+ "normalized_hypothesis": "and tom king patiently endured",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0211",
+ "category": "phonetic_balance",
+ "text": "From that moment his friendship for Belize turns to hatred and jealousy.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0211.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "From that moment his friendship for Belize turns to hatred and jealousy.",
+ "raw": {
+ "normalized_reference": "from that moment his friendship for belize turns to hatred and jealousy.",
+ "normalized_hypothesis": "from that moment his friendship for belize turns to hatred and jealousy.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "from that moment his friendship for belize turns to hatred and jealousy",
+ "normalized_hypothesis": "from that moment his friendship for belize turns to hatred and jealousy",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0212",
+ "category": "phonetic_balance",
+ "text": "Then you don't believe in altruism.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0212.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Then you don't believe in altruism.",
+ "raw": {
+ "normalized_reference": "then you don't believe in altruism.",
+ "normalized_hypothesis": "then you don't believe in altruism.",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "then you do not believe in altruism",
+ "normalized_hypothesis": "then you do not believe in altruism",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0213",
+ "category": "phonetic_balance",
+ "text": "What Pascal glimpsed with the vision of a seer, I have lived.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0213.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "What Pascal glimpsed with the vision of a seer I have lived.",
+ "raw": {
+ "normalized_reference": "what pascal glimpsed with the vision of a seer, i have lived.",
+ "normalized_hypothesis": "what pascal glimpsed with the vision of a seer i have lived.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "what pascal glimpsed with the vision of a seer i have lived",
+ "normalized_hypothesis": "what pascal glimpsed with the vision of a seer i have lived",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0214",
+ "category": "phonetic_balance",
+ "text": "You were making them talk shop, Ruth charged him.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0214.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "You were making them talk shop. Ruth charged him.",
+ "raw": {
+ "normalized_reference": "you were making them talk shop, ruth charged him.",
+ "normalized_hypothesis": "you were making them talk shop. ruth charged him.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "you were making them talk shop ruth charged him",
+ "normalized_hypothesis": "you were making them talk shop ruth charged him",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0215",
+ "category": "phonetic_balance",
+ "text": "Philip snatched at the letter which Gregson held out to him.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0215.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Philip snatched at the letter which Gregson held out to him.",
+ "raw": {
+ "normalized_reference": "philip snatched at the letter which gregson held out to him.",
+ "normalized_hypothesis": "philip snatched at the letter which gregson held out to him.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "philip snatched at the letter which gregson held out to him",
+ "normalized_hypothesis": "philip snatched at the letter which gregson held out to him",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0216",
+ "category": "phonetic_balance",
+ "text": "For a time the exciting thrill of his adventure was gone.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0216.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "For a time the exciting thrill of his adventure was gone.",
+ "raw": {
+ "normalized_reference": "for a time the exciting thrill of his adventure was gone.",
+ "normalized_hypothesis": "for a time the exciting thrill of his adventure was gone.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "for a time the exciting thrill of his adventure was gone",
+ "normalized_hypothesis": "for a time the exciting thrill of his adventure was gone",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0217",
+ "category": "phonetic_balance",
+ "text": "Obviously, it was a disease that could be contracted by contact.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0217.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Obviously, it was a disease that could be contracted by contact.",
+ "raw": {
+ "normalized_reference": "obviously, it was a disease that could be contracted by contact.",
+ "normalized_hypothesis": "obviously, it was a disease that could be contracted by contact.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "obviously it was a disease that could be contracted by contact",
+ "normalized_hypothesis": "obviously it was a disease that could be contracted by contact",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0218",
+ "category": "phonetic_balance",
+ "text": "Blind with rage, he darted in.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0218.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Blind with rage, he darted in.",
+ "raw": {
+ "normalized_reference": "blind with rage, he darted in.",
+ "normalized_hypothesis": "blind with rage, he darted in.",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "blind with rage he darted in",
+ "normalized_hypothesis": "blind with rage he darted in",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0219",
+ "category": "phonetic_balance",
+ "text": "You live on an income which your father earned.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0219.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "You live on an income which your father earned.",
+ "raw": {
+ "normalized_reference": "you live on an income which your father earned.",
+ "normalized_hypothesis": "you live on an income which your father earned.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "you live on an income which your father earned",
+ "normalized_hypothesis": "you live on an income which your father earned",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0220",
+ "category": "phonetic_balance",
+ "text": "Something about it seemed to fascinate him, to challenge his presence.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0220.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Something about it seemed to fascinate him, to challenge his presence.",
+ "raw": {
+ "normalized_reference": "something about it seemed to fascinate him, to challenge his presence.",
+ "normalized_hypothesis": "something about it seemed to fascinate him, to challenge his presence.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "something about it seemed to fascinate him to challenge his presence",
+ "normalized_hypothesis": "something about it seemed to fascinate him to challenge his presence",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0221",
+ "category": "phonetic_balance",
+ "text": "Her beautiful hair was done up in shining coils.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0221.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Her beautiful hair was done up in shining coils.",
+ "raw": {
+ "normalized_reference": "her beautiful hair was done up in shining coils.",
+ "normalized_hypothesis": "her beautiful hair was done up in shining coils.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "her beautiful hair was done up in shining coils",
+ "normalized_hypothesis": "her beautiful hair was done up in shining coils",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0222",
+ "category": "phonetic_balance",
+ "text": "It does, was her audacious answer.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0222.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It does was her audacious answer.",
+ "raw": {
+ "normalized_reference": "it does, was her audacious answer.",
+ "normalized_hypothesis": "it does was her audacious answer.",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it does was her audacious answer",
+ "normalized_hypothesis": "it does was her audacious answer",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0223",
+ "category": "phonetic_balance",
+ "text": "He caught himself with a jerk.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0223.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He caught himself with a jerk.",
+ "raw": {
+ "normalized_reference": "he caught himself with a jerk.",
+ "normalized_hypothesis": "he caught himself with a jerk.",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he caught himself with a jerk",
+ "normalized_hypothesis": "he caught himself with a jerk",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0224",
+ "category": "phonetic_balance",
+ "text": "He did not believe in the burning of daylight for such a luxury.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0224.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He did not believe in the burning of daylight for such a luxury.",
+ "raw": {
+ "normalized_reference": "he did not believe in the burning of daylight for such a luxury.",
+ "normalized_hypothesis": "he did not believe in the burning of daylight for such a luxury.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he did not believe in the burning of daylight for such a luxury",
+ "normalized_hypothesis": "he did not believe in the burning of daylight for such a luxury",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0225",
+ "category": "phonetic_balance",
+ "text": "He had comparatively no advantages at first.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0225.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He had comparatively no advantages at first.",
+ "raw": {
+ "normalized_reference": "he had comparatively no advantages at first.",
+ "normalized_hypothesis": "he had comparatively no advantages at first.",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he had comparatively no advantages at 1st",
+ "normalized_hypothesis": "he had comparatively no advantages at 1st",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0226",
+ "category": "phonetic_balance",
+ "text": "Such men believe, when they come together.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0226.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Such men believe when they come together.",
+ "raw": {
+ "normalized_reference": "such men believe, when they come together.",
+ "normalized_hypothesis": "such men believe when they come together.",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "such men believe when they come together",
+ "normalized_hypothesis": "such men believe when they come together",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0227",
+ "category": "phonetic_balance",
+ "text": "The ship should be in within a week or ten days.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0227.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The ship should be in within a week or ten days.",
+ "raw": {
+ "normalized_reference": "the ship should be in within a week or ten days.",
+ "normalized_hypothesis": "the ship should be in within a week or ten days.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the ship should be in within a week or 10 days",
+ "normalized_hypothesis": "the ship should be in within a week or 10 days",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0228",
+ "category": "phonetic_balance",
+ "text": "Such a dog the wise driver kills, or turns loose.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0228.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Such a dog the wise driver kills or turns loose.",
+ "raw": {
+ "normalized_reference": "such a dog the wise driver kills, or turns loose.",
+ "normalized_hypothesis": "such a dog the wise driver kills or turns loose.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "such a dog the wise driver kills or turns loose",
+ "normalized_hypothesis": "such a dog the wise driver kills or turns loose",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0229",
+ "category": "phonetic_balance",
+ "text": "You have heard always how he was the lover of the Princess Naomi.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0229.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "You have heard always how he was the lover of the princess Naomi.",
+ "raw": {
+ "normalized_reference": "you have heard always how he was the lover of the princess naomi.",
+ "normalized_hypothesis": "you have heard always how he was the lover of the princess naomi.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "you have heard always how he was the lover of the princess naomi",
+ "normalized_hypothesis": "you have heard always how he was the lover of the princess naomi",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0230",
+ "category": "phonetic_balance",
+ "text": "I want to know how all this is possible.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0230.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I want to know how all this is possible.",
+ "raw": {
+ "normalized_reference": "i want to know how all this is possible.",
+ "normalized_hypothesis": "i want to know how all this is possible.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i want to know how all this is possible",
+ "normalized_hypothesis": "i want to know how all this is possible",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0231",
+ "category": "phonetic_balance",
+ "text": "It was my idea to a tee.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0231.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It was my idea to a T.",
+ "raw": {
+ "normalized_reference": "it was my idea to a tee.",
+ "normalized_hypothesis": "it was my idea to a t.",
+ "errors": 1,
+ "words": 7,
+ "wer": 0.14285714285714285
+ },
+ "semantic": {
+ "normalized_reference": "it was my idea to a tee",
+ "normalized_hypothesis": "it was my idea to a t",
+ "errors": 1,
+ "words": 7,
+ "wer": 0.14285714285714285
+ }
+ },
+ {
+ "id": "release500_0232",
+ "category": "phonetic_balance",
+ "text": "They likewise are disinclined to being eaten.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0232.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "They likewise are disinclined to being eaten.",
+ "raw": {
+ "normalized_reference": "they likewise are disinclined to being eaten.",
+ "normalized_hypothesis": "they likewise are disinclined to being eaten.",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "they likewise are disinclined to being eaten",
+ "normalized_hypothesis": "they likewise are disinclined to being eaten",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0233",
+ "category": "phonetic_balance",
+ "text": "This state of mind comes of an undue prominence of the ego.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0233.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "This state of mind comes of an undue prominence of the ego.",
+ "raw": {
+ "normalized_reference": "this state of mind comes of an undue prominence of the ego.",
+ "normalized_hypothesis": "this state of mind comes of an undue prominence of the ego.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "this state of mind comes of an undue prominence of the ego",
+ "normalized_hypothesis": "this state of mind comes of an undue prominence of the ego",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0234",
+ "category": "phonetic_balance",
+ "text": "Then it is as I said, Womble announced with finality.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0234.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Then it is as I said, Womble announced with finality.",
+ "raw": {
+ "normalized_reference": "then it is as i said, womble announced with finality.",
+ "normalized_hypothesis": "then it is as i said, womble announced with finality.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "then it is as i said womble announced with finality",
+ "normalized_hypothesis": "then it is as i said womble announced with finality",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0235",
+ "category": "phonetic_balance",
+ "text": "I can see that knife now.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0235.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I can see that knife now.",
+ "raw": {
+ "normalized_reference": "i can see that knife now.",
+ "normalized_hypothesis": "i can see that knife now.",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i can see that knife now",
+ "normalized_hypothesis": "i can see that knife now",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0236",
+ "category": "phonetic_balance",
+ "text": "If I meet her again I shall apologize, said Eileen.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0236.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "If I meet her again I shall apologize, said Eileen.",
+ "raw": {
+ "normalized_reference": "if i meet her again i shall apologize, said eileen.",
+ "normalized_hypothesis": "if i meet her again i shall apologize, said eileen.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "if i meet her again i shall apologize said eileen",
+ "normalized_hypothesis": "if i meet her again i shall apologize said eileen",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0237",
+ "category": "phonetic_balance",
+ "text": "A wildly exciting time was his during the week preceding Thursday the eighteenth.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0237.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "A wildly exciting time was his during the week preceding Thursday the 18th.",
+ "raw": {
+ "normalized_reference": "a wildly exciting time was his during the week preceding thursday the eighteenth.",
+ "normalized_hypothesis": "a wildly exciting time was his during the week preceding thursday the 18th.",
+ "errors": 1,
+ "words": 13,
+ "wer": 0.07692307692307693
+ },
+ "semantic": {
+ "normalized_reference": "a wildly exciting time was his during the week preceding thursday the 18th",
+ "normalized_hypothesis": "a wildly exciting time was his during the week preceding thursday the 18th",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0238",
+ "category": "phonetic_balance",
+ "text": "In what bucolic school of fence he had been taught was beyond imagining.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0238.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "In what bucolic school of fence he had been taught was beyond imagining.",
+ "raw": {
+ "normalized_reference": "in what bucolic school of fence he had been taught was beyond imagining.",
+ "normalized_hypothesis": "in what bucolic school of fence he had been taught was beyond imagining.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "in what bucolic school of fence he had been taught was beyond imagining",
+ "normalized_hypothesis": "in what bucolic school of fence he had been taught was beyond imagining",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0239",
+ "category": "phonetic_balance",
+ "text": "So she said, the irate skipper dashed on.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0239.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "So she said, the irate skipper dashed on.",
+ "raw": {
+ "normalized_reference": "so she said, the irate skipper dashed on.",
+ "normalized_hypothesis": "so she said, the irate skipper dashed on.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "so she said the irate skipper dashed on",
+ "normalized_hypothesis": "so she said the irate skipper dashed on",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0240",
+ "category": "phonetic_balance",
+ "text": "It is not an attempt to smash the market.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0240.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It is not an attempt to smash the market.",
+ "raw": {
+ "normalized_reference": "it is not an attempt to smash the market.",
+ "normalized_hypothesis": "it is not an attempt to smash the market.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it is not an attempt to smash the market",
+ "normalized_hypothesis": "it is not an attempt to smash the market",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0241",
+ "category": "phonetic_balance",
+ "text": "He destroyed everything that had belonged to the woman.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0241.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He destroyed everything that had belonged to the woman.",
+ "raw": {
+ "normalized_reference": "he destroyed everything that had belonged to the woman.",
+ "normalized_hypothesis": "he destroyed everything that had belonged to the woman.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he destroyed everything that had belonged to the woman",
+ "normalized_hypothesis": "he destroyed everything that had belonged to the woman",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0242",
+ "category": "phonetic_balance",
+ "text": "Did Thorpe go to see any one in Churchill.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0242.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Did Philip go to see anyone in Churchill?",
+ "raw": {
+ "normalized_reference": "did thorpe go to see any one in churchill.",
+ "normalized_hypothesis": "did philip go to see anyone in churchill?",
+ "errors": 3,
+ "words": 9,
+ "wer": 0.3333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "did thorpe go to see any one in churchill",
+ "normalized_hypothesis": "did philip go to see anyone in churchill",
+ "errors": 3,
+ "words": 9,
+ "wer": 0.3333333333333333
+ }
+ },
+ {
+ "id": "release500_0243",
+ "category": "phonetic_balance",
+ "text": "Many other similar disconcerting omissions will be noticed in the Manuscript.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0243.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Many other similar disconcerting omissions will be noticed in the manuscript.",
+ "raw": {
+ "normalized_reference": "many other similar disconcerting omissions will be noticed in the manuscript.",
+ "normalized_hypothesis": "many other similar disconcerting omissions will be noticed in the manuscript.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "many other similar disconcerting omissions will be noticed in the manuscript",
+ "normalized_hypothesis": "many other similar disconcerting omissions will be noticed in the manuscript",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0244",
+ "category": "phonetic_balance",
+ "text": "The ringing of the big bell aroused him.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0244.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The ringing of the big bell aroused him.",
+ "raw": {
+ "normalized_reference": "the ringing of the big bell aroused him.",
+ "normalized_hypothesis": "the ringing of the big bell aroused him.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the ringing of the big bell aroused him",
+ "normalized_hypothesis": "the ringing of the big bell aroused him",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0245",
+ "category": "phonetic_balance",
+ "text": "A sinewy hand, dripping with water, was clutching the rail.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0245.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "A sinee hand, dripping with water, was clutching the rail.",
+ "raw": {
+ "normalized_reference": "a sinewy hand, dripping with water, was clutching the rail.",
+ "normalized_hypothesis": "a sinee hand, dripping with water, was clutching the rail.",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ },
+ "semantic": {
+ "normalized_reference": "a sinewy hand dripping with water was clutching the rail",
+ "normalized_hypothesis": "a sinee hand dripping with water was clutching the rail",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ }
+ },
+ {
+ "id": "release500_0246",
+ "category": "phonetic_balance",
+ "text": "Enters now the psychology of the situation.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0246.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "enters now the psychology of the situation.",
+ "raw": {
+ "normalized_reference": "enters now the psychology of the situation.",
+ "normalized_hypothesis": "enters now the psychology of the situation.",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "enters now the psychology of the situation",
+ "normalized_hypothesis": "enters now the psychology of the situation",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0247",
+ "category": "phonetic_balance",
+ "text": "Wash your hands of me.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0247.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Wash your hands of me.",
+ "raw": {
+ "normalized_reference": "wash your hands of me.",
+ "normalized_hypothesis": "wash your hands of me.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "wash your hands of me",
+ "normalized_hypothesis": "wash your hands of me",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0248",
+ "category": "phonetic_balance",
+ "text": "He confessed that the sketch had startled him.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0248.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He confessed that the sketch had startled him.",
+ "raw": {
+ "normalized_reference": "he confessed that the sketch had startled him.",
+ "normalized_hypothesis": "he confessed that the sketch had startled him.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he confessed that the sketch had startled him",
+ "normalized_hypothesis": "he confessed that the sketch had startled him",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0249",
+ "category": "phonetic_balance",
+ "text": "Her face was against his breast.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0249.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Her face was against his breast.",
+ "raw": {
+ "normalized_reference": "her face was against his breast.",
+ "normalized_hypothesis": "her face was against his breast.",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "her face was against his breast",
+ "normalized_hypothesis": "her face was against his breast",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0250",
+ "category": "phonetic_balance",
+ "text": "He was a pariah; a wanderer without a friend or a home.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0250.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He was a pariah, a wanderer without a friend or a home.",
+ "raw": {
+ "normalized_reference": "he was a pariah; a wanderer without a friend or a home.",
+ "normalized_hypothesis": "he was a pariah, a wanderer without a friend or a home.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he was a pariah a wanderer without a friend or a home",
+ "normalized_hypothesis": "he was a pariah a wanderer without a friend or a home",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0251",
+ "category": "phonetic_balance",
+ "text": "A cry of joy burst from Philip's lips.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0251.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "A cry of joy burst from Philip's lips.",
+ "raw": {
+ "normalized_reference": "a cry of joy burst from philip's lips.",
+ "normalized_hypothesis": "a cry of joy burst from philip's lips.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "a cry of joy burst from philip is lips",
+ "normalized_hypothesis": "a cry of joy burst from philip is lips",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0252",
+ "category": "phonetic_balance",
+ "text": "And the big Persian knew of his existence before he did of hers.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0252.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "And the big Persian knew of his existence before he did of hers.",
+ "raw": {
+ "normalized_reference": "and the big persian knew of his existence before he did of hers.",
+ "normalized_hypothesis": "and the big persian knew of his existence before he did of hers.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "and the big persian knew of his existence before he did of hers",
+ "normalized_hypothesis": "and the big persian knew of his existence before he did of hers",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0253",
+ "category": "phonetic_balance",
+ "text": "The men stared into each other's face.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0253.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The men stared into each other's face.",
+ "raw": {
+ "normalized_reference": "the men stared into each other's face.",
+ "normalized_hypothesis": "the men stared into each other's face.",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the men stared into each other is face",
+ "normalized_hypothesis": "the men stared into each other is face",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0254",
+ "category": "phonetic_balance",
+ "text": "It is the fire, partly, she said.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0254.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It is the fire, partly, she said.",
+ "raw": {
+ "normalized_reference": "it is the fire, partly, she said.",
+ "normalized_hypothesis": "it is the fire, partly, she said.",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it is the fire partly she said",
+ "normalized_hypothesis": "it is the fire partly she said",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0255",
+ "category": "phonetic_balance",
+ "text": "It seemed nearer to him since he had seen and talked with Gregson.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0255.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It seemed nearer to him since he had seen and talked with Gregson.",
+ "raw": {
+ "normalized_reference": "it seemed nearer to him since he had seen and talked with gregson.",
+ "normalized_hypothesis": "it seemed nearer to him since he had seen and talked with gregson.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it seemed nearer to him since he had seen and talked with gregson",
+ "normalized_hypothesis": "it seemed nearer to him since he had seen and talked with gregson",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0256",
+ "category": "phonetic_balance",
+ "text": "Among my minor afflictions, I may mention a new and mysterious one.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0256.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Among my minor afflictions, I may mention a new and mysterious one.",
+ "raw": {
+ "normalized_reference": "among my minor afflictions, i may mention a new and mysterious one.",
+ "normalized_hypothesis": "among my minor afflictions, i may mention a new and mysterious one.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "among my minor afflictions i may mention a new and mysterious one",
+ "normalized_hypothesis": "among my minor afflictions i may mention a new and mysterious one",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0257",
+ "category": "phonetic_balance",
+ "text": "Between him and all domestic animals there must be no hostilities.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0257.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Between him and all domestic animals there must be no hostilities.",
+ "raw": {
+ "normalized_reference": "between him and all domestic animals there must be no hostilities.",
+ "normalized_hypothesis": "between him and all domestic animals there must be no hostilities.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "between him and all domestic animals there must be no hostilities",
+ "normalized_hypothesis": "between him and all domestic animals there must be no hostilities",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0258",
+ "category": "phonetic_balance",
+ "text": "With these arguments he convinced himself that he should go on alone.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0258.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "With these arguments he convinced himself that he should go on alone.",
+ "raw": {
+ "normalized_reference": "with these arguments he convinced himself that he should go on alone.",
+ "normalized_hypothesis": "with these arguments he convinced himself that he should go on alone.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "with these arguments he convinced himself that he should go on alone",
+ "normalized_hypothesis": "with these arguments he convinced himself that he should go on alone",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0259",
+ "category": "phonetic_balance",
+ "text": "We are consumed in our own flesh-pots.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0259.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "We are consumed in our own flesh pots.",
+ "raw": {
+ "normalized_reference": "we are consumed in our own flesh-pots.",
+ "normalized_hypothesis": "we are consumed in our own flesh pots.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "we are consumed in our own flesh pots",
+ "normalized_hypothesis": "we are consumed in our own flesh pots",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0260",
+ "category": "phonetic_balance",
+ "text": "Mercedes screamed, cried, laughed, and manifested the chaotic abandonment of hysteria.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0260.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Mercedes screamed, cried, laughed, and manifested the chaotic abandonment of hysteria.",
+ "raw": {
+ "normalized_reference": "mercedes screamed, cried, laughed, and manifested the chaotic abandonment of hysteria.",
+ "normalized_hypothesis": "mercedes screamed, cried, laughed, and manifested the chaotic abandonment of hysteria.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "mercedes screamed cried laughed and manifested the chaotic abandonment of hysteria",
+ "normalized_hypothesis": "mercedes screamed cried laughed and manifested the chaotic abandonment of hysteria",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0261",
+ "category": "phonetic_balance",
+ "text": "Philip made no effort to follow.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0261.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Philip made no effort to follow.",
+ "raw": {
+ "normalized_reference": "philip made no effort to follow.",
+ "normalized_hypothesis": "philip made no effort to follow.",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "philip made no effort to follow",
+ "normalized_hypothesis": "philip made no effort to follow",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0262",
+ "category": "phonetic_balance",
+ "text": "She said with chattering teeth.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0262.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "She said with chattering teeth.",
+ "raw": {
+ "normalized_reference": "she said with chattering teeth.",
+ "normalized_hypothesis": "she said with chattering teeth.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "she said with chattering teeth",
+ "normalized_hypothesis": "she said with chattering teeth",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0263",
+ "category": "phonetic_balance",
+ "text": "Halfway around the track one donkey got into an argument with its rider.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0263.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Halfway around the track one donkey got into an argument with its rider.",
+ "raw": {
+ "normalized_reference": "halfway around the track one donkey got into an argument with its rider.",
+ "normalized_hypothesis": "halfway around the track one donkey got into an argument with its rider.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "halfway around the track one donkey got into an argument with its rider",
+ "normalized_hypothesis": "halfway around the track one donkey got into an argument with its rider",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0264",
+ "category": "phonetic_balance",
+ "text": "Earth and gravel seemed to fill the pan.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0264.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Earth and gravel seemed to fill the pan.",
+ "raw": {
+ "normalized_reference": "earth and gravel seemed to fill the pan.",
+ "normalized_hypothesis": "earth and gravel seemed to fill the pan.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "earth and gravel seemed to fill the pan",
+ "normalized_hypothesis": "earth and gravel seemed to fill the pan",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0265",
+ "category": "phonetic_balance",
+ "text": "Her achievements with cocoanuts were a revelation.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0265.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Her achievements with coconuts were a revelation.",
+ "raw": {
+ "normalized_reference": "her achievements with cocoanuts were a revelation.",
+ "normalized_hypothesis": "her achievements with coconuts were a revelation.",
+ "errors": 1,
+ "words": 7,
+ "wer": 0.14285714285714285
+ },
+ "semantic": {
+ "normalized_reference": "her achievements with cocoanuts were a revelation",
+ "normalized_hypothesis": "her achievements with coconuts were a revelation",
+ "errors": 1,
+ "words": 7,
+ "wer": 0.14285714285714285
+ }
+ },
+ {
+ "id": "release500_0266",
+ "category": "phonetic_balance",
+ "text": "For a full minute he crouched and listened.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0266.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "For a full minute he crouched and listened.",
+ "raw": {
+ "normalized_reference": "for a full minute he crouched and listened.",
+ "normalized_hypothesis": "for a full minute he crouched and listened.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "for a full minute he crouched and listened",
+ "normalized_hypothesis": "for a full minute he crouched and listened",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0267",
+ "category": "phonetic_balance",
+ "text": "It was edged with ice.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0267.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It was edged with ice.",
+ "raw": {
+ "normalized_reference": "it was edged with ice.",
+ "normalized_hypothesis": "it was edged with ice.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it was edged with ice",
+ "normalized_hypothesis": "it was edged with ice",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0268",
+ "category": "phonetic_balance",
+ "text": "The Portuguese boy passed the Hawaiian.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0268.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The Portuguese boy passed the Hawaiian.",
+ "raw": {
+ "normalized_reference": "the portuguese boy passed the hawaiian.",
+ "normalized_hypothesis": "the portuguese boy passed the hawaiian.",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the portuguese boy passed the hawaiian",
+ "normalized_hypothesis": "the portuguese boy passed the hawaiian",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0269",
+ "category": "phonetic_balance",
+ "text": "But he did not broach it, preferring to mature it carefully.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0269.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "But he did not broach it, preferring to mature it carefully.",
+ "raw": {
+ "normalized_reference": "but he did not broach it, preferring to mature it carefully.",
+ "normalized_hypothesis": "but he did not broach it, preferring to mature it carefully.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "but he did not broach it preferring to mature it carefully",
+ "normalized_hypothesis": "but he did not broach it preferring to mature it carefully",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0270",
+ "category": "phonetic_balance",
+ "text": "Come on, Del Mar challenged.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0270.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Come on, Del Mar challenged.",
+ "raw": {
+ "normalized_reference": "come on, del mar challenged.",
+ "normalized_hypothesis": "come on, del mar challenged.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "come on del mar challenged",
+ "normalized_hypothesis": "come on del mar challenged",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0271",
+ "category": "phonetic_balance",
+ "text": "For a full half minute Jeanne looked at him without speaking.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0271.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "For a full half minute, Gene looked at him without speaking.",
+ "raw": {
+ "normalized_reference": "for a full half minute jeanne looked at him without speaking.",
+ "normalized_hypothesis": "for a full half minute, gene looked at him without speaking.",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ },
+ "semantic": {
+ "normalized_reference": "for a full half minute jeanne looked at him without speaking",
+ "normalized_hypothesis": "for a full half minute gene looked at him without speaking",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ }
+ },
+ {
+ "id": "release500_0272",
+ "category": "phonetic_balance",
+ "text": "He bore no grudges and had few enemies.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0272.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He bore no grudges and had few enemies.",
+ "raw": {
+ "normalized_reference": "he bore no grudges and had few enemies.",
+ "normalized_hypothesis": "he bore no grudges and had few enemies.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he bore no grudges and had few enemies",
+ "normalized_hypothesis": "he bore no grudges and had few enemies",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0273",
+ "category": "phonetic_balance",
+ "text": "I'll go over tomorrow afternoon.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0273.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I'll go over tomorrow afternoon.",
+ "raw": {
+ "normalized_reference": "i'll go over tomorrow afternoon.",
+ "normalized_hypothesis": "i'll go over tomorrow afternoon.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i will go over tomorrow afternoon",
+ "normalized_hypothesis": "i will go over tomorrow afternoon",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0274",
+ "category": "phonetic_balance",
+ "text": "Well, there are better men in Hawaii, that's all.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0274.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Well, there are better men in Hawaii, that's all.",
+ "raw": {
+ "normalized_reference": "well, there are better men in hawaii, that's all.",
+ "normalized_hypothesis": "well, there are better men in hawaii, that's all.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "well there are better men in hawaii that is all",
+ "normalized_hypothesis": "well there are better men in hawaii that is all",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0275",
+ "category": "phonetic_balance",
+ "text": "He was an athlete and a giant.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0275.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He was an athlete and a giant.",
+ "raw": {
+ "normalized_reference": "he was an athlete and a giant.",
+ "normalized_hypothesis": "he was an athlete and a giant.",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he was an athlete and a giant",
+ "normalized_hypothesis": "he was an athlete and a giant",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0276",
+ "category": "phonetic_balance",
+ "text": "McVeigh when he returns from a trip to Honolulu.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0276.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "McVeigh Way returns from a trip to Honolulu.",
+ "raw": {
+ "normalized_reference": "mcveigh when he returns from a trip to honolulu.",
+ "normalized_hypothesis": "mcveigh way returns from a trip to honolulu.",
+ "errors": 2,
+ "words": 9,
+ "wer": 0.2222222222222222
+ },
+ "semantic": {
+ "normalized_reference": "mcveigh when he returns from a trip to honolulu",
+ "normalized_hypothesis": "mcveigh way returns from a trip to honolulu",
+ "errors": 2,
+ "words": 9,
+ "wer": 0.2222222222222222
+ }
+ },
+ {
+ "id": "release500_0277",
+ "category": "phonetic_balance",
+ "text": "You're a devil for fighting, and will surely win.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0277.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "You're a devil for fighting, and will surely win.",
+ "raw": {
+ "normalized_reference": "you're a devil for fighting, and will surely win.",
+ "normalized_hypothesis": "you're a devil for fighting, and will surely win.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "you are a devil for fighting and will surely win",
+ "normalized_hypothesis": "you are a devil for fighting and will surely win",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0278",
+ "category": "phonetic_balance",
+ "text": "All an appearance can know is mirage.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0278.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "All an appearance can know is mirage.",
+ "raw": {
+ "normalized_reference": "all an appearance can know is mirage.",
+ "normalized_hypothesis": "all an appearance can know is mirage.",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "all an appearance can know is mirage",
+ "normalized_hypothesis": "all an appearance can know is mirage",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0279",
+ "category": "phonetic_balance",
+ "text": "You fired me out of your house, in short.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0279.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "You fired me out of your house, ensured.",
+ "raw": {
+ "normalized_reference": "you fired me out of your house, in short.",
+ "normalized_hypothesis": "you fired me out of your house, ensured.",
+ "errors": 2,
+ "words": 9,
+ "wer": 0.2222222222222222
+ },
+ "semantic": {
+ "normalized_reference": "you fired me out of your house in short",
+ "normalized_hypothesis": "you fired me out of your house ensured",
+ "errors": 2,
+ "words": 9,
+ "wer": 0.2222222222222222
+ }
+ },
+ {
+ "id": "release500_0280",
+ "category": "phonetic_balance",
+ "text": "He made sure that the magazine was loaded, and resumed his paddling.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0280.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He made sure that the magazine was loaded and resumed his paddling.",
+ "raw": {
+ "normalized_reference": "he made sure that the magazine was loaded, and resumed his paddling.",
+ "normalized_hypothesis": "he made sure that the magazine was loaded and resumed his paddling.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he made sure that the magazine was loaded and resumed his paddling",
+ "normalized_hypothesis": "he made sure that the magazine was loaded and resumed his paddling",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0281",
+ "category": "phonetic_balance",
+ "text": "Change chairs, Daylight commanded.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0281.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Change chairs, daylight commanded.",
+ "raw": {
+ "normalized_reference": "change chairs, daylight commanded.",
+ "normalized_hypothesis": "change chairs, daylight commanded.",
+ "errors": 0,
+ "words": 4,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "change chairs daylight commanded",
+ "normalized_hypothesis": "change chairs daylight commanded",
+ "errors": 0,
+ "words": 4,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0282",
+ "category": "phonetic_balance",
+ "text": "By that answer my professional medical prestige stood or fell.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0282.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "By that answer my professional medical prestige stood or fell.",
+ "raw": {
+ "normalized_reference": "by that answer my professional medical prestige stood or fell.",
+ "normalized_hypothesis": "by that answer my professional medical prestige stood or fell.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "by that answer my professional medical prestige stood or fell",
+ "normalized_hypothesis": "by that answer my professional medical prestige stood or fell",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0283",
+ "category": "phonetic_balance",
+ "text": "The moon had already begun its westward decline.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0283.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The moon had already begun its westward decline.",
+ "raw": {
+ "normalized_reference": "the moon had already begun its westward decline.",
+ "normalized_hypothesis": "the moon had already begun its westward decline.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the moon had already begun its westward decline",
+ "normalized_hypothesis": "the moon had already begun its westward decline",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0284",
+ "category": "phonetic_balance",
+ "text": "At sea, Monday, March 16, 1908.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0284.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "At sea, Monday, March 16, 1808.",
+ "raw": {
+ "normalized_reference": "at sea, monday, march 16, 1908.",
+ "normalized_hypothesis": "at sea, monday, march 16, 1808.",
+ "errors": 1,
+ "words": 6,
+ "wer": 0.16666666666666666
+ },
+ "semantic": {
+ "normalized_reference": "at sea monday march 16 1908",
+ "normalized_hypothesis": "at sea monday march 16 1808",
+ "errors": 1,
+ "words": 6,
+ "wer": 0.16666666666666666
+ }
+ },
+ {
+ "id": "release500_0285",
+ "category": "phonetic_balance",
+ "text": "And wherever I ranged, the way lay along alcohol-drenched roads.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0285.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "And wherever I ranged, the way lay along alcohol-drenched roads.",
+ "raw": {
+ "normalized_reference": "and wherever i ranged, the way lay along alcohol-drenched roads.",
+ "normalized_hypothesis": "and wherever i ranged, the way lay along alcohol-drenched roads.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "and wherever i ranged the way lay along alcohol drenched roads",
+ "normalized_hypothesis": "and wherever i ranged the way lay along alcohol drenched roads",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0286",
+ "category": "phonetic_balance",
+ "text": "The date was nearly eighteen years old.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0286.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The date was nearly 18 years old.",
+ "raw": {
+ "normalized_reference": "the date was nearly eighteen years old.",
+ "normalized_hypothesis": "the date was nearly 18 years old.",
+ "errors": 1,
+ "words": 7,
+ "wer": 0.14285714285714285
+ },
+ "semantic": {
+ "normalized_reference": "the date was nearly 18 years old",
+ "normalized_hypothesis": "the date was nearly 18 years old",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0287",
+ "category": "phonetic_balance",
+ "text": "Very early in my life, I separated from my mother.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0287.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Very early in my life, I separated from my mother.",
+ "raw": {
+ "normalized_reference": "very early in my life, i separated from my mother.",
+ "normalized_hypothesis": "very early in my life, i separated from my mother.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "very early in my life i separated from my mother",
+ "normalized_hypothesis": "very early in my life i separated from my mother",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0288",
+ "category": "phonetic_balance",
+ "text": "There was pride and strength, the ring of triumph in his voice.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0288.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "There was pride and strength, the ring of triumph in his voice.",
+ "raw": {
+ "normalized_reference": "there was pride and strength, the ring of triumph in his voice.",
+ "normalized_hypothesis": "there was pride and strength, the ring of triumph in his voice.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "there was pride and strength the ring of triumph in his voice",
+ "normalized_hypothesis": "there was pride and strength the ring of triumph in his voice",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0289",
+ "category": "phonetic_balance",
+ "text": "From my earliest recollection my sleep was a period of terror.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0289.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "From my earliest recollection, my sleep was a period of terror.",
+ "raw": {
+ "normalized_reference": "from my earliest recollection my sleep was a period of terror.",
+ "normalized_hypothesis": "from my earliest recollection, my sleep was a period of terror.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "from my earliest recollection my sleep was a period of terror",
+ "normalized_hypothesis": "from my earliest recollection my sleep was a period of terror",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0290",
+ "category": "phonetic_balance",
+ "text": "We'll have to watch our chances.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0290.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "We'll have to watch our chances.",
+ "raw": {
+ "normalized_reference": "we'll have to watch our chances.",
+ "normalized_hypothesis": "we'll have to watch our chances.",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "we will have to watch our chances",
+ "normalized_hypothesis": "we will have to watch our chances",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0291",
+ "category": "phonetic_balance",
+ "text": "Close beside him gleamed the white fangs of the wolf-dog.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0291.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Close beside him gleam the white fangs of the wolf-dog.",
+ "raw": {
+ "normalized_reference": "close beside him gleamed the white fangs of the wolf-dog.",
+ "normalized_hypothesis": "close beside him gleam the white fangs of the wolf-dog.",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ },
+ "semantic": {
+ "normalized_reference": "close beside him gleamed the white fangs of the wolf dog",
+ "normalized_hypothesis": "close beside him gleam the white fangs of the wolf dog",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ }
+ },
+ {
+ "id": "release500_0292",
+ "category": "phonetic_balance",
+ "text": "Would you be satisfied with that one hundredth part of me.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0292.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Would you be satisfied with that 100th part of me?",
+ "raw": {
+ "normalized_reference": "would you be satisfied with that one hundredth part of me.",
+ "normalized_hypothesis": "would you be satisfied with that 100th part of me?",
+ "errors": 2,
+ "words": 11,
+ "wer": 0.18181818181818182
+ },
+ "semantic": {
+ "normalized_reference": "would you be satisfied with that 100th part of me",
+ "normalized_hypothesis": "would you be satisfied with that 100th part of me",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0293",
+ "category": "phonetic_balance",
+ "text": "Two gigantic owls were tearing at the carcass.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0293.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Two gigantic owls were tearing at the carcass.",
+ "raw": {
+ "normalized_reference": "two gigantic owls were tearing at the carcass.",
+ "normalized_hypothesis": "two gigantic owls were tearing at the carcass.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "2 gigantic owls were tearing at the carcass",
+ "normalized_hypothesis": "2 gigantic owls were tearing at the carcass",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0294",
+ "category": "phonetic_balance",
+ "text": "The lush vegetation of that sheltered spot make a natural shield.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0294.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The lush vegetation of that sheltered spot make a natural shield.",
+ "raw": {
+ "normalized_reference": "the lush vegetation of that sheltered spot make a natural shield.",
+ "normalized_hypothesis": "the lush vegetation of that sheltered spot make a natural shield.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the lush vegetation of that sheltered spot make a natural shield",
+ "normalized_hypothesis": "the lush vegetation of that sheltered spot make a natural shield",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0295",
+ "category": "phonetic_balance",
+ "text": "Below him the shadow was broken into a pool of rippling starlight.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0295.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Below him the shadow was broken into a pool of rippling starlight.",
+ "raw": {
+ "normalized_reference": "below him the shadow was broken into a pool of rippling starlight.",
+ "normalized_hypothesis": "below him the shadow was broken into a pool of rippling starlight.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "below him the shadow was broken into a pool of rippling starlight",
+ "normalized_hypothesis": "below him the shadow was broken into a pool of rippling starlight",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0296",
+ "category": "phonetic_balance",
+ "text": "Lots of men take women buggy riding.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0296.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Lots of men take women buddy riding.",
+ "raw": {
+ "normalized_reference": "lots of men take women buggy riding.",
+ "normalized_hypothesis": "lots of men take women buddy riding.",
+ "errors": 1,
+ "words": 7,
+ "wer": 0.14285714285714285
+ },
+ "semantic": {
+ "normalized_reference": "lots of men take women buggy riding",
+ "normalized_hypothesis": "lots of men take women buddy riding",
+ "errors": 1,
+ "words": 7,
+ "wer": 0.14285714285714285
+ }
+ },
+ {
+ "id": "release500_0297",
+ "category": "phonetic_balance",
+ "text": "But it contributed to the smash.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0297.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "But it contributed to the splash.",
+ "raw": {
+ "normalized_reference": "but it contributed to the smash.",
+ "normalized_hypothesis": "but it contributed to the splash.",
+ "errors": 1,
+ "words": 6,
+ "wer": 0.16666666666666666
+ },
+ "semantic": {
+ "normalized_reference": "but it contributed to the smash",
+ "normalized_hypothesis": "but it contributed to the splash",
+ "errors": 1,
+ "words": 6,
+ "wer": 0.16666666666666666
+ }
+ },
+ {
+ "id": "release500_0298",
+ "category": "phonetic_balance",
+ "text": "No man ate of the seal meat or the oil.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0298.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "No man ate of the seal meat or the oil.",
+ "raw": {
+ "normalized_reference": "no man ate of the seal meat or the oil.",
+ "normalized_hypothesis": "no man ate of the seal meat or the oil.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "no man ate of the seal meat or the oil",
+ "normalized_hypothesis": "no man ate of the seal meat or the oil",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0299",
+ "category": "phonetic_balance",
+ "text": "You're going in for grab sharing.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0299.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "You're going in for grad sharing.",
+ "raw": {
+ "normalized_reference": "you're going in for grab sharing.",
+ "normalized_hypothesis": "you're going in for grad sharing.",
+ "errors": 1,
+ "words": 6,
+ "wer": 0.16666666666666666
+ },
+ "semantic": {
+ "normalized_reference": "you are going in for grab sharing",
+ "normalized_hypothesis": "you are going in for grad sharing",
+ "errors": 1,
+ "words": 7,
+ "wer": 0.14285714285714285
+ }
+ },
+ {
+ "id": "release500_0300",
+ "category": "phonetic_balance",
+ "text": "He it was that lived to found the family of the Patino.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0300.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It was that lift to found the family of the Patino.",
+ "raw": {
+ "normalized_reference": "he it was that lived to found the family of the patino.",
+ "normalized_hypothesis": "it was that lift to found the family of the patino.",
+ "errors": 2,
+ "words": 12,
+ "wer": 0.16666666666666666
+ },
+ "semantic": {
+ "normalized_reference": "he it was that lived to found the family of the patino",
+ "normalized_hypothesis": "it was that lift to found the family of the patino",
+ "errors": 2,
+ "words": 12,
+ "wer": 0.16666666666666666
+ }
+ },
+ {
+ "id": "release500_0301",
+ "category": "phonetic_balance",
+ "text": "For such countries nothing remained but reorganization.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0301.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "For such countries nothing remained but reorganization.",
+ "raw": {
+ "normalized_reference": "for such countries nothing remained but reorganization.",
+ "normalized_hypothesis": "for such countries nothing remained but reorganization.",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "for such countries nothing remained but reorganization",
+ "normalized_hypothesis": "for such countries nothing remained but reorganization",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0302",
+ "category": "phonetic_balance",
+ "text": "Stand off butcher and baker and all the rest.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0302.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Stand off Butcher and Baker and all the rest.",
+ "raw": {
+ "normalized_reference": "stand off butcher and baker and all the rest.",
+ "normalized_hypothesis": "stand off butcher and baker and all the rest.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "stand off butcher and baker and all the rest",
+ "normalized_hypothesis": "stand off butcher and baker and all the rest",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0303",
+ "category": "phonetic_balance",
+ "text": "All eyes, however, were staring at him in certitude of expectancy.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0303.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "All eyes, however, were staring at him in certitude of expectancy.",
+ "raw": {
+ "normalized_reference": "all eyes, however, were staring at him in certitude of expectancy.",
+ "normalized_hypothesis": "all eyes, however, were staring at him in certitude of expectancy.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "all eyes however were staring at him in certitude of expectancy",
+ "normalized_hypothesis": "all eyes however were staring at him in certitude of expectancy",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0304",
+ "category": "phonetic_balance",
+ "text": "Two days ago Jeanne learned where her father's men were hiding.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0304.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Two days ago, Jean learned where her father's men were hiding.",
+ "raw": {
+ "normalized_reference": "two days ago jeanne learned where her father's men were hiding.",
+ "normalized_hypothesis": "two days ago, jean learned where her father's men were hiding.",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ },
+ "semantic": {
+ "normalized_reference": "2 days ago jeanne learned where her father is men were hiding",
+ "normalized_hypothesis": "2 days ago jean learned where her father is men were hiding",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ }
+ },
+ {
+ "id": "release500_0305",
+ "category": "phonetic_balance",
+ "text": "His previous wives had never lived long enough to bear him children.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0305.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "as previous wives had never lived long enough to bear him children.",
+ "raw": {
+ "normalized_reference": "his previous wives had never lived long enough to bear him children.",
+ "normalized_hypothesis": "as previous wives had never lived long enough to bear him children.",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "his previous wives had never lived long enough to bear him children",
+ "normalized_hypothesis": "as previous wives had never lived long enough to bear him children",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ }
+ },
+ {
+ "id": "release500_0306",
+ "category": "phonetic_balance",
+ "text": "But this little defect did not worry him.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0306.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "But this little defect did not worry him.",
+ "raw": {
+ "normalized_reference": "but this little defect did not worry him.",
+ "normalized_hypothesis": "but this little defect did not worry him.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "but this little defect did not worry him",
+ "normalized_hypothesis": "but this little defect did not worry him",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0307",
+ "category": "phonetic_balance",
+ "text": "Eli Harding asked, as Shunk started to follow.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0307.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Eli Harding asked as Shunk started to follow.",
+ "raw": {
+ "normalized_reference": "eli harding asked, as shunk started to follow.",
+ "normalized_hypothesis": "eli harding asked as shunk started to follow.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "eli harding asked as shunk started to follow",
+ "normalized_hypothesis": "eli harding asked as shunk started to follow",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0308",
+ "category": "phonetic_balance",
+ "text": "And this was their sole conversation throughout the meal.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0308.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "And this was their sole conversation throughout the meal.",
+ "raw": {
+ "normalized_reference": "and this was their sole conversation throughout the meal.",
+ "normalized_hypothesis": "and this was their sole conversation throughout the meal.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "and this was their sole conversation throughout the meal",
+ "normalized_hypothesis": "and this was their sole conversation throughout the meal",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0309",
+ "category": "phonetic_balance",
+ "text": "Then, again, Tudor had such an irritating way about him.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0309.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Then, again, Tudor had such an irritating way about him.",
+ "raw": {
+ "normalized_reference": "then, again, tudor had such an irritating way about him.",
+ "normalized_hypothesis": "then, again, tudor had such an irritating way about him.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "then again tudor had such an irritating way about him",
+ "normalized_hypothesis": "then again tudor had such an irritating way about him",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0310",
+ "category": "phonetic_balance",
+ "text": "In it there was something that was almost tragedy.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0310.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "In it there was something that was almost tragedy.",
+ "raw": {
+ "normalized_reference": "in it there was something that was almost tragedy.",
+ "normalized_hypothesis": "in it there was something that was almost tragedy.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "in it there was something that was almost tragedy",
+ "normalized_hypothesis": "in it there was something that was almost tragedy",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0311",
+ "category": "phonetic_balance",
+ "text": "Thirty pounds, said the captain with finality.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0311.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Thirty pounds, said the captain with finality.",
+ "raw": {
+ "normalized_reference": "thirty pounds, said the captain with finality.",
+ "normalized_hypothesis": "thirty pounds, said the captain with finality.",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "\u00a330 said the captain with finality",
+ "normalized_hypothesis": "\u00a330 said the captain with finality",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0312",
+ "category": "phonetic_balance",
+ "text": "It was over when he made his way through the ring of spectators.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0312.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It was over when he made his way through the ring of spectators.",
+ "raw": {
+ "normalized_reference": "it was over when he made his way through the ring of spectators.",
+ "normalized_hypothesis": "it was over when he made his way through the ring of spectators.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it was over when he made his way through the ring of spectators",
+ "normalized_hypothesis": "it was over when he made his way through the ring of spectators",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0313",
+ "category": "phonetic_balance",
+ "text": "She had died from cold and starvation.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0313.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "She had died from cold and starvation.",
+ "raw": {
+ "normalized_reference": "she had died from cold and starvation.",
+ "normalized_hypothesis": "she had died from cold and starvation.",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "she had died from cold and starvation",
+ "normalized_hypothesis": "she had died from cold and starvation",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0314",
+ "category": "phonetic_balance",
+ "text": "We fished sharks on Niihau together.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0314.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "We fished sharks on Neha together.",
+ "raw": {
+ "normalized_reference": "we fished sharks on niihau together.",
+ "normalized_hypothesis": "we fished sharks on neha together.",
+ "errors": 1,
+ "words": 6,
+ "wer": 0.16666666666666666
+ },
+ "semantic": {
+ "normalized_reference": "we fished sharks on niihau together",
+ "normalized_hypothesis": "we fished sharks on neha together",
+ "errors": 1,
+ "words": 6,
+ "wer": 0.16666666666666666
+ }
+ },
+ {
+ "id": "release500_0315",
+ "category": "phonetic_balance",
+ "text": "His slim hands gripped the edges of the table.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0315.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "A slim hand gripped the edges of the table.",
+ "raw": {
+ "normalized_reference": "his slim hands gripped the edges of the table.",
+ "normalized_hypothesis": "a slim hand gripped the edges of the table.",
+ "errors": 2,
+ "words": 9,
+ "wer": 0.2222222222222222
+ },
+ "semantic": {
+ "normalized_reference": "his slim hands gripped the edges of the table",
+ "normalized_hypothesis": "a slim hand gripped the edges of the table",
+ "errors": 2,
+ "words": 9,
+ "wer": 0.2222222222222222
+ }
+ },
+ {
+ "id": "release500_0316",
+ "category": "phonetic_balance",
+ "text": "We had been chased by them ourselves, more than once.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0316.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "We had been chased by them ourselves more than once.",
+ "raw": {
+ "normalized_reference": "we had been chased by them ourselves, more than once.",
+ "normalized_hypothesis": "we had been chased by them ourselves more than once.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "we had been chased by them ourselves more than once",
+ "normalized_hypothesis": "we had been chased by them ourselves more than once",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0317",
+ "category": "phonetic_balance",
+ "text": "Or have they already devised one.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0317.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Or have they already devised one?",
+ "raw": {
+ "normalized_reference": "or have they already devised one.",
+ "normalized_hypothesis": "or have they already devised one?",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "or have they already devised one",
+ "normalized_hypothesis": "or have they already devised one",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0318",
+ "category": "phonetic_balance",
+ "text": "I cannot follow you, she said.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0318.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I cannot follow you, she said.",
+ "raw": {
+ "normalized_reference": "i cannot follow you, she said.",
+ "normalized_hypothesis": "i cannot follow you, she said.",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i cannot follow you she said",
+ "normalized_hypothesis": "i cannot follow you she said",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0319",
+ "category": "phonetic_balance",
+ "text": "He went down in midstream, searching the shadows of both shores.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0319.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He went down in midstream, searching the shadows of both shores.",
+ "raw": {
+ "normalized_reference": "he went down in midstream, searching the shadows of both shores.",
+ "normalized_hypothesis": "he went down in midstream, searching the shadows of both shores.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he went down in midstream searching the shadows of both shores",
+ "normalized_hypothesis": "he went down in midstream searching the shadows of both shores",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0320",
+ "category": "phonetic_balance",
+ "text": "I was sick once -- typhoid.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0320.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I was sick once typhoid.",
+ "raw": {
+ "normalized_reference": "i was sick once -- typhoid.",
+ "normalized_hypothesis": "i was sick once typhoid.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i was sick once typhoid",
+ "normalized_hypothesis": "i was sick once typhoid",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0321",
+ "category": "phonetic_balance",
+ "text": "Why not like any railroad station or ferry depot.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0321.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Why not like any railroad station or ferry depot?",
+ "raw": {
+ "normalized_reference": "why not like any railroad station or ferry depot.",
+ "normalized_hypothesis": "why not like any railroad station or ferry depot?",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "why not like any railroad station or ferry depot",
+ "normalized_hypothesis": "why not like any railroad station or ferry depot",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0322",
+ "category": "phonetic_balance",
+ "text": "Red-Eye swung back and forth on the branch farther down.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0322.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Red-Eye swung back and forth on the branch farther down.",
+ "raw": {
+ "normalized_reference": "red-eye swung back and forth on the branch farther down.",
+ "normalized_hypothesis": "red-eye swung back and forth on the branch farther down.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "red eye swung back and forth on the branch farther down",
+ "normalized_hypothesis": "red eye swung back and forth on the branch farther down",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0323",
+ "category": "phonetic_balance",
+ "text": "Within himself he called it no longer his own.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0323.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Within himself he called it no longer his own.",
+ "raw": {
+ "normalized_reference": "within himself he called it no longer his own.",
+ "normalized_hypothesis": "within himself he called it no longer his own.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "within himself he called it no longer his own",
+ "normalized_hypothesis": "within himself he called it no longer his own",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0324",
+ "category": "phonetic_balance",
+ "text": "It lasted as a deterrent for two days.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0324.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It lasted as a deterrent for two days.",
+ "raw": {
+ "normalized_reference": "it lasted as a deterrent for two days.",
+ "normalized_hypothesis": "it lasted as a deterrent for two days.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it lasted as a deterrent for 2 days",
+ "normalized_hypothesis": "it lasted as a deterrent for 2 days",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0325",
+ "category": "phonetic_balance",
+ "text": "He loved to play Chinese lottery.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0325.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He loved to play Chinese lottery.",
+ "raw": {
+ "normalized_reference": "he loved to play chinese lottery.",
+ "normalized_hypothesis": "he loved to play chinese lottery.",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he loved to play chinese lottery",
+ "normalized_hypothesis": "he loved to play chinese lottery",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0326",
+ "category": "phonetic_balance",
+ "text": "Lord, but I'm glad to see you again, Phil.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0326.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Lord, but I'm glad to see you again, Phil.",
+ "raw": {
+ "normalized_reference": "lord, but i'm glad to see you again, phil.",
+ "normalized_hypothesis": "lord, but i'm glad to see you again, phil.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "lord but i am glad to see you again phil",
+ "normalized_hypothesis": "lord but i am glad to see you again phil",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0327",
+ "category": "phonetic_balance",
+ "text": "There is not an iota of truth in it, certainly not.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0327.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "There is not an iota of truth in it. Certainly not.",
+ "raw": {
+ "normalized_reference": "there is not an iota of truth in it, certainly not.",
+ "normalized_hypothesis": "there is not an iota of truth in it. certainly not.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "there is not an iota of truth in it certainly not",
+ "normalized_hypothesis": "there is not an iota of truth in it certainly not",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0328",
+ "category": "phonetic_balance",
+ "text": "She added, with genuine sympathy in her face and voice.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0328.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "She added, with genuine sympathy in her face and voice.",
+ "raw": {
+ "normalized_reference": "she added, with genuine sympathy in her face and voice.",
+ "normalized_hypothesis": "she added, with genuine sympathy in her face and voice.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "she added with genuine sympathy in her face and voice",
+ "normalized_hypothesis": "she added with genuine sympathy in her face and voice",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0329",
+ "category": "phonetic_balance",
+ "text": "We have plenty of capital ourselves, and yet we want more.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0329.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "We have plenty of capital ourselves, and yet we want more.",
+ "raw": {
+ "normalized_reference": "we have plenty of capital ourselves, and yet we want more.",
+ "normalized_hypothesis": "we have plenty of capital ourselves, and yet we want more.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "we have plenty of capital ourselves and yet we want more",
+ "normalized_hypothesis": "we have plenty of capital ourselves and yet we want more",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0330",
+ "category": "phonetic_balance",
+ "text": "And now put yourself in my place for a moment.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0330.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "And now put yourself in my place for a moment.",
+ "raw": {
+ "normalized_reference": "and now put yourself in my place for a moment.",
+ "normalized_hypothesis": "and now put yourself in my place for a moment.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "and now put yourself in my place for a moment",
+ "normalized_hypothesis": "and now put yourself in my place for a moment",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0331",
+ "category": "phonetic_balance",
+ "text": "Fast, but endure.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0331.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Fast, but endure.",
+ "raw": {
+ "normalized_reference": "fast, but endure.",
+ "normalized_hypothesis": "fast, but endure.",
+ "errors": 0,
+ "words": 3,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "fast but endure",
+ "normalized_hypothesis": "fast but endure",
+ "errors": 0,
+ "words": 3,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0332",
+ "category": "phonetic_balance",
+ "text": "You were looking squeamish this afternoon, he began.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0332.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "You were looking squeamish this afternoon, he began.",
+ "raw": {
+ "normalized_reference": "you were looking squeamish this afternoon, he began.",
+ "normalized_hypothesis": "you were looking squeamish this afternoon, he began.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "you were looking squeamish this afternoon he began",
+ "normalized_hypothesis": "you were looking squeamish this afternoon he began",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0333",
+ "category": "phonetic_balance",
+ "text": "It is the nearest refuge.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0333.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It is the nearest refuge.",
+ "raw": {
+ "normalized_reference": "it is the nearest refuge.",
+ "normalized_hypothesis": "it is the nearest refuge.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it is the nearest refuge",
+ "normalized_hypothesis": "it is the nearest refuge",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0334",
+ "category": "phonetic_balance",
+ "text": "I want to die in it.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0334.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I want to die in it.",
+ "raw": {
+ "normalized_reference": "i want to die in it.",
+ "normalized_hypothesis": "i want to die in it.",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i want to die in it",
+ "normalized_hypothesis": "i want to die in it",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0335",
+ "category": "phonetic_balance",
+ "text": "He stopped, and Philip nodded at the horrified question in his eyes.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0335.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He stopped, and Illa nodded at the horrified question in his eyes.",
+ "raw": {
+ "normalized_reference": "he stopped, and philip nodded at the horrified question in his eyes.",
+ "normalized_hypothesis": "he stopped, and illa nodded at the horrified question in his eyes.",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "he stopped and philip nodded at the horrified question in his eyes",
+ "normalized_hypothesis": "he stopped and illa nodded at the horrified question in his eyes",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ }
+ },
+ {
+ "id": "release500_0336",
+ "category": "phonetic_balance",
+ "text": "She turned, fearing that Jacques might see what was in her face.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0336.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "She turned, fearing that Jack might see what was in her face.",
+ "raw": {
+ "normalized_reference": "she turned, fearing that jacques might see what was in her face.",
+ "normalized_hypothesis": "she turned, fearing that jack might see what was in her face.",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "she turned fearing that jacques might see what was in her face",
+ "normalized_hypothesis": "she turned fearing that jack might see what was in her face",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ }
+ },
+ {
+ "id": "release500_0337",
+ "category": "phonetic_balance",
+ "text": "It was a curious coincidence.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0337.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It was a curious coincidence.",
+ "raw": {
+ "normalized_reference": "it was a curious coincidence.",
+ "normalized_hypothesis": "it was a curious coincidence.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it was a curious coincidence",
+ "normalized_hypothesis": "it was a curious coincidence",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0338",
+ "category": "phonetic_balance",
+ "text": "Daylight made no answer, and the door closed behind him.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0338.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Daylight made no answer, and the door closed behind him.",
+ "raw": {
+ "normalized_reference": "daylight made no answer, and the door closed behind him.",
+ "normalized_hypothesis": "daylight made no answer, and the door closed behind him.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "daylight made no answer and the door closed behind him",
+ "normalized_hypothesis": "daylight made no answer and the door closed behind him",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0339",
+ "category": "phonetic_balance",
+ "text": "And he did hurt my arm.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0339.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "And he did hurt my arm.",
+ "raw": {
+ "normalized_reference": "and he did hurt my arm.",
+ "normalized_hypothesis": "and he did hurt my arm.",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "and he did hurt my arm",
+ "normalized_hypothesis": "and he did hurt my arm",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0340",
+ "category": "phonetic_balance",
+ "text": "Dunham, can your boy go along with Jesse.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0340.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Dunham, can your boy go along with Jesse?",
+ "raw": {
+ "normalized_reference": "dunham, can your boy go along with jesse.",
+ "normalized_hypothesis": "dunham, can your boy go along with jesse?",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "dunham can your boy go along with jesse",
+ "normalized_hypothesis": "dunham can your boy go along with jesse",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0341",
+ "category": "phonetic_balance",
+ "text": "The creative joy, I murmured.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0341.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The creative joy, I murmured.",
+ "raw": {
+ "normalized_reference": "the creative joy, i murmured.",
+ "normalized_hypothesis": "the creative joy, i murmured.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the creative joy i murmured",
+ "normalized_hypothesis": "the creative joy i murmured",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0342",
+ "category": "phonetic_balance",
+ "text": "He pulled, and the log crashed down to break his back.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0342.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He pulled, and the log crashed down to break his back.",
+ "raw": {
+ "normalized_reference": "he pulled, and the log crashed down to break his back.",
+ "normalized_hypothesis": "he pulled, and the log crashed down to break his back.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he pulled and the log crashed down to break his back",
+ "normalized_hypothesis": "he pulled and the log crashed down to break his back",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0343",
+ "category": "phonetic_balance",
+ "text": "By golly, the boy wins.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0343.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I golly, the boy wins.",
+ "raw": {
+ "normalized_reference": "by golly, the boy wins.",
+ "normalized_hypothesis": "i golly, the boy wins.",
+ "errors": 1,
+ "words": 5,
+ "wer": 0.2
+ },
+ "semantic": {
+ "normalized_reference": "by golly the boy wins",
+ "normalized_hypothesis": "i golly the boy wins",
+ "errors": 1,
+ "words": 5,
+ "wer": 0.2
+ }
+ },
+ {
+ "id": "release500_0344",
+ "category": "phonetic_balance",
+ "text": "The boy threw back his head with pride.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0344.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The boy threw back his head with pride.",
+ "raw": {
+ "normalized_reference": "the boy threw back his head with pride.",
+ "normalized_hypothesis": "the boy threw back his head with pride.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the boy threw back his head with pride",
+ "normalized_hypothesis": "the boy threw back his head with pride",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0345",
+ "category": "phonetic_balance",
+ "text": "The last one I knew was an overseer.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0345.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The last one I knew was an Overseer.",
+ "raw": {
+ "normalized_reference": "the last one i knew was an overseer.",
+ "normalized_hypothesis": "the last one i knew was an overseer.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the last one i knew was an overseer",
+ "normalized_hypothesis": "the last one i knew was an overseer",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0346",
+ "category": "phonetic_balance",
+ "text": "Two years ago I gave up civilization for this.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0346.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Two years ago I gave up civilization for this.",
+ "raw": {
+ "normalized_reference": "two years ago i gave up civilization for this.",
+ "normalized_hypothesis": "two years ago i gave up civilization for this.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "2 years ago i gave up civilization for this",
+ "normalized_hypothesis": "2 years ago i gave up civilization for this",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0347",
+ "category": "phonetic_balance",
+ "text": "The river bared its bosom, and snorting steamboats challenged the wilderness.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0347.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The river bared its bosom, and snorting steamboats challenged the wilderness.",
+ "raw": {
+ "normalized_reference": "the river bared its bosom, and snorting steamboats challenged the wilderness.",
+ "normalized_hypothesis": "the river bared its bosom, and snorting steamboats challenged the wilderness.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the river bared its bosom and snorting steamboats challenged the wilderness",
+ "normalized_hypothesis": "the river bared its bosom and snorting steamboats challenged the wilderness",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0348",
+ "category": "phonetic_balance",
+ "text": "Mab, she said.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0348.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Mad, she said.",
+ "raw": {
+ "normalized_reference": "mab, she said.",
+ "normalized_hypothesis": "mad, she said.",
+ "errors": 1,
+ "words": 3,
+ "wer": 0.3333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "mab she said",
+ "normalized_hypothesis": "mad she said",
+ "errors": 1,
+ "words": 3,
+ "wer": 0.3333333333333333
+ }
+ },
+ {
+ "id": "release500_0349",
+ "category": "phonetic_balance",
+ "text": "The regret in his voice was provocative of a second burst of laughter.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0349.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The regret in his voice was provocative of a second burst of laughter.",
+ "raw": {
+ "normalized_reference": "the regret in his voice was provocative of a second burst of laughter.",
+ "normalized_hypothesis": "the regret in his voice was provocative of a second burst of laughter.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the regret in his voice was provocative of a 2nd burst of laughter",
+ "normalized_hypothesis": "the regret in his voice was provocative of a 2nd burst of laughter",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0350",
+ "category": "phonetic_balance",
+ "text": "And each year something happened, and I did not go.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0350.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "And each year something happened and I did not go.",
+ "raw": {
+ "normalized_reference": "and each year something happened, and i did not go.",
+ "normalized_hypothesis": "and each year something happened and i did not go.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "and each year something happened and i did not go",
+ "normalized_hypothesis": "and each year something happened and i did not go",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0351",
+ "category": "phonetic_balance",
+ "text": "Gad, your letter came just in time.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0351.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Gad, your ladder came just in time.",
+ "raw": {
+ "normalized_reference": "gad, your letter came just in time.",
+ "normalized_hypothesis": "gad, your ladder came just in time.",
+ "errors": 1,
+ "words": 7,
+ "wer": 0.14285714285714285
+ },
+ "semantic": {
+ "normalized_reference": "gad your letter came just in time",
+ "normalized_hypothesis": "gad your ladder came just in time",
+ "errors": 1,
+ "words": 7,
+ "wer": 0.14285714285714285
+ }
+ },
+ {
+ "id": "release500_0352",
+ "category": "phonetic_balance",
+ "text": "But if Pierre did not return, until tomorrow.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0352.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "But if Pierre did not return until tomorrow,",
+ "raw": {
+ "normalized_reference": "but if pierre did not return, until tomorrow.",
+ "normalized_hypothesis": "but if pierre did not return until tomorrow,",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "but if pierre did not return until tomorrow",
+ "normalized_hypothesis": "but if pierre did not return until tomorrow",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0353",
+ "category": "phonetic_balance",
+ "text": "Billinger may arrive in time.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0353.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Billinger may arrive in time.",
+ "raw": {
+ "normalized_reference": "billinger may arrive in time.",
+ "normalized_hypothesis": "billinger may arrive in time.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "billinger may arrive in time",
+ "normalized_hypothesis": "billinger may arrive in time",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0354",
+ "category": "phonetic_balance",
+ "text": "In such a tumbling of values was no time to sell.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0354.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "In such a tumbling of values was no time to sell.",
+ "raw": {
+ "normalized_reference": "in such a tumbling of values was no time to sell.",
+ "normalized_hypothesis": "in such a tumbling of values was no time to sell.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "in such a tumbling of values was no time to sell",
+ "normalized_hypothesis": "in such a tumbling of values was no time to sell",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0355",
+ "category": "phonetic_balance",
+ "text": "There's not an iota of truth in it.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0355.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "There's not an iota of truth in it.",
+ "raw": {
+ "normalized_reference": "there's not an iota of truth in it.",
+ "normalized_hypothesis": "there's not an iota of truth in it.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "there is not an iota of truth in it",
+ "normalized_hypothesis": "there is not an iota of truth in it",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0356",
+ "category": "phonetic_balance",
+ "text": "It is dog eat dog, and you ate them up.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0356.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It is dog eat dog, and you eat them up.",
+ "raw": {
+ "normalized_reference": "it is dog eat dog, and you ate them up.",
+ "normalized_hypothesis": "it is dog eat dog, and you eat them up.",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ },
+ "semantic": {
+ "normalized_reference": "it is dog eat dog and you ate them up",
+ "normalized_hypothesis": "it is dog eat dog and you eat them up",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ }
+ },
+ {
+ "id": "release500_0357",
+ "category": "phonetic_balance",
+ "text": "You can take a vacation on pay.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0357.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "You can take a vacation on pay.",
+ "raw": {
+ "normalized_reference": "you can take a vacation on pay.",
+ "normalized_hypothesis": "you can take a vacation on pay.",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "you can take a vacation on pay",
+ "normalized_hypothesis": "you can take a vacation on pay",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0358",
+ "category": "phonetic_balance",
+ "text": "I have seen myself that one man contemplated by Pascal's philosophic eye.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0358.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I have seen myself that one man contemplated by Pascal's philosophic eye.",
+ "raw": {
+ "normalized_reference": "i have seen myself that one man contemplated by pascal's philosophic eye.",
+ "normalized_hypothesis": "i have seen myself that one man contemplated by pascal's philosophic eye.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i have seen myself that one man contemplated by pascal is philosophic eye",
+ "normalized_hypothesis": "i have seen myself that one man contemplated by pascal is philosophic eye",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0359",
+ "category": "phonetic_balance",
+ "text": "Ah, I had forgotten, he exclaimed.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0359.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I had forgotten, he exclaimed.",
+ "raw": {
+ "normalized_reference": "ah, i had forgotten, he exclaimed.",
+ "normalized_hypothesis": "i had forgotten, he exclaimed.",
+ "errors": 1,
+ "words": 6,
+ "wer": 0.16666666666666666
+ },
+ "semantic": {
+ "normalized_reference": "ah i had forgotten he exclaimed",
+ "normalized_hypothesis": "i had forgotten he exclaimed",
+ "errors": 1,
+ "words": 6,
+ "wer": 0.16666666666666666
+ }
+ },
+ {
+ "id": "release500_0360",
+ "category": "phonetic_balance",
+ "text": "Providence had delivered him through the maelstrom.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0360.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Providence had delivered him through the maelstrom.",
+ "raw": {
+ "normalized_reference": "providence had delivered him through the maelstrom.",
+ "normalized_hypothesis": "providence had delivered him through the maelstrom.",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "providence had delivered him through the maelstrom",
+ "normalized_hypothesis": "providence had delivered him through the maelstrom",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0361",
+ "category": "phonetic_balance",
+ "text": "She was his now, forever.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0361.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "She was his now, forever.",
+ "raw": {
+ "normalized_reference": "she was his now, forever.",
+ "normalized_hypothesis": "she was his now, forever.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "she was his now forever",
+ "normalized_hypothesis": "she was his now forever",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0362",
+ "category": "phonetic_balance",
+ "text": "It was unobtrusive, yet it was there.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0362.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It was unobtrusive, yet it was there.",
+ "raw": {
+ "normalized_reference": "it was unobtrusive, yet it was there.",
+ "normalized_hypothesis": "it was unobtrusive, yet it was there.",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it was unobtrusive yet it was there",
+ "normalized_hypothesis": "it was unobtrusive yet it was there",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0363",
+ "category": "phonetic_balance",
+ "text": "Men like Joe Goose dated existence from drunk to drunk.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0363.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Men like Joe Goose dated existence from drunk to drunk.",
+ "raw": {
+ "normalized_reference": "men like joe goose dated existence from drunk to drunk.",
+ "normalized_hypothesis": "men like joe goose dated existence from drunk to drunk.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "men like joe goose dated existence from drunk to drunk",
+ "normalized_hypothesis": "men like joe goose dated existence from drunk to drunk",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0364",
+ "category": "phonetic_balance",
+ "text": "I was completely lost in my work.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0364.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I was completely lost in my work.",
+ "raw": {
+ "normalized_reference": "i was completely lost in my work.",
+ "normalized_hypothesis": "i was completely lost in my work.",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i was completely lost in my work",
+ "normalized_hypothesis": "i was completely lost in my work",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0365",
+ "category": "phonetic_balance",
+ "text": "The very thing, Ernest agreed.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0365.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The very thing, Ernest agreed.",
+ "raw": {
+ "normalized_reference": "the very thing, ernest agreed.",
+ "normalized_hypothesis": "the very thing, ernest agreed.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the very thing ernest agreed",
+ "normalized_hypothesis": "the very thing ernest agreed",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0366",
+ "category": "phonetic_balance",
+ "text": "He was the leader, and Tudor was his lieutenant.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0366.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He was the leader, and Tudor was his lieutenant.",
+ "raw": {
+ "normalized_reference": "he was the leader, and tudor was his lieutenant.",
+ "normalized_hypothesis": "he was the leader, and tudor was his lieutenant.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he was the leader and tudor was his lieutenant",
+ "normalized_hypothesis": "he was the leader and tudor was his lieutenant",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0367",
+ "category": "phonetic_balance",
+ "text": "The Fire-Men wore animal skins around their waists and across their shoulders.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0367.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The firemen wore animal skins around their waists and across their shoulders.",
+ "raw": {
+ "normalized_reference": "the fire-men wore animal skins around their waists and across their shoulders.",
+ "normalized_hypothesis": "the firemen wore animal skins around their waists and across their shoulders.",
+ "errors": 2,
+ "words": 13,
+ "wer": 0.15384615384615385
+ },
+ "semantic": {
+ "normalized_reference": "the fire men wore animal skins around their waists and across their shoulders",
+ "normalized_hypothesis": "the firemen wore animal skins around their waists and across their shoulders",
+ "errors": 2,
+ "words": 13,
+ "wer": 0.15384615384615385
+ }
+ },
+ {
+ "id": "release500_0368",
+ "category": "phonetic_balance",
+ "text": "We could throw stones with our feet.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0368.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "We could throw stones with our feet.",
+ "raw": {
+ "normalized_reference": "we could throw stones with our feet.",
+ "normalized_hypothesis": "we could throw stones with our feet.",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "we could throw stones with our feet",
+ "normalized_hypothesis": "we could throw stones with our feet",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0369",
+ "category": "phonetic_balance",
+ "text": "He turned sharply, and faced Gregson across the table.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0369.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He turned sharply and faced Gregson across the table.",
+ "raw": {
+ "normalized_reference": "he turned sharply, and faced gregson across the table.",
+ "normalized_hypothesis": "he turned sharply and faced gregson across the table.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he turned sharply and faced gregson across the table",
+ "normalized_hypothesis": "he turned sharply and faced gregson across the table",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0370",
+ "category": "phonetic_balance",
+ "text": "They are to attack your camp tomorrow night.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0370.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "They are to attack your camp tomorrow night.",
+ "raw": {
+ "normalized_reference": "they are to attack your camp tomorrow night.",
+ "normalized_hypothesis": "they are to attack your camp tomorrow night.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "they are to attack your camp tomorrow night",
+ "normalized_hypothesis": "they are to attack your camp tomorrow night",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0371",
+ "category": "phonetic_balance",
+ "text": "And watch out for wet feet, was his parting advice.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0371.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "And watch out for what feet was his parting advice.",
+ "raw": {
+ "normalized_reference": "and watch out for wet feet, was his parting advice.",
+ "normalized_hypothesis": "and watch out for what feet was his parting advice.",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ },
+ "semantic": {
+ "normalized_reference": "and watch out for wet feet was his parting advice",
+ "normalized_hypothesis": "and watch out for what feet was his parting advice",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ }
+ },
+ {
+ "id": "release500_0372",
+ "category": "phonetic_balance",
+ "text": "Two weeks passed, and in that time Thorpe left camp three times.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0372.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Two weeks passed, and in that time Thorpe left camp three times.",
+ "raw": {
+ "normalized_reference": "two weeks passed, and in that time thorpe left camp three times.",
+ "normalized_hypothesis": "two weeks passed, and in that time thorpe left camp three times.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "2 weeks passed and in that time thorpe left camp 3 times",
+ "normalized_hypothesis": "2 weeks passed and in that time thorpe left camp 3 times",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0373",
+ "category": "phonetic_balance",
+ "text": "But ever his gaze returned to that Crouched Venus on the piano.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0373.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "But ever his gaze returned to that crouched Venus on the piano.",
+ "raw": {
+ "normalized_reference": "but ever his gaze returned to that crouched venus on the piano.",
+ "normalized_hypothesis": "but ever his gaze returned to that crouched venus on the piano.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "but ever his gaze returned to that crouched venus on the piano",
+ "normalized_hypothesis": "but ever his gaze returned to that crouched venus on the piano",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0374",
+ "category": "phonetic_balance",
+ "text": "But she had become an automaton.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0374.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "But she had become an automaton.",
+ "raw": {
+ "normalized_reference": "but she had become an automaton.",
+ "normalized_hypothesis": "but she had become an automaton.",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "but she had become an automaton",
+ "normalized_hypothesis": "but she had become an automaton",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0375",
+ "category": "phonetic_balance",
+ "text": "I do not blame you for anything; remember that.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0375.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I do not blame you for anything. Remember that.",
+ "raw": {
+ "normalized_reference": "i do not blame you for anything; remember that.",
+ "normalized_hypothesis": "i do not blame you for anything. remember that.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i do not blame you for anything remember that",
+ "normalized_hypothesis": "i do not blame you for anything remember that",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0376",
+ "category": "phonetic_balance",
+ "text": "The next thing to watch out for is bed sores.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0376.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The next thing to watch out for is bedsores.",
+ "raw": {
+ "normalized_reference": "the next thing to watch out for is bed sores.",
+ "normalized_hypothesis": "the next thing to watch out for is bedsores.",
+ "errors": 2,
+ "words": 10,
+ "wer": 0.2
+ },
+ "semantic": {
+ "normalized_reference": "the next thing to watch out for is bed sores",
+ "normalized_hypothesis": "the next thing to watch out for is bedsores",
+ "errors": 2,
+ "words": 10,
+ "wer": 0.2
+ }
+ },
+ {
+ "id": "release500_0377",
+ "category": "phonetic_balance",
+ "text": "He will follow us soon.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0377.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He will follow us soon.",
+ "raw": {
+ "normalized_reference": "he will follow us soon.",
+ "normalized_hypothesis": "he will follow us soon.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he will follow us soon",
+ "normalized_hypothesis": "he will follow us soon",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0378",
+ "category": "phonetic_balance",
+ "text": "His beady black eyes saw bargains where other men saw bankruptcy.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0378.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "His beady black eyes saw bargains where other men saw bankruptcy.",
+ "raw": {
+ "normalized_reference": "his beady black eyes saw bargains where other men saw bankruptcy.",
+ "normalized_hypothesis": "his beady black eyes saw bargains where other men saw bankruptcy.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "his beady black eyes saw bargains where other men saw bankruptcy",
+ "normalized_hypothesis": "his beady black eyes saw bargains where other men saw bankruptcy",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0379",
+ "category": "phonetic_balance",
+ "text": "They robbed me a few years later.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0379.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "They robbed me a few years later.",
+ "raw": {
+ "normalized_reference": "they robbed me a few years later.",
+ "normalized_hypothesis": "they robbed me a few years later.",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "they robbed me a few years later",
+ "normalized_hypothesis": "they robbed me a few years later",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0380",
+ "category": "phonetic_balance",
+ "text": "Gregson held a lighted match until it burnt his fingertips.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0380.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Gregson held a lighted match until it burned his fingertips.",
+ "raw": {
+ "normalized_reference": "gregson held a lighted match until it burnt his fingertips.",
+ "normalized_hypothesis": "gregson held a lighted match until it burned his fingertips.",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ },
+ "semantic": {
+ "normalized_reference": "gregson held a lighted match until it burnt his fingertips",
+ "normalized_hypothesis": "gregson held a lighted match until it burned his fingertips",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ }
+ },
+ {
+ "id": "release500_0381",
+ "category": "phonetic_balance",
+ "text": "A burst of laughter was his reward.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0381.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "A burst of laughter was his reward.",
+ "raw": {
+ "normalized_reference": "a burst of laughter was his reward.",
+ "normalized_hypothesis": "a burst of laughter was his reward.",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "a burst of laughter was his reward",
+ "normalized_hypothesis": "a burst of laughter was his reward",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0382",
+ "category": "phonetic_balance",
+ "text": "He cried, and swung the club wildly.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0382.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He cried and swung the club wildly.",
+ "raw": {
+ "normalized_reference": "he cried, and swung the club wildly.",
+ "normalized_hypothesis": "he cried and swung the club wildly.",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he cried and swung the club wildly",
+ "normalized_hypothesis": "he cried and swung the club wildly",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0383",
+ "category": "phonetic_balance",
+ "text": "Sandel would never become a world champion.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0383.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Sandal would never become a world champion.",
+ "raw": {
+ "normalized_reference": "sandel would never become a world champion.",
+ "normalized_hypothesis": "sandal would never become a world champion.",
+ "errors": 1,
+ "words": 7,
+ "wer": 0.14285714285714285
+ },
+ "semantic": {
+ "normalized_reference": "sandel would never become a world champion",
+ "normalized_hypothesis": "sandal would never become a world champion",
+ "errors": 1,
+ "words": 7,
+ "wer": 0.14285714285714285
+ }
+ },
+ {
+ "id": "release500_0384",
+ "category": "phonetic_balance",
+ "text": "And after the bath a shave would not be bad.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0384.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "And after the bath a shave would not be bad.",
+ "raw": {
+ "normalized_reference": "and after the bath a shave would not be bad.",
+ "normalized_hypothesis": "and after the bath a shave would not be bad.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "and after the bath a shave would not be bad",
+ "normalized_hypothesis": "and after the bath a shave would not be bad",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0385",
+ "category": "phonetic_balance",
+ "text": "They saw each other for the first time in Boston.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0385.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "They saw each other for the first time in Boston.",
+ "raw": {
+ "normalized_reference": "they saw each other for the first time in boston.",
+ "normalized_hypothesis": "they saw each other for the first time in boston.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "they saw each other for the 1st time in boston",
+ "normalized_hypothesis": "they saw each other for the 1st time in boston",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0386",
+ "category": "phonetic_balance",
+ "text": "I had been born with no organic, chemical predisposition toward alcohol.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0386.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I had been born with no organic, chemical predisposition toward alcohol.",
+ "raw": {
+ "normalized_reference": "i had been born with no organic, chemical predisposition toward alcohol.",
+ "normalized_hypothesis": "i had been born with no organic, chemical predisposition toward alcohol.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i had been born with no organic chemical predisposition toward alcohol",
+ "normalized_hypothesis": "i had been born with no organic chemical predisposition toward alcohol",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0387",
+ "category": "phonetic_balance",
+ "text": "They had no fixed values, to be altered by adjectives and adverbs.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0387.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "They had no fixed values to be altered by adjectives and adverbs.",
+ "raw": {
+ "normalized_reference": "they had no fixed values, to be altered by adjectives and adverbs.",
+ "normalized_hypothesis": "they had no fixed values to be altered by adjectives and adverbs.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "they had no fixed values to be altered by adjectives and adverbs",
+ "normalized_hypothesis": "they had no fixed values to be altered by adjectives and adverbs",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0388",
+ "category": "phonetic_balance",
+ "text": "Nor was Elam Harnish an exception.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0388.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Nor was the lamb Harnish an exception.",
+ "raw": {
+ "normalized_reference": "nor was elam harnish an exception.",
+ "normalized_hypothesis": "nor was the lamb harnish an exception.",
+ "errors": 2,
+ "words": 6,
+ "wer": 0.3333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "nor was elam harnish an exception",
+ "normalized_hypothesis": "nor was the lamb harnish an exception",
+ "errors": 2,
+ "words": 6,
+ "wer": 0.3333333333333333
+ }
+ },
+ {
+ "id": "release500_0389",
+ "category": "phonetic_balance",
+ "text": "He was a wise hyena.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0389.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He was a wise hyena.",
+ "raw": {
+ "normalized_reference": "he was a wise hyena.",
+ "normalized_hypothesis": "he was a wise hyena.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he was a wise hyena",
+ "normalized_hypothesis": "he was a wise hyena",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0390",
+ "category": "phonetic_balance",
+ "text": "And now behold the perversity of things.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0390.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "And now behold the perversity of things.",
+ "raw": {
+ "normalized_reference": "and now behold the perversity of things.",
+ "normalized_hypothesis": "and now behold the perversity of things.",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "and now behold the perversity of things",
+ "normalized_hypothesis": "and now behold the perversity of things",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0391",
+ "category": "phonetic_balance",
+ "text": "The Fire People, like ourselves, lived in caves.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0391.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The fire people, like ourselves, lived in caves.",
+ "raw": {
+ "normalized_reference": "the fire people, like ourselves, lived in caves.",
+ "normalized_hypothesis": "the fire people, like ourselves, lived in caves.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the fire people like ourselves lived in caves",
+ "normalized_hypothesis": "the fire people like ourselves lived in caves",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0392",
+ "category": "phonetic_balance",
+ "text": "To her the bridge was tambo, which is the native for taboo.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0392.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "To her the bridge was Tambo, which is the native for Taboo.",
+ "raw": {
+ "normalized_reference": "to her the bridge was tambo, which is the native for taboo.",
+ "normalized_hypothesis": "to her the bridge was tambo, which is the native for taboo.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "to her the bridge was tambo which is the native for taboo",
+ "normalized_hypothesis": "to her the bridge was tambo which is the native for taboo",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0393",
+ "category": "phonetic_balance",
+ "text": "The voyage was our idea of a good time.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0393.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The voyage was our idea of a good time.",
+ "raw": {
+ "normalized_reference": "the voyage was our idea of a good time.",
+ "normalized_hypothesis": "the voyage was our idea of a good time.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the voyage was our idea of a good time",
+ "normalized_hypothesis": "the voyage was our idea of a good time",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0394",
+ "category": "phonetic_balance",
+ "text": "Of course much grumbling went on, and little outbursts were continually occurring.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0394.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Of course much grumbling went on, and little outbursts were continually occurring.",
+ "raw": {
+ "normalized_reference": "of course much grumbling went on, and little outbursts were continually occurring.",
+ "normalized_hypothesis": "of course much grumbling went on, and little outbursts were continually occurring.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "of course much grumbling went on and little outbursts were continually occurring",
+ "normalized_hypothesis": "of course much grumbling went on and little outbursts were continually occurring",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0395",
+ "category": "phonetic_balance",
+ "text": "Get down and dig in.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0395.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Get down and dig in.",
+ "raw": {
+ "normalized_reference": "get down and dig in.",
+ "normalized_hypothesis": "get down and dig in.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "get down and dig in",
+ "normalized_hypothesis": "get down and dig in",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0396",
+ "category": "phonetic_balance",
+ "text": "It was put together in a casual, helter-skelter sort of way.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0396.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It was put together in a casual, helter-skelter sort of way.",
+ "raw": {
+ "normalized_reference": "it was put together in a casual, helter-skelter sort of way.",
+ "normalized_hypothesis": "it was put together in a casual, helter-skelter sort of way.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it was put together in a casual helter skelter sort of way",
+ "normalized_hypothesis": "it was put together in a casual helter skelter sort of way",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0397",
+ "category": "phonetic_balance",
+ "text": "He deluged me, overwhelmed me with argument.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0397.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He deluged me, overwhelmed me with argument.",
+ "raw": {
+ "normalized_reference": "he deluged me, overwhelmed me with argument.",
+ "normalized_hypothesis": "he deluged me, overwhelmed me with argument.",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he deluged me overwhelmed me with argument",
+ "normalized_hypothesis": "he deluged me overwhelmed me with argument",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0398",
+ "category": "phonetic_balance",
+ "text": "The girl faced him, her eyes shining with sudden fear.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0398.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The girl faced him, her eyes shining with sudden fear.",
+ "raw": {
+ "normalized_reference": "the girl faced him, her eyes shining with sudden fear.",
+ "normalized_hypothesis": "the girl faced him, her eyes shining with sudden fear.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the girl faced him her eyes shining with sudden fear",
+ "normalized_hypothesis": "the girl faced him her eyes shining with sudden fear",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0399",
+ "category": "phonetic_balance",
+ "text": "It will break our hearts and our backs to hoist anchor by hand.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0399.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It will break our hearts and our backs to hoist anchor by hand.",
+ "raw": {
+ "normalized_reference": "it will break our hearts and our backs to hoist anchor by hand.",
+ "normalized_hypothesis": "it will break our hearts and our backs to hoist anchor by hand.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it will break our hearts and our backs to hoist anchor by hand",
+ "normalized_hypothesis": "it will break our hearts and our backs to hoist anchor by hand",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0400",
+ "category": "phonetic_balance",
+ "text": "He made no reply as he waited for Whittemore to continue.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0400.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He made no reply as he waited for Whittemore to continue.",
+ "raw": {
+ "normalized_reference": "he made no reply as he waited for whittemore to continue.",
+ "normalized_hypothesis": "he made no reply as he waited for whittemore to continue.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he made no reply as he waited for whittemore to continue",
+ "normalized_hypothesis": "he made no reply as he waited for whittemore to continue",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0401",
+ "category": "phonetic_balance",
+ "text": "Who the devil gave it to you to be judge and jury.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0401.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Who the devil gave it to you to be judge and jury?",
+ "raw": {
+ "normalized_reference": "who the devil gave it to you to be judge and jury.",
+ "normalized_hypothesis": "who the devil gave it to you to be judge and jury?",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "who the devil gave it to you to be judge and jury",
+ "normalized_hypothesis": "who the devil gave it to you to be judge and jury",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0402",
+ "category": "phonetic_balance",
+ "text": "Daylight found himself charmed and made curious by the little man.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0402.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Daylight found himself charmed and made curious by the little man.",
+ "raw": {
+ "normalized_reference": "daylight found himself charmed and made curious by the little man.",
+ "normalized_hypothesis": "daylight found himself charmed and made curious by the little man.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "daylight found himself charmed and made curious by the little man",
+ "normalized_hypothesis": "daylight found himself charmed and made curious by the little man",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0403",
+ "category": "phonetic_balance",
+ "text": "It's a Yankee, Joan cried.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0403.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It's a Yankee, Joan cried.",
+ "raw": {
+ "normalized_reference": "it's a yankee, joan cried.",
+ "normalized_hypothesis": "it's a yankee, joan cried.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it is a yankee joan cried",
+ "normalized_hypothesis": "it is a yankee joan cried",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0404",
+ "category": "phonetic_balance",
+ "text": "But what they want with your toothbrush is more than I can imagine.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0404.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "But what they want with your toothbrush is more than I can imagine.",
+ "raw": {
+ "normalized_reference": "but what they want with your toothbrush is more than i can imagine.",
+ "normalized_hypothesis": "but what they want with your toothbrush is more than i can imagine.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "but what they want with your toothbrush is more than i can imagine",
+ "normalized_hypothesis": "but what they want with your toothbrush is more than i can imagine",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0405",
+ "category": "phonetic_balance",
+ "text": "They were babbling and chattering all together.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0405.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "They were babbling and chattering all together.",
+ "raw": {
+ "normalized_reference": "they were babbling and chattering all together.",
+ "normalized_hypothesis": "they were babbling and chattering all together.",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "they were babbling and chattering all together",
+ "normalized_hypothesis": "they were babbling and chattering all together",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0406",
+ "category": "phonetic_balance",
+ "text": "Men of Selden's stamp don't stop at women and children.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0406.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Men of Selden's stamp don't stop at women and children.",
+ "raw": {
+ "normalized_reference": "men of selden's stamp don't stop at women and children.",
+ "normalized_hypothesis": "men of selden's stamp don't stop at women and children.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "men of selden is stamp do not stop at women and children",
+ "normalized_hypothesis": "men of selden is stamp do not stop at women and children",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0407",
+ "category": "phonetic_balance",
+ "text": "Cameron looked at his hands with their long, sinewy fingers.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0407.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Cameron looked at his hands with their long, sinewy fingers.",
+ "raw": {
+ "normalized_reference": "cameron looked at his hands with their long, sinewy fingers.",
+ "normalized_hypothesis": "cameron looked at his hands with their long, sinewy fingers.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "cameron looked at his hands with their long sinewy fingers",
+ "normalized_hypothesis": "cameron looked at his hands with their long sinewy fingers",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0408",
+ "category": "phonetic_balance",
+ "text": "Saxon waited, for she knew a fresh idea had struck Billy.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0408.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Saxon waited, for she knew a fresh idea had struck Billy.",
+ "raw": {
+ "normalized_reference": "saxon waited, for she knew a fresh idea had struck billy.",
+ "normalized_hypothesis": "saxon waited, for she knew a fresh idea had struck billy.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "saxon waited for she knew a fresh idea had struck billy",
+ "normalized_hypothesis": "saxon waited for she knew a fresh idea had struck billy",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0409",
+ "category": "phonetic_balance",
+ "text": "He saw only the effect in a general, sketchy way.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0409.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He saw only the effect in a general, sketchy way.",
+ "raw": {
+ "normalized_reference": "he saw only the effect in a general, sketchy way.",
+ "normalized_hypothesis": "he saw only the effect in a general, sketchy way.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he saw only the effect in a general sketchy way",
+ "normalized_hypothesis": "he saw only the effect in a general sketchy way",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0410",
+ "category": "phonetic_balance",
+ "text": "He began to follow the footprints of the dog.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0410.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He began to follow the footprints of the dog.",
+ "raw": {
+ "normalized_reference": "he began to follow the footprints of the dog.",
+ "normalized_hypothesis": "he began to follow the footprints of the dog.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he began to follow the footprints of the dog",
+ "normalized_hypothesis": "he began to follow the footprints of the dog",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0411",
+ "category": "phonetic_balance",
+ "text": "Do you value your hide.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0411.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Do you value your hide?",
+ "raw": {
+ "normalized_reference": "do you value your hide.",
+ "normalized_hypothesis": "do you value your hide?",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "do you value your hide",
+ "normalized_hypothesis": "do you value your hide",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0412",
+ "category": "phonetic_balance",
+ "text": "A rising tide of fat had submerged them.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0412.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "A rising tide of fat hands submerged them.",
+ "raw": {
+ "normalized_reference": "a rising tide of fat had submerged them.",
+ "normalized_hypothesis": "a rising tide of fat hands submerged them.",
+ "errors": 1,
+ "words": 8,
+ "wer": 0.125
+ },
+ "semantic": {
+ "normalized_reference": "a rising tide of fat had submerged them",
+ "normalized_hypothesis": "a rising tide of fat hands submerged them",
+ "errors": 1,
+ "words": 8,
+ "wer": 0.125
+ }
+ },
+ {
+ "id": "release500_0413",
+ "category": "phonetic_balance",
+ "text": "There was nothing on the rock.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0413.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "There was nothing on the rock.",
+ "raw": {
+ "normalized_reference": "there was nothing on the rock.",
+ "normalized_hypothesis": "there was nothing on the rock.",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "there was nothing on the rock",
+ "normalized_hypothesis": "there was nothing on the rock",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0414",
+ "category": "phonetic_balance",
+ "text": "Whatever he guessed he locked away in the taboo room of Naomi.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0414.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Whatever he guessed he locked away in the taboo room of Naomi.",
+ "raw": {
+ "normalized_reference": "whatever he guessed he locked away in the taboo room of naomi.",
+ "normalized_hypothesis": "whatever he guessed he locked away in the taboo room of naomi.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "whatever he guessed he locked away in the taboo room of naomi",
+ "normalized_hypothesis": "whatever he guessed he locked away in the taboo room of naomi",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0415",
+ "category": "phonetic_balance",
+ "text": "Will we ever forget it.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0415.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Will we ever forget it?",
+ "raw": {
+ "normalized_reference": "will we ever forget it.",
+ "normalized_hypothesis": "will we ever forget it?",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "will we ever forget it",
+ "normalized_hypothesis": "will we ever forget it",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0416",
+ "category": "phonetic_balance",
+ "text": "Jeanne and Pierre both gazed toward the great rock.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0416.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Jean and Pierre both gaze toward the great rock.",
+ "raw": {
+ "normalized_reference": "jeanne and pierre both gazed toward the great rock.",
+ "normalized_hypothesis": "jean and pierre both gaze toward the great rock.",
+ "errors": 2,
+ "words": 9,
+ "wer": 0.2222222222222222
+ },
+ "semantic": {
+ "normalized_reference": "jeanne and pierre both gazed toward the great rock",
+ "normalized_hypothesis": "jean and pierre both gaze toward the great rock",
+ "errors": 2,
+ "words": 9,
+ "wer": 0.2222222222222222
+ }
+ },
+ {
+ "id": "release500_0417",
+ "category": "phonetic_balance",
+ "text": "Does the old boy often go off at half-cock that way.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0417.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Does the old boy often go off at half-cock that way?",
+ "raw": {
+ "normalized_reference": "does the old boy often go off at half-cock that way.",
+ "normalized_hypothesis": "does the old boy often go off at half-cock that way?",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "does the old boy often go off at half cock that way",
+ "normalized_hypothesis": "does the old boy often go off at half cock that way",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0418",
+ "category": "phonetic_balance",
+ "text": "The temperature dropped to fifty below zero and remained there the whole trip.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0418.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The temperature dropped to 50 below zero and remained there the whole trip.",
+ "raw": {
+ "normalized_reference": "the temperature dropped to fifty below zero and remained there the whole trip.",
+ "normalized_hypothesis": "the temperature dropped to 50 below zero and remained there the whole trip.",
+ "errors": 1,
+ "words": 13,
+ "wer": 0.07692307692307693
+ },
+ "semantic": {
+ "normalized_reference": "the temperature dropped to 50 below 0 and remained there the whole trip",
+ "normalized_hypothesis": "the temperature dropped to 50 below 0 and remained there the whole trip",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0419",
+ "category": "phonetic_balance",
+ "text": "It's worth eight dollars.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0419.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It's worth $8.",
+ "raw": {
+ "normalized_reference": "it's worth eight dollars.",
+ "normalized_hypothesis": "it's worth $8.",
+ "errors": 2,
+ "words": 4,
+ "wer": 0.5
+ },
+ "semantic": {
+ "normalized_reference": "it is worth $8",
+ "normalized_hypothesis": "it is worth $8",
+ "errors": 0,
+ "words": 4,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0420",
+ "category": "phonetic_balance",
+ "text": "But I did not enjoy it long.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0420.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "But I did not enjoy it long.",
+ "raw": {
+ "normalized_reference": "but i did not enjoy it long.",
+ "normalized_hypothesis": "but i did not enjoy it long.",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "but i did not enjoy it long",
+ "normalized_hypothesis": "but i did not enjoy it long",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0421",
+ "category": "phonetic_balance",
+ "text": "Shorty turned to their employers.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0421.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Shorty turned to their employers.",
+ "raw": {
+ "normalized_reference": "shorty turned to their employers.",
+ "normalized_hypothesis": "shorty turned to their employers.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "shorty turned to their employers",
+ "normalized_hypothesis": "shorty turned to their employers",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0422",
+ "category": "phonetic_balance",
+ "text": "It was the third or fourth time that Philip had heard MacDougall swear.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0422.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It was the third or fourth time that Philip had heard MacDouble swear.",
+ "raw": {
+ "normalized_reference": "it was the third or fourth time that philip had heard macdougall swear.",
+ "normalized_hypothesis": "it was the third or fourth time that philip had heard macdouble swear.",
+ "errors": 1,
+ "words": 13,
+ "wer": 0.07692307692307693
+ },
+ "semantic": {
+ "normalized_reference": "it was the 3rd or 4th time that philip had heard macdougall swear",
+ "normalized_hypothesis": "it was the 3rd or 4th time that philip had heard macdouble swear",
+ "errors": 1,
+ "words": 13,
+ "wer": 0.07692307692307693
+ }
+ },
+ {
+ "id": "release500_0423",
+ "category": "phonetic_balance",
+ "text": "Ernest saw in the affair the most sinister import.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0423.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Ernest saw in the affair the most sinister import.",
+ "raw": {
+ "normalized_reference": "ernest saw in the affair the most sinister import.",
+ "normalized_hypothesis": "ernest saw in the affair the most sinister import.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "ernest saw in the affair the most sinister import",
+ "normalized_hypothesis": "ernest saw in the affair the most sinister import",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0424",
+ "category": "phonetic_balance",
+ "text": "Also, at regular intervals, he would mutter.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0424.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Also, at regular intervals, he would mutter,",
+ "raw": {
+ "normalized_reference": "also, at regular intervals, he would mutter.",
+ "normalized_hypothesis": "also, at regular intervals, he would mutter,",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "also at regular intervals he would mutter",
+ "normalized_hypothesis": "also at regular intervals he would mutter",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0425",
+ "category": "phonetic_balance",
+ "text": "He may anticipate the day of his death.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0425.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He may anticipate the day of his death.",
+ "raw": {
+ "normalized_reference": "he may anticipate the day of his death.",
+ "normalized_hypothesis": "he may anticipate the day of his death.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he may anticipate the day of his death",
+ "normalized_hypothesis": "he may anticipate the day of his death",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0426",
+ "category": "phonetic_balance",
+ "text": "I defy any man to get a Solomon Island sore in California.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0426.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I defy any man to get a Solomon Island soar in California.",
+ "raw": {
+ "normalized_reference": "i defy any man to get a solomon island sore in california.",
+ "normalized_hypothesis": "i defy any man to get a solomon island soar in california.",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "i defy any man to get a solomon island sore in california",
+ "normalized_hypothesis": "i defy any man to get a solomon island soar in california",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ }
+ },
+ {
+ "id": "release500_0427",
+ "category": "phonetic_balance",
+ "text": "Now our figuring was all right.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0427.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Now our figuring was all right.",
+ "raw": {
+ "normalized_reference": "now our figuring was all right.",
+ "normalized_hypothesis": "now our figuring was all right.",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "now our figuring was all right",
+ "normalized_hypothesis": "now our figuring was all right",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0428",
+ "category": "phonetic_balance",
+ "text": "MacDougall tapped his forehead suspiciously with a stubby forefinger.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0428.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "MacDougall tapped his forehead suspiciously with a stubby forefinger.",
+ "raw": {
+ "normalized_reference": "macdougall tapped his forehead suspiciously with a stubby forefinger.",
+ "normalized_hypothesis": "macdougall tapped his forehead suspiciously with a stubby forefinger.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "macdougall tapped his forehead suspiciously with a stubby forefinger",
+ "normalized_hypothesis": "macdougall tapped his forehead suspiciously with a stubby forefinger",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0429",
+ "category": "phonetic_balance",
+ "text": "O'Brien took off his coat and bared his right arm.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0429.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "O'Brien took off his coat and bared his right arm.",
+ "raw": {
+ "normalized_reference": "o'brien took off his coat and bared his right arm.",
+ "normalized_hypothesis": "o'brien took off his coat and bared his right arm.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "0 brien took off his coat and bared his right arm",
+ "normalized_hypothesis": "0 brien took off his coat and bared his right arm",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0430",
+ "category": "phonetic_balance",
+ "text": "But it won't continue, she said with easy confidence.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0430.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "But it won't continue, she said with easy confidence.",
+ "raw": {
+ "normalized_reference": "but it won't continue, she said with easy confidence.",
+ "normalized_hypothesis": "but it won't continue, she said with easy confidence.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "but it will not continue she said with easy confidence",
+ "normalized_hypothesis": "but it will not continue she said with easy confidence",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0431",
+ "category": "phonetic_balance",
+ "text": "I was still weak from my prolonged immersion.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0431.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I was still weak from my prolonged immersion.",
+ "raw": {
+ "normalized_reference": "i was still weak from my prolonged immersion.",
+ "normalized_hypothesis": "i was still weak from my prolonged immersion.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i was still weak from my prolonged immersion",
+ "normalized_hypothesis": "i was still weak from my prolonged immersion",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0432",
+ "category": "phonetic_balance",
+ "text": "Oolong was two hundred and fifty miles from the nearest land.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0432.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "One was 250 miles from the nearest land.",
+ "raw": {
+ "normalized_reference": "oolong was two hundred and fifty miles from the nearest land.",
+ "normalized_hypothesis": "one was 250 miles from the nearest land.",
+ "errors": 5,
+ "words": 11,
+ "wer": 0.45454545454545453
+ },
+ "semantic": {
+ "normalized_reference": "oolong was 250 miles from the nearest land",
+ "normalized_hypothesis": "one was 250 miles from the nearest land",
+ "errors": 1,
+ "words": 8,
+ "wer": 0.125
+ }
+ },
+ {
+ "id": "release500_0433",
+ "category": "phonetic_balance",
+ "text": "I just do appreciate it without being able to express my feelings.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0433.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I just do appreciate it without being able to express my feelings.",
+ "raw": {
+ "normalized_reference": "i just do appreciate it without being able to express my feelings.",
+ "normalized_hypothesis": "i just do appreciate it without being able to express my feelings.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i just do appreciate it without being able to express my feelings",
+ "normalized_hypothesis": "i just do appreciate it without being able to express my feelings",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0434",
+ "category": "phonetic_balance",
+ "text": "Now just what do you want to know.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0434.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Now just what do you want to know?",
+ "raw": {
+ "normalized_reference": "now just what do you want to know.",
+ "normalized_hypothesis": "now just what do you want to know?",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "now just what do you want to know",
+ "normalized_hypothesis": "now just what do you want to know",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0435",
+ "category": "phonetic_balance",
+ "text": "But she swung obediently on her heel into the wind.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0435.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "But she swung obediently on her heel into the wind.",
+ "raw": {
+ "normalized_reference": "but she swung obediently on her heel into the wind.",
+ "normalized_hypothesis": "but she swung obediently on her heel into the wind.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "but she swung obediently on her heel into the wind",
+ "normalized_hypothesis": "but she swung obediently on her heel into the wind",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0436",
+ "category": "phonetic_balance",
+ "text": "Ask him, I laughed, then turned to Pasquini.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0436.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Ask him, I laughed, then turned to Pasquini.",
+ "raw": {
+ "normalized_reference": "ask him, i laughed, then turned to pasquini.",
+ "normalized_hypothesis": "ask him, i laughed, then turned to pasquini.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "ask him i laughed then turned to pasquini",
+ "normalized_hypothesis": "ask him i laughed then turned to pasquini",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0437",
+ "category": "phonetic_balance",
+ "text": "He did not rush in.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0437.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He did not rush in.",
+ "raw": {
+ "normalized_reference": "he did not rush in.",
+ "normalized_hypothesis": "he did not rush in.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he did not rush in",
+ "normalized_hypothesis": "he did not rush in",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0438",
+ "category": "phonetic_balance",
+ "text": "They obeyed him, and went here and there at his commands.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0438.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "They obeyed him, and went here and there at his commands.",
+ "raw": {
+ "normalized_reference": "they obeyed him, and went here and there at his commands.",
+ "normalized_hypothesis": "they obeyed him, and went here and there at his commands.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "they obeyed him and went here and there at his commands",
+ "normalized_hypothesis": "they obeyed him and went here and there at his commands",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0439",
+ "category": "phonetic_balance",
+ "text": "Tom Spink has a harpoon.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0439.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Tom Spink has a harpoon.",
+ "raw": {
+ "normalized_reference": "tom spink has a harpoon.",
+ "normalized_hypothesis": "tom spink has a harpoon.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "tom spink has a harpoon",
+ "normalized_hypothesis": "tom spink has a harpoon",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0440",
+ "category": "phonetic_balance",
+ "text": "It was my reports from the north which chiefly induced people to buy.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0440.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It was my reports from the north which chiefly induced people to buy.",
+ "raw": {
+ "normalized_reference": "it was my reports from the north which chiefly induced people to buy.",
+ "normalized_hypothesis": "it was my reports from the north which chiefly induced people to buy.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it was my reports from the north which chiefly induced people to buy",
+ "normalized_hypothesis": "it was my reports from the north which chiefly induced people to buy",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0441",
+ "category": "phonetic_balance",
+ "text": "Cherokee identified himself with his instinct.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0441.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Cherokee identified himself with his instinct.",
+ "raw": {
+ "normalized_reference": "cherokee identified himself with his instinct.",
+ "normalized_hypothesis": "cherokee identified himself with his instinct.",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "cherokee identified himself with his instinct",
+ "normalized_hypothesis": "cherokee identified himself with his instinct",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0442",
+ "category": "phonetic_balance",
+ "text": "I graduated last of my class.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0442.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I graduated last of my class.",
+ "raw": {
+ "normalized_reference": "i graduated last of my class.",
+ "normalized_hypothesis": "i graduated last of my class.",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i graduated last of my class",
+ "normalized_hypothesis": "i graduated last of my class",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0443",
+ "category": "phonetic_balance",
+ "text": "It was a gigantic inadequacy.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0443.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It was a gigantic inadequacy.",
+ "raw": {
+ "normalized_reference": "it was a gigantic inadequacy.",
+ "normalized_hypothesis": "it was a gigantic inadequacy.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it was a gigantic inadequacy",
+ "normalized_hypothesis": "it was a gigantic inadequacy",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0444",
+ "category": "phonetic_balance",
+ "text": "Did I possess too much vitality.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0444.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Did I possess too much fatality?",
+ "raw": {
+ "normalized_reference": "did i possess too much vitality.",
+ "normalized_hypothesis": "did i possess too much fatality?",
+ "errors": 1,
+ "words": 6,
+ "wer": 0.16666666666666666
+ },
+ "semantic": {
+ "normalized_reference": "did i possess too much vitality",
+ "normalized_hypothesis": "did i possess too much fatality",
+ "errors": 1,
+ "words": 6,
+ "wer": 0.16666666666666666
+ }
+ },
+ {
+ "id": "release500_0445",
+ "category": "phonetic_balance",
+ "text": "Therefore, hurrah for the game.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0445.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Therefore, hurrah for the game.",
+ "raw": {
+ "normalized_reference": "therefore, hurrah for the game.",
+ "normalized_hypothesis": "therefore, hurrah for the game.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "therefore hurrah for the game",
+ "normalized_hypothesis": "therefore hurrah for the game",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0446",
+ "category": "phonetic_balance",
+ "text": "He had fulfilled his duty and paid properly.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0446.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He had fulfilled his duty and paid properly.",
+ "raw": {
+ "normalized_reference": "he had fulfilled his duty and paid properly.",
+ "normalized_hypothesis": "he had fulfilled his duty and paid properly.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he had fulfilled his duty and paid properly",
+ "normalized_hypothesis": "he had fulfilled his duty and paid properly",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0447",
+ "category": "phonetic_balance",
+ "text": "How can you manage all alone, Mr Young.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0447.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "How can you manage all alone, Mr. Young?",
+ "raw": {
+ "normalized_reference": "how can you manage all alone, mr young.",
+ "normalized_hypothesis": "how can you manage all alone, mr. young?",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "how can you manage all alone mister young",
+ "normalized_hypothesis": "how can you manage all alone mister young",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0448",
+ "category": "phonetic_balance",
+ "text": "Red-Eye never committed a more outrageous deed.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0448.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Red Eye never committed a more outrageous deed.",
+ "raw": {
+ "normalized_reference": "red-eye never committed a more outrageous deed.",
+ "normalized_hypothesis": "red eye never committed a more outrageous deed.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "red eye never committed a more outrageous deed",
+ "normalized_hypothesis": "red eye never committed a more outrageous deed",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0449",
+ "category": "phonetic_balance",
+ "text": "He would keep his faith with Gregson for the promised day or two.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0449.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He would keep his faith with Gregson for the promised day or two.",
+ "raw": {
+ "normalized_reference": "he would keep his faith with gregson for the promised day or two.",
+ "normalized_hypothesis": "he would keep his faith with gregson for the promised day or two.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he would keep his faith with gregson for the promised day or 2",
+ "normalized_hypothesis": "he would keep his faith with gregson for the promised day or 2",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0450",
+ "category": "phonetic_balance",
+ "text": "There was one chance, and only one, of saving Jeanne.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0450.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "There was one chance, and only one, of saving Gene.",
+ "raw": {
+ "normalized_reference": "there was one chance, and only one, of saving jeanne.",
+ "normalized_hypothesis": "there was one chance, and only one, of saving gene.",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ },
+ "semantic": {
+ "normalized_reference": "there was one chance and only one of saving jeanne",
+ "normalized_hypothesis": "there was one chance and only one of saving gene",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ }
+ },
+ {
+ "id": "release500_0451",
+ "category": "phonetic_balance",
+ "text": "I was beneath the water, suffocating and drowning.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0451.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I was beneath the water, suffocating and drowning.",
+ "raw": {
+ "normalized_reference": "i was beneath the water, suffocating and drowning.",
+ "normalized_hypothesis": "i was beneath the water, suffocating and drowning.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i was beneath the water suffocating and drowning",
+ "normalized_hypothesis": "i was beneath the water suffocating and drowning",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0452",
+ "category": "phonetic_balance",
+ "text": "Give them their choice between a fine or an official whipping.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0452.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Give them their choice between a final or an official whipping.",
+ "raw": {
+ "normalized_reference": "give them their choice between a fine or an official whipping.",
+ "normalized_hypothesis": "give them their choice between a final or an official whipping.",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ },
+ "semantic": {
+ "normalized_reference": "give them their choice between a fine or an official whipping",
+ "normalized_hypothesis": "give them their choice between a final or an official whipping",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ }
+ },
+ {
+ "id": "release500_0453",
+ "category": "phonetic_balance",
+ "text": "They must have been swept away by the chaotic currents.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0453.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "They must have been swept away by the chaotic currents.",
+ "raw": {
+ "normalized_reference": "they must have been swept away by the chaotic currents.",
+ "normalized_hypothesis": "they must have been swept away by the chaotic currents.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "they must have been swept away by the chaotic currents",
+ "normalized_hypothesis": "they must have been swept away by the chaotic currents",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0454",
+ "category": "phonetic_balance",
+ "text": "You yellow giant thing of the frost.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0454.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "You yellow giant thing of the frost.",
+ "raw": {
+ "normalized_reference": "you yellow giant thing of the frost.",
+ "normalized_hypothesis": "you yellow giant thing of the frost.",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "you yellow giant thing of the frost",
+ "normalized_hypothesis": "you yellow giant thing of the frost",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0455",
+ "category": "phonetic_balance",
+ "text": "He had peeled off his shirt and was wildly waving it.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0455.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He had peeled off his shirt and was wildly waving it.",
+ "raw": {
+ "normalized_reference": "he had peeled off his shirt and was wildly waving it.",
+ "normalized_hypothesis": "he had peeled off his shirt and was wildly waving it.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he had peeled off his shirt and was wildly waving it",
+ "normalized_hypothesis": "he had peeled off his shirt and was wildly waving it",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0456",
+ "category": "phonetic_balance",
+ "text": "To these he gave castor oil.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0456.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The lease he gave Castor oil.",
+ "raw": {
+ "normalized_reference": "to these he gave castor oil.",
+ "normalized_hypothesis": "the lease he gave castor oil.",
+ "errors": 2,
+ "words": 6,
+ "wer": 0.3333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "to these he gave castor oil",
+ "normalized_hypothesis": "the lease he gave castor oil",
+ "errors": 2,
+ "words": 6,
+ "wer": 0.3333333333333333
+ }
+ },
+ {
+ "id": "release500_0457",
+ "category": "phonetic_balance",
+ "text": "Tudor surveyed him with withering disgust.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0457.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Tudor surveyed him with withering disgust.",
+ "raw": {
+ "normalized_reference": "tudor surveyed him with withering disgust.",
+ "normalized_hypothesis": "tudor surveyed him with withering disgust.",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "tudor surveyed him with withering disgust",
+ "normalized_hypothesis": "tudor surveyed him with withering disgust",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0458",
+ "category": "phonetic_balance",
+ "text": "After all, it was simply a mistake in judgment.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0458.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "After all, it was simply a mistaken judgment.",
+ "raw": {
+ "normalized_reference": "after all, it was simply a mistake in judgment.",
+ "normalized_hypothesis": "after all, it was simply a mistaken judgment.",
+ "errors": 2,
+ "words": 9,
+ "wer": 0.2222222222222222
+ },
+ "semantic": {
+ "normalized_reference": "after all it was simply a mistake in judgment",
+ "normalized_hypothesis": "after all it was simply a mistaken judgment",
+ "errors": 2,
+ "words": 9,
+ "wer": 0.2222222222222222
+ }
+ },
+ {
+ "id": "release500_0459",
+ "category": "phonetic_balance",
+ "text": "That is the strange part of it.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0459.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "That is the strange part of it.",
+ "raw": {
+ "normalized_reference": "that is the strange part of it.",
+ "normalized_hypothesis": "that is the strange part of it.",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "that is the strange part of it",
+ "normalized_hypothesis": "that is the strange part of it",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0460",
+ "category": "phonetic_balance",
+ "text": "Not a wheel moved in his empire.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0460.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Not a wheel moved in his empire.",
+ "raw": {
+ "normalized_reference": "not a wheel moved in his empire.",
+ "normalized_hypothesis": "not a wheel moved in his empire.",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "not a wheel moved in his empire",
+ "normalized_hypothesis": "not a wheel moved in his empire",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0461",
+ "category": "phonetic_balance",
+ "text": "Its diameter was not more than two hundred yards.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0461.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Its diameter was not more than 200 yards.",
+ "raw": {
+ "normalized_reference": "its diameter was not more than two hundred yards.",
+ "normalized_hypothesis": "its diameter was not more than 200 yards.",
+ "errors": 2,
+ "words": 9,
+ "wer": 0.2222222222222222
+ },
+ "semantic": {
+ "normalized_reference": "its diameter was not more than 200 yards",
+ "normalized_hypothesis": "its diameter was not more than 200 yards",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0462",
+ "category": "phonetic_balance",
+ "text": "Nimrod replied, with a slight manifestation of sensitiveness.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0462.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Nimrod replied with a slight manifestation of sensitiveness.",
+ "raw": {
+ "normalized_reference": "nimrod replied, with a slight manifestation of sensitiveness.",
+ "normalized_hypothesis": "nimrod replied with a slight manifestation of sensitiveness.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "nimrod replied with a slight manifestation of sensitiveness",
+ "normalized_hypothesis": "nimrod replied with a slight manifestation of sensitiveness",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0463",
+ "category": "phonetic_balance",
+ "text": "I tried to read George Moore last night, and was dreadfully bored.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0463.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I tried to read George Moore last night and was dreadfully bored.",
+ "raw": {
+ "normalized_reference": "i tried to read george moore last night, and was dreadfully bored.",
+ "normalized_hypothesis": "i tried to read george moore last night and was dreadfully bored.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i tried to read george moore last night and was dreadfully bored",
+ "normalized_hypothesis": "i tried to read george moore last night and was dreadfully bored",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0464",
+ "category": "phonetic_balance",
+ "text": "Yes, it was a man who asked, a stranger.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0464.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Yes, it was a man who asked a stranger.",
+ "raw": {
+ "normalized_reference": "yes, it was a man who asked, a stranger.",
+ "normalized_hypothesis": "yes, it was a man who asked a stranger.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "yes it was a man who asked a stranger",
+ "normalized_hypothesis": "yes it was a man who asked a stranger",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0465",
+ "category": "phonetic_balance",
+ "text": "We were now good friends.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0465.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "We were now good friends.",
+ "raw": {
+ "normalized_reference": "we were now good friends.",
+ "normalized_hypothesis": "we were now good friends.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "we were now good friends",
+ "normalized_hypothesis": "we were now good friends",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0466",
+ "category": "phonetic_balance",
+ "text": "Yet, in accordance with Ernest's test of truth, it worked.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0466.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Yet, in accordance with Ernest's test of truth, it worked.",
+ "raw": {
+ "normalized_reference": "yet, in accordance with ernest's test of truth, it worked.",
+ "normalized_hypothesis": "yet, in accordance with ernest's test of truth, it worked.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "yet in accordance with ernest is test of truth it worked",
+ "normalized_hypothesis": "yet in accordance with ernest is test of truth it worked",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0467",
+ "category": "phonetic_balance",
+ "text": "It's the strap hangers that'll keep us from going under.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0467.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It's the strap hangers that'll keep us from going under.",
+ "raw": {
+ "normalized_reference": "it's the strap hangers that'll keep us from going under.",
+ "normalized_hypothesis": "it's the strap hangers that'll keep us from going under.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it is the strap hangers that will keep us from going under",
+ "normalized_hypothesis": "it is the strap hangers that will keep us from going under",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0468",
+ "category": "phonetic_balance",
+ "text": "I never allow what can't be changed to annoy me.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0468.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I never allow what can't be changed to annoy me.",
+ "raw": {
+ "normalized_reference": "i never allow what can't be changed to annoy me.",
+ "normalized_hypothesis": "i never allow what can't be changed to annoy me.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i never allow what can not be changed to annoy me",
+ "normalized_hypothesis": "i never allow what can not be changed to annoy me",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0469",
+ "category": "phonetic_balance",
+ "text": "After all superfluous flesh is gone what is left is stringy and resistant.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0469.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "After all superfluous, flesh has become what is left his stringy and resistant.",
+ "raw": {
+ "normalized_reference": "after all superfluous flesh is gone what is left is stringy and resistant.",
+ "normalized_hypothesis": "after all superfluous, flesh has become what is left his stringy and resistant.",
+ "errors": 3,
+ "words": 13,
+ "wer": 0.23076923076923078
+ },
+ "semantic": {
+ "normalized_reference": "after all superfluous flesh is gone what is left is stringy and resistant",
+ "normalized_hypothesis": "after all superfluous flesh has become what is left his stringy and resistant",
+ "errors": 3,
+ "words": 13,
+ "wer": 0.23076923076923078
+ }
+ },
+ {
+ "id": "release500_0470",
+ "category": "phonetic_balance",
+ "text": "Then he shouted, Shut up.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0470.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Then he shouted, shut up.",
+ "raw": {
+ "normalized_reference": "then he shouted, shut up.",
+ "normalized_hypothesis": "then he shouted, shut up.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "then he shouted shut up",
+ "normalized_hypothesis": "then he shouted shut up",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0471",
+ "category": "phonetic_balance",
+ "text": "He had worshiped her, as Dante might have worshiped Beatrice.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0471.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He had worshipped her as Dante might have worshipped Beatrice.",
+ "raw": {
+ "normalized_reference": "he had worshiped her, as dante might have worshiped beatrice.",
+ "normalized_hypothesis": "he had worshipped her as dante might have worshipped beatrice.",
+ "errors": 2,
+ "words": 10,
+ "wer": 0.2
+ },
+ "semantic": {
+ "normalized_reference": "he had worshiped her as dante might have worshiped beatrice",
+ "normalized_hypothesis": "he had worshipped her as dante might have worshipped beatrice",
+ "errors": 2,
+ "words": 10,
+ "wer": 0.2
+ }
+ },
+ {
+ "id": "release500_0472",
+ "category": "phonetic_balance",
+ "text": "He looked like one who had passed through an uncomfortable hour or two.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0472.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He looked like one who had passed through an uncomfortable hour or two.",
+ "raw": {
+ "normalized_reference": "he looked like one who had passed through an uncomfortable hour or two.",
+ "normalized_hypothesis": "he looked like one who had passed through an uncomfortable hour or two.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he looked like one who had passed through an uncomfortable hour or 2",
+ "normalized_hypothesis": "he looked like one who had passed through an uncomfortable hour or 2",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0473",
+ "category": "phonetic_balance",
+ "text": "The history of our westward-faring race is written in it.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0473.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The history of our westward-faring race is written in it.",
+ "raw": {
+ "normalized_reference": "the history of our westward-faring race is written in it.",
+ "normalized_hypothesis": "the history of our westward-faring race is written in it.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the history of our westward faring race is written in it",
+ "normalized_hypothesis": "the history of our westward faring race is written in it",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0474",
+ "category": "phonetic_balance",
+ "text": "He may desire to escape pain, or to enjoy pleasure.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0474.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He may desire to escape pain or to enjoy pleasure.",
+ "raw": {
+ "normalized_reference": "he may desire to escape pain, or to enjoy pleasure.",
+ "normalized_hypothesis": "he may desire to escape pain or to enjoy pleasure.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he may desire to escape pain or to enjoy pleasure",
+ "normalized_hypothesis": "he may desire to escape pain or to enjoy pleasure",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0475",
+ "category": "phonetic_balance",
+ "text": "And the air was growing chilly.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0475.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "And the air was growing chilly.",
+ "raw": {
+ "normalized_reference": "and the air was growing chilly.",
+ "normalized_hypothesis": "and the air was growing chilly.",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "and the air was growing chilly",
+ "normalized_hypothesis": "and the air was growing chilly",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0476",
+ "category": "phonetic_balance",
+ "text": "Dig in; you're sure good, was Daylight's answer.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0476.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Dig in. You're sure good, was Daylight's answer.",
+ "raw": {
+ "normalized_reference": "dig in; you're sure good, was daylight's answer.",
+ "normalized_hypothesis": "dig in. you're sure good, was daylight's answer.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "dig in you are sure good was daylight is answer",
+ "normalized_hypothesis": "dig in you are sure good was daylight is answer",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0477",
+ "category": "phonetic_balance",
+ "text": "My idea was that he would have more influence over the natives.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0477.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "My idea was that he would have more influence over the natives.",
+ "raw": {
+ "normalized_reference": "my idea was that he would have more influence over the natives.",
+ "normalized_hypothesis": "my idea was that he would have more influence over the natives.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "my idea was that he would have more influence over the natives",
+ "normalized_hypothesis": "my idea was that he would have more influence over the natives",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0478",
+ "category": "phonetic_balance",
+ "text": "Now run along, and tell them to hurry.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0478.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Now run along and tell them to hurry.",
+ "raw": {
+ "normalized_reference": "now run along, and tell them to hurry.",
+ "normalized_hypothesis": "now run along and tell them to hurry.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "now run along and tell them to hurry",
+ "normalized_hypothesis": "now run along and tell them to hurry",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0479",
+ "category": "phonetic_balance",
+ "text": "I have no idea, replied Philip.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0479.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I have no idea, replied Philip.",
+ "raw": {
+ "normalized_reference": "i have no idea, replied philip.",
+ "normalized_hypothesis": "i have no idea, replied philip.",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i have no idea replied philip",
+ "normalized_hypothesis": "i have no idea replied philip",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0480",
+ "category": "phonetic_balance",
+ "text": "For the rest, he was a mere automaton.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0480.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "For the rest, he was a mere automaton.",
+ "raw": {
+ "normalized_reference": "for the rest, he was a mere automaton.",
+ "normalized_hypothesis": "for the rest, he was a mere automaton.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "for the rest he was a mere automaton",
+ "normalized_hypothesis": "for the rest he was a mere automaton",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0481",
+ "category": "phonetic_balance",
+ "text": "Why, the average review is more nauseating than cod liver oil.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0481.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Why, the average review is more nauseating than cod liver oil.",
+ "raw": {
+ "normalized_reference": "why, the average review is more nauseating than cod liver oil.",
+ "normalized_hypothesis": "why, the average review is more nauseating than cod liver oil.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "why the average review is more nauseating than cod liver oil",
+ "normalized_hypothesis": "why the average review is more nauseating than cod liver oil",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0482",
+ "category": "phonetic_balance",
+ "text": "I have been robbed, sir, I amended.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0482.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I have been robbed, sir. I am mended.",
+ "raw": {
+ "normalized_reference": "i have been robbed, sir, i amended.",
+ "normalized_hypothesis": "i have been robbed, sir. i am mended.",
+ "errors": 2,
+ "words": 7,
+ "wer": 0.2857142857142857
+ },
+ "semantic": {
+ "normalized_reference": "i have been robbed sir i amended",
+ "normalized_hypothesis": "i have been robbed sir i am mended",
+ "errors": 2,
+ "words": 7,
+ "wer": 0.2857142857142857
+ }
+ },
+ {
+ "id": "release500_0483",
+ "category": "phonetic_balance",
+ "text": "He moved away as quietly as he had come.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0483.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He moved away as quietly as he had come.",
+ "raw": {
+ "normalized_reference": "he moved away as quietly as he had come.",
+ "normalized_hypothesis": "he moved away as quietly as he had come.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he moved away as quietly as he had come",
+ "normalized_hypothesis": "he moved away as quietly as he had come",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0484",
+ "category": "phonetic_balance",
+ "text": "A trickle of fresh blood ran over his face.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0484.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "A trickle of fresh blood ran over his face.",
+ "raw": {
+ "normalized_reference": "a trickle of fresh blood ran over his face.",
+ "normalized_hypothesis": "a trickle of fresh blood ran over his face.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "a trickle of fresh blood ran over his face",
+ "normalized_hypothesis": "a trickle of fresh blood ran over his face",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0485",
+ "category": "phonetic_balance",
+ "text": "Suppose you saw me at work through the window.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0485.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Suppose you saw me at work through the window.",
+ "raw": {
+ "normalized_reference": "suppose you saw me at work through the window.",
+ "normalized_hypothesis": "suppose you saw me at work through the window.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "suppose you saw me at work through the window",
+ "normalized_hypothesis": "suppose you saw me at work through the window",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0486",
+ "category": "phonetic_balance",
+ "text": "But who was Eileen's double.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0486.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "But who was Eileen's double?",
+ "raw": {
+ "normalized_reference": "but who was eileen's double.",
+ "normalized_hypothesis": "but who was eileen's double?",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "but who was eileen is double",
+ "normalized_hypothesis": "but who was eileen is double",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0487",
+ "category": "phonetic_balance",
+ "text": "Darkness hid him from Jeanne.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0487.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Darkness hid him from Gene.",
+ "raw": {
+ "normalized_reference": "darkness hid him from jeanne.",
+ "normalized_hypothesis": "darkness hid him from gene.",
+ "errors": 1,
+ "words": 5,
+ "wer": 0.2
+ },
+ "semantic": {
+ "normalized_reference": "darkness hid him from jeanne",
+ "normalized_hypothesis": "darkness hid him from gene",
+ "errors": 1,
+ "words": 5,
+ "wer": 0.2
+ }
+ },
+ {
+ "id": "release500_0488",
+ "category": "phonetic_balance",
+ "text": "By this time Charley was as enraged as the Greek.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0488.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "By this time Charlie was as enraged as the great.",
+ "raw": {
+ "normalized_reference": "by this time charley was as enraged as the greek.",
+ "normalized_hypothesis": "by this time charlie was as enraged as the great.",
+ "errors": 2,
+ "words": 10,
+ "wer": 0.2
+ },
+ "semantic": {
+ "normalized_reference": "by this time charley was as enraged as the greek",
+ "normalized_hypothesis": "by this time charlie was as enraged as the great",
+ "errors": 2,
+ "words": 10,
+ "wer": 0.2
+ }
+ },
+ {
+ "id": "release500_0489",
+ "category": "phonetic_balance",
+ "text": "I am writing these lines in Honolulu, Hawaii.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0489.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I'm writing these lines in Honolulu, Hawaii.",
+ "raw": {
+ "normalized_reference": "i am writing these lines in honolulu, hawaii.",
+ "normalized_hypothesis": "i'm writing these lines in honolulu, hawaii.",
+ "errors": 2,
+ "words": 8,
+ "wer": 0.25
+ },
+ "semantic": {
+ "normalized_reference": "i am writing these lines in honolulu hawaii",
+ "normalized_hypothesis": "i am writing these lines in honolulu hawaii",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0490",
+ "category": "phonetic_balance",
+ "text": "Nobody knows how the natives got them.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0490.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Nobody knows how the natives got them.",
+ "raw": {
+ "normalized_reference": "nobody knows how the natives got them.",
+ "normalized_hypothesis": "nobody knows how the natives got them.",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "nobody knows how the natives got them",
+ "normalized_hypothesis": "nobody knows how the natives got them",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0491",
+ "category": "phonetic_balance",
+ "text": "And how would we ever find ourselves.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0491.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "And how would we ever find ourselves?",
+ "raw": {
+ "normalized_reference": "and how would we ever find ourselves.",
+ "normalized_hypothesis": "and how would we ever find ourselves?",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "and how would we ever find ourselves",
+ "normalized_hypothesis": "and how would we ever find ourselves",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0492",
+ "category": "phonetic_balance",
+ "text": "They are coming ashore, whoever they are.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0492.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "They are coming ashore, to wherever they are.",
+ "raw": {
+ "normalized_reference": "they are coming ashore, whoever they are.",
+ "normalized_hypothesis": "they are coming ashore, to wherever they are.",
+ "errors": 2,
+ "words": 7,
+ "wer": 0.2857142857142857
+ },
+ "semantic": {
+ "normalized_reference": "they are coming ashore whoever they are",
+ "normalized_hypothesis": "they are coming ashore to wherever they are",
+ "errors": 2,
+ "words": 7,
+ "wer": 0.2857142857142857
+ }
+ },
+ {
+ "id": "release500_0493",
+ "category": "phonetic_balance",
+ "text": "This is my fifth voyage.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0493.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "This is my fifth voyage.",
+ "raw": {
+ "normalized_reference": "this is my fifth voyage.",
+ "normalized_hypothesis": "this is my fifth voyage.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "this is my 5th voyage",
+ "normalized_hypothesis": "this is my 5th voyage",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0494",
+ "category": "phonetic_balance",
+ "text": "And yet if she came he had no words to say.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0494.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "And yet if she came, he had no words to say.",
+ "raw": {
+ "normalized_reference": "and yet if she came he had no words to say.",
+ "normalized_hypothesis": "and yet if she came, he had no words to say.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "and yet if she came he had no words to say",
+ "normalized_hypothesis": "and yet if she came he had no words to say",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0495",
+ "category": "phonetic_balance",
+ "text": "You must sleep, he urged.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0495.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "You must sleep, he urged.",
+ "raw": {
+ "normalized_reference": "you must sleep, he urged.",
+ "normalized_hypothesis": "you must sleep, he urged.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "you must sleep he urged",
+ "normalized_hypothesis": "you must sleep he urged",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0496",
+ "category": "phonetic_balance",
+ "text": "Could the incident have anything to do with Jeanne and Pierre.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0496.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Could the incident have anything to do with Jean and Pierre?",
+ "raw": {
+ "normalized_reference": "could the incident have anything to do with jeanne and pierre.",
+ "normalized_hypothesis": "could the incident have anything to do with jean and pierre?",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ },
+ "semantic": {
+ "normalized_reference": "could the incident have anything to do with jeanne and pierre",
+ "normalized_hypothesis": "could the incident have anything to do with jean and pierre",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ }
+ },
+ {
+ "id": "release500_0497",
+ "category": "phonetic_balance",
+ "text": "Each day she became a more vital part of him.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0497.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Each day she became a more vital part of him.",
+ "raw": {
+ "normalized_reference": "each day she became a more vital part of him.",
+ "normalized_hypothesis": "each day she became a more vital part of him.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "each day she became a more vital part of him",
+ "normalized_hypothesis": "each day she became a more vital part of him",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0498",
+ "category": "phonetic_balance",
+ "text": "But why continue the tirade, for tirade it was.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0498.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "But why continue the tirade? For tirade it was.",
+ "raw": {
+ "normalized_reference": "but why continue the tirade, for tirade it was.",
+ "normalized_hypothesis": "but why continue the tirade? for tirade it was.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "but why continue the tirade for tirade it was",
+ "normalized_hypothesis": "but why continue the tirade for tirade it was",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0499",
+ "category": "phonetic_balance",
+ "text": "That is what distinguishes all of us from the lower animals.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0499.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "That is what distinguishes all of us from the lower animals.",
+ "raw": {
+ "normalized_reference": "that is what distinguishes all of us from the lower animals.",
+ "normalized_hypothesis": "that is what distinguishes all of us from the lower animals.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "that is what distinguishes all of us from the lower animals",
+ "normalized_hypothesis": "that is what distinguishes all of us from the lower animals",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0500",
+ "category": "phonetic_balance",
+ "text": "There was something pathetic in the girl's attitude now.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0500.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "There was something pathetic in the girl's attitude now.",
+ "raw": {
+ "normalized_reference": "there was something pathetic in the girl's attitude now.",
+ "normalized_hypothesis": "there was something pathetic in the girl's attitude now.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "there was something pathetic in the girl is attitude now",
+ "normalized_hypothesis": "there was something pathetic in the girl is attitude now",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0001",
+ "category": "conversational",
+ "text": "It wasn't until later that I realized what had actually happened.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0001.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It wasn't until later that I realized what had actually happened.",
+ "raw": {
+ "normalized_reference": "it wasn't until later that i realized what had actually happened.",
+ "normalized_hypothesis": "it wasn't until later that i realized what had actually happened.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it was not until later that i realized what had actually happened",
+ "normalized_hypothesis": "it was not until later that i realized what had actually happened",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0002",
+ "category": "descriptive",
+ "text": "The light at that time of afternoon is perfect for about twenty minutes.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0002.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The light at that time of afternoon is perfect for about 20 minutes.",
+ "raw": {
+ "normalized_reference": "the light at that time of afternoon is perfect for about twenty minutes.",
+ "normalized_hypothesis": "the light at that time of afternoon is perfect for about 20 minutes.",
+ "errors": 1,
+ "words": 13,
+ "wer": 0.07692307692307693
+ },
+ "semantic": {
+ "normalized_reference": "the light at that time of afternoon is perfect for about 20 minutes",
+ "normalized_hypothesis": "the light at that time of afternoon is perfect for about 20 minutes",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0003",
+ "category": "emotional",
+ "text": "That was the scariest thing I've ever done, and also maybe the best.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0003.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "That was the scariest thing I've ever done, and also maybe the best.",
+ "raw": {
+ "normalized_reference": "that was the scariest thing i've ever done, and also maybe the best.",
+ "normalized_hypothesis": "that was the scariest thing i've ever done, and also maybe the best.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "that was the scariest thing i have ever done and also maybe the best",
+ "normalized_hypothesis": "that was the scariest thing i have ever done and also maybe the best",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0004",
+ "category": "homographs",
+ "text": "After I record the record, please present the present to the bass player.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0004.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "After I record the record, please present the present to the bass player.",
+ "raw": {
+ "normalized_reference": "after i record the record, please present the present to the bass player.",
+ "normalized_hypothesis": "after i record the record, please present the present to the bass player.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "after i record the record please present the present to the bass player",
+ "normalized_hypothesis": "after i record the record please present the present to the bass player",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0005",
+ "category": "long",
+ "text": "If I'm being completely honest, I had no idea what I was doing for most of it, and I think that might be the most normal thing about the whole experience.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0005.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "If I'm being completely honest, I had no idea what I was doing for most of it, and I think that might be the most normal thing about the whole experience.",
+ "raw": {
+ "normalized_reference": "if i'm being completely honest, i had no idea what i was doing for most of it, and i think that might be the most normal thing about the whole experience.",
+ "normalized_hypothesis": "if i'm being completely honest, i had no idea what i was doing for most of it, and i think that might be the most normal thing about the whole experience.",
+ "errors": 0,
+ "words": 31,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "if i am being completely honest i had no idea what i was doing for most of it and i think that might be the most normal thing about the whole experience",
+ "normalized_hypothesis": "if i am being completely honest i had no idea what i was doing for most of it and i think that might be the most normal thing about the whole experience",
+ "errors": 0,
+ "words": 32,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0006",
+ "category": "long_range",
+ "text": "The committee postponed its decision after discovering that three pages of the report had been printed in the wrong order.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0006.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The committee postponed its decision after discovering that three pages of the report had been printed in the wrong order.",
+ "raw": {
+ "normalized_reference": "the committee postponed its decision after discovering that three pages of the report had been printed in the wrong order.",
+ "normalized_hypothesis": "the committee postponed its decision after discovering that three pages of the report had been printed in the wrong order.",
+ "errors": 0,
+ "words": 20,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the committee postponed its decision after discovering that 3 pages of the report had been printed in the wrong order",
+ "normalized_hypothesis": "the committee postponed its decision after discovering that 3 pages of the report had been printed in the wrong order",
+ "errors": 0,
+ "words": 20,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0007",
+ "category": "names_places",
+ "text": "Gwendolyn photographed the eucalyptus trees outside Ljubljana.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0007.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Gwendolyn photographed the eucalyptus trees outside Ljubljana.",
+ "raw": {
+ "normalized_reference": "gwendolyn photographed the eucalyptus trees outside ljubljana.",
+ "normalized_hypothesis": "gwendolyn photographed the eucalyptus trees outside ljubljana.",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "gwendolyn photographed the eucalyptus trees outside ljubljana",
+ "normalized_hypothesis": "gwendolyn photographed the eucalyptus trees outside ljubljana",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0008",
+ "category": "narrative",
+ "text": "He had this theory, and the more time passes, the more I think he was right.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0008.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He had this theory, and the more time passes, the more I think he was right.",
+ "raw": {
+ "normalized_reference": "he had this theory, and the more time passes, the more i think he was right.",
+ "normalized_hypothesis": "he had this theory, and the more time passes, the more i think he was right.",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he had this theory and the more time passes the more i think he was right",
+ "normalized_hypothesis": "he had this theory and the more time passes the more i think he was right",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0009",
+ "category": "numbers",
+ "text": "The package weighs twelve point six kilograms and arrived on July twenty-first.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0009.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The package weighs 12.6 kilograms and arrived on July 21st.",
+ "raw": {
+ "normalized_reference": "the package weighs twelve point six kilograms and arrived on july twenty-first.",
+ "normalized_hypothesis": "the package weighs 12.6 kilograms and arrived on july 21st.",
+ "errors": 5,
+ "words": 13,
+ "wer": 0.38461538461538464
+ },
+ "semantic": {
+ "normalized_reference": "the package weighs 12.6 kilograms and arrived on july 21st",
+ "normalized_hypothesis": "the package weighs 12.6 kilograms and arrived on july 21st",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0010",
+ "category": "punctuation",
+ "text": "First, close the window; second, turn off the lamp; finally, lock the door.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0010.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "First, close the window. Second, turn off the lamp. Finally, lock the door.",
+ "raw": {
+ "normalized_reference": "first, close the window; second, turn off the lamp; finally, lock the door.",
+ "normalized_hypothesis": "first, close the window. second, turn off the lamp. finally, lock the door.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "1st close the window 2nd turn off the lamp finally lock the door",
+ "normalized_hypothesis": "1st close the window 2nd turn off the lamp finally lock the door",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0011",
+ "category": "question",
+ "text": "When did you first realize something was wrong?",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0011.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "When did you first realize something was wrong?",
+ "raw": {
+ "normalized_reference": "when did you first realize something was wrong?",
+ "normalized_hypothesis": "when did you first realize something was wrong?",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "when did you 1st realize something was wrong",
+ "normalized_hypothesis": "when did you 1st realize something was wrong",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0012",
+ "category": "rare_words",
+ "text": "The archaeologist catalogued a Byzantine reliquary and several anachronistic fragments.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0012.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The archaeologist catalogued the Byzantine reliquary and several anachronistic fragments.",
+ "raw": {
+ "normalized_reference": "the archaeologist catalogued a byzantine reliquary and several anachronistic fragments.",
+ "normalized_hypothesis": "the archaeologist catalogued the byzantine reliquary and several anachronistic fragments.",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ },
+ "semantic": {
+ "normalized_reference": "the archaeologist catalogued a byzantine reliquary and several anachronistic fragments",
+ "normalized_hypothesis": "the archaeologist catalogued the byzantine reliquary and several anachronistic fragments",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ }
+ },
+ {
+ "id": "release500_0013",
+ "category": "short",
+ "text": "I have no idea.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0013.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I have no idea.",
+ "raw": {
+ "normalized_reference": "i have no idea.",
+ "normalized_hypothesis": "i have no idea.",
+ "errors": 0,
+ "words": 4,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i have no idea",
+ "normalized_hypothesis": "i have no idea",
+ "errors": 0,
+ "words": 4,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0014",
+ "category": "technical",
+ "text": "The system runs on three core components that all have to stay in sync.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0014.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The system runs on three core components that all have to stay in sync.",
+ "raw": {
+ "normalized_reference": "the system runs on three core components that all have to stay in sync.",
+ "normalized_hypothesis": "the system runs on three core components that all have to stay in sync.",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the system runs on 3 core components that all have to stay in sync",
+ "normalized_hypothesis": "the system runs on 3 core components that all have to stay in sync",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0015",
+ "category": "conversational",
+ "text": "We improvised the entire second half and somehow nobody noticed.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0015.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "We improvised the entire second half and somehow nobody noticed.",
+ "raw": {
+ "normalized_reference": "we improvised the entire second half and somehow nobody noticed.",
+ "normalized_hypothesis": "we improvised the entire second half and somehow nobody noticed.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "we improvised the entire 2nd half and somehow nobody noticed",
+ "normalized_hypothesis": "we improvised the entire 2nd half and somehow nobody noticed",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0016",
+ "category": "descriptive",
+ "text": "There's a smell in old buildings that's impossible to describe but immediately familiar.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0016.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "There's a smell in old buildings that's impossible to describe but immediately familiar.",
+ "raw": {
+ "normalized_reference": "there's a smell in old buildings that's impossible to describe but immediately familiar.",
+ "normalized_hypothesis": "there's a smell in old buildings that's impossible to describe but immediately familiar.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "there is a smell in old buildings that is impossible to describe but immediately familiar",
+ "normalized_hypothesis": "there is a smell in old buildings that is impossible to describe but immediately familiar",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0017",
+ "category": "emotional",
+ "text": "I specifically asked for one thing. One thing. And it still didn't happen.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0017.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I specifically asked for one thing, one thing, and it still didn't happen.",
+ "raw": {
+ "normalized_reference": "i specifically asked for one thing. one thing. and it still didn't happen.",
+ "normalized_hypothesis": "i specifically asked for one thing, one thing, and it still didn't happen.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i specifically asked for one thing one thing and it still did not happen",
+ "normalized_hypothesis": "i specifically asked for one thing one thing and it still did not happen",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0018",
+ "category": "homographs",
+ "text": "They refuse to place the refuse beside the produce aisle.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0018.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "They refused to place the refuse beside the produce aisle.",
+ "raw": {
+ "normalized_reference": "they refuse to place the refuse beside the produce aisle.",
+ "normalized_hypothesis": "they refused to place the refuse beside the produce aisle.",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ },
+ "semantic": {
+ "normalized_reference": "they refuse to place the refuse beside the produce aisle",
+ "normalized_hypothesis": "they refused to place the refuse beside the produce aisle",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ }
+ },
+ {
+ "id": "release500_0019",
+ "category": "long",
+ "text": "There were days last year when I genuinely didn't know how I was going to get through it, and somehow I did, and I still don't fully understand how.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0019.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "There were days last year when I genuinely didn't know how I was going to get through it, and somehow I did, and I still don't fully understand how.",
+ "raw": {
+ "normalized_reference": "there were days last year when i genuinely didn't know how i was going to get through it, and somehow i did, and i still don't fully understand how.",
+ "normalized_hypothesis": "there were days last year when i genuinely didn't know how i was going to get through it, and somehow i did, and i still don't fully understand how.",
+ "errors": 0,
+ "words": 29,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "there were days last year when i genuinely did not know how i was going to get through it and somehow i did and i still do not fully understand how",
+ "normalized_hypothesis": "there were days last year when i genuinely did not know how i was going to get through it and somehow i did and i still do not fully understand how",
+ "errors": 0,
+ "words": 31,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0020",
+ "category": "long_range",
+ "text": "Before announcing the winner, the judge thanked the volunteers who had worked through the storm to repair the damaged stage.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0020.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Before announcing the winner, the judge thanked the volunteers who had worked through this storm to repair the damaged stage.",
+ "raw": {
+ "normalized_reference": "before announcing the winner, the judge thanked the volunteers who had worked through the storm to repair the damaged stage.",
+ "normalized_hypothesis": "before announcing the winner, the judge thanked the volunteers who had worked through this storm to repair the damaged stage.",
+ "errors": 1,
+ "words": 20,
+ "wer": 0.05
+ },
+ "semantic": {
+ "normalized_reference": "before announcing the winner the judge thanked the volunteers who had worked through the storm to repair the damaged stage",
+ "normalized_hypothesis": "before announcing the winner the judge thanked the volunteers who had worked through this storm to repair the damaged stage",
+ "errors": 1,
+ "words": 20,
+ "wer": 0.05
+ }
+ },
+ {
+ "id": "release500_0021",
+ "category": "names_places",
+ "text": "Professor Nguyen mailed the fluorescent poster to Reykjavik on Thursday.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0021.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Professor Newman mailed the fluorescent poster to Reykjavik on Thursday.",
+ "raw": {
+ "normalized_reference": "professor nguyen mailed the fluorescent poster to reykjavik on thursday.",
+ "normalized_hypothesis": "professor newman mailed the fluorescent poster to reykjavik on thursday.",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ },
+ "semantic": {
+ "normalized_reference": "professor nguyen mailed the fluorescent poster to reykjavik on thursday",
+ "normalized_hypothesis": "professor newman mailed the fluorescent poster to reykjavik on thursday",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ }
+ },
+ {
+ "id": "release500_0022",
+ "category": "narrative",
+ "text": "The plan was simple. The execution was anything but.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0022.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The plan was simple. The execution was anything but simple.",
+ "raw": {
+ "normalized_reference": "the plan was simple. the execution was anything but.",
+ "normalized_hypothesis": "the plan was simple. the execution was anything but simple.",
+ "errors": 1,
+ "words": 9,
+ "wer": 0.1111111111111111
+ },
+ "semantic": {
+ "normalized_reference": "the plan was simple the execution was anything but",
+ "normalized_hypothesis": "the plan was simple the execution was anything but simple",
+ "errors": 1,
+ "words": 9,
+ "wer": 0.1111111111111111
+ }
+ },
+ {
+ "id": "release500_0023",
+ "category": "numbers",
+ "text": "Her score improved from eighty-two percent to ninety-seven percent in two weeks.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0023.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Her score improved from 82% to 97% in two weeks.",
+ "raw": {
+ "normalized_reference": "her score improved from eighty-two percent to ninety-seven percent in two weeks.",
+ "normalized_hypothesis": "her score improved from 82% to 97% in two weeks.",
+ "errors": 6,
+ "words": 14,
+ "wer": 0.42857142857142855
+ },
+ "semantic": {
+ "normalized_reference": "her score improved from 82% to 97% in 2 weeks",
+ "normalized_hypothesis": "her score improved from 82% to 97% in 2 weeks",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0024",
+ "category": "punctuation",
+ "text": "Why would anyone pack spoons, batteries, and a violin for one short trip?",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0024.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Why would anyone pack spoons, batteries, and a violin for one short trip?",
+ "raw": {
+ "normalized_reference": "why would anyone pack spoons, batteries, and a violin for one short trip?",
+ "normalized_hypothesis": "why would anyone pack spoons, batteries, and a violin for one short trip?",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "why would anyone pack spoons batteries and a violin for one short trip",
+ "normalized_hypothesis": "why would anyone pack spoons batteries and a violin for one short trip",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0025",
+ "category": "question",
+ "text": "How long have you been sitting on this?",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0025.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "How long have you been sitting on this?",
+ "raw": {
+ "normalized_reference": "how long have you been sitting on this?",
+ "normalized_hypothesis": "how long have you been sitting on this?",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "how long have you been sitting on this",
+ "normalized_hypothesis": "how long have you been sitting on this",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0026",
+ "category": "rare_words",
+ "text": "A mischievous chameleon rested beside the chrysanthemum and the bougainvillea.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0026.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "A mischievous chameleon rested beside the chrysanthemum and the bagging villia.",
+ "raw": {
+ "normalized_reference": "a mischievous chameleon rested beside the chrysanthemum and the bougainvillea.",
+ "normalized_hypothesis": "a mischievous chameleon rested beside the chrysanthemum and the bagging villia.",
+ "errors": 2,
+ "words": 10,
+ "wer": 0.2
+ },
+ "semantic": {
+ "normalized_reference": "a mischievous chameleon rested beside the chrysanthemum and the bougainvillea",
+ "normalized_hypothesis": "a mischievous chameleon rested beside the chrysanthemum and the bagging villia",
+ "errors": 2,
+ "words": 10,
+ "wer": 0.2
+ }
+ },
+ {
+ "id": "release500_0027",
+ "category": "short",
+ "text": "We don't talk about that.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0027.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "We don't talk about that.",
+ "raw": {
+ "normalized_reference": "we don't talk about that.",
+ "normalized_hypothesis": "we don't talk about that.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "we do not talk about that",
+ "normalized_hypothesis": "we do not talk about that",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0028",
+ "category": "technical",
+ "text": "Hash the UTF eight payload with S H A two fifty-six before opening the checkpoint.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0028.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Ash the UTF-8 payload with SH-256 before opening the checkpoint.",
+ "raw": {
+ "normalized_reference": "hash the utf eight payload with s h a two fifty-six before opening the checkpoint.",
+ "normalized_hypothesis": "ash the utf-8 payload with sh-256 before opening the checkpoint.",
+ "errors": 8,
+ "words": 16,
+ "wer": 0.5
+ },
+ "semantic": {
+ "normalized_reference": "hash the utf 8 payload with s h a 256 before opening the checkpoint",
+ "normalized_hypothesis": "ash the utf 8 payload with sh 256 before opening the checkpoint",
+ "errors": 4,
+ "words": 14,
+ "wer": 0.2857142857142857
+ }
+ },
+ {
+ "id": "release500_0029",
+ "category": "conversational",
+ "text": "I thought I had it all figured out, and then this happened.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0029.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I thought I had it all figured out, and then this happened.",
+ "raw": {
+ "normalized_reference": "i thought i had it all figured out, and then this happened.",
+ "normalized_hypothesis": "i thought i had it all figured out, and then this happened.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i thought i had it all figured out and then this happened",
+ "normalized_hypothesis": "i thought i had it all figured out and then this happened",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0030",
+ "category": "descriptive",
+ "text": "The kind of quiet you only get at three in the morning in a city.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0030.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The kind of quiet you only get at three in the morning in a city.",
+ "raw": {
+ "normalized_reference": "the kind of quiet you only get at three in the morning in a city.",
+ "normalized_hypothesis": "the kind of quiet you only get at three in the morning in a city.",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the kind of quiet you only get at 3 in the morning in a city",
+ "normalized_hypothesis": "the kind of quiet you only get at 3 in the morning in a city",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0031",
+ "category": "emotional",
+ "text": "The joy of it was almost overwhelming, like too much good all at once.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0031.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The joy of it was almost overwhelming, like too much good all at once.",
+ "raw": {
+ "normalized_reference": "the joy of it was almost overwhelming, like too much good all at once.",
+ "normalized_hypothesis": "the joy of it was almost overwhelming, like too much good all at once.",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the joy of it was almost overwhelming like too much good all at once",
+ "normalized_hypothesis": "the joy of it was almost overwhelming like too much good all at once",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0032",
+ "category": "homographs",
+ "text": "Please close the door while I stand close to the microphone.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0032.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Please close the door while I stand close to the microphone.",
+ "raw": {
+ "normalized_reference": "please close the door while i stand close to the microphone.",
+ "normalized_hypothesis": "please close the door while i stand close to the microphone.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "please close the door while i stand close to the microphone",
+ "normalized_hypothesis": "please close the door while i stand close to the microphone",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0033",
+ "category": "long",
+ "text": "I think the most underrated skill in the world is knowing when to stop explaining yourself and just let people draw their own conclusions.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0033.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I think the most underrated skill in the world is knowing when to stop explaining yourself and just let people draw their own conclusions.",
+ "raw": {
+ "normalized_reference": "i think the most underrated skill in the world is knowing when to stop explaining yourself and just let people draw their own conclusions.",
+ "normalized_hypothesis": "i think the most underrated skill in the world is knowing when to stop explaining yourself and just let people draw their own conclusions.",
+ "errors": 0,
+ "words": 24,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i think the most underrated skill in the world is knowing when to stop explaining yourself and just let people draw their own conclusions",
+ "normalized_hypothesis": "i think the most underrated skill in the world is knowing when to stop explaining yourself and just let people draw their own conclusions",
+ "errors": 0,
+ "words": 24,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0034",
+ "category": "long_range",
+ "text": "When the final bus disappeared around the corner, Nadia remembered that her umbrella, notebook, and lunch were still inside the quiet office.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0034.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "When the final bus disappeared around the corner, Nadia remembered that her umbrella, notebook, and lunch were still inside the quiet office.",
+ "raw": {
+ "normalized_reference": "when the final bus disappeared around the corner, nadia remembered that her umbrella, notebook, and lunch were still inside the quiet office.",
+ "normalized_hypothesis": "when the final bus disappeared around the corner, nadia remembered that her umbrella, notebook, and lunch were still inside the quiet office.",
+ "errors": 0,
+ "words": 22,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "when the final bus disappeared around the corner nadia remembered that her umbrella notebook and lunch were still inside the quiet office",
+ "normalized_hypothesis": "when the final bus disappeared around the corner nadia remembered that her umbrella notebook and lunch were still inside the quiet office",
+ "errors": 0,
+ "words": 22,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0035",
+ "category": "names_places",
+ "text": "Aurelia and Joaquin crossed the Schuylkill River before noon.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0035.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Aurelia and Joaquin crossed the Chioco River before noon.",
+ "raw": {
+ "normalized_reference": "aurelia and joaquin crossed the schuylkill river before noon.",
+ "normalized_hypothesis": "aurelia and joaquin crossed the chioco river before noon.",
+ "errors": 1,
+ "words": 9,
+ "wer": 0.1111111111111111
+ },
+ "semantic": {
+ "normalized_reference": "aurelia and joaquin crossed the schuylkill river before noon",
+ "normalized_hypothesis": "aurelia and joaquin crossed the chioco river before noon",
+ "errors": 1,
+ "words": 9,
+ "wer": 0.1111111111111111
+ }
+ },
+ {
+ "id": "release500_0036",
+ "category": "narrative",
+ "text": "The village had a name, but nobody used it. Everyone just called it home.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0036.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The village had a name, but nobody used it and everyone just called it home.",
+ "raw": {
+ "normalized_reference": "the village had a name, but nobody used it. everyone just called it home.",
+ "normalized_hypothesis": "the village had a name, but nobody used it and everyone just called it home.",
+ "errors": 1,
+ "words": 14,
+ "wer": 0.07142857142857142
+ },
+ "semantic": {
+ "normalized_reference": "the village had a name but nobody used it everyone just called it home",
+ "normalized_hypothesis": "the village had a name but nobody used it and everyone just called it home",
+ "errors": 1,
+ "words": 14,
+ "wer": 0.07142857142857142
+ }
+ },
+ {
+ "id": "release500_0037",
+ "category": "numbers",
+ "text": "Increase the gain by three decibels, wait five seconds, then reduce it by one.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0037.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Increase the gain by 3 decibels, wait 5 seconds, then reduce it by 1.",
+ "raw": {
+ "normalized_reference": "increase the gain by three decibels, wait five seconds, then reduce it by one.",
+ "normalized_hypothesis": "increase the gain by 3 decibels, wait 5 seconds, then reduce it by 1.",
+ "errors": 3,
+ "words": 14,
+ "wer": 0.21428571428571427
+ },
+ "semantic": {
+ "normalized_reference": "increase the gain by 3 decibels wait 5 seconds then reduce it by one",
+ "normalized_hypothesis": "increase the gain by 3 decibels wait 5 seconds then reduce it by one",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0038",
+ "category": "punctuation",
+ "text": "The result was unexpected: clear, stable, and surprisingly natural.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0038.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The result was unexpected, clear, stable, and surprisingly natural.",
+ "raw": {
+ "normalized_reference": "the result was unexpected: clear, stable, and surprisingly natural.",
+ "normalized_hypothesis": "the result was unexpected, clear, stable, and surprisingly natural.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the result was unexpected clear stable and surprisingly natural",
+ "normalized_hypothesis": "the result was unexpected clear stable and surprisingly natural",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0039",
+ "category": "question",
+ "text": "Who told you that was the right way to do it?",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0039.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Who told you that was the right way to do it?",
+ "raw": {
+ "normalized_reference": "who told you that was the right way to do it?",
+ "normalized_hypothesis": "who told you that was the right way to do it?",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "who told you that was the right way to do it",
+ "normalized_hypothesis": "who told you that was the right way to do it",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0040",
+ "category": "rare_words",
+ "text": "The meticulous cartographer examined an iridescent archipelago through binoculars.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0040.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The meticulous cartographer examined an iridescent archipelago through binoculars.",
+ "raw": {
+ "normalized_reference": "the meticulous cartographer examined an iridescent archipelago through binoculars.",
+ "normalized_hypothesis": "the meticulous cartographer examined an iridescent archipelago through binoculars.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the meticulous cartographer examined an iridescent archipelago through binoculars",
+ "normalized_hypothesis": "the meticulous cartographer examined an iridescent archipelago through binoculars",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0041",
+ "category": "short",
+ "text": "I mean. Yeah. Kind of.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0041.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I mean that yet I kind of...",
+ "raw": {
+ "normalized_reference": "i mean. yeah. kind of.",
+ "normalized_hypothesis": "i mean that yet i kind of...",
+ "errors": 3,
+ "words": 5,
+ "wer": 0.6
+ },
+ "semantic": {
+ "normalized_reference": "i mean yeah kind of",
+ "normalized_hypothesis": "i mean that yet i kind of .",
+ "errors": 3,
+ "words": 5,
+ "wer": 0.6
+ }
+ },
+ {
+ "id": "release500_0042",
+ "category": "technical",
+ "text": "The logs are your best friend when something behaves unexpectedly.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0042.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The logs are your best friend when something behaves unexpectedly.",
+ "raw": {
+ "normalized_reference": "the logs are your best friend when something behaves unexpectedly.",
+ "normalized_hypothesis": "the logs are your best friend when something behaves unexpectedly.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the logs are your best friend when something behaves unexpectedly",
+ "normalized_hypothesis": "the logs are your best friend when something behaves unexpectedly",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0043",
+ "category": "conversational",
+ "text": "He kept talking and I kept nodding and neither of us was listening.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0043.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He kept talking, and I kept nodding, and neither of us was listening.",
+ "raw": {
+ "normalized_reference": "he kept talking and i kept nodding and neither of us was listening.",
+ "normalized_hypothesis": "he kept talking, and i kept nodding, and neither of us was listening.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he kept talking and i kept nodding and neither of us was listening",
+ "normalized_hypothesis": "he kept talking and i kept nodding and neither of us was listening",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0044",
+ "category": "descriptive",
+ "text": "It's one of those places where everyone seems to know each other already.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0044.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It's one of those places where everyone seems to know each other already.",
+ "raw": {
+ "normalized_reference": "it's one of those places where everyone seems to know each other already.",
+ "normalized_hypothesis": "it's one of those places where everyone seems to know each other already.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it is one of those places where everyone seems to know each other already",
+ "normalized_hypothesis": "it is one of those places where everyone seems to know each other already",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0045",
+ "category": "emotional",
+ "text": "He said he was proud of me and nobody had ever said that to me before.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0045.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He said he was proud of me, and nobody had ever said that to me before.",
+ "raw": {
+ "normalized_reference": "he said he was proud of me and nobody had ever said that to me before.",
+ "normalized_hypothesis": "he said he was proud of me, and nobody had ever said that to me before.",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he said he was proud of me and nobody had ever said that to me before",
+ "normalized_hypothesis": "he said he was proud of me and nobody had ever said that to me before",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0046",
+ "category": "homographs",
+ "text": "The minute detail took a minute to explain clearly.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0046.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The minute detail took a minute to explain clearly.",
+ "raw": {
+ "normalized_reference": "the minute detail took a minute to explain clearly.",
+ "normalized_hypothesis": "the minute detail took a minute to explain clearly.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the minute detail took a minute to explain clearly",
+ "normalized_hypothesis": "the minute detail took a minute to explain clearly",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0047",
+ "category": "long",
+ "text": "The strange thing about really wanting something is that getting it never quite feels the way you imagined it would, and not getting it never quite feels as bad.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0047.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The strange thing about really wanting something is that getting it never quite feels the way you imagined it would, and not getting it never quite feels as bad.",
+ "raw": {
+ "normalized_reference": "the strange thing about really wanting something is that getting it never quite feels the way you imagined it would, and not getting it never quite feels as bad.",
+ "normalized_hypothesis": "the strange thing about really wanting something is that getting it never quite feels the way you imagined it would, and not getting it never quite feels as bad.",
+ "errors": 0,
+ "words": 29,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the strange thing about really wanting something is that getting it never quite feels the way you imagined it would and not getting it never quite feels as bad",
+ "normalized_hypothesis": "the strange thing about really wanting something is that getting it never quite feels the way you imagined it would and not getting it never quite feels as bad",
+ "errors": 0,
+ "words": 29,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0048",
+ "category": "long_range",
+ "text": "Even though the recipe looked simple at first, preparing the sauce required patience, careful timing, and a much larger pan.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0048.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Even though the recipe looked simple at first, preparing the sauce required patience, careful timing, and a much larger pan.",
+ "raw": {
+ "normalized_reference": "even though the recipe looked simple at first, preparing the sauce required patience, careful timing, and a much larger pan.",
+ "normalized_hypothesis": "even though the recipe looked simple at first, preparing the sauce required patience, careful timing, and a much larger pan.",
+ "errors": 0,
+ "words": 20,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "even though the recipe looked simple at 1st preparing the sauce required patience careful timing and a much larger pan",
+ "normalized_hypothesis": "even though the recipe looked simple at 1st preparing the sauce required patience careful timing and a much larger pan",
+ "errors": 0,
+ "words": 20,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0049",
+ "category": "names_places",
+ "text": "Xiomara met Siobhan near the old library in Worcestershire.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0049.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Damara met Siobhan near the old library in Worcestershire.",
+ "raw": {
+ "normalized_reference": "xiomara met siobhan near the old library in worcestershire.",
+ "normalized_hypothesis": "damara met siobhan near the old library in worcestershire.",
+ "errors": 1,
+ "words": 9,
+ "wer": 0.1111111111111111
+ },
+ "semantic": {
+ "normalized_reference": "xiomara met siobhan near the old library in worcestershire",
+ "normalized_hypothesis": "damara met siobhan near the old library in worcestershire",
+ "errors": 1,
+ "words": 9,
+ "wer": 0.1111111111111111
+ }
+ },
+ {
+ "id": "release500_0050",
+ "category": "narrative",
+ "text": "She walked into the room and everyone went quiet, not because she was loud, but because she wasn't.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0050.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "She walked into the room and everyone went quiet, not because she was loud, but because she wasn't.",
+ "raw": {
+ "normalized_reference": "she walked into the room and everyone went quiet, not because she was loud, but because she wasn't.",
+ "normalized_hypothesis": "she walked into the room and everyone went quiet, not because she was loud, but because she wasn't.",
+ "errors": 0,
+ "words": 18,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "she walked into the room and everyone went quiet not because she was loud but because she was not",
+ "normalized_hypothesis": "she walked into the room and everyone went quiet not because she was loud but because she was not",
+ "errors": 0,
+ "words": 19,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0051",
+ "category": "numbers",
+ "text": "At exactly seven forty-five, Morgan placed eighteen dollars and fifty cents beside the notebook.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0051.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Exactly $7.45. Morgan placed $18.50 beside the notebook.",
+ "raw": {
+ "normalized_reference": "at exactly seven forty-five, morgan placed eighteen dollars and fifty cents beside the notebook.",
+ "normalized_hypothesis": "exactly $7.45. morgan placed $18.50 beside the notebook.",
+ "errors": 9,
+ "words": 15,
+ "wer": 0.6
+ },
+ "semantic": {
+ "normalized_reference": "at exactly 745 morgan placed $18.50 beside the notebook",
+ "normalized_hypothesis": "exactly $7.45 morgan placed $18.50 beside the notebook",
+ "errors": 3,
+ "words": 10,
+ "wer": 0.3
+ }
+ },
+ {
+ "id": "release500_0052",
+ "category": "punctuation",
+ "text": "Does punctuation work? Yes: commas, semicolons, and periods should create sensible boundaries.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0052.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Does punctuation work? Yes. Commas, semicolons, and periods should create sensible boundaries.",
+ "raw": {
+ "normalized_reference": "does punctuation work? yes: commas, semicolons, and periods should create sensible boundaries.",
+ "normalized_hypothesis": "does punctuation work? yes. commas, semicolons, and periods should create sensible boundaries.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "does punctuation work yes commas semicolons and periods should create sensible boundaries",
+ "normalized_hypothesis": "does punctuation work yes commas semicolons and periods should create sensible boundaries",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0053",
+ "category": "question",
+ "text": "Have you ever just stopped and thought about how strange all of this is?",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0053.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Have you ever just stopped and thought about how strange all of this is?",
+ "raw": {
+ "normalized_reference": "have you ever just stopped and thought about how strange all of this is?",
+ "normalized_hypothesis": "have you ever just stopped and thought about how strange all of this is?",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "have you ever just stopped and thought about how strange all of this is",
+ "normalized_hypothesis": "have you ever just stopped and thought about how strange all of this is",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0054",
+ "category": "rare_words",
+ "text": "Her idiosyncratic explanation sounded plausible but remained fundamentally ambiguous.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0054.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Her idiosyncratic explanation sounded plausible, but remained fundamentally ambiguous.",
+ "raw": {
+ "normalized_reference": "her idiosyncratic explanation sounded plausible but remained fundamentally ambiguous.",
+ "normalized_hypothesis": "her idiosyncratic explanation sounded plausible, but remained fundamentally ambiguous.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "her idiosyncratic explanation sounded plausible but remained fundamentally ambiguous",
+ "normalized_hypothesis": "her idiosyncratic explanation sounded plausible but remained fundamentally ambiguous",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0055",
+ "category": "short",
+ "text": "That tracks.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0055.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "That tracks.",
+ "raw": {
+ "normalized_reference": "that tracks.",
+ "normalized_hypothesis": "that tracks.",
+ "errors": 0,
+ "words": 2,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "that tracks",
+ "normalized_hypothesis": "that tracks",
+ "errors": 0,
+ "words": 2,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0056",
+ "category": "technical",
+ "text": "The key thing to remember is that order matters here more than you'd think.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0056.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The key thing to remember is that order matters here more than you'd think.",
+ "raw": {
+ "normalized_reference": "the key thing to remember is that order matters here more than you'd think.",
+ "normalized_hypothesis": "the key thing to remember is that order matters here more than you'd think.",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the key thing to remember is that order matters here more than you would think",
+ "normalized_hypothesis": "the key thing to remember is that order matters here more than you would think",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0057",
+ "category": "conversational",
+ "text": "I would not say that again in this room if I were you.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0057.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I would not say that again in this room if I were you.",
+ "raw": {
+ "normalized_reference": "i would not say that again in this room if i were you.",
+ "normalized_hypothesis": "i would not say that again in this room if i were you.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i would not say that again in this room if i were you",
+ "normalized_hypothesis": "i would not say that again in this room if i were you",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0058",
+ "category": "descriptive",
+ "text": "The room was filled with the quiet that means something is about to change.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0058.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The room was filled with a quiet that means something is about to change.",
+ "raw": {
+ "normalized_reference": "the room was filled with the quiet that means something is about to change.",
+ "normalized_hypothesis": "the room was filled with a quiet that means something is about to change.",
+ "errors": 1,
+ "words": 14,
+ "wer": 0.07142857142857142
+ },
+ "semantic": {
+ "normalized_reference": "the room was filled with the quiet that means something is about to change",
+ "normalized_hypothesis": "the room was filled with a quiet that means something is about to change",
+ "errors": 1,
+ "words": 14,
+ "wer": 0.07142857142857142
+ }
+ },
+ {
+ "id": "release500_0059",
+ "category": "emotional",
+ "text": "I didn't expect to feel that way about it, but I did.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0059.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I didn't expect to feel that way about it, but I did.",
+ "raw": {
+ "normalized_reference": "i didn't expect to feel that way about it, but i did.",
+ "normalized_hypothesis": "i didn't expect to feel that way about it, but i did.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i did not expect to feel that way about it but i did",
+ "normalized_hypothesis": "i did not expect to feel that way about it but i did",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0060",
+ "category": "homographs",
+ "text": "I object to the strange object that was left near the entrance.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0060.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I object to the strange object that was left near the entrance.",
+ "raw": {
+ "normalized_reference": "i object to the strange object that was left near the entrance.",
+ "normalized_hypothesis": "i object to the strange object that was left near the entrance.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i object to the strange object that was left near the entrance",
+ "normalized_hypothesis": "i object to the strange object that was left near the entrance",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0061",
+ "category": "long",
+ "text": "I've learned more from the things that went wrong than from the things that worked, which is a frustrating way to have to learn things but apparently the only way that stuck.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0061.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I've learned more from the things that went wrong than from the things that worked, which is a frustrating way to have to learn things but apparently the only way that's to...",
+ "raw": {
+ "normalized_reference": "i've learned more from the things that went wrong than from the things that worked, which is a frustrating way to have to learn things but apparently the only way that stuck.",
+ "normalized_hypothesis": "i've learned more from the things that went wrong than from the things that worked, which is a frustrating way to have to learn things but apparently the only way that's to...",
+ "errors": 2,
+ "words": 32,
+ "wer": 0.0625
+ },
+ "semantic": {
+ "normalized_reference": "i have learned more from the things that went wrong than from the things that worked which is a frustrating way to have to learn things but apparently the only way that stuck",
+ "normalized_hypothesis": "i have learned more from the things that went wrong than from the things that worked which is a frustrating way to have to learn things but apparently the only way that is to .",
+ "errors": 2,
+ "words": 33,
+ "wer": 0.06060606060606061
+ }
+ },
+ {
+ "id": "release500_0062",
+ "category": "long_range",
+ "text": "Because every nearby hotel was full, the exhausted hikers followed the narrow road until they found a farmhouse with one empty room.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0062.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Because every nearby hotel was full, the exhausted hikers followed the narrow road until they found a farmhouse with one empty room.",
+ "raw": {
+ "normalized_reference": "because every nearby hotel was full, the exhausted hikers followed the narrow road until they found a farmhouse with one empty room.",
+ "normalized_hypothesis": "because every nearby hotel was full, the exhausted hikers followed the narrow road until they found a farmhouse with one empty room.",
+ "errors": 0,
+ "words": 22,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "because every nearby hotel was full the exhausted hikers followed the narrow road until they found a farmhouse with one empty room",
+ "normalized_hypothesis": "because every nearby hotel was full the exhausted hikers followed the narrow road until they found a farmhouse with one empty room",
+ "errors": 0,
+ "words": 22,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0063",
+ "category": "names_places",
+ "text": "The itinerary sends Beatrice from Albuquerque to Des Moines and then to Boise.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0063.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The itinerary sends Beatrice from Albuquerque to Des Moines, and then to Boise.",
+ "raw": {
+ "normalized_reference": "the itinerary sends beatrice from albuquerque to des moines and then to boise.",
+ "normalized_hypothesis": "the itinerary sends beatrice from albuquerque to des moines, and then to boise.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the itinerary sends beatrice from albuquerque to des moines and then to boise",
+ "normalized_hypothesis": "the itinerary sends beatrice from albuquerque to des moines and then to boise",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0064",
+ "category": "narrative",
+ "text": "We had been driving for six hours when we finally saw the sign for the town.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0064.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "We had been driving for six hours when we finally saw the sign for the town.",
+ "raw": {
+ "normalized_reference": "we had been driving for six hours when we finally saw the sign for the town.",
+ "normalized_hypothesis": "we had been driving for six hours when we finally saw the sign for the town.",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "we had been driving for 6 hours when we finally saw the sign for the town",
+ "normalized_hypothesis": "we had been driving for 6 hours when we finally saw the sign for the town",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0065",
+ "category": "numbers",
+ "text": "Call extension zero four nine after eight thirty in the morning.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0065.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Call extension 049 after 8.30 in the morning.",
+ "raw": {
+ "normalized_reference": "call extension zero four nine after eight thirty in the morning.",
+ "normalized_hypothesis": "call extension 049 after 8.30 in the morning.",
+ "errors": 5,
+ "words": 11,
+ "wer": 0.45454545454545453
+ },
+ "semantic": {
+ "normalized_reference": "call extension 049 after 830 in the morning",
+ "normalized_hypothesis": "call extension 49 after 8.30 in the morning",
+ "errors": 3,
+ "words": 8,
+ "wer": 0.375
+ }
+ },
+ {
+ "id": "release500_0066",
+ "category": "punctuation",
+ "text": "Stop - listen carefully - then answer the question without guessing.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0066.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Stop, listen carefully, then answer the question without guessing.",
+ "raw": {
+ "normalized_reference": "stop - listen carefully - then answer the question without guessing.",
+ "normalized_hypothesis": "stop, listen carefully, then answer the question without guessing.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "stop listen carefully then answer the question without guessing",
+ "normalized_hypothesis": "stop listen carefully then answer the question without guessing",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0067",
+ "category": "question",
+ "text": "Does it matter at this point, honestly?",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0067.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Does it matter at this point? Honestly?",
+ "raw": {
+ "normalized_reference": "does it matter at this point, honestly?",
+ "normalized_hypothesis": "does it matter at this point? honestly?",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "does it matter at this point honestly",
+ "normalized_hypothesis": "does it matter at this point honestly",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0068",
+ "category": "rare_words",
+ "text": "An entrepreneurial meteorologist described the unprecedented atmospheric phenomenon.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0068.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "An entrepreneurial meteorologist described the unprecedented atmospheric phenomenon.",
+ "raw": {
+ "normalized_reference": "an entrepreneurial meteorologist described the unprecedented atmospheric phenomenon.",
+ "normalized_hypothesis": "an entrepreneurial meteorologist described the unprecedented atmospheric phenomenon.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "an entrepreneurial meteorologist described the unprecedented atmospheric phenomenon",
+ "normalized_hypothesis": "an entrepreneurial meteorologist described the unprecedented atmospheric phenomenon",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0069",
+ "category": "short",
+ "text": "That's not what I expected.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0069.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "That's not what I expected.",
+ "raw": {
+ "normalized_reference": "that's not what i expected.",
+ "normalized_hypothesis": "that's not what i expected.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "that is not what i expected",
+ "normalized_hypothesis": "that is not what i expected",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0070",
+ "category": "technical",
+ "text": "There's a known issue with that version that causes failures under load.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0070.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "There's a known issue with that version that causes failures under load.",
+ "raw": {
+ "normalized_reference": "there's a known issue with that version that causes failures under load.",
+ "normalized_hypothesis": "there's a known issue with that version that causes failures under load.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "there is a known issue with that version that causes failures under load",
+ "normalized_hypothesis": "there is a known issue with that version that causes failures under load",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0071",
+ "category": "conversational",
+ "text": "Don't look now but I think that's him over there.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0071.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Don't look now, but I think that's him over there.",
+ "raw": {
+ "normalized_reference": "don't look now but i think that's him over there.",
+ "normalized_hypothesis": "don't look now, but i think that's him over there.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "do not look now but i think that is him over there",
+ "normalized_hypothesis": "do not look now but i think that is him over there",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0072",
+ "category": "descriptive",
+ "text": "The city at night looks completely different from how it feels during the day.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0072.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The city at night looks completely different from how it feels during the day.",
+ "raw": {
+ "normalized_reference": "the city at night looks completely different from how it feels during the day.",
+ "normalized_hypothesis": "the city at night looks completely different from how it feels during the day.",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the city at night looks completely different from how it feels during the day",
+ "normalized_hypothesis": "the city at night looks completely different from how it feels during the day",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0073",
+ "category": "emotional",
+ "text": "It's the kind of happy that makes you nervous because you know it can't last.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0073.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It's the kind of happy that makes you nervous because you know it can't last.",
+ "raw": {
+ "normalized_reference": "it's the kind of happy that makes you nervous because you know it can't last.",
+ "normalized_hypothesis": "it's the kind of happy that makes you nervous because you know it can't last.",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it is the kind of happy that makes you nervous because you know it can not last",
+ "normalized_hypothesis": "it is the kind of happy that makes you nervous because you know it can not last",
+ "errors": 0,
+ "words": 17,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0074",
+ "category": "homographs",
+ "text": "The dove dove beneath the branch while the wind began to wind the loose cable.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0074.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The dub-dub beneath the branch while the wind began to wind the loose cable.",
+ "raw": {
+ "normalized_reference": "the dove dove beneath the branch while the wind began to wind the loose cable.",
+ "normalized_hypothesis": "the dub-dub beneath the branch while the wind began to wind the loose cable.",
+ "errors": 2,
+ "words": 15,
+ "wer": 0.13333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "the dove dove beneath the branch while the wind began to wind the loose cable",
+ "normalized_hypothesis": "the dub dub beneath the branch while the wind began to wind the loose cable",
+ "errors": 2,
+ "words": 15,
+ "wer": 0.13333333333333333
+ }
+ },
+ {
+ "id": "release500_0075",
+ "category": "long",
+ "text": "I've thought a lot about what I would do differently if I could go back, and the honest answer is probably not as much as I'd like to think.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0075.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I've thought a lot about what I would do differently if I could go back, and the honest answer is probably not as much as I'd like to think.",
+ "raw": {
+ "normalized_reference": "i've thought a lot about what i would do differently if i could go back, and the honest answer is probably not as much as i'd like to think.",
+ "normalized_hypothesis": "i've thought a lot about what i would do differently if i could go back, and the honest answer is probably not as much as i'd like to think.",
+ "errors": 0,
+ "words": 29,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i have thought a lot about what i would do differently if i could go back and the honest answer is probably not as much as i would like to think",
+ "normalized_hypothesis": "i have thought a lot about what i would do differently if i could go back and the honest answer is probably not as much as i would like to think",
+ "errors": 0,
+ "words": 31,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0076",
+ "category": "long_range",
+ "text": "Although the weather changed without warning, the musicians finished packing before anyone opened the enormous glass door.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0076.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Although the weather changed without warning, the musicians finished packing before anyone opened the enormous glass door.",
+ "raw": {
+ "normalized_reference": "although the weather changed without warning, the musicians finished packing before anyone opened the enormous glass door.",
+ "normalized_hypothesis": "although the weather changed without warning, the musicians finished packing before anyone opened the enormous glass door.",
+ "errors": 0,
+ "words": 17,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "although the weather changed without warning the musicians finished packing before anyone opened the enormous glass door",
+ "normalized_hypothesis": "although the weather changed without warning the musicians finished packing before anyone opened the enormous glass door",
+ "errors": 0,
+ "words": 17,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0077",
+ "category": "names_places",
+ "text": "Saskatchewan looked unusually bright beneath the pale February sunrise.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0077.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Saskatchewan looked unusually bright beneath the pale February sunrise.",
+ "raw": {
+ "normalized_reference": "saskatchewan looked unusually bright beneath the pale february sunrise.",
+ "normalized_hypothesis": "saskatchewan looked unusually bright beneath the pale february sunrise.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "saskatchewan looked unusually bright beneath the pale february sunrise",
+ "normalized_hypothesis": "saskatchewan looked unusually bright beneath the pale february sunrise",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0078",
+ "category": "narrative",
+ "text": "She handled it with a grace I genuinely did not expect.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0078.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "She handled it with a grace I genuinely did not expect.",
+ "raw": {
+ "normalized_reference": "she handled it with a grace i genuinely did not expect.",
+ "normalized_hypothesis": "she handled it with a grace i genuinely did not expect.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "she handled it with a grace i genuinely did not expect",
+ "normalized_hypothesis": "she handled it with a grace i genuinely did not expect",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0079",
+ "category": "numbers",
+ "text": "Apartment four hundred seven is across from room twenty-nine on the third floor.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0079.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Apartment 407 is across from room 29 on the third floor.",
+ "raw": {
+ "normalized_reference": "apartment four hundred seven is across from room twenty-nine on the third floor.",
+ "normalized_hypothesis": "apartment 407 is across from room 29 on the third floor.",
+ "errors": 5,
+ "words": 14,
+ "wer": 0.35714285714285715
+ },
+ "semantic": {
+ "normalized_reference": "apartment 407 is across from room 29 on the 3rd floor",
+ "normalized_hypothesis": "apartment 407 is across from room 29 on the 3rd floor",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0080",
+ "category": "punctuation",
+ "text": "\"Not yet,\" Rowan whispered. \"Give me another minute... please.\"",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0080.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Not yet, Rowan whispered. Give me another minute, please.",
+ "raw": {
+ "normalized_reference": "\"not yet,\" rowan whispered. \"give me another minute... please.\"",
+ "normalized_hypothesis": "not yet, rowan whispered. give me another minute, please.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "not yet rowan whispered give me another minute please",
+ "normalized_hypothesis": "not yet rowan whispered give me another minute please",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0081",
+ "category": "question",
+ "text": "What are we actually trying to accomplish here?",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0081.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "What are we actually trying to accomplish here?",
+ "raw": {
+ "normalized_reference": "what are we actually trying to accomplish here?",
+ "normalized_hypothesis": "what are we actually trying to accomplish here?",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "what are we actually trying to accomplish here",
+ "normalized_hypothesis": "what are we actually trying to accomplish here",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0082",
+ "category": "rare_words",
+ "text": "The physician recommended a noninvasive procedure for the recurring neurological symptoms.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0082.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The physician recommended a non-invasive procedure for the recurring neurological symptoms.",
+ "raw": {
+ "normalized_reference": "the physician recommended a noninvasive procedure for the recurring neurological symptoms.",
+ "normalized_hypothesis": "the physician recommended a non-invasive procedure for the recurring neurological symptoms.",
+ "errors": 2,
+ "words": 11,
+ "wer": 0.18181818181818182
+ },
+ "semantic": {
+ "normalized_reference": "the physician recommended a noninvasive procedure for the recurring neurological symptoms",
+ "normalized_hypothesis": "the physician recommended a non invasive procedure for the recurring neurological symptoms",
+ "errors": 2,
+ "words": 11,
+ "wer": 0.18181818181818182
+ }
+ },
+ {
+ "id": "release500_0083",
+ "category": "short",
+ "text": "Are you serious right now?",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0083.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Are you serious right now?",
+ "raw": {
+ "normalized_reference": "are you serious right now?",
+ "normalized_hypothesis": "are you serious right now?",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "are you serious right now",
+ "normalized_hypothesis": "are you serious right now",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0084",
+ "category": "technical",
+ "text": "The performance improvement is real, but it comes with tradeoffs.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0084.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The performance improvement is real, but it comes with trade-offs.",
+ "raw": {
+ "normalized_reference": "the performance improvement is real, but it comes with tradeoffs.",
+ "normalized_hypothesis": "the performance improvement is real, but it comes with trade-offs.",
+ "errors": 2,
+ "words": 10,
+ "wer": 0.2
+ },
+ "semantic": {
+ "normalized_reference": "the performance improvement is real but it comes with tradeoffs",
+ "normalized_hypothesis": "the performance improvement is real but it comes with trade offs",
+ "errors": 2,
+ "words": 10,
+ "wer": 0.2
+ }
+ },
+ {
+ "id": "release500_0085",
+ "category": "conversational",
+ "text": "I don't know. It's just been one of those weeks, you know?",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0085.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I don't know. It's just been one of those weeks, you know?",
+ "raw": {
+ "normalized_reference": "i don't know. it's just been one of those weeks, you know?",
+ "normalized_hypothesis": "i don't know. it's just been one of those weeks, you know?",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i do not know it is just been one of those weeks you know",
+ "normalized_hypothesis": "i do not know it is just been one of those weeks you know",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0086",
+ "category": "descriptive",
+ "text": "The moment you stop expecting perfection, things get a lot more interesting.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0086.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The moment you stop expecting perfection, things get a lot more interesting.",
+ "raw": {
+ "normalized_reference": "the moment you stop expecting perfection, things get a lot more interesting.",
+ "normalized_hypothesis": "the moment you stop expecting perfection, things get a lot more interesting.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the moment you stop expecting perfection things get a lot more interesting",
+ "normalized_hypothesis": "the moment you stop expecting perfection things get a lot more interesting",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0087",
+ "category": "emotional",
+ "text": "It's the kind of thing that stays with you even years later.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0087.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It's the kind of thing that stays with you even years later.",
+ "raw": {
+ "normalized_reference": "it's the kind of thing that stays with you even years later.",
+ "normalized_hypothesis": "it's the kind of thing that stays with you even years later.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it is the kind of thing that stays with you even years later",
+ "normalized_hypothesis": "it is the kind of thing that stays with you even years later",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0088",
+ "category": "long",
+ "text": "She told me once that the most important skill she'd developed was learning to disagree with someone without making them feel they were wrong.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0088.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "She told me once that the most important skill she'd developed was learning to disagree with someone without making them feel they were wrong.",
+ "raw": {
+ "normalized_reference": "she told me once that the most important skill she'd developed was learning to disagree with someone without making them feel they were wrong.",
+ "normalized_hypothesis": "she told me once that the most important skill she'd developed was learning to disagree with someone without making them feel they were wrong.",
+ "errors": 0,
+ "words": 24,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "she told me once that the most important skill she would developed was learning to disagree with someone without making them feel they were wrong",
+ "normalized_hypothesis": "she told me once that the most important skill she would developed was learning to disagree with someone without making them feel they were wrong",
+ "errors": 0,
+ "words": 25,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0089",
+ "category": "narrative",
+ "text": "He stood at the edge of the cliff and wondered, not for the first time, how he'd gotten here.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0089.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He stood at the edge of the cliff and wondered, not for the first time, how he'd gotten here.",
+ "raw": {
+ "normalized_reference": "he stood at the edge of the cliff and wondered, not for the first time, how he'd gotten here.",
+ "normalized_hypothesis": "he stood at the edge of the cliff and wondered, not for the first time, how he'd gotten here.",
+ "errors": 0,
+ "words": 19,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he stood at the edge of the cliff and wondered not for the 1st time how he would gotten here",
+ "normalized_hypothesis": "he stood at the edge of the cliff and wondered not for the 1st time how he would gotten here",
+ "errors": 0,
+ "words": 20,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0090",
+ "category": "question",
+ "text": "Why does asking for help feel so hard when you actually need it?",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0090.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Why does asking for help feel so hard when you actually need it?",
+ "raw": {
+ "normalized_reference": "why does asking for help feel so hard when you actually need it?",
+ "normalized_hypothesis": "why does asking for help feel so hard when you actually need it?",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "why does asking for help feel so hard when you actually need it",
+ "normalized_hypothesis": "why does asking for help feel so hard when you actually need it",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0091",
+ "category": "short",
+ "text": "That's fair.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0091.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "That's fair.",
+ "raw": {
+ "normalized_reference": "that's fair.",
+ "normalized_hypothesis": "that's fair.",
+ "errors": 0,
+ "words": 2,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "that is fair",
+ "normalized_hypothesis": "that is fair",
+ "errors": 0,
+ "words": 3,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0092",
+ "category": "technical",
+ "text": "If you run into that error, the fix is usually just restarting the service.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0092.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "If you run into that error, the fix is usually just restarting the service.",
+ "raw": {
+ "normalized_reference": "if you run into that error, the fix is usually just restarting the service.",
+ "normalized_hypothesis": "if you run into that error, the fix is usually just restarting the service.",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "if you run into that error the fix is usually just restarting the service",
+ "normalized_hypothesis": "if you run into that error the fix is usually just restarting the service",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0093",
+ "category": "conversational",
+ "text": "It was a disaster in the best possible way.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0093.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It was a disaster in the best possible way.",
+ "raw": {
+ "normalized_reference": "it was a disaster in the best possible way.",
+ "normalized_hypothesis": "it was a disaster in the best possible way.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it was a disaster in the best possible way",
+ "normalized_hypothesis": "it was a disaster in the best possible way",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0094",
+ "category": "descriptive",
+ "text": "It had the look of a place that used to mean something to someone.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0094.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It had the look of a place that used to mean something to someone.",
+ "raw": {
+ "normalized_reference": "it had the look of a place that used to mean something to someone.",
+ "normalized_hypothesis": "it had the look of a place that used to mean something to someone.",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it had the look of a place that used to mean something to someone",
+ "normalized_hypothesis": "it had the look of a place that used to mean something to someone",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0095",
+ "category": "emotional",
+ "text": "There's this feeling you get when something finally works after weeks of failure.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0095.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "There's this feeling you get when something finally works after weeks of failure.",
+ "raw": {
+ "normalized_reference": "there's this feeling you get when something finally works after weeks of failure.",
+ "normalized_hypothesis": "there's this feeling you get when something finally works after weeks of failure.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "there is this feeling you get when something finally works after weeks of failure",
+ "normalized_hypothesis": "there is this feeling you get when something finally works after weeks of failure",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0096",
+ "category": "long",
+ "text": "He had a theory that most arguments aren't actually about what they seem to be about, and the longer I live the more I think he was onto something.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0096.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He had a theory that most arguments aren't actually about what they seem to be about, and the longer I lived the more I think he was onto something.",
+ "raw": {
+ "normalized_reference": "he had a theory that most arguments aren't actually about what they seem to be about, and the longer i live the more i think he was onto something.",
+ "normalized_hypothesis": "he had a theory that most arguments aren't actually about what they seem to be about, and the longer i lived the more i think he was onto something.",
+ "errors": 1,
+ "words": 29,
+ "wer": 0.034482758620689655
+ },
+ "semantic": {
+ "normalized_reference": "he had a theory that most arguments are not actually about what they seem to be about and the longer i live the more i think he was onto something",
+ "normalized_hypothesis": "he had a theory that most arguments are not actually about what they seem to be about and the longer i lived the more i think he was onto something",
+ "errors": 1,
+ "words": 30,
+ "wer": 0.03333333333333333
+ }
+ },
+ {
+ "id": "release500_0097",
+ "category": "narrative",
+ "text": "The thing nobody tells you about starting over is how ordinary most of it feels.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0097.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The thing nobody tells you about starting over is how ordinary most of it feels.",
+ "raw": {
+ "normalized_reference": "the thing nobody tells you about starting over is how ordinary most of it feels.",
+ "normalized_hypothesis": "the thing nobody tells you about starting over is how ordinary most of it feels.",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the thing nobody tells you about starting over is how ordinary most of it feels",
+ "normalized_hypothesis": "the thing nobody tells you about starting over is how ordinary most of it feels",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0098",
+ "category": "question",
+ "text": "What exactly are we waiting for at this point?",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0098.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "What exactly are we waiting for at this point?",
+ "raw": {
+ "normalized_reference": "what exactly are we waiting for at this point?",
+ "normalized_hypothesis": "what exactly are we waiting for at this point?",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "what exactly are we waiting for at this point",
+ "normalized_hypothesis": "what exactly are we waiting for at this point",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0099",
+ "category": "short",
+ "text": "I stand by what I said.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0099.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I stand by what I said.",
+ "raw": {
+ "normalized_reference": "i stand by what i said.",
+ "normalized_hypothesis": "i stand by what i said.",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i stand by what i said",
+ "normalized_hypothesis": "i stand by what i said",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0100",
+ "category": "technical",
+ "text": "Double-check that value before you push anything to production.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0100.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Double check that value before you push anything to production.",
+ "raw": {
+ "normalized_reference": "double-check that value before you push anything to production.",
+ "normalized_hypothesis": "double check that value before you push anything to production.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "double check that value before you push anything to production",
+ "normalized_hypothesis": "double check that value before you push anything to production",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0101",
+ "category": "conversational",
+ "text": "I understand your point; I just do not agree with the conclusion.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0101.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I understand your point. I just do not agree with the conclusion.",
+ "raw": {
+ "normalized_reference": "i understand your point; i just do not agree with the conclusion.",
+ "normalized_hypothesis": "i understand your point. i just do not agree with the conclusion.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i understand your point i just do not agree with the conclusion",
+ "normalized_hypothesis": "i understand your point i just do not agree with the conclusion",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0102",
+ "category": "descriptive",
+ "text": "The sky that day was a strange shade of orange I'd never seen before.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0102.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The sky that day was a strange shade of orange I'd never seen before.",
+ "raw": {
+ "normalized_reference": "the sky that day was a strange shade of orange i'd never seen before.",
+ "normalized_hypothesis": "the sky that day was a strange shade of orange i'd never seen before.",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the sky that day was a strange shade of orange i would never seen before",
+ "normalized_hypothesis": "the sky that day was a strange shade of orange i would never seen before",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0103",
+ "category": "emotional",
+ "text": "It hurt more than I expected, and I thought I was prepared.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0103.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It hurt more than I expected, and I thought I was prepared.",
+ "raw": {
+ "normalized_reference": "it hurt more than i expected, and i thought i was prepared.",
+ "normalized_hypothesis": "it hurt more than i expected, and i thought i was prepared.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it hurt more than i expected and i thought i was prepared",
+ "normalized_hypothesis": "it hurt more than i expected and i thought i was prepared",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0104",
+ "category": "long",
+ "text": "The hardest part wasn't making the decision, it was sitting with the uncertainty of not knowing whether it was the right one for months afterward.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0104.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The hardest part wasn't making the decision. It was sitting with the uncertainty of not knowing whether it was the right one for months afterward.",
+ "raw": {
+ "normalized_reference": "the hardest part wasn't making the decision, it was sitting with the uncertainty of not knowing whether it was the right one for months afterward.",
+ "normalized_hypothesis": "the hardest part wasn't making the decision. it was sitting with the uncertainty of not knowing whether it was the right one for months afterward.",
+ "errors": 0,
+ "words": 25,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the hardest part was not making the decision it was sitting with the uncertainty of not knowing whether it was the right one for months afterward",
+ "normalized_hypothesis": "the hardest part was not making the decision it was sitting with the uncertainty of not knowing whether it was the right one for months afterward",
+ "errors": 0,
+ "words": 26,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0105",
+ "category": "narrative",
+ "text": "There was this one day last summer when everything just clicked into place.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0105.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "There was this one day last summer when everything just clicked into place.",
+ "raw": {
+ "normalized_reference": "there was this one day last summer when everything just clicked into place.",
+ "normalized_hypothesis": "there was this one day last summer when everything just clicked into place.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "there was this one day last summer when everything just clicked into place",
+ "normalized_hypothesis": "there was this one day last summer when everything just clicked into place",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0106",
+ "category": "question",
+ "text": "Have you ever thought about why we keep doing things the same way when they stop working?",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0106.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Have you ever thought about why we keep doing things the same way when they stop working?",
+ "raw": {
+ "normalized_reference": "have you ever thought about why we keep doing things the same way when they stop working?",
+ "normalized_hypothesis": "have you ever thought about why we keep doing things the same way when they stop working?",
+ "errors": 0,
+ "words": 17,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "have you ever thought about why we keep doing things the same way when they stop working",
+ "normalized_hypothesis": "have you ever thought about why we keep doing things the same way when they stop working",
+ "errors": 0,
+ "words": 17,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0107",
+ "category": "short",
+ "text": "I can't believe you just said that.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0107.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I can't believe you just said that.",
+ "raw": {
+ "normalized_reference": "i can't believe you just said that.",
+ "normalized_hypothesis": "i can't believe you just said that.",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i can not believe you just said that",
+ "normalized_hypothesis": "i can not believe you just said that",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0108",
+ "category": "technical",
+ "text": "You'll want to isolate that variable before drawing any conclusions.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0108.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "You'll want to isolate that variable before drawing any conclusions.",
+ "raw": {
+ "normalized_reference": "you'll want to isolate that variable before drawing any conclusions.",
+ "normalized_hypothesis": "you'll want to isolate that variable before drawing any conclusions.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "you will want to isolate that variable before drawing any conclusions",
+ "normalized_hypothesis": "you will want to isolate that variable before drawing any conclusions",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0109",
+ "category": "conversational",
+ "text": "The meeting that could have been an email lasted two hours.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0109.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The meeting that could have been an email lasted two hours.",
+ "raw": {
+ "normalized_reference": "the meeting that could have been an email lasted two hours.",
+ "normalized_hypothesis": "the meeting that could have been an email lasted two hours.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the meeting that could have been an email lasted 2 hours",
+ "normalized_hypothesis": "the meeting that could have been an email lasted 2 hours",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0110",
+ "category": "descriptive",
+ "text": "There's a specific kind of tired that sleep doesn't fix.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0110.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "There's a specific kind of tired that sleep doesn't fix.",
+ "raw": {
+ "normalized_reference": "there's a specific kind of tired that sleep doesn't fix.",
+ "normalized_hypothesis": "there's a specific kind of tired that sleep doesn't fix.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "there is a specific kind of tired that sleep does not fix",
+ "normalized_hypothesis": "there is a specific kind of tired that sleep does not fix",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0111",
+ "category": "emotional",
+ "text": "Something about that moment broke something open in me that I hadn't realized was closed.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0111.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Something about that moment broke something open in me that I hadn't realized was closed.",
+ "raw": {
+ "normalized_reference": "something about that moment broke something open in me that i hadn't realized was closed.",
+ "normalized_hypothesis": "something about that moment broke something open in me that i hadn't realized was closed.",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "something about that moment broke something open in me that i had not realized was closed",
+ "normalized_hypothesis": "something about that moment broke something open in me that i had not realized was closed",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0112",
+ "category": "long",
+ "text": "The thing about making something good is that you don't usually know it's good while you're making it. You just know you're not done yet.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0112.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The thing about making something good is that you don't usually know it's good while you're making it. You just know you're not done yet.",
+ "raw": {
+ "normalized_reference": "the thing about making something good is that you don't usually know it's good while you're making it. you just know you're not done yet.",
+ "normalized_hypothesis": "the thing about making something good is that you don't usually know it's good while you're making it. you just know you're not done yet.",
+ "errors": 0,
+ "words": 25,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the thing about making something good is that you do not usually know it is good while you are making it you just know you are not done yet",
+ "normalized_hypothesis": "the thing about making something good is that you do not usually know it is good while you are making it you just know you are not done yet",
+ "errors": 0,
+ "words": 29,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0113",
+ "category": "narrative",
+ "text": "She opened the door slowly, as if she expected to find something she wasn't ready for.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0113.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "She opened the door slowly, as if she expected to find something she wasn't ready for.",
+ "raw": {
+ "normalized_reference": "she opened the door slowly, as if she expected to find something she wasn't ready for.",
+ "normalized_hypothesis": "she opened the door slowly, as if she expected to find something she wasn't ready for.",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "she opened the door slowly as if she expected to find something she was not ready for",
+ "normalized_hypothesis": "she opened the door slowly as if she expected to find something she was not ready for",
+ "errors": 0,
+ "words": 17,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0114",
+ "category": "question",
+ "text": "What would you do differently if you could go back?",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0114.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "What would you do differently if you could go back?",
+ "raw": {
+ "normalized_reference": "what would you do differently if you could go back?",
+ "normalized_hypothesis": "what would you do differently if you could go back?",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "what would you do differently if you could go back",
+ "normalized_hypothesis": "what would you do differently if you could go back",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0115",
+ "category": "short",
+ "text": "Okay. Okay okay okay.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0115.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Okay, that okay, okay, okay.",
+ "raw": {
+ "normalized_reference": "okay. okay okay okay.",
+ "normalized_hypothesis": "okay, that okay, okay, okay.",
+ "errors": 1,
+ "words": 4,
+ "wer": 0.25
+ },
+ "semantic": {
+ "normalized_reference": "okay okay okay okay",
+ "normalized_hypothesis": "okay that okay okay okay",
+ "errors": 1,
+ "words": 4,
+ "wer": 0.25
+ }
+ },
+ {
+ "id": "release500_0116",
+ "category": "technical",
+ "text": "The latency goes up significantly at scale, which is expected.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0116.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The latency goes up significantly at scale, which is expected.",
+ "raw": {
+ "normalized_reference": "the latency goes up significantly at scale, which is expected.",
+ "normalized_hypothesis": "the latency goes up significantly at scale, which is expected.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the latency goes up significantly at scale which is expected",
+ "normalized_hypothesis": "the latency goes up significantly at scale which is expected",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0117",
+ "category": "conversational",
+ "text": "We ended up staying until two in the morning talking about nothing.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0117.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "We ended up staying until two in the morning talking about nothing.",
+ "raw": {
+ "normalized_reference": "we ended up staying until two in the morning talking about nothing.",
+ "normalized_hypothesis": "we ended up staying until two in the morning talking about nothing.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "we ended up staying until 2 in the morning talking about nothing",
+ "normalized_hypothesis": "we ended up staying until 2 in the morning talking about nothing",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0118",
+ "category": "descriptive",
+ "text": "The rain started slowly, and then all at once, the way it usually does.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0118.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The rain started slowly, and then all at once, the way it usually does.",
+ "raw": {
+ "normalized_reference": "the rain started slowly, and then all at once, the way it usually does.",
+ "normalized_hypothesis": "the rain started slowly, and then all at once, the way it usually does.",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the rain started slowly and then all at once the way it usually does",
+ "normalized_hypothesis": "the rain started slowly and then all at once the way it usually does",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0119",
+ "category": "emotional",
+ "text": "I honestly did not think I was going to make it through that year.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0119.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I honestly did not think I was going to make it through that year.",
+ "raw": {
+ "normalized_reference": "i honestly did not think i was going to make it through that year.",
+ "normalized_hypothesis": "i honestly did not think i was going to make it through that year.",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i honestly did not think i was going to make it through that year",
+ "normalized_hypothesis": "i honestly did not think i was going to make it through that year",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0120",
+ "category": "long",
+ "text": "There's an assumption that people who seem confident all the time actually feel confident, and in my experience that's almost never true.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0120.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "There's an assumption that people who seem confident all the time actually feel confident, and in my experience that's almost never true.",
+ "raw": {
+ "normalized_reference": "there's an assumption that people who seem confident all the time actually feel confident, and in my experience that's almost never true.",
+ "normalized_hypothesis": "there's an assumption that people who seem confident all the time actually feel confident, and in my experience that's almost never true.",
+ "errors": 0,
+ "words": 22,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "there is an assumption that people who seem confident all the time actually feel confident and in my experience that is almost never true",
+ "normalized_hypothesis": "there is an assumption that people who seem confident all the time actually feel confident and in my experience that is almost never true",
+ "errors": 0,
+ "words": 24,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0121",
+ "category": "narrative",
+ "text": "The moment I opened the door I knew something was different, even if I couldn't say what.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0121.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The moment I opened the door I knew something was different, even if I couldn't say what.",
+ "raw": {
+ "normalized_reference": "the moment i opened the door i knew something was different, even if i couldn't say what.",
+ "normalized_hypothesis": "the moment i opened the door i knew something was different, even if i couldn't say what.",
+ "errors": 0,
+ "words": 17,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the moment i opened the door i knew something was different even if i could not say what",
+ "normalized_hypothesis": "the moment i opened the door i knew something was different even if i could not say what",
+ "errors": 0,
+ "words": 18,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0122",
+ "category": "question",
+ "text": "How do you explain something that doesn't make sense even to you?",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0122.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "How do you explain something that doesn't make sense even to you?",
+ "raw": {
+ "normalized_reference": "how do you explain something that doesn't make sense even to you?",
+ "normalized_hypothesis": "how do you explain something that doesn't make sense even to you?",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "how do you explain something that does not make sense even to you",
+ "normalized_hypothesis": "how do you explain something that does not make sense even to you",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0123",
+ "category": "short",
+ "text": "Hold on a second.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0123.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Hold on a second.",
+ "raw": {
+ "normalized_reference": "hold on a second.",
+ "normalized_hypothesis": "hold on a second.",
+ "errors": 0,
+ "words": 4,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "hold on a 2nd",
+ "normalized_hypothesis": "hold on a 2nd",
+ "errors": 0,
+ "words": 4,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0124",
+ "category": "technical",
+ "text": "A deterministic split prevents duplicate normalized text from leaking into evaluation.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0124.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "A deterministic split prevents duplicate normalized text from leaking into evaluation.",
+ "raw": {
+ "normalized_reference": "a deterministic split prevents duplicate normalized text from leaking into evaluation.",
+ "normalized_hypothesis": "a deterministic split prevents duplicate normalized text from leaking into evaluation.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "a deterministic split prevents duplicate normalized text from leaking into evaluation",
+ "normalized_hypothesis": "a deterministic split prevents duplicate normalized text from leaking into evaluation",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0125",
+ "category": "conversational",
+ "text": "Honestly, that was much better than I expected, but the ending still felt rushed.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0125.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Honestly, that was much better than I expected, but the ending still felt rushed.",
+ "raw": {
+ "normalized_reference": "honestly, that was much better than i expected, but the ending still felt rushed.",
+ "normalized_hypothesis": "honestly, that was much better than i expected, but the ending still felt rushed.",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "honestly that was much better than i expected but the ending still felt rushed",
+ "normalized_hypothesis": "honestly that was much better than i expected but the ending still felt rushed",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0126",
+ "category": "descriptive",
+ "text": "She had a habit of tilting her head slightly when thinking hard about something.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0126.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "She had a habit of tilting her head slightly when thinking hard about something.",
+ "raw": {
+ "normalized_reference": "she had a habit of tilting her head slightly when thinking hard about something.",
+ "normalized_hypothesis": "she had a habit of tilting her head slightly when thinking hard about something.",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "she had a habit of tilting her head slightly when thinking hard about something",
+ "normalized_hypothesis": "she had a habit of tilting her head slightly when thinking hard about something",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0127",
+ "category": "emotional",
+ "text": "I cannot stress enough how much I did not want to be right about this.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0127.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I cannot stress enough how much I did not want to be right about this.",
+ "raw": {
+ "normalized_reference": "i cannot stress enough how much i did not want to be right about this.",
+ "normalized_hypothesis": "i cannot stress enough how much i did not want to be right about this.",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i cannot stress enough how much i did not want to be right about this",
+ "normalized_hypothesis": "i cannot stress enough how much i did not want to be right about this",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0128",
+ "category": "long",
+ "text": "I used to think the goal was to get to a point where nothing was hard anymore, and now I think that point doesn't exist and I'm not sure I'd want it to.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0128.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I used to think the goal was to get to a point where nothing was hard anymore, and now I think that point doesn't exist, and I'm not sure I'd want it to.",
+ "raw": {
+ "normalized_reference": "i used to think the goal was to get to a point where nothing was hard anymore, and now i think that point doesn't exist and i'm not sure i'd want it to.",
+ "normalized_hypothesis": "i used to think the goal was to get to a point where nothing was hard anymore, and now i think that point doesn't exist, and i'm not sure i'd want it to.",
+ "errors": 0,
+ "words": 33,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i used to think the goal was to get to a point where nothing was hard anymore and now i think that point does not exist and i am not sure i would want it to",
+ "normalized_hypothesis": "i used to think the goal was to get to a point where nothing was hard anymore and now i think that point does not exist and i am not sure i would want it to",
+ "errors": 0,
+ "words": 36,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0129",
+ "category": "narrative",
+ "text": "So this is the part of the story where everything goes completely sideways.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0129.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "So this is the part of the story where everything goes completely sideways.",
+ "raw": {
+ "normalized_reference": "so this is the part of the story where everything goes completely sideways.",
+ "normalized_hypothesis": "so this is the part of the story where everything goes completely sideways.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "so this is the part of the story where everything goes completely sideways",
+ "normalized_hypothesis": "so this is the part of the story where everything goes completely sideways",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0130",
+ "category": "question",
+ "text": "Is there a version of this where it actually works out okay?",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0130.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Is there a version of this where it actually works out okay?",
+ "raw": {
+ "normalized_reference": "is there a version of this where it actually works out okay?",
+ "normalized_hypothesis": "is there a version of this where it actually works out okay?",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "is there a version of this where it actually works out okay",
+ "normalized_hypothesis": "is there a version of this where it actually works out okay",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0131",
+ "category": "short",
+ "text": "Oh no. Oh no no no.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0131.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Oh, no, no, no, no, no.",
+ "raw": {
+ "normalized_reference": "oh no. oh no no no.",
+ "normalized_hypothesis": "oh, no, no, no, no, no.",
+ "errors": 1,
+ "words": 6,
+ "wer": 0.16666666666666666
+ },
+ "semantic": {
+ "normalized_reference": "0 no 0 no no no",
+ "normalized_hypothesis": "0 no no no no no",
+ "errors": 1,
+ "words": 6,
+ "wer": 0.16666666666666666
+ }
+ },
+ {
+ "id": "release500_0132",
+ "category": "technical",
+ "text": "Always test on a copy before modifying the original.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0132.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Always test on a copy before modifying the original.",
+ "raw": {
+ "normalized_reference": "always test on a copy before modifying the original.",
+ "normalized_hypothesis": "always test on a copy before modifying the original.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "always test on a copy before modifying the original",
+ "normalized_hypothesis": "always test on a copy before modifying the original",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0133",
+ "category": "conversational",
+ "text": "I wasn't even supposed to be there. That's the wild part.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0133.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I wasn't even supposed to be there. That's the wild part.",
+ "raw": {
+ "normalized_reference": "i wasn't even supposed to be there. that's the wild part.",
+ "normalized_hypothesis": "i wasn't even supposed to be there. that's the wild part.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i was not even supposed to be there that is the wild part",
+ "normalized_hypothesis": "i was not even supposed to be there that is the wild part",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0134",
+ "category": "descriptive",
+ "text": "It's a small town. Not the kind you leave. The kind you leave and come back to.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0134.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It's a small town, not the kind you leave. The kind you leave and come back to.",
+ "raw": {
+ "normalized_reference": "it's a small town. not the kind you leave. the kind you leave and come back to.",
+ "normalized_hypothesis": "it's a small town, not the kind you leave. the kind you leave and come back to.",
+ "errors": 0,
+ "words": 17,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it is a small town not the kind you leave the kind you leave and come back to",
+ "normalized_hypothesis": "it is a small town not the kind you leave the kind you leave and come back to",
+ "errors": 0,
+ "words": 18,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0135",
+ "category": "emotional",
+ "text": "She cried. I cried. We all just sort of stood there crying.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0135.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "She cried, I cried, but we all just sort of stood there crying.",
+ "raw": {
+ "normalized_reference": "she cried. i cried. we all just sort of stood there crying.",
+ "normalized_hypothesis": "she cried, i cried, but we all just sort of stood there crying.",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "she cried i cried we all just sort of stood there crying",
+ "normalized_hypothesis": "she cried i cried but we all just sort of stood there crying",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ }
+ },
+ {
+ "id": "release500_0136",
+ "category": "long",
+ "text": "There's a version of being busy that's actually just a very elaborate way of avoiding the one thing you know you should be doing.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0136.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "There's a version of being busy that's actually just a very elaborate way of avoiding the one thing you know you should be doing.",
+ "raw": {
+ "normalized_reference": "there's a version of being busy that's actually just a very elaborate way of avoiding the one thing you know you should be doing.",
+ "normalized_hypothesis": "there's a version of being busy that's actually just a very elaborate way of avoiding the one thing you know you should be doing.",
+ "errors": 0,
+ "words": 24,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "there is a version of being busy that is actually just a very elaborate way of avoiding the one thing you know you should be doing",
+ "normalized_hypothesis": "there is a version of being busy that is actually just a very elaborate way of avoiding the one thing you know you should be doing",
+ "errors": 0,
+ "words": 26,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0137",
+ "category": "narrative",
+ "text": "The letter arrived on a Tuesday, which felt appropriate somehow.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0137.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The letter arrived on a Tuesday, which felt appropriate somehow.",
+ "raw": {
+ "normalized_reference": "the letter arrived on a tuesday, which felt appropriate somehow.",
+ "normalized_hypothesis": "the letter arrived on a tuesday, which felt appropriate somehow.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the letter arrived on a tuesday which felt appropriate somehow",
+ "normalized_hypothesis": "the letter arrived on a tuesday which felt appropriate somehow",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0138",
+ "category": "question",
+ "text": "What exactly did she say when you told her?",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0138.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "What exactly did she say when you told her?",
+ "raw": {
+ "normalized_reference": "what exactly did she say when you told her?",
+ "normalized_hypothesis": "what exactly did she say when you told her?",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "what exactly did she say when you told her",
+ "normalized_hypothesis": "what exactly did she say when you told her",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0139",
+ "category": "short",
+ "text": "No way. Absolutely not.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0139.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "No way to absolutely not.",
+ "raw": {
+ "normalized_reference": "no way. absolutely not.",
+ "normalized_hypothesis": "no way to absolutely not.",
+ "errors": 1,
+ "words": 4,
+ "wer": 0.25
+ },
+ "semantic": {
+ "normalized_reference": "no way absolutely not",
+ "normalized_hypothesis": "no way to absolutely not",
+ "errors": 1,
+ "words": 4,
+ "wer": 0.25
+ }
+ },
+ {
+ "id": "release500_0140",
+ "category": "technical",
+ "text": "The first thing you want to do is make sure everything is backed up.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0140.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The first thing you want to do is make sure everything is backed up.",
+ "raw": {
+ "normalized_reference": "the first thing you want to do is make sure everything is backed up.",
+ "normalized_hypothesis": "the first thing you want to do is make sure everything is backed up.",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the 1st thing you want to do is make sure everything is backed up",
+ "normalized_hypothesis": "the 1st thing you want to do is make sure everything is backed up",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0141",
+ "category": "conversational",
+ "text": "I have the most insane idea and you need to just hear me out.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0141.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I have the most insane idea and you need to just hear me out.",
+ "raw": {
+ "normalized_reference": "i have the most insane idea and you need to just hear me out.",
+ "normalized_hypothesis": "i have the most insane idea and you need to just hear me out.",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i have the most insane idea and you need to just hear me out",
+ "normalized_hypothesis": "i have the most insane idea and you need to just hear me out",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0142",
+ "category": "descriptive",
+ "text": "There's a coffee shop on the corner that's been there longer than anyone can remember.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0142.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "There's a coffee shop on the corner that's been there longer than anyone can remember.",
+ "raw": {
+ "normalized_reference": "there's a coffee shop on the corner that's been there longer than anyone can remember.",
+ "normalized_hypothesis": "there's a coffee shop on the corner that's been there longer than anyone can remember.",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "there is a coffee shop on the corner that has been there longer than anyone can remember",
+ "normalized_hypothesis": "there is a coffee shop on the corner that has been there longer than anyone can remember",
+ "errors": 0,
+ "words": 17,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0143",
+ "category": "emotional",
+ "text": "I kept it together until I got to the car, and then I completely fell apart.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0143.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I kept it together until I got to the car, and then I completely fell apart.",
+ "raw": {
+ "normalized_reference": "i kept it together until i got to the car, and then i completely fell apart.",
+ "normalized_hypothesis": "i kept it together until i got to the car, and then i completely fell apart.",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i kept it together until i got to the car and then i completely fell apart",
+ "normalized_hypothesis": "i kept it together until i got to the car and then i completely fell apart",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0144",
+ "category": "long",
+ "text": "The older I get, the more I think that most of what we call wisdom is just pattern recognition built from having been wrong enough times to start noticing.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0144.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The older I get, the more I think that most of what we call wisdom is just pattern recognition built from having been wrong enough times to start noticing.",
+ "raw": {
+ "normalized_reference": "the older i get, the more i think that most of what we call wisdom is just pattern recognition built from having been wrong enough times to start noticing.",
+ "normalized_hypothesis": "the older i get, the more i think that most of what we call wisdom is just pattern recognition built from having been wrong enough times to start noticing.",
+ "errors": 0,
+ "words": 29,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the older i get the more i think that most of what we call wisdom is just pattern recognition built from having been wrong enough times to start noticing",
+ "normalized_hypothesis": "the older i get the more i think that most of what we call wisdom is just pattern recognition built from having been wrong enough times to start noticing",
+ "errors": 0,
+ "words": 29,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0145",
+ "category": "narrative",
+ "text": "I remember the exact moment I realized I had made a huge mistake.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0145.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I remember the exact moment I realized I had made a huge mistake.",
+ "raw": {
+ "normalized_reference": "i remember the exact moment i realized i had made a huge mistake.",
+ "normalized_hypothesis": "i remember the exact moment i realized i had made a huge mistake.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i remember the exact moment i realized i had made a huge mistake",
+ "normalized_hypothesis": "i remember the exact moment i realized i had made a huge mistake",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0146",
+ "category": "question",
+ "text": "Do you think it would have gone differently if we'd left earlier?",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0146.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Do you think it would have come differently if we'd left earlier?",
+ "raw": {
+ "normalized_reference": "do you think it would have gone differently if we'd left earlier?",
+ "normalized_hypothesis": "do you think it would have come differently if we'd left earlier?",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "do you think it would have gone differently if we would left earlier",
+ "normalized_hypothesis": "do you think it would have come differently if we would left earlier",
+ "errors": 1,
+ "words": 13,
+ "wer": 0.07692307692307693
+ }
+ },
+ {
+ "id": "release500_0147",
+ "category": "short",
+ "text": "You have to be kidding me.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0147.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "You have to be kidding me.",
+ "raw": {
+ "normalized_reference": "you have to be kidding me.",
+ "normalized_hypothesis": "you have to be kidding me.",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "you have to be kidding me",
+ "normalized_hypothesis": "you have to be kidding me",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0148",
+ "category": "technical",
+ "text": "If the checksum fails, restart the worker process without deleting the cached tensors.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0148.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "If the checksum fails, restart the worker process without deleting the cache tensors.",
+ "raw": {
+ "normalized_reference": "if the checksum fails, restart the worker process without deleting the cached tensors.",
+ "normalized_hypothesis": "if the checksum fails, restart the worker process without deleting the cache tensors.",
+ "errors": 1,
+ "words": 13,
+ "wer": 0.07692307692307693
+ },
+ "semantic": {
+ "normalized_reference": "if the checksum fails restart the worker process without deleting the cached tensors",
+ "normalized_hypothesis": "if the checksum fails restart the worker process without deleting the cache tensors",
+ "errors": 1,
+ "words": 13,
+ "wer": 0.07692307692307693
+ }
+ },
+ {
+ "id": "release500_0149",
+ "category": "conversational",
+ "text": "I need you to understand how close that was to not working.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0149.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I need you to understand how close that was to not working.",
+ "raw": {
+ "normalized_reference": "i need you to understand how close that was to not working.",
+ "normalized_hypothesis": "i need you to understand how close that was to not working.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i need you to understand how close that was to not working",
+ "normalized_hypothesis": "i need you to understand how close that was to not working",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0150",
+ "category": "descriptive",
+ "text": "It was the sort of morning where the whole world looks rinsed clean.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0150.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It was the sort of morning where the whole world looks rinsed clean.",
+ "raw": {
+ "normalized_reference": "it was the sort of morning where the whole world looks rinsed clean.",
+ "normalized_hypothesis": "it was the sort of morning where the whole world looks rinsed clean.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it was the sort of morning where the whole world looks rinsed clean",
+ "normalized_hypothesis": "it was the sort of morning where the whole world looks rinsed clean",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0151",
+ "category": "emotional",
+ "text": "She started laughing and I started laughing and we couldn't stop for a full minute.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0151.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "She started laughing, and I started laughing, and we couldn't stop for a full minute.",
+ "raw": {
+ "normalized_reference": "she started laughing and i started laughing and we couldn't stop for a full minute.",
+ "normalized_hypothesis": "she started laughing, and i started laughing, and we couldn't stop for a full minute.",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "she started laughing and i started laughing and we could not stop for a full minute",
+ "normalized_hypothesis": "she started laughing and i started laughing and we could not stop for a full minute",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0152",
+ "category": "long",
+ "text": "We spent three years building something we were genuinely proud of, and then the market shifted and none of it mattered the way we thought it would.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0152.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "We spent three years building something we were genuinely proud of, and then the market shifted and none of it mattered the way we thought it would.",
+ "raw": {
+ "normalized_reference": "we spent three years building something we were genuinely proud of, and then the market shifted and none of it mattered the way we thought it would.",
+ "normalized_hypothesis": "we spent three years building something we were genuinely proud of, and then the market shifted and none of it mattered the way we thought it would.",
+ "errors": 0,
+ "words": 27,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "we spent 3 years building something we were genuinely proud of and then the market shifted and none of it mattered the way we thought it would",
+ "normalized_hypothesis": "we spent 3 years building something we were genuinely proud of and then the market shifted and none of it mattered the way we thought it would",
+ "errors": 0,
+ "words": 27,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0153",
+ "category": "narrative",
+ "text": "And somehow, despite all of that, it actually turned out okay.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0153.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "And somehow, despite all of that, it actually turned out okay.",
+ "raw": {
+ "normalized_reference": "and somehow, despite all of that, it actually turned out okay.",
+ "normalized_hypothesis": "and somehow, despite all of that, it actually turned out okay.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "and somehow despite all of that it actually turned out okay",
+ "normalized_hypothesis": "and somehow despite all of that it actually turned out okay",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0154",
+ "category": "question",
+ "text": "Is that something you actually want, or something you think you should want?",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0154.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Is that something you actually want or something you think you should want?",
+ "raw": {
+ "normalized_reference": "is that something you actually want, or something you think you should want?",
+ "normalized_hypothesis": "is that something you actually want or something you think you should want?",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "is that something you actually want or something you think you should want",
+ "normalized_hypothesis": "is that something you actually want or something you think you should want",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0155",
+ "category": "short",
+ "text": "This is fine.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0155.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "This is fine.",
+ "raw": {
+ "normalized_reference": "this is fine.",
+ "normalized_hypothesis": "this is fine.",
+ "errors": 0,
+ "words": 3,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "this is fine",
+ "normalized_hypothesis": "this is fine",
+ "errors": 0,
+ "words": 3,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0156",
+ "category": "technical",
+ "text": "The tricky part isn't the setup, it's making sure all dependencies line up.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0156.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The tricky part isn't the setup. It's making sure all dependencies line up.",
+ "raw": {
+ "normalized_reference": "the tricky part isn't the setup, it's making sure all dependencies line up.",
+ "normalized_hypothesis": "the tricky part isn't the setup. it's making sure all dependencies line up.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the tricky part is not the setup it is making sure all dependencies line up",
+ "normalized_hypothesis": "the tricky part is not the setup it is making sure all dependencies line up",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0157",
+ "category": "conversational",
+ "text": "He called it a minor setback. I called it a disaster.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0157.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He called it a minor setback that I called it a disaster.",
+ "raw": {
+ "normalized_reference": "he called it a minor setback. i called it a disaster.",
+ "normalized_hypothesis": "he called it a minor setback that i called it a disaster.",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ },
+ "semantic": {
+ "normalized_reference": "he called it a minor setback i called it a disaster",
+ "normalized_hypothesis": "he called it a minor setback that i called it a disaster",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ }
+ },
+ {
+ "id": "release500_0158",
+ "category": "descriptive",
+ "text": "The whole place smelled like pine and old wood and cold air.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0158.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The whole place smelled like pine and old wood and cold air.",
+ "raw": {
+ "normalized_reference": "the whole place smelled like pine and old wood and cold air.",
+ "normalized_hypothesis": "the whole place smelled like pine and old wood and cold air.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the whole place smelled like pine and old wood and cold air",
+ "normalized_hypothesis": "the whole place smelled like pine and old wood and cold air",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0159",
+ "category": "emotional",
+ "text": "I've never been so relieved in my entire life.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0159.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I've never been so relieved in my entire life.",
+ "raw": {
+ "normalized_reference": "i've never been so relieved in my entire life.",
+ "normalized_hypothesis": "i've never been so relieved in my entire life.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i have never been so relieved in my entire life",
+ "normalized_hypothesis": "i have never been so relieved in my entire life",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0160",
+ "category": "narrative",
+ "text": "I wrote it all down that night so I wouldn't forget, and I still haven't looked at it.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0160.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I wrote it all down that night so I wouldn't forget, and I still haven't looked at it.",
+ "raw": {
+ "normalized_reference": "i wrote it all down that night so i wouldn't forget, and i still haven't looked at it.",
+ "normalized_hypothesis": "i wrote it all down that night so i wouldn't forget, and i still haven't looked at it.",
+ "errors": 0,
+ "words": 18,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i wrote it all down that night so i would not forget and i still have not looked at it",
+ "normalized_hypothesis": "i wrote it all down that night so i would not forget and i still have not looked at it",
+ "errors": 0,
+ "words": 20,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0161",
+ "category": "short",
+ "text": "That's kind of wild actually.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0161.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "That's kind of wild, actually.",
+ "raw": {
+ "normalized_reference": "that's kind of wild actually.",
+ "normalized_hypothesis": "that's kind of wild, actually.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "that is kind of wild actually",
+ "normalized_hypothesis": "that is kind of wild actually",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0162",
+ "category": "technical",
+ "text": "Version pinning will save you a lot of pain later down the line.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0162.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Version pinning will save you a lot of pain later down the line.",
+ "raw": {
+ "normalized_reference": "version pinning will save you a lot of pain later down the line.",
+ "normalized_hypothesis": "version pinning will save you a lot of pain later down the line.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "version pinning will save you a lot of pain later down the line",
+ "normalized_hypothesis": "version pinning will save you a lot of pain later down the line",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0163",
+ "category": "conversational",
+ "text": "At some point you just have to commit and see what happens.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0163.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "At some point you just have to commit to see what happens.",
+ "raw": {
+ "normalized_reference": "at some point you just have to commit and see what happens.",
+ "normalized_hypothesis": "at some point you just have to commit to see what happens.",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "at some point you just have to commit and see what happens",
+ "normalized_hypothesis": "at some point you just have to commit to see what happens",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ }
+ },
+ {
+ "id": "release500_0164",
+ "category": "emotional",
+ "text": "Nobody told me it was going to be this hard, and honestly, I'm grateful.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0164.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Nobody told me it was going to be this hard, and honestly, I'm grateful.",
+ "raw": {
+ "normalized_reference": "nobody told me it was going to be this hard, and honestly, i'm grateful.",
+ "normalized_hypothesis": "nobody told me it was going to be this hard, and honestly, i'm grateful.",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "nobody told me it was going to be this hard and honestly i am grateful",
+ "normalized_hypothesis": "nobody told me it was going to be this hard and honestly i am grateful",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0165",
+ "category": "narrative",
+ "text": "Chapter seven. In which everything that could go wrong did, and a few things that couldn't.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0165.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Chapter 7, in which everything that could go wrong did, and a few things that couldn't.",
+ "raw": {
+ "normalized_reference": "chapter seven. in which everything that could go wrong did, and a few things that couldn't.",
+ "normalized_hypothesis": "chapter 7, in which everything that could go wrong did, and a few things that couldn't.",
+ "errors": 1,
+ "words": 16,
+ "wer": 0.0625
+ },
+ "semantic": {
+ "normalized_reference": "chapter 7 in which everything that could go wrong did and a few things that could not",
+ "normalized_hypothesis": "chapter 7 in which everything that could go wrong did and a few things that could not",
+ "errors": 0,
+ "words": 17,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0166",
+ "category": "short",
+ "text": "Wait, what?",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0166.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Wait, what?",
+ "raw": {
+ "normalized_reference": "wait, what?",
+ "normalized_hypothesis": "wait, what?",
+ "errors": 0,
+ "words": 2,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "wait what",
+ "normalized_hypothesis": "wait what",
+ "errors": 0,
+ "words": 2,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0167",
+ "category": "technical",
+ "text": "The graphics processor remained idle because the data loader exhausted every central processing core.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0167.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The graphics processor remained idle because the data loader exhausted every central processing core.",
+ "raw": {
+ "normalized_reference": "the graphics processor remained idle because the data loader exhausted every central processing core.",
+ "normalized_hypothesis": "the graphics processor remained idle because the data loader exhausted every central processing core.",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the graphics processor remained idle because the data loader exhausted every central processing core",
+ "normalized_hypothesis": "the graphics processor remained idle because the data loader exhausted every central processing core",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0168",
+ "category": "conversational",
+ "text": "Could you speak a little more slowly? I am trying to write every detail down.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0168.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Could you speak a little more slowly? I'm trying to write every detail down.",
+ "raw": {
+ "normalized_reference": "could you speak a little more slowly? i am trying to write every detail down.",
+ "normalized_hypothesis": "could you speak a little more slowly? i'm trying to write every detail down.",
+ "errors": 2,
+ "words": 15,
+ "wer": 0.13333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "could you speak a little more slowly i am trying to write every detail down",
+ "normalized_hypothesis": "could you speak a little more slowly i am trying to write every detail down",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0169",
+ "category": "emotional",
+ "text": "I was so excited I could barely form a complete sentence.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0169.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I was so excited I could barely form a complete sentence.",
+ "raw": {
+ "normalized_reference": "i was so excited i could barely form a complete sentence.",
+ "normalized_hypothesis": "i was so excited i could barely form a complete sentence.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i was so excited i could barely form a complete sentence",
+ "normalized_hypothesis": "i was so excited i could barely form a complete sentence",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0170",
+ "category": "narrative",
+ "text": "He looked at me like I had said something in a language he didn't speak.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0170.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He looked at me like I had said something in a language he didn't speak.",
+ "raw": {
+ "normalized_reference": "he looked at me like i had said something in a language he didn't speak.",
+ "normalized_hypothesis": "he looked at me like i had said something in a language he didn't speak.",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he looked at me like i had said something in a language he did not speak",
+ "normalized_hypothesis": "he looked at me like i had said something in a language he did not speak",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0171",
+ "category": "short",
+ "text": "That's actually hilarious.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0171.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "That's actually hilarious.",
+ "raw": {
+ "normalized_reference": "that's actually hilarious.",
+ "normalized_hypothesis": "that's actually hilarious.",
+ "errors": 0,
+ "words": 3,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "that is actually hilarious",
+ "normalized_hypothesis": "that is actually hilarious",
+ "errors": 0,
+ "words": 4,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0172",
+ "category": "technical",
+ "text": "The encoder maps phonemes, stress, and phrase boundaries into a shared hidden representation.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0172.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The encoder maps phonemes, stress, and phrase boundaries into a shared hidden representation.",
+ "raw": {
+ "normalized_reference": "the encoder maps phonemes, stress, and phrase boundaries into a shared hidden representation.",
+ "normalized_hypothesis": "the encoder maps phonemes, stress, and phrase boundaries into a shared hidden representation.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the encoder maps phonemes stress and phrase boundaries into a shared hidden representation",
+ "normalized_hypothesis": "the encoder maps phonemes stress and phrase boundaries into a shared hidden representation",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0173",
+ "category": "conversational",
+ "text": "She showed up three hours late and acted like nothing had happened.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0173.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "She showed up three hours late and acted like nothing had happened.",
+ "raw": {
+ "normalized_reference": "she showed up three hours late and acted like nothing had happened.",
+ "normalized_hypothesis": "she showed up three hours late and acted like nothing had happened.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "she showed up 3 hours late and acted like nothing had happened",
+ "normalized_hypothesis": "she showed up 3 hours late and acted like nothing had happened",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0174",
+ "category": "emotional",
+ "text": "Oh my god. This is the best thing I've ever tasted.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0174.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Oh my god. This is the best thing I've ever tasted.",
+ "raw": {
+ "normalized_reference": "oh my god. this is the best thing i've ever tasted.",
+ "normalized_hypothesis": "oh my god. this is the best thing i've ever tasted.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "0 my god this is the best thing i have ever tasted",
+ "normalized_hypothesis": "0 my god this is the best thing i have ever tasted",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0175",
+ "category": "narrative",
+ "text": "The note was short, which was somehow more frightening than a long one would have been.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0175.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The note was short, which was somehow more frightening than a long one would have been.",
+ "raw": {
+ "normalized_reference": "the note was short, which was somehow more frightening than a long one would have been.",
+ "normalized_hypothesis": "the note was short, which was somehow more frightening than a long one would have been.",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the note was short which was somehow more frightening than a long one would have been",
+ "normalized_hypothesis": "the note was short which was somehow more frightening than a long one would have been",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0176",
+ "category": "short",
+ "text": "Noted.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0176.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Noted.",
+ "raw": {
+ "normalized_reference": "noted.",
+ "normalized_hypothesis": "noted.",
+ "errors": 0,
+ "words": 1,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "noted",
+ "normalized_hypothesis": "noted",
+ "errors": 0,
+ "words": 1,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0177",
+ "category": "technical",
+ "text": "It's a simple process once you understand the underlying structure.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0177.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It's a simple process once you understand the underlying structure.",
+ "raw": {
+ "normalized_reference": "it's a simple process once you understand the underlying structure.",
+ "normalized_hypothesis": "it's a simple process once you understand the underlying structure.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it is a simple process once you understand the underlying structure",
+ "normalized_hypothesis": "it is a simple process once you understand the underlying structure",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0178",
+ "category": "conversational",
+ "text": "You have to hear this story because it's almost too good to be true.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0178.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "You have to hear this story because it's almost too good to be true.",
+ "raw": {
+ "normalized_reference": "you have to hear this story because it's almost too good to be true.",
+ "normalized_hypothesis": "you have to hear this story because it's almost too good to be true.",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "you have to hear this story because it is almost too good to be true",
+ "normalized_hypothesis": "you have to hear this story because it is almost too good to be true",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0179",
+ "category": "emotional",
+ "text": "I have never felt more seen in my entire life than in that moment.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0179.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I have never felt more seen in my entire life than in that moment.",
+ "raw": {
+ "normalized_reference": "i have never felt more seen in my entire life than in that moment.",
+ "normalized_hypothesis": "i have never felt more seen in my entire life than in that moment.",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i have never felt more seen in my entire life than in that moment",
+ "normalized_hypothesis": "i have never felt more seen in my entire life than in that moment",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0180",
+ "category": "narrative",
+ "text": "She knew. She had always known. She just waited for the right moment.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0180.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "She knew that she had always known. She just waited for the right moment.",
+ "raw": {
+ "normalized_reference": "she knew. she had always known. she just waited for the right moment.",
+ "normalized_hypothesis": "she knew that she had always known. she just waited for the right moment.",
+ "errors": 1,
+ "words": 13,
+ "wer": 0.07692307692307693
+ },
+ "semantic": {
+ "normalized_reference": "she knew she had always known she just waited for the right moment",
+ "normalized_hypothesis": "she knew that she had always known she just waited for the right moment",
+ "errors": 1,
+ "words": 13,
+ "wer": 0.07692307692307693
+ }
+ },
+ {
+ "id": "release500_0181",
+ "category": "short",
+ "text": "Yeah, no. Hard pass.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0181.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Yeah, no duck hard pass.",
+ "raw": {
+ "normalized_reference": "yeah, no. hard pass.",
+ "normalized_hypothesis": "yeah, no duck hard pass.",
+ "errors": 1,
+ "words": 4,
+ "wer": 0.25
+ },
+ "semantic": {
+ "normalized_reference": "yeah no hard pass",
+ "normalized_hypothesis": "yeah no duck hard pass",
+ "errors": 1,
+ "words": 4,
+ "wer": 0.25
+ }
+ },
+ {
+ "id": "release500_0182",
+ "category": "technical",
+ "text": "Please verify the sample rate, hop length, and mel filter configuration before inference.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0182.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Please verify the sample rate, pop length, and MEL filter configuration before inference.",
+ "raw": {
+ "normalized_reference": "please verify the sample rate, hop length, and mel filter configuration before inference.",
+ "normalized_hypothesis": "please verify the sample rate, pop length, and mel filter configuration before inference.",
+ "errors": 1,
+ "words": 13,
+ "wer": 0.07692307692307693
+ },
+ "semantic": {
+ "normalized_reference": "please verify the sample rate hop length and mel filter configuration before inference",
+ "normalized_hypothesis": "please verify the sample rate pop length and mel filter configuration before inference",
+ "errors": 1,
+ "words": 13,
+ "wer": 0.07692307692307693
+ }
+ },
+ {
+ "id": "release500_0183",
+ "category": "conversational",
+ "text": "That's the worst idea I've ever heard and I'm completely in.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0183.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "That's the worst idea I've ever heard, and I'm completely in.",
+ "raw": {
+ "normalized_reference": "that's the worst idea i've ever heard and i'm completely in.",
+ "normalized_hypothesis": "that's the worst idea i've ever heard, and i'm completely in.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "that is the worst idea i have ever heard and i am completely in",
+ "normalized_hypothesis": "that is the worst idea i have ever heard and i am completely in",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0184",
+ "category": "emotional",
+ "text": "I remember thinking: this is it. This is the thing I've been waiting for.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0184.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I remember thinking, this is it. This is the thing I've been waiting for.",
+ "raw": {
+ "normalized_reference": "i remember thinking: this is it. this is the thing i've been waiting for.",
+ "normalized_hypothesis": "i remember thinking, this is it. this is the thing i've been waiting for.",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i remember thinking this is it this is the thing i have been waiting for",
+ "normalized_hypothesis": "i remember thinking this is it this is the thing i have been waiting for",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0185",
+ "category": "narrative",
+ "text": "We never talked about it again, which I think was the right call.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0185.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "We never talked about it again, which I think was the right call.",
+ "raw": {
+ "normalized_reference": "we never talked about it again, which i think was the right call.",
+ "normalized_hypothesis": "we never talked about it again, which i think was the right call.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "we never talked about it again which i think was the right call",
+ "normalized_hypothesis": "we never talked about it again which i think was the right call",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0186",
+ "category": "short",
+ "text": "In hindsight, obvious.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0186.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "In hindsight, obvious.",
+ "raw": {
+ "normalized_reference": "in hindsight, obvious.",
+ "normalized_hypothesis": "in hindsight, obvious.",
+ "errors": 0,
+ "words": 3,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "in hindsight obvious",
+ "normalized_hypothesis": "in hindsight obvious",
+ "errors": 0,
+ "words": 3,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0187",
+ "category": "technical",
+ "text": "Once it's configured correctly, it basically runs itself.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0187.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Once it's configured correctly, it basically runs itself.",
+ "raw": {
+ "normalized_reference": "once it's configured correctly, it basically runs itself.",
+ "normalized_hypothesis": "once it's configured correctly, it basically runs itself.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "once it is configured correctly it basically runs itself",
+ "normalized_hypothesis": "once it is configured correctly it basically runs itself",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0188",
+ "category": "conversational",
+ "text": "You got engaged? When? How? Tell me everything right now.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0188.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "You got engaged? When? How? Tell me everything right now.",
+ "raw": {
+ "normalized_reference": "you got engaged? when? how? tell me everything right now.",
+ "normalized_hypothesis": "you got engaged? when? how? tell me everything right now.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "you got engaged when how tell me everything right now",
+ "normalized_hypothesis": "you got engaged when how tell me everything right now",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0189",
+ "category": "emotional",
+ "text": "It's hard to explain how much that moment meant to me.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0189.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It's hard to explain how much that moment meant to me.",
+ "raw": {
+ "normalized_reference": "it's hard to explain how much that moment meant to me.",
+ "normalized_hypothesis": "it's hard to explain how much that moment meant to me.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it is hard to explain how much that moment meant to me",
+ "normalized_hypothesis": "it is hard to explain how much that moment meant to me",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0190",
+ "category": "narrative",
+ "text": "He told the story three different times that night and somehow it got better each time.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0190.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He told this story three different times that night, and somehow it got better each time.",
+ "raw": {
+ "normalized_reference": "he told the story three different times that night and somehow it got better each time.",
+ "normalized_hypothesis": "he told this story three different times that night, and somehow it got better each time.",
+ "errors": 1,
+ "words": 16,
+ "wer": 0.0625
+ },
+ "semantic": {
+ "normalized_reference": "he told the story 3 different times that night and somehow it got better each time",
+ "normalized_hypothesis": "he told this story 3 different times that night and somehow it got better each time",
+ "errors": 1,
+ "words": 16,
+ "wer": 0.0625
+ }
+ },
+ {
+ "id": "release500_0191",
+ "category": "short",
+ "text": "Right. Moving on.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0191.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Right at moving on.",
+ "raw": {
+ "normalized_reference": "right. moving on.",
+ "normalized_hypothesis": "right at moving on.",
+ "errors": 1,
+ "words": 3,
+ "wer": 0.3333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "right moving on",
+ "normalized_hypothesis": "right at moving on",
+ "errors": 1,
+ "words": 3,
+ "wer": 0.3333333333333333
+ }
+ },
+ {
+ "id": "release500_0192",
+ "category": "conversational",
+ "text": "No, I did not move your keys; check beside the small ceramic bowl.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0192.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "No, I did not move your keys. Check beside this small ceramic bowl.",
+ "raw": {
+ "normalized_reference": "no, i did not move your keys; check beside the small ceramic bowl.",
+ "normalized_hypothesis": "no, i did not move your keys. check beside this small ceramic bowl.",
+ "errors": 1,
+ "words": 13,
+ "wer": 0.07692307692307693
+ },
+ "semantic": {
+ "normalized_reference": "no i did not move your keys check beside the small ceramic bowl",
+ "normalized_hypothesis": "no i did not move your keys check beside this small ceramic bowl",
+ "errors": 1,
+ "words": 13,
+ "wer": 0.07692307692307693
+ }
+ },
+ {
+ "id": "release500_0193",
+ "category": "emotional",
+ "text": "I was devastated. I'm not going to pretend otherwise.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0193.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I was devastated. I'm not going to pretend otherwise.",
+ "raw": {
+ "normalized_reference": "i was devastated. i'm not going to pretend otherwise.",
+ "normalized_hypothesis": "i was devastated. i'm not going to pretend otherwise.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i was devastated i am not going to pretend otherwise",
+ "normalized_hypothesis": "i was devastated i am not going to pretend otherwise",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0194",
+ "category": "narrative",
+ "text": "That was the last summer before everything changed, though we didn't know it yet.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0194.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "That was the last summer before everything changed, though we didn't know it yet.",
+ "raw": {
+ "normalized_reference": "that was the last summer before everything changed, though we didn't know it yet.",
+ "normalized_hypothesis": "that was the last summer before everything changed, though we didn't know it yet.",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "that was the last summer before everything changed though we did not know it yet",
+ "normalized_hypothesis": "that was the last summer before everything changed though we did not know it yet",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0195",
+ "category": "short",
+ "text": "Bold of you to assume.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0195.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Bold of you to assume.",
+ "raw": {
+ "normalized_reference": "bold of you to assume.",
+ "normalized_hypothesis": "bold of you to assume.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "bold of you to assume",
+ "normalized_hypothesis": "bold of you to assume",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0196",
+ "category": "conversational",
+ "text": "It's one of those decisions that makes total sense in the moment.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0196.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It's one of those decisions that makes total sense in the moment.",
+ "raw": {
+ "normalized_reference": "it's one of those decisions that makes total sense in the moment.",
+ "normalized_hypothesis": "it's one of those decisions that makes total sense in the moment.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it is one of those decisions that makes total sense in the moment",
+ "normalized_hypothesis": "it is one of those decisions that makes total sense in the moment",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0197",
+ "category": "emotional",
+ "text": "I was furious. Not the loud kind. The quiet kind that lasts for days.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0197.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I was furious, not the loud kind, the quiet kind that lasts for days.",
+ "raw": {
+ "normalized_reference": "i was furious. not the loud kind. the quiet kind that lasts for days.",
+ "normalized_hypothesis": "i was furious, not the loud kind, the quiet kind that lasts for days.",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i was furious not the loud kind the quiet kind that lasts for days",
+ "normalized_hypothesis": "i was furious not the loud kind the quiet kind that lasts for days",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0198",
+ "category": "narrative",
+ "text": "That's the part of the story I usually leave out.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0198.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "That's the part of the story I usually leave out.",
+ "raw": {
+ "normalized_reference": "that's the part of the story i usually leave out.",
+ "normalized_hypothesis": "that's the part of the story i usually leave out.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "that is the part of the story i usually leave out",
+ "normalized_hypothesis": "that is the part of the story i usually leave out",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0199",
+ "category": "short",
+ "text": "I've said too much.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0199.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I've said too much.",
+ "raw": {
+ "normalized_reference": "i've said too much.",
+ "normalized_hypothesis": "i've said too much.",
+ "errors": 0,
+ "words": 4,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i have said too much",
+ "normalized_hypothesis": "i have said too much",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0200",
+ "category": "conversational",
+ "text": "I've been faking that I understood this the entire time.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0200.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I've been faking that I understood this the entire time.",
+ "raw": {
+ "normalized_reference": "i've been faking that i understood this the entire time.",
+ "normalized_hypothesis": "i've been faking that i understood this the entire time.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i have been faking that i understood this the entire time",
+ "normalized_hypothesis": "i have been faking that i understood this the entire time",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0201",
+ "category": "phonetic_balance",
+ "text": "Peace be unto you and grace before the Lord.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0201.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Peace be unto you, and grace before the Lord.",
+ "raw": {
+ "normalized_reference": "peace be unto you and grace before the lord.",
+ "normalized_hypothesis": "peace be unto you, and grace before the lord.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "peace be unto you and grace before the lord",
+ "normalized_hypothesis": "peace be unto you and grace before the lord",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0202",
+ "category": "phonetic_balance",
+ "text": "From now on we're pals.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0202.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "From now onward, pals.",
+ "raw": {
+ "normalized_reference": "from now on we're pals.",
+ "normalized_hypothesis": "from now onward, pals.",
+ "errors": 2,
+ "words": 5,
+ "wer": 0.4
+ },
+ "semantic": {
+ "normalized_reference": "from now on we are pals",
+ "normalized_hypothesis": "from now onward pals",
+ "errors": 3,
+ "words": 6,
+ "wer": 0.5
+ }
+ },
+ {
+ "id": "release500_0203",
+ "category": "phonetic_balance",
+ "text": "Here we allow our solicitors to look after our legal work.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0203.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Here we allow our solicitors to look after our legal work.",
+ "raw": {
+ "normalized_reference": "here we allow our solicitors to look after our legal work.",
+ "normalized_hypothesis": "here we allow our solicitors to look after our legal work.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "here we allow our solicitors to look after our legal work",
+ "normalized_hypothesis": "here we allow our solicitors to look after our legal work",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0204",
+ "category": "phonetic_balance",
+ "text": "Her words sent a strange chill through Philip.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0204.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Her words sent a strange chill through Phyllis.",
+ "raw": {
+ "normalized_reference": "her words sent a strange chill through philip.",
+ "normalized_hypothesis": "her words sent a strange chill through phyllis.",
+ "errors": 1,
+ "words": 8,
+ "wer": 0.125
+ },
+ "semantic": {
+ "normalized_reference": "her words sent a strange chill through philip",
+ "normalized_hypothesis": "her words sent a strange chill through phyllis",
+ "errors": 1,
+ "words": 8,
+ "wer": 0.125
+ }
+ },
+ {
+ "id": "release500_0205",
+ "category": "phonetic_balance",
+ "text": "It was introduced by Representative Dick of Ohio.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0205.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It was introduced by Representative Dick of Ohio.",
+ "raw": {
+ "normalized_reference": "it was introduced by representative dick of ohio.",
+ "normalized_hypothesis": "it was introduced by representative dick of ohio.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it was introduced by representative dick of ohio",
+ "normalized_hypothesis": "it was introduced by representative dick of ohio",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0206",
+ "category": "phonetic_balance",
+ "text": "It's the aurora borealis.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0206.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It's the Aurora Borealis.",
+ "raw": {
+ "normalized_reference": "it's the aurora borealis.",
+ "normalized_hypothesis": "it's the aurora borealis.",
+ "errors": 0,
+ "words": 4,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it is the aurora borealis",
+ "normalized_hypothesis": "it is the aurora borealis",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0207",
+ "category": "phonetic_balance",
+ "text": "The boy grew and prospered.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0207.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The boy grew and prospered.",
+ "raw": {
+ "normalized_reference": "the boy grew and prospered.",
+ "normalized_hypothesis": "the boy grew and prospered.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the boy grew and prospered",
+ "normalized_hypothesis": "the boy grew and prospered",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0208",
+ "category": "phonetic_balance",
+ "text": "His voice was passionately rebellious.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0208.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "His voice was passionately rebellious.",
+ "raw": {
+ "normalized_reference": "his voice was passionately rebellious.",
+ "normalized_hypothesis": "his voice was passionately rebellious.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "his voice was passionately rebellious",
+ "normalized_hypothesis": "his voice was passionately rebellious",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0209",
+ "category": "phonetic_balance",
+ "text": "That's where they cut off the Scottish Chiefs and killed all hands.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0209.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "That's where they cut off the Scottish chiefs and killed all hands.",
+ "raw": {
+ "normalized_reference": "that's where they cut off the scottish chiefs and killed all hands.",
+ "normalized_hypothesis": "that's where they cut off the scottish chiefs and killed all hands.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "that is where they cut off the scottish chiefs and killed all hands",
+ "normalized_hypothesis": "that is where they cut off the scottish chiefs and killed all hands",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0210",
+ "category": "phonetic_balance",
+ "text": "And Tom King patiently endured.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0210.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "And Tom King patiently endured.",
+ "raw": {
+ "normalized_reference": "and tom king patiently endured.",
+ "normalized_hypothesis": "and tom king patiently endured.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "and tom king patiently endured",
+ "normalized_hypothesis": "and tom king patiently endured",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0211",
+ "category": "phonetic_balance",
+ "text": "From that moment his friendship for Belize turns to hatred and jealousy.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0211.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "From that moment his friendship for Belize turns to hatred and jealousy.",
+ "raw": {
+ "normalized_reference": "from that moment his friendship for belize turns to hatred and jealousy.",
+ "normalized_hypothesis": "from that moment his friendship for belize turns to hatred and jealousy.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "from that moment his friendship for belize turns to hatred and jealousy",
+ "normalized_hypothesis": "from that moment his friendship for belize turns to hatred and jealousy",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0212",
+ "category": "phonetic_balance",
+ "text": "Then you don't believe in altruism.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0212.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Then you don't believe in altruism.",
+ "raw": {
+ "normalized_reference": "then you don't believe in altruism.",
+ "normalized_hypothesis": "then you don't believe in altruism.",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "then you do not believe in altruism",
+ "normalized_hypothesis": "then you do not believe in altruism",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0213",
+ "category": "phonetic_balance",
+ "text": "What Pascal glimpsed with the vision of a seer, I have lived.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0213.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "What Pascal glimpsed with the vision of a seer, I have lived.",
+ "raw": {
+ "normalized_reference": "what pascal glimpsed with the vision of a seer, i have lived.",
+ "normalized_hypothesis": "what pascal glimpsed with the vision of a seer, i have lived.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "what pascal glimpsed with the vision of a seer i have lived",
+ "normalized_hypothesis": "what pascal glimpsed with the vision of a seer i have lived",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0214",
+ "category": "phonetic_balance",
+ "text": "You were making them talk shop, Ruth charged him.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0214.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "You were making them talk shop. Ruth charged him.",
+ "raw": {
+ "normalized_reference": "you were making them talk shop, ruth charged him.",
+ "normalized_hypothesis": "you were making them talk shop. ruth charged him.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "you were making them talk shop ruth charged him",
+ "normalized_hypothesis": "you were making them talk shop ruth charged him",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0215",
+ "category": "phonetic_balance",
+ "text": "Philip snatched at the letter which Gregson held out to him.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0215.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Philip snatched at the letter which Gregson held out to him.",
+ "raw": {
+ "normalized_reference": "philip snatched at the letter which gregson held out to him.",
+ "normalized_hypothesis": "philip snatched at the letter which gregson held out to him.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "philip snatched at the letter which gregson held out to him",
+ "normalized_hypothesis": "philip snatched at the letter which gregson held out to him",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0216",
+ "category": "phonetic_balance",
+ "text": "For a time the exciting thrill of his adventure was gone.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0216.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "For a time the exciting thrill of his adventure was gone.",
+ "raw": {
+ "normalized_reference": "for a time the exciting thrill of his adventure was gone.",
+ "normalized_hypothesis": "for a time the exciting thrill of his adventure was gone.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "for a time the exciting thrill of his adventure was gone",
+ "normalized_hypothesis": "for a time the exciting thrill of his adventure was gone",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0217",
+ "category": "phonetic_balance",
+ "text": "Obviously, it was a disease that could be contracted by contact.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0217.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Obviously, it was a disease that could be contracted by contact.",
+ "raw": {
+ "normalized_reference": "obviously, it was a disease that could be contracted by contact.",
+ "normalized_hypothesis": "obviously, it was a disease that could be contracted by contact.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "obviously it was a disease that could be contracted by contact",
+ "normalized_hypothesis": "obviously it was a disease that could be contracted by contact",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0218",
+ "category": "phonetic_balance",
+ "text": "Blind with rage, he darted in.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0218.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Blind with rage, he darted in.",
+ "raw": {
+ "normalized_reference": "blind with rage, he darted in.",
+ "normalized_hypothesis": "blind with rage, he darted in.",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "blind with rage he darted in",
+ "normalized_hypothesis": "blind with rage he darted in",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0219",
+ "category": "phonetic_balance",
+ "text": "You live on an income which your father earned.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0219.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "You live on an income which your father earned.",
+ "raw": {
+ "normalized_reference": "you live on an income which your father earned.",
+ "normalized_hypothesis": "you live on an income which your father earned.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "you live on an income which your father earned",
+ "normalized_hypothesis": "you live on an income which your father earned",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0220",
+ "category": "phonetic_balance",
+ "text": "Something about it seemed to fascinate him, to challenge his presence.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0220.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Something about it seemed to fascinate him, to challenge his presence.",
+ "raw": {
+ "normalized_reference": "something about it seemed to fascinate him, to challenge his presence.",
+ "normalized_hypothesis": "something about it seemed to fascinate him, to challenge his presence.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "something about it seemed to fascinate him to challenge his presence",
+ "normalized_hypothesis": "something about it seemed to fascinate him to challenge his presence",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0221",
+ "category": "phonetic_balance",
+ "text": "Her beautiful hair was done up in shining coils.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0221.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Her beautiful hair was done up in shining coils.",
+ "raw": {
+ "normalized_reference": "her beautiful hair was done up in shining coils.",
+ "normalized_hypothesis": "her beautiful hair was done up in shining coils.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "her beautiful hair was done up in shining coils",
+ "normalized_hypothesis": "her beautiful hair was done up in shining coils",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0222",
+ "category": "phonetic_balance",
+ "text": "It does, was her audacious answer.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0222.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It does, was her audacious answer.",
+ "raw": {
+ "normalized_reference": "it does, was her audacious answer.",
+ "normalized_hypothesis": "it does, was her audacious answer.",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it does was her audacious answer",
+ "normalized_hypothesis": "it does was her audacious answer",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0223",
+ "category": "phonetic_balance",
+ "text": "He caught himself with a jerk.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0223.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He caught himself with a jerk.",
+ "raw": {
+ "normalized_reference": "he caught himself with a jerk.",
+ "normalized_hypothesis": "he caught himself with a jerk.",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he caught himself with a jerk",
+ "normalized_hypothesis": "he caught himself with a jerk",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0224",
+ "category": "phonetic_balance",
+ "text": "He did not believe in the burning of daylight for such a luxury.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0224.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He did not believe in the burning of daylight for such a luxury.",
+ "raw": {
+ "normalized_reference": "he did not believe in the burning of daylight for such a luxury.",
+ "normalized_hypothesis": "he did not believe in the burning of daylight for such a luxury.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he did not believe in the burning of daylight for such a luxury",
+ "normalized_hypothesis": "he did not believe in the burning of daylight for such a luxury",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0225",
+ "category": "phonetic_balance",
+ "text": "He had comparatively no advantages at first.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0225.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He had comparatively no advantages at first.",
+ "raw": {
+ "normalized_reference": "he had comparatively no advantages at first.",
+ "normalized_hypothesis": "he had comparatively no advantages at first.",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he had comparatively no advantages at 1st",
+ "normalized_hypothesis": "he had comparatively no advantages at 1st",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0226",
+ "category": "phonetic_balance",
+ "text": "Such men believe, when they come together.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0226.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Such men believe when they come together.",
+ "raw": {
+ "normalized_reference": "such men believe, when they come together.",
+ "normalized_hypothesis": "such men believe when they come together.",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "such men believe when they come together",
+ "normalized_hypothesis": "such men believe when they come together",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0227",
+ "category": "phonetic_balance",
+ "text": "The ship should be in within a week or ten days.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0227.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The ship should be in within a week or ten days.",
+ "raw": {
+ "normalized_reference": "the ship should be in within a week or ten days.",
+ "normalized_hypothesis": "the ship should be in within a week or ten days.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the ship should be in within a week or 10 days",
+ "normalized_hypothesis": "the ship should be in within a week or 10 days",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0228",
+ "category": "phonetic_balance",
+ "text": "Such a dog the wise driver kills, or turns loose.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0228.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Such a dog the wise driver kills or turns loose.",
+ "raw": {
+ "normalized_reference": "such a dog the wise driver kills, or turns loose.",
+ "normalized_hypothesis": "such a dog the wise driver kills or turns loose.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "such a dog the wise driver kills or turns loose",
+ "normalized_hypothesis": "such a dog the wise driver kills or turns loose",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0229",
+ "category": "phonetic_balance",
+ "text": "You have heard always how he was the lover of the Princess Naomi.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0229.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "You have heard always how he was the lover of the princess Naomi.",
+ "raw": {
+ "normalized_reference": "you have heard always how he was the lover of the princess naomi.",
+ "normalized_hypothesis": "you have heard always how he was the lover of the princess naomi.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "you have heard always how he was the lover of the princess naomi",
+ "normalized_hypothesis": "you have heard always how he was the lover of the princess naomi",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0230",
+ "category": "phonetic_balance",
+ "text": "I want to know how all this is possible.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0230.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I want to know how all this is possible.",
+ "raw": {
+ "normalized_reference": "i want to know how all this is possible.",
+ "normalized_hypothesis": "i want to know how all this is possible.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i want to know how all this is possible",
+ "normalized_hypothesis": "i want to know how all this is possible",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0231",
+ "category": "phonetic_balance",
+ "text": "It was my idea to a tee.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0231.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It was my idea to a T.",
+ "raw": {
+ "normalized_reference": "it was my idea to a tee.",
+ "normalized_hypothesis": "it was my idea to a t.",
+ "errors": 1,
+ "words": 7,
+ "wer": 0.14285714285714285
+ },
+ "semantic": {
+ "normalized_reference": "it was my idea to a tee",
+ "normalized_hypothesis": "it was my idea to a t",
+ "errors": 1,
+ "words": 7,
+ "wer": 0.14285714285714285
+ }
+ },
+ {
+ "id": "release500_0232",
+ "category": "phonetic_balance",
+ "text": "They likewise are disinclined to being eaten.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0232.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "They likewise are disinclined to being eaten.",
+ "raw": {
+ "normalized_reference": "they likewise are disinclined to being eaten.",
+ "normalized_hypothesis": "they likewise are disinclined to being eaten.",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "they likewise are disinclined to being eaten",
+ "normalized_hypothesis": "they likewise are disinclined to being eaten",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0233",
+ "category": "phonetic_balance",
+ "text": "This state of mind comes of an undue prominence of the ego.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0233.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "This state of mind comes of an undue prominence of the ego.",
+ "raw": {
+ "normalized_reference": "this state of mind comes of an undue prominence of the ego.",
+ "normalized_hypothesis": "this state of mind comes of an undue prominence of the ego.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "this state of mind comes of an undue prominence of the ego",
+ "normalized_hypothesis": "this state of mind comes of an undue prominence of the ego",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0234",
+ "category": "phonetic_balance",
+ "text": "Then it is as I said, Womble announced with finality.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0234.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Then it is as I said, Womble announced with finality.",
+ "raw": {
+ "normalized_reference": "then it is as i said, womble announced with finality.",
+ "normalized_hypothesis": "then it is as i said, womble announced with finality.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "then it is as i said womble announced with finality",
+ "normalized_hypothesis": "then it is as i said womble announced with finality",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0235",
+ "category": "phonetic_balance",
+ "text": "I can see that knife now.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0235.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I can see that knife now.",
+ "raw": {
+ "normalized_reference": "i can see that knife now.",
+ "normalized_hypothesis": "i can see that knife now.",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i can see that knife now",
+ "normalized_hypothesis": "i can see that knife now",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0236",
+ "category": "phonetic_balance",
+ "text": "If I meet her again I shall apologize, said Eileen.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0236.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "If I meet her again I shall apologize, said Eileen.",
+ "raw": {
+ "normalized_reference": "if i meet her again i shall apologize, said eileen.",
+ "normalized_hypothesis": "if i meet her again i shall apologize, said eileen.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "if i meet her again i shall apologize said eileen",
+ "normalized_hypothesis": "if i meet her again i shall apologize said eileen",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0237",
+ "category": "phonetic_balance",
+ "text": "A wildly exciting time was his during the week preceding Thursday the eighteenth.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0237.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "A wildly exciting time was his during the week preceding Thursday the 18th.",
+ "raw": {
+ "normalized_reference": "a wildly exciting time was his during the week preceding thursday the eighteenth.",
+ "normalized_hypothesis": "a wildly exciting time was his during the week preceding thursday the 18th.",
+ "errors": 1,
+ "words": 13,
+ "wer": 0.07692307692307693
+ },
+ "semantic": {
+ "normalized_reference": "a wildly exciting time was his during the week preceding thursday the 18th",
+ "normalized_hypothesis": "a wildly exciting time was his during the week preceding thursday the 18th",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0238",
+ "category": "phonetic_balance",
+ "text": "In what bucolic school of fence he had been taught was beyond imagining.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0238.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "In what bucolic school of fence he had been taught was beyond imagining.",
+ "raw": {
+ "normalized_reference": "in what bucolic school of fence he had been taught was beyond imagining.",
+ "normalized_hypothesis": "in what bucolic school of fence he had been taught was beyond imagining.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "in what bucolic school of fence he had been taught was beyond imagining",
+ "normalized_hypothesis": "in what bucolic school of fence he had been taught was beyond imagining",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0239",
+ "category": "phonetic_balance",
+ "text": "So she said, the irate skipper dashed on.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0239.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "So she said, the irate skipper dashed on.",
+ "raw": {
+ "normalized_reference": "so she said, the irate skipper dashed on.",
+ "normalized_hypothesis": "so she said, the irate skipper dashed on.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "so she said the irate skipper dashed on",
+ "normalized_hypothesis": "so she said the irate skipper dashed on",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0240",
+ "category": "phonetic_balance",
+ "text": "It is not an attempt to smash the market.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0240.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "is not an attempt to smash the market.",
+ "raw": {
+ "normalized_reference": "it is not an attempt to smash the market.",
+ "normalized_hypothesis": "is not an attempt to smash the market.",
+ "errors": 1,
+ "words": 9,
+ "wer": 0.1111111111111111
+ },
+ "semantic": {
+ "normalized_reference": "it is not an attempt to smash the market",
+ "normalized_hypothesis": "is not an attempt to smash the market",
+ "errors": 1,
+ "words": 9,
+ "wer": 0.1111111111111111
+ }
+ },
+ {
+ "id": "release500_0241",
+ "category": "phonetic_balance",
+ "text": "He destroyed everything that had belonged to the woman.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0241.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He destroyed everything that had belonged to the woman.",
+ "raw": {
+ "normalized_reference": "he destroyed everything that had belonged to the woman.",
+ "normalized_hypothesis": "he destroyed everything that had belonged to the woman.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he destroyed everything that had belonged to the woman",
+ "normalized_hypothesis": "he destroyed everything that had belonged to the woman",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0242",
+ "category": "phonetic_balance",
+ "text": "Did Thorpe go to see any one in Churchill.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0242.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Did Thorpe go to see anyone in Churchill?",
+ "raw": {
+ "normalized_reference": "did thorpe go to see any one in churchill.",
+ "normalized_hypothesis": "did thorpe go to see anyone in churchill?",
+ "errors": 2,
+ "words": 9,
+ "wer": 0.2222222222222222
+ },
+ "semantic": {
+ "normalized_reference": "did thorpe go to see any one in churchill",
+ "normalized_hypothesis": "did thorpe go to see anyone in churchill",
+ "errors": 2,
+ "words": 9,
+ "wer": 0.2222222222222222
+ }
+ },
+ {
+ "id": "release500_0243",
+ "category": "phonetic_balance",
+ "text": "Many other similar disconcerting omissions will be noticed in the Manuscript.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0243.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Many other similar disconcerting omissions will be noticed in the manuscript.",
+ "raw": {
+ "normalized_reference": "many other similar disconcerting omissions will be noticed in the manuscript.",
+ "normalized_hypothesis": "many other similar disconcerting omissions will be noticed in the manuscript.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "many other similar disconcerting omissions will be noticed in the manuscript",
+ "normalized_hypothesis": "many other similar disconcerting omissions will be noticed in the manuscript",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0244",
+ "category": "phonetic_balance",
+ "text": "The ringing of the big bell aroused him.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0244.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The ringing of the big bell aroused him.",
+ "raw": {
+ "normalized_reference": "the ringing of the big bell aroused him.",
+ "normalized_hypothesis": "the ringing of the big bell aroused him.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the ringing of the big bell aroused him",
+ "normalized_hypothesis": "the ringing of the big bell aroused him",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0245",
+ "category": "phonetic_balance",
+ "text": "A sinewy hand, dripping with water, was clutching the rail.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0245.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "A sinewy hand, dripping with water, was clutching the rail.",
+ "raw": {
+ "normalized_reference": "a sinewy hand, dripping with water, was clutching the rail.",
+ "normalized_hypothesis": "a sinewy hand, dripping with water, was clutching the rail.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "a sinewy hand dripping with water was clutching the rail",
+ "normalized_hypothesis": "a sinewy hand dripping with water was clutching the rail",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0246",
+ "category": "phonetic_balance",
+ "text": "Enters now the psychology of the situation.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0246.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Enters now the psychology of the situation.",
+ "raw": {
+ "normalized_reference": "enters now the psychology of the situation.",
+ "normalized_hypothesis": "enters now the psychology of the situation.",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "enters now the psychology of the situation",
+ "normalized_hypothesis": "enters now the psychology of the situation",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0247",
+ "category": "phonetic_balance",
+ "text": "Wash your hands of me.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0247.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Wash your hands of me.",
+ "raw": {
+ "normalized_reference": "wash your hands of me.",
+ "normalized_hypothesis": "wash your hands of me.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "wash your hands of me",
+ "normalized_hypothesis": "wash your hands of me",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0248",
+ "category": "phonetic_balance",
+ "text": "He confessed that the sketch had startled him.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0248.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He confessed that this sketch had startled him.",
+ "raw": {
+ "normalized_reference": "he confessed that the sketch had startled him.",
+ "normalized_hypothesis": "he confessed that this sketch had startled him.",
+ "errors": 1,
+ "words": 8,
+ "wer": 0.125
+ },
+ "semantic": {
+ "normalized_reference": "he confessed that the sketch had startled him",
+ "normalized_hypothesis": "he confessed that this sketch had startled him",
+ "errors": 1,
+ "words": 8,
+ "wer": 0.125
+ }
+ },
+ {
+ "id": "release500_0249",
+ "category": "phonetic_balance",
+ "text": "Her face was against his breast.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0249.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Her face was against his breast.",
+ "raw": {
+ "normalized_reference": "her face was against his breast.",
+ "normalized_hypothesis": "her face was against his breast.",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "her face was against his breast",
+ "normalized_hypothesis": "her face was against his breast",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0250",
+ "category": "phonetic_balance",
+ "text": "He was a pariah; a wanderer without a friend or a home.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0250.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He was a pariah, a wanderer without a friend or a home.",
+ "raw": {
+ "normalized_reference": "he was a pariah; a wanderer without a friend or a home.",
+ "normalized_hypothesis": "he was a pariah, a wanderer without a friend or a home.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he was a pariah a wanderer without a friend or a home",
+ "normalized_hypothesis": "he was a pariah a wanderer without a friend or a home",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0251",
+ "category": "phonetic_balance",
+ "text": "A cry of joy burst from Philip's lips.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0251.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The cry of joy burst from Philip's lips.",
+ "raw": {
+ "normalized_reference": "a cry of joy burst from philip's lips.",
+ "normalized_hypothesis": "the cry of joy burst from philip's lips.",
+ "errors": 1,
+ "words": 8,
+ "wer": 0.125
+ },
+ "semantic": {
+ "normalized_reference": "a cry of joy burst from philip is lips",
+ "normalized_hypothesis": "the cry of joy burst from philip is lips",
+ "errors": 1,
+ "words": 9,
+ "wer": 0.1111111111111111
+ }
+ },
+ {
+ "id": "release500_0252",
+ "category": "phonetic_balance",
+ "text": "And the big Persian knew of his existence before he did of hers.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0252.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "And the big Persian knew of his existence before he did of hers.",
+ "raw": {
+ "normalized_reference": "and the big persian knew of his existence before he did of hers.",
+ "normalized_hypothesis": "and the big persian knew of his existence before he did of hers.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "and the big persian knew of his existence before he did of hers",
+ "normalized_hypothesis": "and the big persian knew of his existence before he did of hers",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0253",
+ "category": "phonetic_balance",
+ "text": "The men stared into each other's face.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0253.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The men stared into each other's face.",
+ "raw": {
+ "normalized_reference": "the men stared into each other's face.",
+ "normalized_hypothesis": "the men stared into each other's face.",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the men stared into each other is face",
+ "normalized_hypothesis": "the men stared into each other is face",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0254",
+ "category": "phonetic_balance",
+ "text": "It is the fire, partly, she said.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0254.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Is the fire partly, she said.",
+ "raw": {
+ "normalized_reference": "it is the fire, partly, she said.",
+ "normalized_hypothesis": "is the fire partly, she said.",
+ "errors": 1,
+ "words": 7,
+ "wer": 0.14285714285714285
+ },
+ "semantic": {
+ "normalized_reference": "it is the fire partly she said",
+ "normalized_hypothesis": "is the fire partly she said",
+ "errors": 1,
+ "words": 7,
+ "wer": 0.14285714285714285
+ }
+ },
+ {
+ "id": "release500_0255",
+ "category": "phonetic_balance",
+ "text": "It seemed nearer to him since he had seen and talked with Gregson.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0255.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It seemed nearer to him since he had seen and talked with Gregson.",
+ "raw": {
+ "normalized_reference": "it seemed nearer to him since he had seen and talked with gregson.",
+ "normalized_hypothesis": "it seemed nearer to him since he had seen and talked with gregson.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it seemed nearer to him since he had seen and talked with gregson",
+ "normalized_hypothesis": "it seemed nearer to him since he had seen and talked with gregson",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0256",
+ "category": "phonetic_balance",
+ "text": "Among my minor afflictions, I may mention a new and mysterious one.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0256.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Among my minor afflictions, I may mention a new and mysterious one.",
+ "raw": {
+ "normalized_reference": "among my minor afflictions, i may mention a new and mysterious one.",
+ "normalized_hypothesis": "among my minor afflictions, i may mention a new and mysterious one.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "among my minor afflictions i may mention a new and mysterious one",
+ "normalized_hypothesis": "among my minor afflictions i may mention a new and mysterious one",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0257",
+ "category": "phonetic_balance",
+ "text": "Between him and all domestic animals there must be no hostilities.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0257.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Between them and all domestic animals there must be no hostilities.",
+ "raw": {
+ "normalized_reference": "between him and all domestic animals there must be no hostilities.",
+ "normalized_hypothesis": "between them and all domestic animals there must be no hostilities.",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ },
+ "semantic": {
+ "normalized_reference": "between him and all domestic animals there must be no hostilities",
+ "normalized_hypothesis": "between them and all domestic animals there must be no hostilities",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ }
+ },
+ {
+ "id": "release500_0258",
+ "category": "phonetic_balance",
+ "text": "With these arguments he convinced himself that he should go on alone.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0258.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "With these arguments he convinced himself that he should go on alone.",
+ "raw": {
+ "normalized_reference": "with these arguments he convinced himself that he should go on alone.",
+ "normalized_hypothesis": "with these arguments he convinced himself that he should go on alone.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "with these arguments he convinced himself that he should go on alone",
+ "normalized_hypothesis": "with these arguments he convinced himself that he should go on alone",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0259",
+ "category": "phonetic_balance",
+ "text": "We are consumed in our own flesh-pots.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0259.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "We are consumed in our own fleshbots.",
+ "raw": {
+ "normalized_reference": "we are consumed in our own flesh-pots.",
+ "normalized_hypothesis": "we are consumed in our own fleshbots.",
+ "errors": 2,
+ "words": 8,
+ "wer": 0.25
+ },
+ "semantic": {
+ "normalized_reference": "we are consumed in our own flesh pots",
+ "normalized_hypothesis": "we are consumed in our own fleshbots",
+ "errors": 2,
+ "words": 8,
+ "wer": 0.25
+ }
+ },
+ {
+ "id": "release500_0260",
+ "category": "phonetic_balance",
+ "text": "Mercedes screamed, cried, laughed, and manifested the chaotic abandonment of hysteria.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0260.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Mercedes screamed, cried, laughed, and manifested the chaotic abandonment of hysteria.",
+ "raw": {
+ "normalized_reference": "mercedes screamed, cried, laughed, and manifested the chaotic abandonment of hysteria.",
+ "normalized_hypothesis": "mercedes screamed, cried, laughed, and manifested the chaotic abandonment of hysteria.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "mercedes screamed cried laughed and manifested the chaotic abandonment of hysteria",
+ "normalized_hypothesis": "mercedes screamed cried laughed and manifested the chaotic abandonment of hysteria",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0261",
+ "category": "phonetic_balance",
+ "text": "Philip made no effort to follow.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0261.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Philip made no effort to follow.",
+ "raw": {
+ "normalized_reference": "philip made no effort to follow.",
+ "normalized_hypothesis": "philip made no effort to follow.",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "philip made no effort to follow",
+ "normalized_hypothesis": "philip made no effort to follow",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0262",
+ "category": "phonetic_balance",
+ "text": "She said with chattering teeth.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0262.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "She said with chattering teeth.",
+ "raw": {
+ "normalized_reference": "she said with chattering teeth.",
+ "normalized_hypothesis": "she said with chattering teeth.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "she said with chattering teeth",
+ "normalized_hypothesis": "she said with chattering teeth",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0263",
+ "category": "phonetic_balance",
+ "text": "Halfway around the track one donkey got into an argument with its rider.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0263.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Halfway around mid-track one donkey got into an argument with its rider.",
+ "raw": {
+ "normalized_reference": "halfway around the track one donkey got into an argument with its rider.",
+ "normalized_hypothesis": "halfway around mid-track one donkey got into an argument with its rider.",
+ "errors": 1,
+ "words": 13,
+ "wer": 0.07692307692307693
+ },
+ "semantic": {
+ "normalized_reference": "halfway around the track one donkey got into an argument with its rider",
+ "normalized_hypothesis": "halfway around mid track one donkey got into an argument with its rider",
+ "errors": 1,
+ "words": 13,
+ "wer": 0.07692307692307693
+ }
+ },
+ {
+ "id": "release500_0264",
+ "category": "phonetic_balance",
+ "text": "Earth and gravel seemed to fill the pan.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0264.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Earth and gravel seemed to fill the pan.",
+ "raw": {
+ "normalized_reference": "earth and gravel seemed to fill the pan.",
+ "normalized_hypothesis": "earth and gravel seemed to fill the pan.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "earth and gravel seemed to fill the pan",
+ "normalized_hypothesis": "earth and gravel seemed to fill the pan",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0265",
+ "category": "phonetic_balance",
+ "text": "Her achievements with cocoanuts were a revelation.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0265.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Her achievements with Coconuts were a revelation.",
+ "raw": {
+ "normalized_reference": "her achievements with cocoanuts were a revelation.",
+ "normalized_hypothesis": "her achievements with coconuts were a revelation.",
+ "errors": 1,
+ "words": 7,
+ "wer": 0.14285714285714285
+ },
+ "semantic": {
+ "normalized_reference": "her achievements with cocoanuts were a revelation",
+ "normalized_hypothesis": "her achievements with coconuts were a revelation",
+ "errors": 1,
+ "words": 7,
+ "wer": 0.14285714285714285
+ }
+ },
+ {
+ "id": "release500_0266",
+ "category": "phonetic_balance",
+ "text": "For a full minute he crouched and listened.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0266.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "For a full minute he crouched and listened.",
+ "raw": {
+ "normalized_reference": "for a full minute he crouched and listened.",
+ "normalized_hypothesis": "for a full minute he crouched and listened.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "for a full minute he crouched and listened",
+ "normalized_hypothesis": "for a full minute he crouched and listened",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0267",
+ "category": "phonetic_balance",
+ "text": "It was edged with ice.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0267.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It was edged with ice.",
+ "raw": {
+ "normalized_reference": "it was edged with ice.",
+ "normalized_hypothesis": "it was edged with ice.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it was edged with ice",
+ "normalized_hypothesis": "it was edged with ice",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0268",
+ "category": "phonetic_balance",
+ "text": "The Portuguese boy passed the Hawaiian.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0268.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The Portuguese boy passed the Hawaiian.",
+ "raw": {
+ "normalized_reference": "the portuguese boy passed the hawaiian.",
+ "normalized_hypothesis": "the portuguese boy passed the hawaiian.",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the portuguese boy passed the hawaiian",
+ "normalized_hypothesis": "the portuguese boy passed the hawaiian",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0269",
+ "category": "phonetic_balance",
+ "text": "But he did not broach it, preferring to mature it carefully.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0269.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "But he did not broach it, preferring to mature it carefully.",
+ "raw": {
+ "normalized_reference": "but he did not broach it, preferring to mature it carefully.",
+ "normalized_hypothesis": "but he did not broach it, preferring to mature it carefully.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "but he did not broach it preferring to mature it carefully",
+ "normalized_hypothesis": "but he did not broach it preferring to mature it carefully",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0270",
+ "category": "phonetic_balance",
+ "text": "Come on, Del Mar challenged.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0270.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Come on, Del Mar challenged.",
+ "raw": {
+ "normalized_reference": "come on, del mar challenged.",
+ "normalized_hypothesis": "come on, del mar challenged.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "come on del mar challenged",
+ "normalized_hypothesis": "come on del mar challenged",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0271",
+ "category": "phonetic_balance",
+ "text": "For a full half minute Jeanne looked at him without speaking.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0271.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "For a full half minute, Gene looked at him without speaking.",
+ "raw": {
+ "normalized_reference": "for a full half minute jeanne looked at him without speaking.",
+ "normalized_hypothesis": "for a full half minute, gene looked at him without speaking.",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ },
+ "semantic": {
+ "normalized_reference": "for a full half minute jeanne looked at him without speaking",
+ "normalized_hypothesis": "for a full half minute gene looked at him without speaking",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ }
+ },
+ {
+ "id": "release500_0272",
+ "category": "phonetic_balance",
+ "text": "He bore no grudges and had few enemies.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0272.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He bore no grudges and had few enemies.",
+ "raw": {
+ "normalized_reference": "he bore no grudges and had few enemies.",
+ "normalized_hypothesis": "he bore no grudges and had few enemies.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he bore no grudges and had few enemies",
+ "normalized_hypothesis": "he bore no grudges and had few enemies",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0273",
+ "category": "phonetic_balance",
+ "text": "I'll go over tomorrow afternoon.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0273.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I'll go over tomorrow afternoon.",
+ "raw": {
+ "normalized_reference": "i'll go over tomorrow afternoon.",
+ "normalized_hypothesis": "i'll go over tomorrow afternoon.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i will go over tomorrow afternoon",
+ "normalized_hypothesis": "i will go over tomorrow afternoon",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0274",
+ "category": "phonetic_balance",
+ "text": "Well, there are better men in Hawaii, that's all.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0274.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Well, there are better men in Hawaii, that's all.",
+ "raw": {
+ "normalized_reference": "well, there are better men in hawaii, that's all.",
+ "normalized_hypothesis": "well, there are better men in hawaii, that's all.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "well there are better men in hawaii that is all",
+ "normalized_hypothesis": "well there are better men in hawaii that is all",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0275",
+ "category": "phonetic_balance",
+ "text": "He was an athlete and a giant.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0275.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He was an athlete and a giant.",
+ "raw": {
+ "normalized_reference": "he was an athlete and a giant.",
+ "normalized_hypothesis": "he was an athlete and a giant.",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he was an athlete and a giant",
+ "normalized_hypothesis": "he was an athlete and a giant",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0276",
+ "category": "phonetic_balance",
+ "text": "McVeigh when he returns from a trip to Honolulu.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0276.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "McVeigh when he returns from a trip to Honolulu.",
+ "raw": {
+ "normalized_reference": "mcveigh when he returns from a trip to honolulu.",
+ "normalized_hypothesis": "mcveigh when he returns from a trip to honolulu.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "mcveigh when he returns from a trip to honolulu",
+ "normalized_hypothesis": "mcveigh when he returns from a trip to honolulu",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0277",
+ "category": "phonetic_balance",
+ "text": "You're a devil for fighting, and will surely win.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0277.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "You're a devil for fighting, and will surely win.",
+ "raw": {
+ "normalized_reference": "you're a devil for fighting, and will surely win.",
+ "normalized_hypothesis": "you're a devil for fighting, and will surely win.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "you are a devil for fighting and will surely win",
+ "normalized_hypothesis": "you are a devil for fighting and will surely win",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0278",
+ "category": "phonetic_balance",
+ "text": "All an appearance can know is mirage.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0278.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "All an appearance can know is mirage.",
+ "raw": {
+ "normalized_reference": "all an appearance can know is mirage.",
+ "normalized_hypothesis": "all an appearance can know is mirage.",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "all an appearance can know is mirage",
+ "normalized_hypothesis": "all an appearance can know is mirage",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0279",
+ "category": "phonetic_balance",
+ "text": "You fired me out of your house, in short.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0279.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "You fired me out of your house, insured.",
+ "raw": {
+ "normalized_reference": "you fired me out of your house, in short.",
+ "normalized_hypothesis": "you fired me out of your house, insured.",
+ "errors": 2,
+ "words": 9,
+ "wer": 0.2222222222222222
+ },
+ "semantic": {
+ "normalized_reference": "you fired me out of your house in short",
+ "normalized_hypothesis": "you fired me out of your house insured",
+ "errors": 2,
+ "words": 9,
+ "wer": 0.2222222222222222
+ }
+ },
+ {
+ "id": "release500_0280",
+ "category": "phonetic_balance",
+ "text": "He made sure that the magazine was loaded, and resumed his paddling.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0280.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He made sure that the magazine was loaded and resumed his paddling.",
+ "raw": {
+ "normalized_reference": "he made sure that the magazine was loaded, and resumed his paddling.",
+ "normalized_hypothesis": "he made sure that the magazine was loaded and resumed his paddling.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he made sure that the magazine was loaded and resumed his paddling",
+ "normalized_hypothesis": "he made sure that the magazine was loaded and resumed his paddling",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0281",
+ "category": "phonetic_balance",
+ "text": "Change chairs, Daylight commanded.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0281.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Change chairs, Daylight commanded.",
+ "raw": {
+ "normalized_reference": "change chairs, daylight commanded.",
+ "normalized_hypothesis": "change chairs, daylight commanded.",
+ "errors": 0,
+ "words": 4,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "change chairs daylight commanded",
+ "normalized_hypothesis": "change chairs daylight commanded",
+ "errors": 0,
+ "words": 4,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0282",
+ "category": "phonetic_balance",
+ "text": "By that answer my professional medical prestige stood or fell.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0282.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "By that answer, my professional medical prestige stood or fell.",
+ "raw": {
+ "normalized_reference": "by that answer my professional medical prestige stood or fell.",
+ "normalized_hypothesis": "by that answer, my professional medical prestige stood or fell.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "by that answer my professional medical prestige stood or fell",
+ "normalized_hypothesis": "by that answer my professional medical prestige stood or fell",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0283",
+ "category": "phonetic_balance",
+ "text": "The moon had already begun its westward decline.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0283.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The moon had already begun its westward decline.",
+ "raw": {
+ "normalized_reference": "the moon had already begun its westward decline.",
+ "normalized_hypothesis": "the moon had already begun its westward decline.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the moon had already begun its westward decline",
+ "normalized_hypothesis": "the moon had already begun its westward decline",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0284",
+ "category": "phonetic_balance",
+ "text": "At sea, Monday, March 16, 1908.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0284.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "At sea, Monday, March 16, 1908.",
+ "raw": {
+ "normalized_reference": "at sea, monday, march 16, 1908.",
+ "normalized_hypothesis": "at sea, monday, march 16, 1908.",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "at sea monday march 16 1908",
+ "normalized_hypothesis": "at sea monday march 16 1908",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0285",
+ "category": "phonetic_balance",
+ "text": "And wherever I ranged, the way lay along alcohol-drenched roads.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0285.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "And wherever I ranged, the way lay along alcohol-drenched roads.",
+ "raw": {
+ "normalized_reference": "and wherever i ranged, the way lay along alcohol-drenched roads.",
+ "normalized_hypothesis": "and wherever i ranged, the way lay along alcohol-drenched roads.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "and wherever i ranged the way lay along alcohol drenched roads",
+ "normalized_hypothesis": "and wherever i ranged the way lay along alcohol drenched roads",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0286",
+ "category": "phonetic_balance",
+ "text": "The date was nearly eighteen years old.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0286.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The date was nearly 18 years old.",
+ "raw": {
+ "normalized_reference": "the date was nearly eighteen years old.",
+ "normalized_hypothesis": "the date was nearly 18 years old.",
+ "errors": 1,
+ "words": 7,
+ "wer": 0.14285714285714285
+ },
+ "semantic": {
+ "normalized_reference": "the date was nearly 18 years old",
+ "normalized_hypothesis": "the date was nearly 18 years old",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0287",
+ "category": "phonetic_balance",
+ "text": "Very early in my life, I separated from my mother.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0287.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Very early in my life, I separated from my mother.",
+ "raw": {
+ "normalized_reference": "very early in my life, i separated from my mother.",
+ "normalized_hypothesis": "very early in my life, i separated from my mother.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "very early in my life i separated from my mother",
+ "normalized_hypothesis": "very early in my life i separated from my mother",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0288",
+ "category": "phonetic_balance",
+ "text": "There was pride and strength, the ring of triumph in his voice.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0288.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "There was pride and strength, the ring of triumph in his voice.",
+ "raw": {
+ "normalized_reference": "there was pride and strength, the ring of triumph in his voice.",
+ "normalized_hypothesis": "there was pride and strength, the ring of triumph in his voice.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "there was pride and strength the ring of triumph in his voice",
+ "normalized_hypothesis": "there was pride and strength the ring of triumph in his voice",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0289",
+ "category": "phonetic_balance",
+ "text": "From my earliest recollection my sleep was a period of terror.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0289.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "From my earliest recollection, my sleep was a period of terror.",
+ "raw": {
+ "normalized_reference": "from my earliest recollection my sleep was a period of terror.",
+ "normalized_hypothesis": "from my earliest recollection, my sleep was a period of terror.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "from my earliest recollection my sleep was a period of terror",
+ "normalized_hypothesis": "from my earliest recollection my sleep was a period of terror",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0290",
+ "category": "phonetic_balance",
+ "text": "We'll have to watch our chances.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0290.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "We'll have to watch our chances.",
+ "raw": {
+ "normalized_reference": "we'll have to watch our chances.",
+ "normalized_hypothesis": "we'll have to watch our chances.",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "we will have to watch our chances",
+ "normalized_hypothesis": "we will have to watch our chances",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0291",
+ "category": "phonetic_balance",
+ "text": "Close beside him gleamed the white fangs of the wolf-dog.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0291.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Close beside him gleamed the white fangs of the wolf-dog.",
+ "raw": {
+ "normalized_reference": "close beside him gleamed the white fangs of the wolf-dog.",
+ "normalized_hypothesis": "close beside him gleamed the white fangs of the wolf-dog.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "close beside him gleamed the white fangs of the wolf dog",
+ "normalized_hypothesis": "close beside him gleamed the white fangs of the wolf dog",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0292",
+ "category": "phonetic_balance",
+ "text": "Would you be satisfied with that one hundredth part of me.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0292.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Would you be satisfied with that one hundredth part of me?",
+ "raw": {
+ "normalized_reference": "would you be satisfied with that one hundredth part of me.",
+ "normalized_hypothesis": "would you be satisfied with that one hundredth part of me?",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "would you be satisfied with that 100th part of me",
+ "normalized_hypothesis": "would you be satisfied with that 100th part of me",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0293",
+ "category": "phonetic_balance",
+ "text": "Two gigantic owls were tearing at the carcass.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0293.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Two gigantic owls were tearing at the carcass.",
+ "raw": {
+ "normalized_reference": "two gigantic owls were tearing at the carcass.",
+ "normalized_hypothesis": "two gigantic owls were tearing at the carcass.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "2 gigantic owls were tearing at the carcass",
+ "normalized_hypothesis": "2 gigantic owls were tearing at the carcass",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0294",
+ "category": "phonetic_balance",
+ "text": "The lush vegetation of that sheltered spot make a natural shield.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0294.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The lush vegetation of that sheltered spot make a natural shield.",
+ "raw": {
+ "normalized_reference": "the lush vegetation of that sheltered spot make a natural shield.",
+ "normalized_hypothesis": "the lush vegetation of that sheltered spot make a natural shield.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the lush vegetation of that sheltered spot make a natural shield",
+ "normalized_hypothesis": "the lush vegetation of that sheltered spot make a natural shield",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0295",
+ "category": "phonetic_balance",
+ "text": "Below him the shadow was broken into a pool of rippling starlight.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0295.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Below him, the shadow was broken into a pool of rippling starlight.",
+ "raw": {
+ "normalized_reference": "below him the shadow was broken into a pool of rippling starlight.",
+ "normalized_hypothesis": "below him, the shadow was broken into a pool of rippling starlight.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "below him the shadow was broken into a pool of rippling starlight",
+ "normalized_hypothesis": "below him the shadow was broken into a pool of rippling starlight",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0296",
+ "category": "phonetic_balance",
+ "text": "Lots of men take women buggy riding.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0296.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Lots of men take women buggy riding.",
+ "raw": {
+ "normalized_reference": "lots of men take women buggy riding.",
+ "normalized_hypothesis": "lots of men take women buggy riding.",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "lots of men take women buggy riding",
+ "normalized_hypothesis": "lots of men take women buggy riding",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0297",
+ "category": "phonetic_balance",
+ "text": "But it contributed to the smash.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0297.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "But it contributed to this splash.",
+ "raw": {
+ "normalized_reference": "but it contributed to the smash.",
+ "normalized_hypothesis": "but it contributed to this splash.",
+ "errors": 2,
+ "words": 6,
+ "wer": 0.3333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "but it contributed to the smash",
+ "normalized_hypothesis": "but it contributed to this splash",
+ "errors": 2,
+ "words": 6,
+ "wer": 0.3333333333333333
+ }
+ },
+ {
+ "id": "release500_0298",
+ "category": "phonetic_balance",
+ "text": "No man ate of the seal meat or the oil.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0298.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "No man ate of the seal meat or the oil.",
+ "raw": {
+ "normalized_reference": "no man ate of the seal meat or the oil.",
+ "normalized_hypothesis": "no man ate of the seal meat or the oil.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "no man ate of the seal meat or the oil",
+ "normalized_hypothesis": "no man ate of the seal meat or the oil",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0299",
+ "category": "phonetic_balance",
+ "text": "You're going in for grab sharing.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0299.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "You're going in for grab sharing.",
+ "raw": {
+ "normalized_reference": "you're going in for grab sharing.",
+ "normalized_hypothesis": "you're going in for grab sharing.",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "you are going in for grab sharing",
+ "normalized_hypothesis": "you are going in for grab sharing",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0300",
+ "category": "phonetic_balance",
+ "text": "He it was that lived to found the family of the Patino.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0300.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He it was that lived to found the family of the Patino.",
+ "raw": {
+ "normalized_reference": "he it was that lived to found the family of the patino.",
+ "normalized_hypothesis": "he it was that lived to found the family of the patino.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he it was that lived to found the family of the patino",
+ "normalized_hypothesis": "he it was that lived to found the family of the patino",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0301",
+ "category": "phonetic_balance",
+ "text": "For such countries nothing remained but reorganization.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0301.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "For such countries nothing remained but reorganization.",
+ "raw": {
+ "normalized_reference": "for such countries nothing remained but reorganization.",
+ "normalized_hypothesis": "for such countries nothing remained but reorganization.",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "for such countries nothing remained but reorganization",
+ "normalized_hypothesis": "for such countries nothing remained but reorganization",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0302",
+ "category": "phonetic_balance",
+ "text": "Stand off butcher and baker and all the rest.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0302.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Stand off Butcher and Baker and all the rest.",
+ "raw": {
+ "normalized_reference": "stand off butcher and baker and all the rest.",
+ "normalized_hypothesis": "stand off butcher and baker and all the rest.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "stand off butcher and baker and all the rest",
+ "normalized_hypothesis": "stand off butcher and baker and all the rest",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0303",
+ "category": "phonetic_balance",
+ "text": "All eyes, however, were staring at him in certitude of expectancy.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0303.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "All eyes, however, were staring at him in certitude of expectancy.",
+ "raw": {
+ "normalized_reference": "all eyes, however, were staring at him in certitude of expectancy.",
+ "normalized_hypothesis": "all eyes, however, were staring at him in certitude of expectancy.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "all eyes however were staring at him in certitude of expectancy",
+ "normalized_hypothesis": "all eyes however were staring at him in certitude of expectancy",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0304",
+ "category": "phonetic_balance",
+ "text": "Two days ago Jeanne learned where her father's men were hiding.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0304.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Two days ago, Jean learned where her father's men were hiding.",
+ "raw": {
+ "normalized_reference": "two days ago jeanne learned where her father's men were hiding.",
+ "normalized_hypothesis": "two days ago, jean learned where her father's men were hiding.",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ },
+ "semantic": {
+ "normalized_reference": "2 days ago jeanne learned where her father is men were hiding",
+ "normalized_hypothesis": "2 days ago jean learned where her father is men were hiding",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ }
+ },
+ {
+ "id": "release500_0305",
+ "category": "phonetic_balance",
+ "text": "His previous wives had never lived long enough to bear him children.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0305.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "His previous wives had never lived long enough to bear him children.",
+ "raw": {
+ "normalized_reference": "his previous wives had never lived long enough to bear him children.",
+ "normalized_hypothesis": "his previous wives had never lived long enough to bear him children.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "his previous wives had never lived long enough to bear him children",
+ "normalized_hypothesis": "his previous wives had never lived long enough to bear him children",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0306",
+ "category": "phonetic_balance",
+ "text": "But this little defect did not worry him.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0306.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "But this little defect did not worry him.",
+ "raw": {
+ "normalized_reference": "but this little defect did not worry him.",
+ "normalized_hypothesis": "but this little defect did not worry him.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "but this little defect did not worry him",
+ "normalized_hypothesis": "but this little defect did not worry him",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0307",
+ "category": "phonetic_balance",
+ "text": "Eli Harding asked, as Shunk started to follow.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0307.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Eli Harding asked as Shunk started to follow.",
+ "raw": {
+ "normalized_reference": "eli harding asked, as shunk started to follow.",
+ "normalized_hypothesis": "eli harding asked as shunk started to follow.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "eli harding asked as shunk started to follow",
+ "normalized_hypothesis": "eli harding asked as shunk started to follow",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0308",
+ "category": "phonetic_balance",
+ "text": "And this was their sole conversation throughout the meal.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0308.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "And this was their sole conversation throughout the meal.",
+ "raw": {
+ "normalized_reference": "and this was their sole conversation throughout the meal.",
+ "normalized_hypothesis": "and this was their sole conversation throughout the meal.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "and this was their sole conversation throughout the meal",
+ "normalized_hypothesis": "and this was their sole conversation throughout the meal",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0309",
+ "category": "phonetic_balance",
+ "text": "Then, again, Tudor had such an irritating way about him.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0309.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Then, again, Tudor had such an irritating way about him.",
+ "raw": {
+ "normalized_reference": "then, again, tudor had such an irritating way about him.",
+ "normalized_hypothesis": "then, again, tudor had such an irritating way about him.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "then again tudor had such an irritating way about him",
+ "normalized_hypothesis": "then again tudor had such an irritating way about him",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0310",
+ "category": "phonetic_balance",
+ "text": "In it there was something that was almost tragedy.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0310.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "In it there was something that was almost tragedy.",
+ "raw": {
+ "normalized_reference": "in it there was something that was almost tragedy.",
+ "normalized_hypothesis": "in it there was something that was almost tragedy.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "in it there was something that was almost tragedy",
+ "normalized_hypothesis": "in it there was something that was almost tragedy",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0311",
+ "category": "phonetic_balance",
+ "text": "Thirty pounds, said the captain with finality.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0311.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Thirty pounds. Send the captain with finality.",
+ "raw": {
+ "normalized_reference": "thirty pounds, said the captain with finality.",
+ "normalized_hypothesis": "thirty pounds. send the captain with finality.",
+ "errors": 1,
+ "words": 7,
+ "wer": 0.14285714285714285
+ },
+ "semantic": {
+ "normalized_reference": "\u00a330 said the captain with finality",
+ "normalized_hypothesis": "\u00a330 send the captain with finality",
+ "errors": 1,
+ "words": 6,
+ "wer": 0.16666666666666666
+ }
+ },
+ {
+ "id": "release500_0312",
+ "category": "phonetic_balance",
+ "text": "It was over when he made his way through the ring of spectators.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0312.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It was over when he made his way through the ring of spectators.",
+ "raw": {
+ "normalized_reference": "it was over when he made his way through the ring of spectators.",
+ "normalized_hypothesis": "it was over when he made his way through the ring of spectators.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it was over when he made his way through the ring of spectators",
+ "normalized_hypothesis": "it was over when he made his way through the ring of spectators",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0313",
+ "category": "phonetic_balance",
+ "text": "She had died from cold and starvation.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0313.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "She had died from cold and starvation.",
+ "raw": {
+ "normalized_reference": "she had died from cold and starvation.",
+ "normalized_hypothesis": "she had died from cold and starvation.",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "she had died from cold and starvation",
+ "normalized_hypothesis": "she had died from cold and starvation",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0314",
+ "category": "phonetic_balance",
+ "text": "We fished sharks on Niihau together.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0314.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "We fished sharks on Nihau together.",
+ "raw": {
+ "normalized_reference": "we fished sharks on niihau together.",
+ "normalized_hypothesis": "we fished sharks on nihau together.",
+ "errors": 1,
+ "words": 6,
+ "wer": 0.16666666666666666
+ },
+ "semantic": {
+ "normalized_reference": "we fished sharks on niihau together",
+ "normalized_hypothesis": "we fished sharks on nihau together",
+ "errors": 1,
+ "words": 6,
+ "wer": 0.16666666666666666
+ }
+ },
+ {
+ "id": "release500_0315",
+ "category": "phonetic_balance",
+ "text": "His slim hands gripped the edges of the table.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0315.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "His slim hands gripped the edges of the table.",
+ "raw": {
+ "normalized_reference": "his slim hands gripped the edges of the table.",
+ "normalized_hypothesis": "his slim hands gripped the edges of the table.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "his slim hands gripped the edges of the table",
+ "normalized_hypothesis": "his slim hands gripped the edges of the table",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0316",
+ "category": "phonetic_balance",
+ "text": "We had been chased by them ourselves, more than once.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0316.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "We had been chased by them ourselves more than once.",
+ "raw": {
+ "normalized_reference": "we had been chased by them ourselves, more than once.",
+ "normalized_hypothesis": "we had been chased by them ourselves more than once.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "we had been chased by them ourselves more than once",
+ "normalized_hypothesis": "we had been chased by them ourselves more than once",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0317",
+ "category": "phonetic_balance",
+ "text": "Or have they already devised one.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0317.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Or have they already devised one?",
+ "raw": {
+ "normalized_reference": "or have they already devised one.",
+ "normalized_hypothesis": "or have they already devised one?",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "or have they already devised one",
+ "normalized_hypothesis": "or have they already devised one",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0318",
+ "category": "phonetic_balance",
+ "text": "I cannot follow you, she said.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0318.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I cannot follow you, she said.",
+ "raw": {
+ "normalized_reference": "i cannot follow you, she said.",
+ "normalized_hypothesis": "i cannot follow you, she said.",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i cannot follow you she said",
+ "normalized_hypothesis": "i cannot follow you she said",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0319",
+ "category": "phonetic_balance",
+ "text": "He went down in midstream, searching the shadows of both shores.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0319.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He went down in midstream, searching the shadows of both shores.",
+ "raw": {
+ "normalized_reference": "he went down in midstream, searching the shadows of both shores.",
+ "normalized_hypothesis": "he went down in midstream, searching the shadows of both shores.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he went down in midstream searching the shadows of both shores",
+ "normalized_hypothesis": "he went down in midstream searching the shadows of both shores",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0320",
+ "category": "phonetic_balance",
+ "text": "I was sick once -- typhoid.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0320.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I was sick once, Typhoid.",
+ "raw": {
+ "normalized_reference": "i was sick once -- typhoid.",
+ "normalized_hypothesis": "i was sick once, typhoid.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i was sick once typhoid",
+ "normalized_hypothesis": "i was sick once typhoid",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0321",
+ "category": "phonetic_balance",
+ "text": "Why not like any railroad station or ferry depot.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0321.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Why not like any railroad station or a furry depot?",
+ "raw": {
+ "normalized_reference": "why not like any railroad station or ferry depot.",
+ "normalized_hypothesis": "why not like any railroad station or a furry depot?",
+ "errors": 2,
+ "words": 9,
+ "wer": 0.2222222222222222
+ },
+ "semantic": {
+ "normalized_reference": "why not like any railroad station or ferry depot",
+ "normalized_hypothesis": "why not like any railroad station or a furry depot",
+ "errors": 2,
+ "words": 9,
+ "wer": 0.2222222222222222
+ }
+ },
+ {
+ "id": "release500_0322",
+ "category": "phonetic_balance",
+ "text": "Red-Eye swung back and forth on the branch farther down.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0322.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Red Eye swung back and forth on the branch farther down.",
+ "raw": {
+ "normalized_reference": "red-eye swung back and forth on the branch farther down.",
+ "normalized_hypothesis": "red eye swung back and forth on the branch farther down.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "red eye swung back and forth on the branch farther down",
+ "normalized_hypothesis": "red eye swung back and forth on the branch farther down",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0323",
+ "category": "phonetic_balance",
+ "text": "Within himself he called it no longer his own.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0323.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Within himself he called it no longer his own.",
+ "raw": {
+ "normalized_reference": "within himself he called it no longer his own.",
+ "normalized_hypothesis": "within himself he called it no longer his own.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "within himself he called it no longer his own",
+ "normalized_hypothesis": "within himself he called it no longer his own",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0324",
+ "category": "phonetic_balance",
+ "text": "It lasted as a deterrent for two days.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0324.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It lasted as a deterrent for two days.",
+ "raw": {
+ "normalized_reference": "it lasted as a deterrent for two days.",
+ "normalized_hypothesis": "it lasted as a deterrent for two days.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it lasted as a deterrent for 2 days",
+ "normalized_hypothesis": "it lasted as a deterrent for 2 days",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0325",
+ "category": "phonetic_balance",
+ "text": "He loved to play Chinese lottery.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0325.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He loved to play Chinese lottery.",
+ "raw": {
+ "normalized_reference": "he loved to play chinese lottery.",
+ "normalized_hypothesis": "he loved to play chinese lottery.",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he loved to play chinese lottery",
+ "normalized_hypothesis": "he loved to play chinese lottery",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0326",
+ "category": "phonetic_balance",
+ "text": "Lord, but I'm glad to see you again, Phil.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0326.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Ward, but I'm glad to see you again, Phil.",
+ "raw": {
+ "normalized_reference": "lord, but i'm glad to see you again, phil.",
+ "normalized_hypothesis": "ward, but i'm glad to see you again, phil.",
+ "errors": 1,
+ "words": 9,
+ "wer": 0.1111111111111111
+ },
+ "semantic": {
+ "normalized_reference": "lord but i am glad to see you again phil",
+ "normalized_hypothesis": "ward but i am glad to see you again phil",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ }
+ },
+ {
+ "id": "release500_0327",
+ "category": "phonetic_balance",
+ "text": "There is not an iota of truth in it, certainly not.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0327.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "There is not an iota of truth in it, certainly not.",
+ "raw": {
+ "normalized_reference": "there is not an iota of truth in it, certainly not.",
+ "normalized_hypothesis": "there is not an iota of truth in it, certainly not.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "there is not an iota of truth in it certainly not",
+ "normalized_hypothesis": "there is not an iota of truth in it certainly not",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0328",
+ "category": "phonetic_balance",
+ "text": "She added, with genuine sympathy in her face and voice.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0328.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "She added, with genuine sympathy in her face and voice.",
+ "raw": {
+ "normalized_reference": "she added, with genuine sympathy in her face and voice.",
+ "normalized_hypothesis": "she added, with genuine sympathy in her face and voice.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "she added with genuine sympathy in her face and voice",
+ "normalized_hypothesis": "she added with genuine sympathy in her face and voice",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0329",
+ "category": "phonetic_balance",
+ "text": "We have plenty of capital ourselves, and yet we want more.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0329.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "We have plenty of capital ourselves, and yet we want more.",
+ "raw": {
+ "normalized_reference": "we have plenty of capital ourselves, and yet we want more.",
+ "normalized_hypothesis": "we have plenty of capital ourselves, and yet we want more.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "we have plenty of capital ourselves and yet we want more",
+ "normalized_hypothesis": "we have plenty of capital ourselves and yet we want more",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0330",
+ "category": "phonetic_balance",
+ "text": "And now put yourself in my place for a moment.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0330.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "And now put yourself in my place for a moment.",
+ "raw": {
+ "normalized_reference": "and now put yourself in my place for a moment.",
+ "normalized_hypothesis": "and now put yourself in my place for a moment.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "and now put yourself in my place for a moment",
+ "normalized_hypothesis": "and now put yourself in my place for a moment",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0331",
+ "category": "phonetic_balance",
+ "text": "Fast, but endure.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0331.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Fast, but Endure.",
+ "raw": {
+ "normalized_reference": "fast, but endure.",
+ "normalized_hypothesis": "fast, but endure.",
+ "errors": 0,
+ "words": 3,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "fast but endure",
+ "normalized_hypothesis": "fast but endure",
+ "errors": 0,
+ "words": 3,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0332",
+ "category": "phonetic_balance",
+ "text": "You were looking squeamish this afternoon, he began.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0332.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "You were looking squeamish this afternoon, he began.",
+ "raw": {
+ "normalized_reference": "you were looking squeamish this afternoon, he began.",
+ "normalized_hypothesis": "you were looking squeamish this afternoon, he began.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "you were looking squeamish this afternoon he began",
+ "normalized_hypothesis": "you were looking squeamish this afternoon he began",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0333",
+ "category": "phonetic_balance",
+ "text": "It is the nearest refuge.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0333.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "is the nearest refuge.",
+ "raw": {
+ "normalized_reference": "it is the nearest refuge.",
+ "normalized_hypothesis": "is the nearest refuge.",
+ "errors": 1,
+ "words": 5,
+ "wer": 0.2
+ },
+ "semantic": {
+ "normalized_reference": "it is the nearest refuge",
+ "normalized_hypothesis": "is the nearest refuge",
+ "errors": 1,
+ "words": 5,
+ "wer": 0.2
+ }
+ },
+ {
+ "id": "release500_0334",
+ "category": "phonetic_balance",
+ "text": "I want to die in it.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0334.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I want to die in it.",
+ "raw": {
+ "normalized_reference": "i want to die in it.",
+ "normalized_hypothesis": "i want to die in it.",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i want to die in it",
+ "normalized_hypothesis": "i want to die in it",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0335",
+ "category": "phonetic_balance",
+ "text": "He stopped, and Philip nodded at the horrified question in his eyes.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0335.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Key stopped, and Philip nodded at the horrified question in his eyes.",
+ "raw": {
+ "normalized_reference": "he stopped, and philip nodded at the horrified question in his eyes.",
+ "normalized_hypothesis": "key stopped, and philip nodded at the horrified question in his eyes.",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "he stopped and philip nodded at the horrified question in his eyes",
+ "normalized_hypothesis": "key stopped and philip nodded at the horrified question in his eyes",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ }
+ },
+ {
+ "id": "release500_0336",
+ "category": "phonetic_balance",
+ "text": "She turned, fearing that Jacques might see what was in her face.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0336.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "She turned, fearing that Jack might see what was in her face.",
+ "raw": {
+ "normalized_reference": "she turned, fearing that jacques might see what was in her face.",
+ "normalized_hypothesis": "she turned, fearing that jack might see what was in her face.",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "she turned fearing that jacques might see what was in her face",
+ "normalized_hypothesis": "she turned fearing that jack might see what was in her face",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ }
+ },
+ {
+ "id": "release500_0337",
+ "category": "phonetic_balance",
+ "text": "It was a curious coincidence.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0337.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It was a curious coincidence.",
+ "raw": {
+ "normalized_reference": "it was a curious coincidence.",
+ "normalized_hypothesis": "it was a curious coincidence.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it was a curious coincidence",
+ "normalized_hypothesis": "it was a curious coincidence",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0338",
+ "category": "phonetic_balance",
+ "text": "Daylight made no answer, and the door closed behind him.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0338.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Daylight made no answer, and the door closed behind him.",
+ "raw": {
+ "normalized_reference": "daylight made no answer, and the door closed behind him.",
+ "normalized_hypothesis": "daylight made no answer, and the door closed behind him.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "daylight made no answer and the door closed behind him",
+ "normalized_hypothesis": "daylight made no answer and the door closed behind him",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0339",
+ "category": "phonetic_balance",
+ "text": "And he did hurt my arm.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0339.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "And he did hurt my arm.",
+ "raw": {
+ "normalized_reference": "and he did hurt my arm.",
+ "normalized_hypothesis": "and he did hurt my arm.",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "and he did hurt my arm",
+ "normalized_hypothesis": "and he did hurt my arm",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0340",
+ "category": "phonetic_balance",
+ "text": "Dunham, can your boy go along with Jesse.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0340.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Dunham, can your boy go along with Jesse?",
+ "raw": {
+ "normalized_reference": "dunham, can your boy go along with jesse.",
+ "normalized_hypothesis": "dunham, can your boy go along with jesse?",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "dunham can your boy go along with jesse",
+ "normalized_hypothesis": "dunham can your boy go along with jesse",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0341",
+ "category": "phonetic_balance",
+ "text": "The creative joy, I murmured.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0341.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The creative joy, I murmured.",
+ "raw": {
+ "normalized_reference": "the creative joy, i murmured.",
+ "normalized_hypothesis": "the creative joy, i murmured.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the creative joy i murmured",
+ "normalized_hypothesis": "the creative joy i murmured",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0342",
+ "category": "phonetic_balance",
+ "text": "He pulled, and the log crashed down to break his back.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0342.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He pulled, and the log crashed down to break his back.",
+ "raw": {
+ "normalized_reference": "he pulled, and the log crashed down to break his back.",
+ "normalized_hypothesis": "he pulled, and the log crashed down to break his back.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he pulled and the log crashed down to break his back",
+ "normalized_hypothesis": "he pulled and the log crashed down to break his back",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0343",
+ "category": "phonetic_balance",
+ "text": "By golly, the boy wins.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0343.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Ay, Kali, the boy wins.",
+ "raw": {
+ "normalized_reference": "by golly, the boy wins.",
+ "normalized_hypothesis": "ay, kali, the boy wins.",
+ "errors": 2,
+ "words": 5,
+ "wer": 0.4
+ },
+ "semantic": {
+ "normalized_reference": "by golly the boy wins",
+ "normalized_hypothesis": "ay kali the boy wins",
+ "errors": 2,
+ "words": 5,
+ "wer": 0.4
+ }
+ },
+ {
+ "id": "release500_0344",
+ "category": "phonetic_balance",
+ "text": "The boy threw back his head with pride.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0344.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The boy threw back his head with pride.",
+ "raw": {
+ "normalized_reference": "the boy threw back his head with pride.",
+ "normalized_hypothesis": "the boy threw back his head with pride.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the boy threw back his head with pride",
+ "normalized_hypothesis": "the boy threw back his head with pride",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0345",
+ "category": "phonetic_balance",
+ "text": "The last one I knew was an overseer.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0345.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The last one I knew was an overseer.",
+ "raw": {
+ "normalized_reference": "the last one i knew was an overseer.",
+ "normalized_hypothesis": "the last one i knew was an overseer.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the last one i knew was an overseer",
+ "normalized_hypothesis": "the last one i knew was an overseer",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0346",
+ "category": "phonetic_balance",
+ "text": "Two years ago I gave up civilization for this.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0346.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Two years ago I gave up civilization for this.",
+ "raw": {
+ "normalized_reference": "two years ago i gave up civilization for this.",
+ "normalized_hypothesis": "two years ago i gave up civilization for this.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "2 years ago i gave up civilization for this",
+ "normalized_hypothesis": "2 years ago i gave up civilization for this",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0347",
+ "category": "phonetic_balance",
+ "text": "The river bared its bosom, and snorting steamboats challenged the wilderness.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0347.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The river bared its bosom, and snorting steamboats challenged the wilderness.",
+ "raw": {
+ "normalized_reference": "the river bared its bosom, and snorting steamboats challenged the wilderness.",
+ "normalized_hypothesis": "the river bared its bosom, and snorting steamboats challenged the wilderness.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the river bared its bosom and snorting steamboats challenged the wilderness",
+ "normalized_hypothesis": "the river bared its bosom and snorting steamboats challenged the wilderness",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0348",
+ "category": "phonetic_balance",
+ "text": "Mab, she said.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0348.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Mab, she said.",
+ "raw": {
+ "normalized_reference": "mab, she said.",
+ "normalized_hypothesis": "mab, she said.",
+ "errors": 0,
+ "words": 3,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "mab she said",
+ "normalized_hypothesis": "mab she said",
+ "errors": 0,
+ "words": 3,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0349",
+ "category": "phonetic_balance",
+ "text": "The regret in his voice was provocative of a second burst of laughter.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0349.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The regret in his voice was provocative of a second burst of laughter.",
+ "raw": {
+ "normalized_reference": "the regret in his voice was provocative of a second burst of laughter.",
+ "normalized_hypothesis": "the regret in his voice was provocative of a second burst of laughter.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the regret in his voice was provocative of a 2nd burst of laughter",
+ "normalized_hypothesis": "the regret in his voice was provocative of a 2nd burst of laughter",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0350",
+ "category": "phonetic_balance",
+ "text": "And each year something happened, and I did not go.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0350.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "And each year something happened, and I did not go.",
+ "raw": {
+ "normalized_reference": "and each year something happened, and i did not go.",
+ "normalized_hypothesis": "and each year something happened, and i did not go.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "and each year something happened and i did not go",
+ "normalized_hypothesis": "and each year something happened and i did not go",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0351",
+ "category": "phonetic_balance",
+ "text": "Gad, your letter came just in time.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0351.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Gad, your letter came just in time.",
+ "raw": {
+ "normalized_reference": "gad, your letter came just in time.",
+ "normalized_hypothesis": "gad, your letter came just in time.",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "gad your letter came just in time",
+ "normalized_hypothesis": "gad your letter came just in time",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0352",
+ "category": "phonetic_balance",
+ "text": "But if Pierre did not return, until tomorrow.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0352.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "But if Pierre did not return until tomorrow,",
+ "raw": {
+ "normalized_reference": "but if pierre did not return, until tomorrow.",
+ "normalized_hypothesis": "but if pierre did not return until tomorrow,",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "but if pierre did not return until tomorrow",
+ "normalized_hypothesis": "but if pierre did not return until tomorrow",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0353",
+ "category": "phonetic_balance",
+ "text": "Billinger may arrive in time.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0353.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Billinger may arrive in time.",
+ "raw": {
+ "normalized_reference": "billinger may arrive in time.",
+ "normalized_hypothesis": "billinger may arrive in time.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "billinger may arrive in time",
+ "normalized_hypothesis": "billinger may arrive in time",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0354",
+ "category": "phonetic_balance",
+ "text": "In such a tumbling of values was no time to sell.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0354.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "And such a tumbling of values was no time to sell.",
+ "raw": {
+ "normalized_reference": "in such a tumbling of values was no time to sell.",
+ "normalized_hypothesis": "and such a tumbling of values was no time to sell.",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ },
+ "semantic": {
+ "normalized_reference": "in such a tumbling of values was no time to sell",
+ "normalized_hypothesis": "and such a tumbling of values was no time to sell",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ }
+ },
+ {
+ "id": "release500_0355",
+ "category": "phonetic_balance",
+ "text": "There's not an iota of truth in it.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0355.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "There's not an iota of truth in it.",
+ "raw": {
+ "normalized_reference": "there's not an iota of truth in it.",
+ "normalized_hypothesis": "there's not an iota of truth in it.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "there is not an iota of truth in it",
+ "normalized_hypothesis": "there is not an iota of truth in it",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0356",
+ "category": "phonetic_balance",
+ "text": "It is dog eat dog, and you ate them up.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0356.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "This dog eat dog, and you ate them up.",
+ "raw": {
+ "normalized_reference": "it is dog eat dog, and you ate them up.",
+ "normalized_hypothesis": "this dog eat dog, and you ate them up.",
+ "errors": 2,
+ "words": 10,
+ "wer": 0.2
+ },
+ "semantic": {
+ "normalized_reference": "it is dog eat dog and you ate them up",
+ "normalized_hypothesis": "this dog eat dog and you ate them up",
+ "errors": 2,
+ "words": 10,
+ "wer": 0.2
+ }
+ },
+ {
+ "id": "release500_0357",
+ "category": "phonetic_balance",
+ "text": "You can take a vacation on pay.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0357.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "You can take a vacation on pay.",
+ "raw": {
+ "normalized_reference": "you can take a vacation on pay.",
+ "normalized_hypothesis": "you can take a vacation on pay.",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "you can take a vacation on pay",
+ "normalized_hypothesis": "you can take a vacation on pay",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0358",
+ "category": "phonetic_balance",
+ "text": "I have seen myself that one man contemplated by Pascal's philosophic eye.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0358.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I have seen myself that one man contemplated by Pascal's philosophic eye.",
+ "raw": {
+ "normalized_reference": "i have seen myself that one man contemplated by pascal's philosophic eye.",
+ "normalized_hypothesis": "i have seen myself that one man contemplated by pascal's philosophic eye.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i have seen myself that one man contemplated by pascal is philosophic eye",
+ "normalized_hypothesis": "i have seen myself that one man contemplated by pascal is philosophic eye",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0359",
+ "category": "phonetic_balance",
+ "text": "Ah, I had forgotten, he exclaimed.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0359.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Ah, I had forgotten, he explained.",
+ "raw": {
+ "normalized_reference": "ah, i had forgotten, he exclaimed.",
+ "normalized_hypothesis": "ah, i had forgotten, he explained.",
+ "errors": 1,
+ "words": 6,
+ "wer": 0.16666666666666666
+ },
+ "semantic": {
+ "normalized_reference": "ah i had forgotten he exclaimed",
+ "normalized_hypothesis": "ah i had forgotten he explained",
+ "errors": 1,
+ "words": 6,
+ "wer": 0.16666666666666666
+ }
+ },
+ {
+ "id": "release500_0360",
+ "category": "phonetic_balance",
+ "text": "Providence had delivered him through the maelstrom.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0360.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Providence had delivered him through the maelstrom.",
+ "raw": {
+ "normalized_reference": "providence had delivered him through the maelstrom.",
+ "normalized_hypothesis": "providence had delivered him through the maelstrom.",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "providence had delivered him through the maelstrom",
+ "normalized_hypothesis": "providence had delivered him through the maelstrom",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0361",
+ "category": "phonetic_balance",
+ "text": "She was his now, forever.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0361.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "She was his now, forever.",
+ "raw": {
+ "normalized_reference": "she was his now, forever.",
+ "normalized_hypothesis": "she was his now, forever.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "she was his now forever",
+ "normalized_hypothesis": "she was his now forever",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0362",
+ "category": "phonetic_balance",
+ "text": "It was unobtrusive, yet it was there.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0362.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It was unobtrusive, yet it was there.",
+ "raw": {
+ "normalized_reference": "it was unobtrusive, yet it was there.",
+ "normalized_hypothesis": "it was unobtrusive, yet it was there.",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it was unobtrusive yet it was there",
+ "normalized_hypothesis": "it was unobtrusive yet it was there",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0363",
+ "category": "phonetic_balance",
+ "text": "Men like Joe Goose dated existence from drunk to drunk.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0363.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Men like Joe Goose dated existence from drunk to drunk.",
+ "raw": {
+ "normalized_reference": "men like joe goose dated existence from drunk to drunk.",
+ "normalized_hypothesis": "men like joe goose dated existence from drunk to drunk.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "men like joe goose dated existence from drunk to drunk",
+ "normalized_hypothesis": "men like joe goose dated existence from drunk to drunk",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0364",
+ "category": "phonetic_balance",
+ "text": "I was completely lost in my work.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0364.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I was completely lost in my work.",
+ "raw": {
+ "normalized_reference": "i was completely lost in my work.",
+ "normalized_hypothesis": "i was completely lost in my work.",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i was completely lost in my work",
+ "normalized_hypothesis": "i was completely lost in my work",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0365",
+ "category": "phonetic_balance",
+ "text": "The very thing, Ernest agreed.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0365.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The very thing, Ernest agreed.",
+ "raw": {
+ "normalized_reference": "the very thing, ernest agreed.",
+ "normalized_hypothesis": "the very thing, ernest agreed.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the very thing ernest agreed",
+ "normalized_hypothesis": "the very thing ernest agreed",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0366",
+ "category": "phonetic_balance",
+ "text": "He was the leader, and Tudor was his lieutenant.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0366.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He was the leader, and Tudor was his lieutenant.",
+ "raw": {
+ "normalized_reference": "he was the leader, and tudor was his lieutenant.",
+ "normalized_hypothesis": "he was the leader, and tudor was his lieutenant.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he was the leader and tudor was his lieutenant",
+ "normalized_hypothesis": "he was the leader and tudor was his lieutenant",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0367",
+ "category": "phonetic_balance",
+ "text": "The Fire-Men wore animal skins around their waists and across their shoulders.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0367.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The firemen wore animal skins around their waists and across their shoulders.",
+ "raw": {
+ "normalized_reference": "the fire-men wore animal skins around their waists and across their shoulders.",
+ "normalized_hypothesis": "the firemen wore animal skins around their waists and across their shoulders.",
+ "errors": 2,
+ "words": 13,
+ "wer": 0.15384615384615385
+ },
+ "semantic": {
+ "normalized_reference": "the fire men wore animal skins around their waists and across their shoulders",
+ "normalized_hypothesis": "the firemen wore animal skins around their waists and across their shoulders",
+ "errors": 2,
+ "words": 13,
+ "wer": 0.15384615384615385
+ }
+ },
+ {
+ "id": "release500_0368",
+ "category": "phonetic_balance",
+ "text": "We could throw stones with our feet.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0368.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "We could throw stones with our feet.",
+ "raw": {
+ "normalized_reference": "we could throw stones with our feet.",
+ "normalized_hypothesis": "we could throw stones with our feet.",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "we could throw stones with our feet",
+ "normalized_hypothesis": "we could throw stones with our feet",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0369",
+ "category": "phonetic_balance",
+ "text": "He turned sharply, and faced Gregson across the table.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0369.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He turned sharply and faced Gregson across the table.",
+ "raw": {
+ "normalized_reference": "he turned sharply, and faced gregson across the table.",
+ "normalized_hypothesis": "he turned sharply and faced gregson across the table.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he turned sharply and faced gregson across the table",
+ "normalized_hypothesis": "he turned sharply and faced gregson across the table",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0370",
+ "category": "phonetic_balance",
+ "text": "They are to attack your camp tomorrow night.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0370.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "They are to attack your camp tomorrow night.",
+ "raw": {
+ "normalized_reference": "they are to attack your camp tomorrow night.",
+ "normalized_hypothesis": "they are to attack your camp tomorrow night.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "they are to attack your camp tomorrow night",
+ "normalized_hypothesis": "they are to attack your camp tomorrow night",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0371",
+ "category": "phonetic_balance",
+ "text": "And watch out for wet feet, was his parting advice.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0371.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "And watch out for what feet was his parting advice.",
+ "raw": {
+ "normalized_reference": "and watch out for wet feet, was his parting advice.",
+ "normalized_hypothesis": "and watch out for what feet was his parting advice.",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ },
+ "semantic": {
+ "normalized_reference": "and watch out for wet feet was his parting advice",
+ "normalized_hypothesis": "and watch out for what feet was his parting advice",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ }
+ },
+ {
+ "id": "release500_0372",
+ "category": "phonetic_balance",
+ "text": "Two weeks passed, and in that time Thorpe left camp three times.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0372.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Two weeks passed, and in that time Thorpe left camp three times.",
+ "raw": {
+ "normalized_reference": "two weeks passed, and in that time thorpe left camp three times.",
+ "normalized_hypothesis": "two weeks passed, and in that time thorpe left camp three times.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "2 weeks passed and in that time thorpe left camp 3 times",
+ "normalized_hypothesis": "2 weeks passed and in that time thorpe left camp 3 times",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0373",
+ "category": "phonetic_balance",
+ "text": "But ever his gaze returned to that Crouched Venus on the piano.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0373.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "But ever his gaze returned to that crouched Venus on the piano.",
+ "raw": {
+ "normalized_reference": "but ever his gaze returned to that crouched venus on the piano.",
+ "normalized_hypothesis": "but ever his gaze returned to that crouched venus on the piano.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "but ever his gaze returned to that crouched venus on the piano",
+ "normalized_hypothesis": "but ever his gaze returned to that crouched venus on the piano",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0374",
+ "category": "phonetic_balance",
+ "text": "But she had become an automaton.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0374.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "But she had become an automaton.",
+ "raw": {
+ "normalized_reference": "but she had become an automaton.",
+ "normalized_hypothesis": "but she had become an automaton.",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "but she had become an automaton",
+ "normalized_hypothesis": "but she had become an automaton",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0375",
+ "category": "phonetic_balance",
+ "text": "I do not blame you for anything; remember that.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0375.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I do not blame you for anything. Remember that.",
+ "raw": {
+ "normalized_reference": "i do not blame you for anything; remember that.",
+ "normalized_hypothesis": "i do not blame you for anything. remember that.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i do not blame you for anything remember that",
+ "normalized_hypothesis": "i do not blame you for anything remember that",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0376",
+ "category": "phonetic_balance",
+ "text": "The next thing to watch out for is bed sores.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0376.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The next thing to watch out for is bedsores.",
+ "raw": {
+ "normalized_reference": "the next thing to watch out for is bed sores.",
+ "normalized_hypothesis": "the next thing to watch out for is bedsores.",
+ "errors": 2,
+ "words": 10,
+ "wer": 0.2
+ },
+ "semantic": {
+ "normalized_reference": "the next thing to watch out for is bed sores",
+ "normalized_hypothesis": "the next thing to watch out for is bedsores",
+ "errors": 2,
+ "words": 10,
+ "wer": 0.2
+ }
+ },
+ {
+ "id": "release500_0377",
+ "category": "phonetic_balance",
+ "text": "He will follow us soon.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0377.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He will follow us soon.",
+ "raw": {
+ "normalized_reference": "he will follow us soon.",
+ "normalized_hypothesis": "he will follow us soon.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he will follow us soon",
+ "normalized_hypothesis": "he will follow us soon",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0378",
+ "category": "phonetic_balance",
+ "text": "His beady black eyes saw bargains where other men saw bankruptcy.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0378.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "His beady black eyes saw bargains where other men saw bankruptcy.",
+ "raw": {
+ "normalized_reference": "his beady black eyes saw bargains where other men saw bankruptcy.",
+ "normalized_hypothesis": "his beady black eyes saw bargains where other men saw bankruptcy.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "his beady black eyes saw bargains where other men saw bankruptcy",
+ "normalized_hypothesis": "his beady black eyes saw bargains where other men saw bankruptcy",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0379",
+ "category": "phonetic_balance",
+ "text": "They robbed me a few years later.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0379.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "They robbed me a few years later.",
+ "raw": {
+ "normalized_reference": "they robbed me a few years later.",
+ "normalized_hypothesis": "they robbed me a few years later.",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "they robbed me a few years later",
+ "normalized_hypothesis": "they robbed me a few years later",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0380",
+ "category": "phonetic_balance",
+ "text": "Gregson held a lighted match until it burnt his fingertips.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0380.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Gregson held a lighted match until it burned his fingertips.",
+ "raw": {
+ "normalized_reference": "gregson held a lighted match until it burnt his fingertips.",
+ "normalized_hypothesis": "gregson held a lighted match until it burned his fingertips.",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ },
+ "semantic": {
+ "normalized_reference": "gregson held a lighted match until it burnt his fingertips",
+ "normalized_hypothesis": "gregson held a lighted match until it burned his fingertips",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ }
+ },
+ {
+ "id": "release500_0381",
+ "category": "phonetic_balance",
+ "text": "A burst of laughter was his reward.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0381.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The burst of laughter was his reward.",
+ "raw": {
+ "normalized_reference": "a burst of laughter was his reward.",
+ "normalized_hypothesis": "the burst of laughter was his reward.",
+ "errors": 1,
+ "words": 7,
+ "wer": 0.14285714285714285
+ },
+ "semantic": {
+ "normalized_reference": "a burst of laughter was his reward",
+ "normalized_hypothesis": "the burst of laughter was his reward",
+ "errors": 1,
+ "words": 7,
+ "wer": 0.14285714285714285
+ }
+ },
+ {
+ "id": "release500_0382",
+ "category": "phonetic_balance",
+ "text": "He cried, and swung the club wildly.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0382.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He cried and swung the club wildly.",
+ "raw": {
+ "normalized_reference": "he cried, and swung the club wildly.",
+ "normalized_hypothesis": "he cried and swung the club wildly.",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he cried and swung the club wildly",
+ "normalized_hypothesis": "he cried and swung the club wildly",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0383",
+ "category": "phonetic_balance",
+ "text": "Sandel would never become a world champion.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0383.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Sandal would never become a world champion.",
+ "raw": {
+ "normalized_reference": "sandel would never become a world champion.",
+ "normalized_hypothesis": "sandal would never become a world champion.",
+ "errors": 1,
+ "words": 7,
+ "wer": 0.14285714285714285
+ },
+ "semantic": {
+ "normalized_reference": "sandel would never become a world champion",
+ "normalized_hypothesis": "sandal would never become a world champion",
+ "errors": 1,
+ "words": 7,
+ "wer": 0.14285714285714285
+ }
+ },
+ {
+ "id": "release500_0384",
+ "category": "phonetic_balance",
+ "text": "And after the bath a shave would not be bad.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0384.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "And after the bath a shave would not be bad.",
+ "raw": {
+ "normalized_reference": "and after the bath a shave would not be bad.",
+ "normalized_hypothesis": "and after the bath a shave would not be bad.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "and after the bath a shave would not be bad",
+ "normalized_hypothesis": "and after the bath a shave would not be bad",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0385",
+ "category": "phonetic_balance",
+ "text": "They saw each other for the first time in Boston.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0385.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "They saw each other for the first time in Boston.",
+ "raw": {
+ "normalized_reference": "they saw each other for the first time in boston.",
+ "normalized_hypothesis": "they saw each other for the first time in boston.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "they saw each other for the 1st time in boston",
+ "normalized_hypothesis": "they saw each other for the 1st time in boston",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0386",
+ "category": "phonetic_balance",
+ "text": "I had been born with no organic, chemical predisposition toward alcohol.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0386.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I had been born with no organic, chemical predisposition toward alcohol.",
+ "raw": {
+ "normalized_reference": "i had been born with no organic, chemical predisposition toward alcohol.",
+ "normalized_hypothesis": "i had been born with no organic, chemical predisposition toward alcohol.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i had been born with no organic chemical predisposition toward alcohol",
+ "normalized_hypothesis": "i had been born with no organic chemical predisposition toward alcohol",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0387",
+ "category": "phonetic_balance",
+ "text": "They had no fixed values, to be altered by adjectives and adverbs.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0387.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "They had no fixed values to be altered by adjectives and adverbs.",
+ "raw": {
+ "normalized_reference": "they had no fixed values, to be altered by adjectives and adverbs.",
+ "normalized_hypothesis": "they had no fixed values to be altered by adjectives and adverbs.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "they had no fixed values to be altered by adjectives and adverbs",
+ "normalized_hypothesis": "they had no fixed values to be altered by adjectives and adverbs",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0388",
+ "category": "phonetic_balance",
+ "text": "Nor was Elam Harnish an exception.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0388.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Nor was the lamb Harnish an exception.",
+ "raw": {
+ "normalized_reference": "nor was elam harnish an exception.",
+ "normalized_hypothesis": "nor was the lamb harnish an exception.",
+ "errors": 2,
+ "words": 6,
+ "wer": 0.3333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "nor was elam harnish an exception",
+ "normalized_hypothesis": "nor was the lamb harnish an exception",
+ "errors": 2,
+ "words": 6,
+ "wer": 0.3333333333333333
+ }
+ },
+ {
+ "id": "release500_0389",
+ "category": "phonetic_balance",
+ "text": "He was a wise hyena.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0389.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He was a wise hyena.",
+ "raw": {
+ "normalized_reference": "he was a wise hyena.",
+ "normalized_hypothesis": "he was a wise hyena.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he was a wise hyena",
+ "normalized_hypothesis": "he was a wise hyena",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0390",
+ "category": "phonetic_balance",
+ "text": "And now behold the perversity of things.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0390.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "And now behold the perversity of things.",
+ "raw": {
+ "normalized_reference": "and now behold the perversity of things.",
+ "normalized_hypothesis": "and now behold the perversity of things.",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "and now behold the perversity of things",
+ "normalized_hypothesis": "and now behold the perversity of things",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0391",
+ "category": "phonetic_balance",
+ "text": "The Fire People, like ourselves, lived in caves.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0391.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The fire people, like ourselves, lived in caves.",
+ "raw": {
+ "normalized_reference": "the fire people, like ourselves, lived in caves.",
+ "normalized_hypothesis": "the fire people, like ourselves, lived in caves.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the fire people like ourselves lived in caves",
+ "normalized_hypothesis": "the fire people like ourselves lived in caves",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0392",
+ "category": "phonetic_balance",
+ "text": "To her the bridge was tambo, which is the native for taboo.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0392.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "To her the bridge was Tambo, which is the native for Taboo.",
+ "raw": {
+ "normalized_reference": "to her the bridge was tambo, which is the native for taboo.",
+ "normalized_hypothesis": "to her the bridge was tambo, which is the native for taboo.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "to her the bridge was tambo which is the native for taboo",
+ "normalized_hypothesis": "to her the bridge was tambo which is the native for taboo",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0393",
+ "category": "phonetic_balance",
+ "text": "The voyage was our idea of a good time.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0393.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The voyage was our idea of a good time.",
+ "raw": {
+ "normalized_reference": "the voyage was our idea of a good time.",
+ "normalized_hypothesis": "the voyage was our idea of a good time.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the voyage was our idea of a good time",
+ "normalized_hypothesis": "the voyage was our idea of a good time",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0394",
+ "category": "phonetic_balance",
+ "text": "Of course much grumbling went on, and little outbursts were continually occurring.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0394.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Of course much grumbling went on, and little outbursts were continually occurring.",
+ "raw": {
+ "normalized_reference": "of course much grumbling went on, and little outbursts were continually occurring.",
+ "normalized_hypothesis": "of course much grumbling went on, and little outbursts were continually occurring.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "of course much grumbling went on and little outbursts were continually occurring",
+ "normalized_hypothesis": "of course much grumbling went on and little outbursts were continually occurring",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0395",
+ "category": "phonetic_balance",
+ "text": "Get down and dig in.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0395.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Get down and dig in.",
+ "raw": {
+ "normalized_reference": "get down and dig in.",
+ "normalized_hypothesis": "get down and dig in.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "get down and dig in",
+ "normalized_hypothesis": "get down and dig in",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0396",
+ "category": "phonetic_balance",
+ "text": "It was put together in a casual, helter-skelter sort of way.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0396.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It was put together in a casual, kelter-skelter sort of way.",
+ "raw": {
+ "normalized_reference": "it was put together in a casual, helter-skelter sort of way.",
+ "normalized_hypothesis": "it was put together in a casual, kelter-skelter sort of way.",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "it was put together in a casual helter skelter sort of way",
+ "normalized_hypothesis": "it was put together in a casual kelter skelter sort of way",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ }
+ },
+ {
+ "id": "release500_0397",
+ "category": "phonetic_balance",
+ "text": "He deluged me, overwhelmed me with argument.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0397.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He deluged me, overwhelmed me with argument.",
+ "raw": {
+ "normalized_reference": "he deluged me, overwhelmed me with argument.",
+ "normalized_hypothesis": "he deluged me, overwhelmed me with argument.",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he deluged me overwhelmed me with argument",
+ "normalized_hypothesis": "he deluged me overwhelmed me with argument",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0398",
+ "category": "phonetic_balance",
+ "text": "The girl faced him, her eyes shining with sudden fear.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0398.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The girl faced him, her eyes shining with sudden fear.",
+ "raw": {
+ "normalized_reference": "the girl faced him, her eyes shining with sudden fear.",
+ "normalized_hypothesis": "the girl faced him, her eyes shining with sudden fear.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the girl faced him her eyes shining with sudden fear",
+ "normalized_hypothesis": "the girl faced him her eyes shining with sudden fear",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0399",
+ "category": "phonetic_balance",
+ "text": "It will break our hearts and our backs to hoist anchor by hand.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0399.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It will break our hearts and our backs to hoist anchor by hand.",
+ "raw": {
+ "normalized_reference": "it will break our hearts and our backs to hoist anchor by hand.",
+ "normalized_hypothesis": "it will break our hearts and our backs to hoist anchor by hand.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it will break our hearts and our backs to hoist anchor by hand",
+ "normalized_hypothesis": "it will break our hearts and our backs to hoist anchor by hand",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0400",
+ "category": "phonetic_balance",
+ "text": "He made no reply as he waited for Whittemore to continue.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0400.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He made no reply as he waited for Whittemore to continue.",
+ "raw": {
+ "normalized_reference": "he made no reply as he waited for whittemore to continue.",
+ "normalized_hypothesis": "he made no reply as he waited for whittemore to continue.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he made no reply as he waited for whittemore to continue",
+ "normalized_hypothesis": "he made no reply as he waited for whittemore to continue",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0401",
+ "category": "phonetic_balance",
+ "text": "Who the devil gave it to you to be judge and jury.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0401.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Who the devil gave it to you to be judge and jury?",
+ "raw": {
+ "normalized_reference": "who the devil gave it to you to be judge and jury.",
+ "normalized_hypothesis": "who the devil gave it to you to be judge and jury?",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "who the devil gave it to you to be judge and jury",
+ "normalized_hypothesis": "who the devil gave it to you to be judge and jury",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0402",
+ "category": "phonetic_balance",
+ "text": "Daylight found himself charmed and made curious by the little man.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0402.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Daylight found himself charmed and made curious by the little man.",
+ "raw": {
+ "normalized_reference": "daylight found himself charmed and made curious by the little man.",
+ "normalized_hypothesis": "daylight found himself charmed and made curious by the little man.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "daylight found himself charmed and made curious by the little man",
+ "normalized_hypothesis": "daylight found himself charmed and made curious by the little man",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0403",
+ "category": "phonetic_balance",
+ "text": "It's a Yankee, Joan cried.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0403.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It's a Yankee, Joan cried.",
+ "raw": {
+ "normalized_reference": "it's a yankee, joan cried.",
+ "normalized_hypothesis": "it's a yankee, joan cried.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it is a yankee joan cried",
+ "normalized_hypothesis": "it is a yankee joan cried",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0404",
+ "category": "phonetic_balance",
+ "text": "But what they want with your toothbrush is more than I can imagine.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0404.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "But what they want with you too fresh is more than I can imagine.",
+ "raw": {
+ "normalized_reference": "but what they want with your toothbrush is more than i can imagine.",
+ "normalized_hypothesis": "but what they want with you too fresh is more than i can imagine.",
+ "errors": 3,
+ "words": 13,
+ "wer": 0.23076923076923078
+ },
+ "semantic": {
+ "normalized_reference": "but what they want with your toothbrush is more than i can imagine",
+ "normalized_hypothesis": "but what they want with you too fresh is more than i can imagine",
+ "errors": 3,
+ "words": 13,
+ "wer": 0.23076923076923078
+ }
+ },
+ {
+ "id": "release500_0405",
+ "category": "phonetic_balance",
+ "text": "They were babbling and chattering all together.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0405.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "They were babbling and chattering all together.",
+ "raw": {
+ "normalized_reference": "they were babbling and chattering all together.",
+ "normalized_hypothesis": "they were babbling and chattering all together.",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "they were babbling and chattering all together",
+ "normalized_hypothesis": "they were babbling and chattering all together",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0406",
+ "category": "phonetic_balance",
+ "text": "Men of Selden's stamp don't stop at women and children.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0406.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Men of Selden's stamp don't stop at women and children.",
+ "raw": {
+ "normalized_reference": "men of selden's stamp don't stop at women and children.",
+ "normalized_hypothesis": "men of selden's stamp don't stop at women and children.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "men of selden is stamp do not stop at women and children",
+ "normalized_hypothesis": "men of selden is stamp do not stop at women and children",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0407",
+ "category": "phonetic_balance",
+ "text": "Cameron looked at his hands with their long, sinewy fingers.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0407.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Cameron looked at his hands with their long, sinewy fingers.",
+ "raw": {
+ "normalized_reference": "cameron looked at his hands with their long, sinewy fingers.",
+ "normalized_hypothesis": "cameron looked at his hands with their long, sinewy fingers.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "cameron looked at his hands with their long sinewy fingers",
+ "normalized_hypothesis": "cameron looked at his hands with their long sinewy fingers",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0408",
+ "category": "phonetic_balance",
+ "text": "Saxon waited, for she knew a fresh idea had struck Billy.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0408.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Saxon waited, for she knew a fresh idea had struck Billy.",
+ "raw": {
+ "normalized_reference": "saxon waited, for she knew a fresh idea had struck billy.",
+ "normalized_hypothesis": "saxon waited, for she knew a fresh idea had struck billy.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "saxon waited for she knew a fresh idea had struck billy",
+ "normalized_hypothesis": "saxon waited for she knew a fresh idea had struck billy",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0409",
+ "category": "phonetic_balance",
+ "text": "He saw only the effect in a general, sketchy way.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0409.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Keyes saw only the effect in a general, sketchy way.",
+ "raw": {
+ "normalized_reference": "he saw only the effect in a general, sketchy way.",
+ "normalized_hypothesis": "keyes saw only the effect in a general, sketchy way.",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ },
+ "semantic": {
+ "normalized_reference": "he saw only the effect in a general sketchy way",
+ "normalized_hypothesis": "keyes saw only the effect in a general sketchy way",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ }
+ },
+ {
+ "id": "release500_0410",
+ "category": "phonetic_balance",
+ "text": "He began to follow the footprints of the dog.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0410.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He began to follow the footprints of the dog.",
+ "raw": {
+ "normalized_reference": "he began to follow the footprints of the dog.",
+ "normalized_hypothesis": "he began to follow the footprints of the dog.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he began to follow the footprints of the dog",
+ "normalized_hypothesis": "he began to follow the footprints of the dog",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0411",
+ "category": "phonetic_balance",
+ "text": "Do you value your hide.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0411.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Do you value your hide?",
+ "raw": {
+ "normalized_reference": "do you value your hide.",
+ "normalized_hypothesis": "do you value your hide?",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "do you value your hide",
+ "normalized_hypothesis": "do you value your hide",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0412",
+ "category": "phonetic_balance",
+ "text": "A rising tide of fat had submerged them.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0412.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The rising tide of fat had submerged them.",
+ "raw": {
+ "normalized_reference": "a rising tide of fat had submerged them.",
+ "normalized_hypothesis": "the rising tide of fat had submerged them.",
+ "errors": 1,
+ "words": 8,
+ "wer": 0.125
+ },
+ "semantic": {
+ "normalized_reference": "a rising tide of fat had submerged them",
+ "normalized_hypothesis": "the rising tide of fat had submerged them",
+ "errors": 1,
+ "words": 8,
+ "wer": 0.125
+ }
+ },
+ {
+ "id": "release500_0413",
+ "category": "phonetic_balance",
+ "text": "There was nothing on the rock.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0413.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "There was nothing on the rock.",
+ "raw": {
+ "normalized_reference": "there was nothing on the rock.",
+ "normalized_hypothesis": "there was nothing on the rock.",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "there was nothing on the rock",
+ "normalized_hypothesis": "there was nothing on the rock",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0414",
+ "category": "phonetic_balance",
+ "text": "Whatever he guessed he locked away in the taboo room of Naomi.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0414.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Whatever he guessed he locked away in the taboo room of Naomi.",
+ "raw": {
+ "normalized_reference": "whatever he guessed he locked away in the taboo room of naomi.",
+ "normalized_hypothesis": "whatever he guessed he locked away in the taboo room of naomi.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "whatever he guessed he locked away in the taboo room of naomi",
+ "normalized_hypothesis": "whatever he guessed he locked away in the taboo room of naomi",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0415",
+ "category": "phonetic_balance",
+ "text": "Will we ever forget it.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0415.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Will we ever forget it?",
+ "raw": {
+ "normalized_reference": "will we ever forget it.",
+ "normalized_hypothesis": "will we ever forget it?",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "will we ever forget it",
+ "normalized_hypothesis": "will we ever forget it",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0416",
+ "category": "phonetic_balance",
+ "text": "Jeanne and Pierre both gazed toward the great rock.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0416.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Jean and Pierre both gazed toward the great rock.",
+ "raw": {
+ "normalized_reference": "jeanne and pierre both gazed toward the great rock.",
+ "normalized_hypothesis": "jean and pierre both gazed toward the great rock.",
+ "errors": 1,
+ "words": 9,
+ "wer": 0.1111111111111111
+ },
+ "semantic": {
+ "normalized_reference": "jeanne and pierre both gazed toward the great rock",
+ "normalized_hypothesis": "jean and pierre both gazed toward the great rock",
+ "errors": 1,
+ "words": 9,
+ "wer": 0.1111111111111111
+ }
+ },
+ {
+ "id": "release500_0417",
+ "category": "phonetic_balance",
+ "text": "Does the old boy often go off at half-cock that way.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0417.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Does the old boy often go off at half-cock that way?",
+ "raw": {
+ "normalized_reference": "does the old boy often go off at half-cock that way.",
+ "normalized_hypothesis": "does the old boy often go off at half-cock that way?",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "does the old boy often go off at half cock that way",
+ "normalized_hypothesis": "does the old boy often go off at half cock that way",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0418",
+ "category": "phonetic_balance",
+ "text": "The temperature dropped to fifty below zero and remained there the whole trip.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0418.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The temperature dropped to 50 below zero and remained there the whole trip.",
+ "raw": {
+ "normalized_reference": "the temperature dropped to fifty below zero and remained there the whole trip.",
+ "normalized_hypothesis": "the temperature dropped to 50 below zero and remained there the whole trip.",
+ "errors": 1,
+ "words": 13,
+ "wer": 0.07692307692307693
+ },
+ "semantic": {
+ "normalized_reference": "the temperature dropped to 50 below 0 and remained there the whole trip",
+ "normalized_hypothesis": "the temperature dropped to 50 below 0 and remained there the whole trip",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0419",
+ "category": "phonetic_balance",
+ "text": "It's worth eight dollars.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0419.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It's worth $8.",
+ "raw": {
+ "normalized_reference": "it's worth eight dollars.",
+ "normalized_hypothesis": "it's worth $8.",
+ "errors": 2,
+ "words": 4,
+ "wer": 0.5
+ },
+ "semantic": {
+ "normalized_reference": "it is worth $8",
+ "normalized_hypothesis": "it is worth $8",
+ "errors": 0,
+ "words": 4,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0420",
+ "category": "phonetic_balance",
+ "text": "But I did not enjoy it long.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0420.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "But I did not enjoy it long.",
+ "raw": {
+ "normalized_reference": "but i did not enjoy it long.",
+ "normalized_hypothesis": "but i did not enjoy it long.",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "but i did not enjoy it long",
+ "normalized_hypothesis": "but i did not enjoy it long",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0421",
+ "category": "phonetic_balance",
+ "text": "Shorty turned to their employers.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0421.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Shorty turned to their employers.",
+ "raw": {
+ "normalized_reference": "shorty turned to their employers.",
+ "normalized_hypothesis": "shorty turned to their employers.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "shorty turned to their employers",
+ "normalized_hypothesis": "shorty turned to their employers",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0422",
+ "category": "phonetic_balance",
+ "text": "It was the third or fourth time that Philip had heard MacDougall swear.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0422.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It was the third or fourth time that Philip had heard Mac Duvel swear.",
+ "raw": {
+ "normalized_reference": "it was the third or fourth time that philip had heard macdougall swear.",
+ "normalized_hypothesis": "it was the third or fourth time that philip had heard mac duvel swear.",
+ "errors": 2,
+ "words": 13,
+ "wer": 0.15384615384615385
+ },
+ "semantic": {
+ "normalized_reference": "it was the 3rd or 4th time that philip had heard macdougall swear",
+ "normalized_hypothesis": "it was the 3rd or 4th time that philip had heard mac duvel swear",
+ "errors": 2,
+ "words": 13,
+ "wer": 0.15384615384615385
+ }
+ },
+ {
+ "id": "release500_0423",
+ "category": "phonetic_balance",
+ "text": "Ernest saw in the affair the most sinister import.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0423.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Ernest saw in the affair the most sinister import.",
+ "raw": {
+ "normalized_reference": "ernest saw in the affair the most sinister import.",
+ "normalized_hypothesis": "ernest saw in the affair the most sinister import.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "ernest saw in the affair the most sinister import",
+ "normalized_hypothesis": "ernest saw in the affair the most sinister import",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0424",
+ "category": "phonetic_balance",
+ "text": "Also, at regular intervals, he would mutter.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0424.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Also, at regular intervals, he would mutter.",
+ "raw": {
+ "normalized_reference": "also, at regular intervals, he would mutter.",
+ "normalized_hypothesis": "also, at regular intervals, he would mutter.",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "also at regular intervals he would mutter",
+ "normalized_hypothesis": "also at regular intervals he would mutter",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0425",
+ "category": "phonetic_balance",
+ "text": "He may anticipate the day of his death.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0425.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He may anticipate the day of his death.",
+ "raw": {
+ "normalized_reference": "he may anticipate the day of his death.",
+ "normalized_hypothesis": "he may anticipate the day of his death.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he may anticipate the day of his death",
+ "normalized_hypothesis": "he may anticipate the day of his death",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0426",
+ "category": "phonetic_balance",
+ "text": "I defy any man to get a Solomon Island sore in California.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0426.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I defy any man to get a Solomon Islands or in California.",
+ "raw": {
+ "normalized_reference": "i defy any man to get a solomon island sore in california.",
+ "normalized_hypothesis": "i defy any man to get a solomon islands or in california.",
+ "errors": 2,
+ "words": 12,
+ "wer": 0.16666666666666666
+ },
+ "semantic": {
+ "normalized_reference": "i defy any man to get a solomon island sore in california",
+ "normalized_hypothesis": "i defy any man to get a solomon islands or in california",
+ "errors": 2,
+ "words": 12,
+ "wer": 0.16666666666666666
+ }
+ },
+ {
+ "id": "release500_0427",
+ "category": "phonetic_balance",
+ "text": "Now our figuring was all right.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0427.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Now our figuring was all right.",
+ "raw": {
+ "normalized_reference": "now our figuring was all right.",
+ "normalized_hypothesis": "now our figuring was all right.",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "now our figuring was all right",
+ "normalized_hypothesis": "now our figuring was all right",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0428",
+ "category": "phonetic_balance",
+ "text": "MacDougall tapped his forehead suspiciously with a stubby forefinger.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0428.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "MacDougall tapped his forehead suspiciously with his stubby forefinger.",
+ "raw": {
+ "normalized_reference": "macdougall tapped his forehead suspiciously with a stubby forefinger.",
+ "normalized_hypothesis": "macdougall tapped his forehead suspiciously with his stubby forefinger.",
+ "errors": 1,
+ "words": 9,
+ "wer": 0.1111111111111111
+ },
+ "semantic": {
+ "normalized_reference": "macdougall tapped his forehead suspiciously with a stubby forefinger",
+ "normalized_hypothesis": "macdougall tapped his forehead suspiciously with his stubby forefinger",
+ "errors": 1,
+ "words": 9,
+ "wer": 0.1111111111111111
+ }
+ },
+ {
+ "id": "release500_0429",
+ "category": "phonetic_balance",
+ "text": "O'Brien took off his coat and bared his right arm.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0429.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "O'Brien took off his coat and buried his right arm.",
+ "raw": {
+ "normalized_reference": "o'brien took off his coat and bared his right arm.",
+ "normalized_hypothesis": "o'brien took off his coat and buried his right arm.",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ },
+ "semantic": {
+ "normalized_reference": "0 brien took off his coat and bared his right arm",
+ "normalized_hypothesis": "0 brien took off his coat and buried his right arm",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ }
+ },
+ {
+ "id": "release500_0430",
+ "category": "phonetic_balance",
+ "text": "But it won't continue, she said with easy confidence.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0430.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "But it won't continue, she said with easy confidence.",
+ "raw": {
+ "normalized_reference": "but it won't continue, she said with easy confidence.",
+ "normalized_hypothesis": "but it won't continue, she said with easy confidence.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "but it will not continue she said with easy confidence",
+ "normalized_hypothesis": "but it will not continue she said with easy confidence",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0431",
+ "category": "phonetic_balance",
+ "text": "I was still weak from my prolonged immersion.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0431.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I was still weak from my prolonged immersion.",
+ "raw": {
+ "normalized_reference": "i was still weak from my prolonged immersion.",
+ "normalized_hypothesis": "i was still weak from my prolonged immersion.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i was still weak from my prolonged immersion",
+ "normalized_hypothesis": "i was still weak from my prolonged immersion",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0432",
+ "category": "phonetic_balance",
+ "text": "Oolong was two hundred and fifty miles from the nearest land.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0432.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The one was 250 miles from the nearest land.",
+ "raw": {
+ "normalized_reference": "oolong was two hundred and fifty miles from the nearest land.",
+ "normalized_hypothesis": "the one was 250 miles from the nearest land.",
+ "errors": 6,
+ "words": 11,
+ "wer": 0.5454545454545454
+ },
+ "semantic": {
+ "normalized_reference": "oolong was 250 miles from the nearest land",
+ "normalized_hypothesis": "the one was 250 miles from the nearest land",
+ "errors": 2,
+ "words": 8,
+ "wer": 0.25
+ }
+ },
+ {
+ "id": "release500_0433",
+ "category": "phonetic_balance",
+ "text": "I just do appreciate it without being able to express my feelings.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0433.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I just do appreciate it without being able to express my feelings.",
+ "raw": {
+ "normalized_reference": "i just do appreciate it without being able to express my feelings.",
+ "normalized_hypothesis": "i just do appreciate it without being able to express my feelings.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i just do appreciate it without being able to express my feelings",
+ "normalized_hypothesis": "i just do appreciate it without being able to express my feelings",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0434",
+ "category": "phonetic_balance",
+ "text": "Now just what do you want to know.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0434.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Now just what do you want to know?",
+ "raw": {
+ "normalized_reference": "now just what do you want to know.",
+ "normalized_hypothesis": "now just what do you want to know?",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "now just what do you want to know",
+ "normalized_hypothesis": "now just what do you want to know",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0435",
+ "category": "phonetic_balance",
+ "text": "But she swung obediently on her heel into the wind.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0435.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "But she swung obediently on her heel into the wind.",
+ "raw": {
+ "normalized_reference": "but she swung obediently on her heel into the wind.",
+ "normalized_hypothesis": "but she swung obediently on her heel into the wind.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "but she swung obediently on her heel into the wind",
+ "normalized_hypothesis": "but she swung obediently on her heel into the wind",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0436",
+ "category": "phonetic_balance",
+ "text": "Ask him, I laughed, then turned to Pasquini.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0436.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Ask him, I laughed, then turned to Pasquini.",
+ "raw": {
+ "normalized_reference": "ask him, i laughed, then turned to pasquini.",
+ "normalized_hypothesis": "ask him, i laughed, then turned to pasquini.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "ask him i laughed then turned to pasquini",
+ "normalized_hypothesis": "ask him i laughed then turned to pasquini",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0437",
+ "category": "phonetic_balance",
+ "text": "He did not rush in.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0437.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He did not rush in.",
+ "raw": {
+ "normalized_reference": "he did not rush in.",
+ "normalized_hypothesis": "he did not rush in.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he did not rush in",
+ "normalized_hypothesis": "he did not rush in",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0438",
+ "category": "phonetic_balance",
+ "text": "They obeyed him, and went here and there at his commands.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0438.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "They obeyed him, and went here and there at his commands.",
+ "raw": {
+ "normalized_reference": "they obeyed him, and went here and there at his commands.",
+ "normalized_hypothesis": "they obeyed him, and went here and there at his commands.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "they obeyed him and went here and there at his commands",
+ "normalized_hypothesis": "they obeyed him and went here and there at his commands",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0439",
+ "category": "phonetic_balance",
+ "text": "Tom Spink has a harpoon.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0439.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Tom's Spin-Casa Harpoon",
+ "raw": {
+ "normalized_reference": "tom spink has a harpoon.",
+ "normalized_hypothesis": "tom's spin-casa harpoon",
+ "errors": 4,
+ "words": 5,
+ "wer": 0.8
+ },
+ "semantic": {
+ "normalized_reference": "tom spink has a harpoon",
+ "normalized_hypothesis": "tom is spin casa harpoon",
+ "errors": 3,
+ "words": 5,
+ "wer": 0.6
+ }
+ },
+ {
+ "id": "release500_0440",
+ "category": "phonetic_balance",
+ "text": "It was my reports from the north which chiefly induced people to buy.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0440.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It was my reports from the north which chiefly induced people to buy.",
+ "raw": {
+ "normalized_reference": "it was my reports from the north which chiefly induced people to buy.",
+ "normalized_hypothesis": "it was my reports from the north which chiefly induced people to buy.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it was my reports from the north which chiefly induced people to buy",
+ "normalized_hypothesis": "it was my reports from the north which chiefly induced people to buy",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0441",
+ "category": "phonetic_balance",
+ "text": "Cherokee identified himself with his instinct.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0441.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Cherokee identified himself with his instinct.",
+ "raw": {
+ "normalized_reference": "cherokee identified himself with his instinct.",
+ "normalized_hypothesis": "cherokee identified himself with his instinct.",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "cherokee identified himself with his instinct",
+ "normalized_hypothesis": "cherokee identified himself with his instinct",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0442",
+ "category": "phonetic_balance",
+ "text": "I graduated last of my class.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0442.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I graduated last of my class.",
+ "raw": {
+ "normalized_reference": "i graduated last of my class.",
+ "normalized_hypothesis": "i graduated last of my class.",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i graduated last of my class",
+ "normalized_hypothesis": "i graduated last of my class",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0443",
+ "category": "phonetic_balance",
+ "text": "It was a gigantic inadequacy.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0443.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It was a gigantic inadequacy.",
+ "raw": {
+ "normalized_reference": "it was a gigantic inadequacy.",
+ "normalized_hypothesis": "it was a gigantic inadequacy.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it was a gigantic inadequacy",
+ "normalized_hypothesis": "it was a gigantic inadequacy",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0444",
+ "category": "phonetic_balance",
+ "text": "Did I possess too much vitality.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0444.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Did I possess too much vitality?",
+ "raw": {
+ "normalized_reference": "did i possess too much vitality.",
+ "normalized_hypothesis": "did i possess too much vitality?",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "did i possess too much vitality",
+ "normalized_hypothesis": "did i possess too much vitality",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0445",
+ "category": "phonetic_balance",
+ "text": "Therefore, hurrah for the game.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0445.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Therefore, hurrah for the game.",
+ "raw": {
+ "normalized_reference": "therefore, hurrah for the game.",
+ "normalized_hypothesis": "therefore, hurrah for the game.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "therefore hurrah for the game",
+ "normalized_hypothesis": "therefore hurrah for the game",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0446",
+ "category": "phonetic_balance",
+ "text": "He had fulfilled his duty and paid properly.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0446.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He had fulfilled his duty and paid properly.",
+ "raw": {
+ "normalized_reference": "he had fulfilled his duty and paid properly.",
+ "normalized_hypothesis": "he had fulfilled his duty and paid properly.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he had fulfilled his duty and paid properly",
+ "normalized_hypothesis": "he had fulfilled his duty and paid properly",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0447",
+ "category": "phonetic_balance",
+ "text": "How can you manage all alone, Mr Young.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0447.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "How can you manage all alone, Mr. Young?",
+ "raw": {
+ "normalized_reference": "how can you manage all alone, mr young.",
+ "normalized_hypothesis": "how can you manage all alone, mr. young?",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "how can you manage all alone mister young",
+ "normalized_hypothesis": "how can you manage all alone mister young",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0448",
+ "category": "phonetic_balance",
+ "text": "Red-Eye never committed a more outrageous deed.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0448.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Red-Eye never committed a more outrageous deed.",
+ "raw": {
+ "normalized_reference": "red-eye never committed a more outrageous deed.",
+ "normalized_hypothesis": "red-eye never committed a more outrageous deed.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "red eye never committed a more outrageous deed",
+ "normalized_hypothesis": "red eye never committed a more outrageous deed",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0449",
+ "category": "phonetic_balance",
+ "text": "He would keep his faith with Gregson for the promised day or two.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0449.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He would keep his faith with Gregson for the promised day or two.",
+ "raw": {
+ "normalized_reference": "he would keep his faith with gregson for the promised day or two.",
+ "normalized_hypothesis": "he would keep his faith with gregson for the promised day or two.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he would keep his faith with gregson for the promised day or 2",
+ "normalized_hypothesis": "he would keep his faith with gregson for the promised day or 2",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0450",
+ "category": "phonetic_balance",
+ "text": "There was one chance, and only one, of saving Jeanne.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0450.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "There was one chance, and only one, of saving Gene.",
+ "raw": {
+ "normalized_reference": "there was one chance, and only one, of saving jeanne.",
+ "normalized_hypothesis": "there was one chance, and only one, of saving gene.",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ },
+ "semantic": {
+ "normalized_reference": "there was one chance and only one of saving jeanne",
+ "normalized_hypothesis": "there was one chance and only one of saving gene",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ }
+ },
+ {
+ "id": "release500_0451",
+ "category": "phonetic_balance",
+ "text": "I was beneath the water, suffocating and drowning.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0451.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I was beneath the water, suffocating and drowning.",
+ "raw": {
+ "normalized_reference": "i was beneath the water, suffocating and drowning.",
+ "normalized_hypothesis": "i was beneath the water, suffocating and drowning.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i was beneath the water suffocating and drowning",
+ "normalized_hypothesis": "i was beneath the water suffocating and drowning",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0452",
+ "category": "phonetic_balance",
+ "text": "Give them their choice between a fine or an official whipping.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0452.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Give them their choice between a fine or an official whipping.",
+ "raw": {
+ "normalized_reference": "give them their choice between a fine or an official whipping.",
+ "normalized_hypothesis": "give them their choice between a fine or an official whipping.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "give them their choice between a fine or an official whipping",
+ "normalized_hypothesis": "give them their choice between a fine or an official whipping",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0453",
+ "category": "phonetic_balance",
+ "text": "They must have been swept away by the chaotic currents.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0453.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "They must have been swept away by the chaotic currents.",
+ "raw": {
+ "normalized_reference": "they must have been swept away by the chaotic currents.",
+ "normalized_hypothesis": "they must have been swept away by the chaotic currents.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "they must have been swept away by the chaotic currents",
+ "normalized_hypothesis": "they must have been swept away by the chaotic currents",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0454",
+ "category": "phonetic_balance",
+ "text": "You yellow giant thing of the frost.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0454.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "You yellow giant thing of the frost.",
+ "raw": {
+ "normalized_reference": "you yellow giant thing of the frost.",
+ "normalized_hypothesis": "you yellow giant thing of the frost.",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "you yellow giant thing of the frost",
+ "normalized_hypothesis": "you yellow giant thing of the frost",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0455",
+ "category": "phonetic_balance",
+ "text": "He had peeled off his shirt and was wildly waving it.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0455.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He had peeled off his shirt and was wildly waving it.",
+ "raw": {
+ "normalized_reference": "he had peeled off his shirt and was wildly waving it.",
+ "normalized_hypothesis": "he had peeled off his shirt and was wildly waving it.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he had peeled off his shirt and was wildly waving it",
+ "normalized_hypothesis": "he had peeled off his shirt and was wildly waving it",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0456",
+ "category": "phonetic_balance",
+ "text": "To these he gave castor oil.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0456.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Did these he gave cast your oil?",
+ "raw": {
+ "normalized_reference": "to these he gave castor oil.",
+ "normalized_hypothesis": "did these he gave cast your oil?",
+ "errors": 3,
+ "words": 6,
+ "wer": 0.5
+ },
+ "semantic": {
+ "normalized_reference": "to these he gave castor oil",
+ "normalized_hypothesis": "did these he gave cast your oil",
+ "errors": 3,
+ "words": 6,
+ "wer": 0.5
+ }
+ },
+ {
+ "id": "release500_0457",
+ "category": "phonetic_balance",
+ "text": "Tudor surveyed him with withering disgust.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0457.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Tudor surveyed him with withering disgust.",
+ "raw": {
+ "normalized_reference": "tudor surveyed him with withering disgust.",
+ "normalized_hypothesis": "tudor surveyed him with withering disgust.",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "tudor surveyed him with withering disgust",
+ "normalized_hypothesis": "tudor surveyed him with withering disgust",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0458",
+ "category": "phonetic_balance",
+ "text": "After all, it was simply a mistake in judgment.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0458.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "After all, it was simply a mistake in judgment.",
+ "raw": {
+ "normalized_reference": "after all, it was simply a mistake in judgment.",
+ "normalized_hypothesis": "after all, it was simply a mistake in judgment.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "after all it was simply a mistake in judgment",
+ "normalized_hypothesis": "after all it was simply a mistake in judgment",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0459",
+ "category": "phonetic_balance",
+ "text": "That is the strange part of it.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0459.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "That is the strange part of it.",
+ "raw": {
+ "normalized_reference": "that is the strange part of it.",
+ "normalized_hypothesis": "that is the strange part of it.",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "that is the strange part of it",
+ "normalized_hypothesis": "that is the strange part of it",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0460",
+ "category": "phonetic_balance",
+ "text": "Not a wheel moved in his empire.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0460.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Not a wheel moved in his empire.",
+ "raw": {
+ "normalized_reference": "not a wheel moved in his empire.",
+ "normalized_hypothesis": "not a wheel moved in his empire.",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "not a wheel moved in his empire",
+ "normalized_hypothesis": "not a wheel moved in his empire",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0461",
+ "category": "phonetic_balance",
+ "text": "Its diameter was not more than two hundred yards.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0461.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Its diameter was not more than 200 yards.",
+ "raw": {
+ "normalized_reference": "its diameter was not more than two hundred yards.",
+ "normalized_hypothesis": "its diameter was not more than 200 yards.",
+ "errors": 2,
+ "words": 9,
+ "wer": 0.2222222222222222
+ },
+ "semantic": {
+ "normalized_reference": "its diameter was not more than 200 yards",
+ "normalized_hypothesis": "its diameter was not more than 200 yards",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0462",
+ "category": "phonetic_balance",
+ "text": "Nimrod replied, with a slight manifestation of sensitiveness.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0462.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Nimrod replied with a slight manifestation of sensitiveness.",
+ "raw": {
+ "normalized_reference": "nimrod replied, with a slight manifestation of sensitiveness.",
+ "normalized_hypothesis": "nimrod replied with a slight manifestation of sensitiveness.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "nimrod replied with a slight manifestation of sensitiveness",
+ "normalized_hypothesis": "nimrod replied with a slight manifestation of sensitiveness",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0463",
+ "category": "phonetic_balance",
+ "text": "I tried to read George Moore last night, and was dreadfully bored.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0463.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I tried to read George more last night and was dreadfully bored.",
+ "raw": {
+ "normalized_reference": "i tried to read george moore last night, and was dreadfully bored.",
+ "normalized_hypothesis": "i tried to read george more last night and was dreadfully bored.",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "i tried to read george moore last night and was dreadfully bored",
+ "normalized_hypothesis": "i tried to read george more last night and was dreadfully bored",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ }
+ },
+ {
+ "id": "release500_0464",
+ "category": "phonetic_balance",
+ "text": "Yes, it was a man who asked, a stranger.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0464.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Yes, it was a man who asked, a stranger.",
+ "raw": {
+ "normalized_reference": "yes, it was a man who asked, a stranger.",
+ "normalized_hypothesis": "yes, it was a man who asked, a stranger.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "yes it was a man who asked a stranger",
+ "normalized_hypothesis": "yes it was a man who asked a stranger",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0465",
+ "category": "phonetic_balance",
+ "text": "We were now good friends.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0465.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "We were now good friends.",
+ "raw": {
+ "normalized_reference": "we were now good friends.",
+ "normalized_hypothesis": "we were now good friends.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "we were now good friends",
+ "normalized_hypothesis": "we were now good friends",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0466",
+ "category": "phonetic_balance",
+ "text": "Yet, in accordance with Ernest's test of truth, it worked.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0466.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Yet, in accordance with Ernest's test of truth, it worked.",
+ "raw": {
+ "normalized_reference": "yet, in accordance with ernest's test of truth, it worked.",
+ "normalized_hypothesis": "yet, in accordance with ernest's test of truth, it worked.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "yet in accordance with ernest is test of truth it worked",
+ "normalized_hypothesis": "yet in accordance with ernest is test of truth it worked",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0467",
+ "category": "phonetic_balance",
+ "text": "It's the strap hangers that'll keep us from going under.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0467.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It's the strap hangers that'll keep us from going under.",
+ "raw": {
+ "normalized_reference": "it's the strap hangers that'll keep us from going under.",
+ "normalized_hypothesis": "it's the strap hangers that'll keep us from going under.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it is the strap hangers that will keep us from going under",
+ "normalized_hypothesis": "it is the strap hangers that will keep us from going under",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0468",
+ "category": "phonetic_balance",
+ "text": "I never allow what can't be changed to annoy me.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0468.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I never allow what can't be changed to annoy me.",
+ "raw": {
+ "normalized_reference": "i never allow what can't be changed to annoy me.",
+ "normalized_hypothesis": "i never allow what can't be changed to annoy me.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i never allow what can not be changed to annoy me",
+ "normalized_hypothesis": "i never allow what can not be changed to annoy me",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0469",
+ "category": "phonetic_balance",
+ "text": "After all superfluous flesh is gone what is left is stringy and resistant.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0469.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "After all superfluous flesh has come what is left is stringy and resistant.",
+ "raw": {
+ "normalized_reference": "after all superfluous flesh is gone what is left is stringy and resistant.",
+ "normalized_hypothesis": "after all superfluous flesh has come what is left is stringy and resistant.",
+ "errors": 2,
+ "words": 13,
+ "wer": 0.15384615384615385
+ },
+ "semantic": {
+ "normalized_reference": "after all superfluous flesh is gone what is left is stringy and resistant",
+ "normalized_hypothesis": "after all superfluous flesh has come what is left is stringy and resistant",
+ "errors": 2,
+ "words": 13,
+ "wer": 0.15384615384615385
+ }
+ },
+ {
+ "id": "release500_0470",
+ "category": "phonetic_balance",
+ "text": "Then he shouted, Shut up.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0470.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Then he shouted, shut up.",
+ "raw": {
+ "normalized_reference": "then he shouted, shut up.",
+ "normalized_hypothesis": "then he shouted, shut up.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "then he shouted shut up",
+ "normalized_hypothesis": "then he shouted shut up",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0471",
+ "category": "phonetic_balance",
+ "text": "He had worshiped her, as Dante might have worshiped Beatrice.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0471.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Kihan worshipped her as Dante and Mike have worshipped Beatrice.",
+ "raw": {
+ "normalized_reference": "he had worshiped her, as dante might have worshiped beatrice.",
+ "normalized_hypothesis": "kihan worshipped her as dante and mike have worshipped beatrice.",
+ "errors": 6,
+ "words": 10,
+ "wer": 0.6
+ },
+ "semantic": {
+ "normalized_reference": "he had worshiped her as dante might have worshiped beatrice",
+ "normalized_hypothesis": "kihan worshipped her as dante and mike have worshipped beatrice",
+ "errors": 6,
+ "words": 10,
+ "wer": 0.6
+ }
+ },
+ {
+ "id": "release500_0472",
+ "category": "phonetic_balance",
+ "text": "He looked like one who had passed through an uncomfortable hour or two.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0472.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He looked like one who had passed through an uncomfortable hour or two.",
+ "raw": {
+ "normalized_reference": "he looked like one who had passed through an uncomfortable hour or two.",
+ "normalized_hypothesis": "he looked like one who had passed through an uncomfortable hour or two.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he looked like one who had passed through an uncomfortable hour or 2",
+ "normalized_hypothesis": "he looked like one who had passed through an uncomfortable hour or 2",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0473",
+ "category": "phonetic_balance",
+ "text": "The history of our westward-faring race is written in it.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0473.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The history of our westward-faring race is written in it.",
+ "raw": {
+ "normalized_reference": "the history of our westward-faring race is written in it.",
+ "normalized_hypothesis": "the history of our westward-faring race is written in it.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the history of our westward faring race is written in it",
+ "normalized_hypothesis": "the history of our westward faring race is written in it",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0474",
+ "category": "phonetic_balance",
+ "text": "He may desire to escape pain, or to enjoy pleasure.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0474.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He may desire to escape pain or to enjoy pleasure.",
+ "raw": {
+ "normalized_reference": "he may desire to escape pain, or to enjoy pleasure.",
+ "normalized_hypothesis": "he may desire to escape pain or to enjoy pleasure.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he may desire to escape pain or to enjoy pleasure",
+ "normalized_hypothesis": "he may desire to escape pain or to enjoy pleasure",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0475",
+ "category": "phonetic_balance",
+ "text": "And the air was growing chilly.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0475.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "And the air was growing chilly.",
+ "raw": {
+ "normalized_reference": "and the air was growing chilly.",
+ "normalized_hypothesis": "and the air was growing chilly.",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "and the air was growing chilly",
+ "normalized_hypothesis": "and the air was growing chilly",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0476",
+ "category": "phonetic_balance",
+ "text": "Dig in; you're sure good, was Daylight's answer.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0476.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Dig in. You're sure good, was Daylight's answer.",
+ "raw": {
+ "normalized_reference": "dig in; you're sure good, was daylight's answer.",
+ "normalized_hypothesis": "dig in. you're sure good, was daylight's answer.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "dig in you are sure good was daylight is answer",
+ "normalized_hypothesis": "dig in you are sure good was daylight is answer",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0477",
+ "category": "phonetic_balance",
+ "text": "My idea was that he would have more influence over the natives.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0477.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "My idea was that he would have more influence over the natives.",
+ "raw": {
+ "normalized_reference": "my idea was that he would have more influence over the natives.",
+ "normalized_hypothesis": "my idea was that he would have more influence over the natives.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "my idea was that he would have more influence over the natives",
+ "normalized_hypothesis": "my idea was that he would have more influence over the natives",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0478",
+ "category": "phonetic_balance",
+ "text": "Now run along, and tell them to hurry.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0478.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Now run along and tell them to hurry.",
+ "raw": {
+ "normalized_reference": "now run along, and tell them to hurry.",
+ "normalized_hypothesis": "now run along and tell them to hurry.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "now run along and tell them to hurry",
+ "normalized_hypothesis": "now run along and tell them to hurry",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0479",
+ "category": "phonetic_balance",
+ "text": "I have no idea, replied Philip.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0479.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I have no idea, replied Philip.",
+ "raw": {
+ "normalized_reference": "i have no idea, replied philip.",
+ "normalized_hypothesis": "i have no idea, replied philip.",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i have no idea replied philip",
+ "normalized_hypothesis": "i have no idea replied philip",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0480",
+ "category": "phonetic_balance",
+ "text": "For the rest, he was a mere automaton.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0480.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "For the rest, he was a mere automaton.",
+ "raw": {
+ "normalized_reference": "for the rest, he was a mere automaton.",
+ "normalized_hypothesis": "for the rest, he was a mere automaton.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "for the rest he was a mere automaton",
+ "normalized_hypothesis": "for the rest he was a mere automaton",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0481",
+ "category": "phonetic_balance",
+ "text": "Why, the average review is more nauseating than cod liver oil.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0481.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Why, the average review is more nauseating than cod liver oil.",
+ "raw": {
+ "normalized_reference": "why, the average review is more nauseating than cod liver oil.",
+ "normalized_hypothesis": "why, the average review is more nauseating than cod liver oil.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "why the average review is more nauseating than cod liver oil",
+ "normalized_hypothesis": "why the average review is more nauseating than cod liver oil",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0482",
+ "category": "phonetic_balance",
+ "text": "I have been robbed, sir, I amended.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0482.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I have been robbed, sir, I amended.",
+ "raw": {
+ "normalized_reference": "i have been robbed, sir, i amended.",
+ "normalized_hypothesis": "i have been robbed, sir, i amended.",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i have been robbed sir i amended",
+ "normalized_hypothesis": "i have been robbed sir i amended",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0483",
+ "category": "phonetic_balance",
+ "text": "He moved away as quietly as he had come.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0483.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He moved away as quietly as he had come.",
+ "raw": {
+ "normalized_reference": "he moved away as quietly as he had come.",
+ "normalized_hypothesis": "he moved away as quietly as he had come.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he moved away as quietly as he had come",
+ "normalized_hypothesis": "he moved away as quietly as he had come",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0484",
+ "category": "phonetic_balance",
+ "text": "A trickle of fresh blood ran over his face.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0484.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "A trickle of fresh blood ran over his face.",
+ "raw": {
+ "normalized_reference": "a trickle of fresh blood ran over his face.",
+ "normalized_hypothesis": "a trickle of fresh blood ran over his face.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "a trickle of fresh blood ran over his face",
+ "normalized_hypothesis": "a trickle of fresh blood ran over his face",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0485",
+ "category": "phonetic_balance",
+ "text": "Suppose you saw me at work through the window.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0485.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Suppose you saw me at work through the window.",
+ "raw": {
+ "normalized_reference": "suppose you saw me at work through the window.",
+ "normalized_hypothesis": "suppose you saw me at work through the window.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "suppose you saw me at work through the window",
+ "normalized_hypothesis": "suppose you saw me at work through the window",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0486",
+ "category": "phonetic_balance",
+ "text": "But who was Eileen's double.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0486.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "But who was Eileen's double?",
+ "raw": {
+ "normalized_reference": "but who was eileen's double.",
+ "normalized_hypothesis": "but who was eileen's double?",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "but who was eileen is double",
+ "normalized_hypothesis": "but who was eileen is double",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0487",
+ "category": "phonetic_balance",
+ "text": "Darkness hid him from Jeanne.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0487.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Darkus hid him from G.",
+ "raw": {
+ "normalized_reference": "darkness hid him from jeanne.",
+ "normalized_hypothesis": "darkus hid him from g.",
+ "errors": 2,
+ "words": 5,
+ "wer": 0.4
+ },
+ "semantic": {
+ "normalized_reference": "darkness hid him from jeanne",
+ "normalized_hypothesis": "darkus hid him from g",
+ "errors": 2,
+ "words": 5,
+ "wer": 0.4
+ }
+ },
+ {
+ "id": "release500_0488",
+ "category": "phonetic_balance",
+ "text": "By this time Charley was as enraged as the Greek.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0488.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "By this time Charlie was as enraged as the Greek.",
+ "raw": {
+ "normalized_reference": "by this time charley was as enraged as the greek.",
+ "normalized_hypothesis": "by this time charlie was as enraged as the greek.",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ },
+ "semantic": {
+ "normalized_reference": "by this time charley was as enraged as the greek",
+ "normalized_hypothesis": "by this time charlie was as enraged as the greek",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ }
+ },
+ {
+ "id": "release500_0489",
+ "category": "phonetic_balance",
+ "text": "I am writing these lines in Honolulu, Hawaii.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0489.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I'm writing these lines in Honolulu, Kauai.",
+ "raw": {
+ "normalized_reference": "i am writing these lines in honolulu, hawaii.",
+ "normalized_hypothesis": "i'm writing these lines in honolulu, kauai.",
+ "errors": 3,
+ "words": 8,
+ "wer": 0.375
+ },
+ "semantic": {
+ "normalized_reference": "i am writing these lines in honolulu hawaii",
+ "normalized_hypothesis": "i am writing these lines in honolulu kauai",
+ "errors": 1,
+ "words": 8,
+ "wer": 0.125
+ }
+ },
+ {
+ "id": "release500_0490",
+ "category": "phonetic_balance",
+ "text": "Nobody knows how the natives got them.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0490.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Nobody knows how the natives caught them.",
+ "raw": {
+ "normalized_reference": "nobody knows how the natives got them.",
+ "normalized_hypothesis": "nobody knows how the natives caught them.",
+ "errors": 1,
+ "words": 7,
+ "wer": 0.14285714285714285
+ },
+ "semantic": {
+ "normalized_reference": "nobody knows how the natives got them",
+ "normalized_hypothesis": "nobody knows how the natives caught them",
+ "errors": 1,
+ "words": 7,
+ "wer": 0.14285714285714285
+ }
+ },
+ {
+ "id": "release500_0491",
+ "category": "phonetic_balance",
+ "text": "And how would we ever find ourselves.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0491.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "And how would we ever find ourselves?",
+ "raw": {
+ "normalized_reference": "and how would we ever find ourselves.",
+ "normalized_hypothesis": "and how would we ever find ourselves?",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "and how would we ever find ourselves",
+ "normalized_hypothesis": "and how would we ever find ourselves",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0492",
+ "category": "phonetic_balance",
+ "text": "They are coming ashore, whoever they are.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0492.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "They are coming ashore, whoever they are.",
+ "raw": {
+ "normalized_reference": "they are coming ashore, whoever they are.",
+ "normalized_hypothesis": "they are coming ashore, whoever they are.",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "they are coming ashore whoever they are",
+ "normalized_hypothesis": "they are coming ashore whoever they are",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0493",
+ "category": "phonetic_balance",
+ "text": "This is my fifth voyage.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0493.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "This is my fifth voyage.",
+ "raw": {
+ "normalized_reference": "this is my fifth voyage.",
+ "normalized_hypothesis": "this is my fifth voyage.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "this is my 5th voyage",
+ "normalized_hypothesis": "this is my 5th voyage",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0494",
+ "category": "phonetic_balance",
+ "text": "And yet if she came he had no words to say.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0494.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "And yet if she came he had no words to say.",
+ "raw": {
+ "normalized_reference": "and yet if she came he had no words to say.",
+ "normalized_hypothesis": "and yet if she came he had no words to say.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "and yet if she came he had no words to say",
+ "normalized_hypothesis": "and yet if she came he had no words to say",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0495",
+ "category": "phonetic_balance",
+ "text": "You must sleep, he urged.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0495.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "You must sleep, he urged.",
+ "raw": {
+ "normalized_reference": "you must sleep, he urged.",
+ "normalized_hypothesis": "you must sleep, he urged.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "you must sleep he urged",
+ "normalized_hypothesis": "you must sleep he urged",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0496",
+ "category": "phonetic_balance",
+ "text": "Could the incident have anything to do with Jeanne and Pierre.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0496.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Could the incident have anything to do with Jean and Pierre?",
+ "raw": {
+ "normalized_reference": "could the incident have anything to do with jeanne and pierre.",
+ "normalized_hypothesis": "could the incident have anything to do with jean and pierre?",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ },
+ "semantic": {
+ "normalized_reference": "could the incident have anything to do with jeanne and pierre",
+ "normalized_hypothesis": "could the incident have anything to do with jean and pierre",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ }
+ },
+ {
+ "id": "release500_0497",
+ "category": "phonetic_balance",
+ "text": "Each day she became a more vital part of him.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0497.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Each day she became a more vital part of him.",
+ "raw": {
+ "normalized_reference": "each day she became a more vital part of him.",
+ "normalized_hypothesis": "each day she became a more vital part of him.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "each day she became a more vital part of him",
+ "normalized_hypothesis": "each day she became a more vital part of him",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0498",
+ "category": "phonetic_balance",
+ "text": "But why continue the tirade, for tirade it was.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0498.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "But why continue the tirade? For tirade it was.",
+ "raw": {
+ "normalized_reference": "but why continue the tirade, for tirade it was.",
+ "normalized_hypothesis": "but why continue the tirade? for tirade it was.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "but why continue the tirade for tirade it was",
+ "normalized_hypothesis": "but why continue the tirade for tirade it was",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0499",
+ "category": "phonetic_balance",
+ "text": "That is what distinguishes all of us from the lower animals.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0499.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "That is what distinguishes all of us from the lower animals.",
+ "raw": {
+ "normalized_reference": "that is what distinguishes all of us from the lower animals.",
+ "normalized_hypothesis": "that is what distinguishes all of us from the lower animals.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "that is what distinguishes all of us from the lower animals",
+ "normalized_hypothesis": "that is what distinguishes all of us from the lower animals",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0500",
+ "category": "phonetic_balance",
+ "text": "There was something pathetic in the girl's attitude now.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0500.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "There was something pathetic in the girl's attitude now.",
+ "raw": {
+ "normalized_reference": "there was something pathetic in the girl's attitude now.",
+ "normalized_hypothesis": "there was something pathetic in the girl's attitude now.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "there was something pathetic in the girl is attitude now",
+ "normalized_hypothesis": "there was something pathetic in the girl is attitude now",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0001",
+ "category": "conversational",
+ "text": "It wasn't until later that I realized what had actually happened.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0001.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It wasn't until later that I realized what had actually happened.",
+ "raw": {
+ "normalized_reference": "it wasn't until later that i realized what had actually happened.",
+ "normalized_hypothesis": "it wasn't until later that i realized what had actually happened.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it was not until later that i realized what had actually happened",
+ "normalized_hypothesis": "it was not until later that i realized what had actually happened",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0002",
+ "category": "descriptive",
+ "text": "The light at that time of afternoon is perfect for about twenty minutes.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0002.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The light at that time of afternoon is perfect for about 20 minutes.",
+ "raw": {
+ "normalized_reference": "the light at that time of afternoon is perfect for about twenty minutes.",
+ "normalized_hypothesis": "the light at that time of afternoon is perfect for about 20 minutes.",
+ "errors": 1,
+ "words": 13,
+ "wer": 0.07692307692307693
+ },
+ "semantic": {
+ "normalized_reference": "the light at that time of afternoon is perfect for about 20 minutes",
+ "normalized_hypothesis": "the light at that time of afternoon is perfect for about 20 minutes",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0003",
+ "category": "emotional",
+ "text": "That was the scariest thing I've ever done, and also maybe the best.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0003.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "That was the scariest thing I've ever done, and also maybe the best.",
+ "raw": {
+ "normalized_reference": "that was the scariest thing i've ever done, and also maybe the best.",
+ "normalized_hypothesis": "that was the scariest thing i've ever done, and also maybe the best.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "that was the scariest thing i have ever done and also maybe the best",
+ "normalized_hypothesis": "that was the scariest thing i have ever done and also maybe the best",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0004",
+ "category": "homographs",
+ "text": "After I record the record, please present the present to the bass player.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0004.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "After I record the record, please present the present to the bass player.",
+ "raw": {
+ "normalized_reference": "after i record the record, please present the present to the bass player.",
+ "normalized_hypothesis": "after i record the record, please present the present to the bass player.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "after i record the record please present the present to the bass player",
+ "normalized_hypothesis": "after i record the record please present the present to the bass player",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0005",
+ "category": "long",
+ "text": "If I'm being completely honest, I had no idea what I was doing for most of it, and I think that might be the most normal thing about the whole experience.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0005.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "If I'm being completely honest, I had no idea what I was doing for most of it, and I think that might be the most normal thing about the whole experience.",
+ "raw": {
+ "normalized_reference": "if i'm being completely honest, i had no idea what i was doing for most of it, and i think that might be the most normal thing about the whole experience.",
+ "normalized_hypothesis": "if i'm being completely honest, i had no idea what i was doing for most of it, and i think that might be the most normal thing about the whole experience.",
+ "errors": 0,
+ "words": 31,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "if i am being completely honest i had no idea what i was doing for most of it and i think that might be the most normal thing about the whole experience",
+ "normalized_hypothesis": "if i am being completely honest i had no idea what i was doing for most of it and i think that might be the most normal thing about the whole experience",
+ "errors": 0,
+ "words": 32,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0006",
+ "category": "long_range",
+ "text": "The committee postponed its decision after discovering that three pages of the report had been printed in the wrong order.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0006.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The committee postponed its decision after discovering that three pages of the report had been printed in the wrong order.",
+ "raw": {
+ "normalized_reference": "the committee postponed its decision after discovering that three pages of the report had been printed in the wrong order.",
+ "normalized_hypothesis": "the committee postponed its decision after discovering that three pages of the report had been printed in the wrong order.",
+ "errors": 0,
+ "words": 20,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the committee postponed its decision after discovering that 3 pages of the report had been printed in the wrong order",
+ "normalized_hypothesis": "the committee postponed its decision after discovering that 3 pages of the report had been printed in the wrong order",
+ "errors": 0,
+ "words": 20,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0007",
+ "category": "names_places",
+ "text": "Gwendolyn photographed the eucalyptus trees outside Ljubljana.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0007.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Gwendolyn photographed the eucalyptus trees outside Ljubljana.",
+ "raw": {
+ "normalized_reference": "gwendolyn photographed the eucalyptus trees outside ljubljana.",
+ "normalized_hypothesis": "gwendolyn photographed the eucalyptus trees outside ljubljana.",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "gwendolyn photographed the eucalyptus trees outside ljubljana",
+ "normalized_hypothesis": "gwendolyn photographed the eucalyptus trees outside ljubljana",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0008",
+ "category": "narrative",
+ "text": "He had this theory, and the more time passes, the more I think he was right.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0008.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He had this theory and the more time passes, the more I think he was right.",
+ "raw": {
+ "normalized_reference": "he had this theory, and the more time passes, the more i think he was right.",
+ "normalized_hypothesis": "he had this theory and the more time passes, the more i think he was right.",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he had this theory and the more time passes the more i think he was right",
+ "normalized_hypothesis": "he had this theory and the more time passes the more i think he was right",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0009",
+ "category": "numbers",
+ "text": "The package weighs twelve point six kilograms and arrived on July twenty-first.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0009.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The package weighs 12.6 kilograms and arrived on July 21st.",
+ "raw": {
+ "normalized_reference": "the package weighs twelve point six kilograms and arrived on july twenty-first.",
+ "normalized_hypothesis": "the package weighs 12.6 kilograms and arrived on july 21st.",
+ "errors": 5,
+ "words": 13,
+ "wer": 0.38461538461538464
+ },
+ "semantic": {
+ "normalized_reference": "the package weighs 12.6 kilograms and arrived on july 21st",
+ "normalized_hypothesis": "the package weighs 12.6 kilograms and arrived on july 21st",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0010",
+ "category": "punctuation",
+ "text": "First, close the window; second, turn off the lamp; finally, lock the door.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0010.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "First, close the window. Second, turn off the lamp. Finally, lock the door.",
+ "raw": {
+ "normalized_reference": "first, close the window; second, turn off the lamp; finally, lock the door.",
+ "normalized_hypothesis": "first, close the window. second, turn off the lamp. finally, lock the door.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "1st close the window 2nd turn off the lamp finally lock the door",
+ "normalized_hypothesis": "1st close the window 2nd turn off the lamp finally lock the door",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0011",
+ "category": "question",
+ "text": "When did you first realize something was wrong?",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0011.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "When did you first realize something was wrong?",
+ "raw": {
+ "normalized_reference": "when did you first realize something was wrong?",
+ "normalized_hypothesis": "when did you first realize something was wrong?",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "when did you 1st realize something was wrong",
+ "normalized_hypothesis": "when did you 1st realize something was wrong",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0012",
+ "category": "rare_words",
+ "text": "The archaeologist catalogued a Byzantine reliquary and several anachronistic fragments.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0012.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The archaeologist catalogued a Byzantine reliquary and several anachronistic fragments.",
+ "raw": {
+ "normalized_reference": "the archaeologist catalogued a byzantine reliquary and several anachronistic fragments.",
+ "normalized_hypothesis": "the archaeologist catalogued a byzantine reliquary and several anachronistic fragments.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the archaeologist catalogued a byzantine reliquary and several anachronistic fragments",
+ "normalized_hypothesis": "the archaeologist catalogued a byzantine reliquary and several anachronistic fragments",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0013",
+ "category": "short",
+ "text": "I have no idea.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0013.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I have no idea.",
+ "raw": {
+ "normalized_reference": "i have no idea.",
+ "normalized_hypothesis": "i have no idea.",
+ "errors": 0,
+ "words": 4,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i have no idea",
+ "normalized_hypothesis": "i have no idea",
+ "errors": 0,
+ "words": 4,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0014",
+ "category": "technical",
+ "text": "The system runs on three core components that all have to stay in sync.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0014.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The system runs on three core components that all have to stay in sync.",
+ "raw": {
+ "normalized_reference": "the system runs on three core components that all have to stay in sync.",
+ "normalized_hypothesis": "the system runs on three core components that all have to stay in sync.",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the system runs on 3 core components that all have to stay in sync",
+ "normalized_hypothesis": "the system runs on 3 core components that all have to stay in sync",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0015",
+ "category": "conversational",
+ "text": "We improvised the entire second half and somehow nobody noticed.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0015.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "We improvised the entire second half, and somehow nobody noticed.",
+ "raw": {
+ "normalized_reference": "we improvised the entire second half and somehow nobody noticed.",
+ "normalized_hypothesis": "we improvised the entire second half, and somehow nobody noticed.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "we improvised the entire 2nd half and somehow nobody noticed",
+ "normalized_hypothesis": "we improvised the entire 2nd half and somehow nobody noticed",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0016",
+ "category": "descriptive",
+ "text": "There's a smell in old buildings that's impossible to describe but immediately familiar.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0016.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "There's a smell in old buildings that's impossible to describe, but immediately familiar.",
+ "raw": {
+ "normalized_reference": "there's a smell in old buildings that's impossible to describe but immediately familiar.",
+ "normalized_hypothesis": "there's a smell in old buildings that's impossible to describe, but immediately familiar.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "there is a smell in old buildings that is impossible to describe but immediately familiar",
+ "normalized_hypothesis": "there is a smell in old buildings that is impossible to describe but immediately familiar",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0017",
+ "category": "emotional",
+ "text": "I specifically asked for one thing. One thing. And it still didn't happen.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0017.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I specifically asked for one thing. One thing. And it still didn't happen.",
+ "raw": {
+ "normalized_reference": "i specifically asked for one thing. one thing. and it still didn't happen.",
+ "normalized_hypothesis": "i specifically asked for one thing. one thing. and it still didn't happen.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i specifically asked for one thing one thing and it still did not happen",
+ "normalized_hypothesis": "i specifically asked for one thing one thing and it still did not happen",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0018",
+ "category": "homographs",
+ "text": "They refuse to place the refuse beside the produce aisle.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0018.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "They refused to place the refuse beside the produce aisle.",
+ "raw": {
+ "normalized_reference": "they refuse to place the refuse beside the produce aisle.",
+ "normalized_hypothesis": "they refused to place the refuse beside the produce aisle.",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ },
+ "semantic": {
+ "normalized_reference": "they refuse to place the refuse beside the produce aisle",
+ "normalized_hypothesis": "they refused to place the refuse beside the produce aisle",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ }
+ },
+ {
+ "id": "release500_0019",
+ "category": "long",
+ "text": "There were days last year when I genuinely didn't know how I was going to get through it, and somehow I did, and I still don't fully understand how.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0019.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "There were days last year when I genuinely didn't know how I was going to get through it, and somehow I did, and I still don't fully understand how.",
+ "raw": {
+ "normalized_reference": "there were days last year when i genuinely didn't know how i was going to get through it, and somehow i did, and i still don't fully understand how.",
+ "normalized_hypothesis": "there were days last year when i genuinely didn't know how i was going to get through it, and somehow i did, and i still don't fully understand how.",
+ "errors": 0,
+ "words": 29,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "there were days last year when i genuinely did not know how i was going to get through it and somehow i did and i still do not fully understand how",
+ "normalized_hypothesis": "there were days last year when i genuinely did not know how i was going to get through it and somehow i did and i still do not fully understand how",
+ "errors": 0,
+ "words": 31,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0020",
+ "category": "long_range",
+ "text": "Before announcing the winner, the judge thanked the volunteers who had worked through the storm to repair the damaged stage.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0020.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Before announcing the winner, the judge thanked the volunteers who had worked through the storm to repair the damaged stage.",
+ "raw": {
+ "normalized_reference": "before announcing the winner, the judge thanked the volunteers who had worked through the storm to repair the damaged stage.",
+ "normalized_hypothesis": "before announcing the winner, the judge thanked the volunteers who had worked through the storm to repair the damaged stage.",
+ "errors": 0,
+ "words": 20,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "before announcing the winner the judge thanked the volunteers who had worked through the storm to repair the damaged stage",
+ "normalized_hypothesis": "before announcing the winner the judge thanked the volunteers who had worked through the storm to repair the damaged stage",
+ "errors": 0,
+ "words": 20,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0021",
+ "category": "names_places",
+ "text": "Professor Nguyen mailed the fluorescent poster to Reykjavik on Thursday.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0021.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Professor Nguyen mailed the fluorescent poster to Reykjavik on Thursday.",
+ "raw": {
+ "normalized_reference": "professor nguyen mailed the fluorescent poster to reykjavik on thursday.",
+ "normalized_hypothesis": "professor nguyen mailed the fluorescent poster to reykjavik on thursday.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "professor nguyen mailed the fluorescent poster to reykjavik on thursday",
+ "normalized_hypothesis": "professor nguyen mailed the fluorescent poster to reykjavik on thursday",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0022",
+ "category": "narrative",
+ "text": "The plan was simple. The execution was anything but.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0022.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The plan was simple. The execution was anything but.",
+ "raw": {
+ "normalized_reference": "the plan was simple. the execution was anything but.",
+ "normalized_hypothesis": "the plan was simple. the execution was anything but.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the plan was simple the execution was anything but",
+ "normalized_hypothesis": "the plan was simple the execution was anything but",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0023",
+ "category": "numbers",
+ "text": "Her score improved from eighty-two percent to ninety-seven percent in two weeks.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0023.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Her score improved from 82% to 97% in two weeks.",
+ "raw": {
+ "normalized_reference": "her score improved from eighty-two percent to ninety-seven percent in two weeks.",
+ "normalized_hypothesis": "her score improved from 82% to 97% in two weeks.",
+ "errors": 6,
+ "words": 14,
+ "wer": 0.42857142857142855
+ },
+ "semantic": {
+ "normalized_reference": "her score improved from 82% to 97% in 2 weeks",
+ "normalized_hypothesis": "her score improved from 82% to 97% in 2 weeks",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0024",
+ "category": "punctuation",
+ "text": "Why would anyone pack spoons, batteries, and a violin for one short trip?",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0024.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Why would anyone pack spoons, batteries, and a violin for one short trip?",
+ "raw": {
+ "normalized_reference": "why would anyone pack spoons, batteries, and a violin for one short trip?",
+ "normalized_hypothesis": "why would anyone pack spoons, batteries, and a violin for one short trip?",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "why would anyone pack spoons batteries and a violin for one short trip",
+ "normalized_hypothesis": "why would anyone pack spoons batteries and a violin for one short trip",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0025",
+ "category": "question",
+ "text": "How long have you been sitting on this?",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0025.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "How long have you been sitting on this?",
+ "raw": {
+ "normalized_reference": "how long have you been sitting on this?",
+ "normalized_hypothesis": "how long have you been sitting on this?",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "how long have you been sitting on this",
+ "normalized_hypothesis": "how long have you been sitting on this",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0026",
+ "category": "rare_words",
+ "text": "A mischievous chameleon rested beside the chrysanthemum and the bougainvillea.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0026.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "A mischievous chameleon rested beside the Chrysanthemum and the Bougainvillea.",
+ "raw": {
+ "normalized_reference": "a mischievous chameleon rested beside the chrysanthemum and the bougainvillea.",
+ "normalized_hypothesis": "a mischievous chameleon rested beside the chrysanthemum and the bougainvillea.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "a mischievous chameleon rested beside the chrysanthemum and the bougainvillea",
+ "normalized_hypothesis": "a mischievous chameleon rested beside the chrysanthemum and the bougainvillea",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0027",
+ "category": "short",
+ "text": "We don't talk about that.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0027.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "We don't talk about that.",
+ "raw": {
+ "normalized_reference": "we don't talk about that.",
+ "normalized_hypothesis": "we don't talk about that.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "we do not talk about that",
+ "normalized_hypothesis": "we do not talk about that",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0028",
+ "category": "technical",
+ "text": "Hash the UTF eight payload with S H A two fifty-six before opening the checkpoint.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0028.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Hash the UTF-8 payload with SH-256 before opening the checkpoint.",
+ "raw": {
+ "normalized_reference": "hash the utf eight payload with s h a two fifty-six before opening the checkpoint.",
+ "normalized_hypothesis": "hash the utf-8 payload with sh-256 before opening the checkpoint.",
+ "errors": 7,
+ "words": 16,
+ "wer": 0.4375
+ },
+ "semantic": {
+ "normalized_reference": "hash the utf 8 payload with s h a 256 before opening the checkpoint",
+ "normalized_hypothesis": "hash the utf 8 payload with sh 256 before opening the checkpoint",
+ "errors": 3,
+ "words": 14,
+ "wer": 0.21428571428571427
+ }
+ },
+ {
+ "id": "release500_0029",
+ "category": "conversational",
+ "text": "I thought I had it all figured out, and then this happened.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0029.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I thought I had it all figured out and then this happened.",
+ "raw": {
+ "normalized_reference": "i thought i had it all figured out, and then this happened.",
+ "normalized_hypothesis": "i thought i had it all figured out and then this happened.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i thought i had it all figured out and then this happened",
+ "normalized_hypothesis": "i thought i had it all figured out and then this happened",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0030",
+ "category": "descriptive",
+ "text": "The kind of quiet you only get at three in the morning in a city.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0030.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The kind of quiet you only get at three in the morning in a city.",
+ "raw": {
+ "normalized_reference": "the kind of quiet you only get at three in the morning in a city.",
+ "normalized_hypothesis": "the kind of quiet you only get at three in the morning in a city.",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the kind of quiet you only get at 3 in the morning in a city",
+ "normalized_hypothesis": "the kind of quiet you only get at 3 in the morning in a city",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0031",
+ "category": "emotional",
+ "text": "The joy of it was almost overwhelming, like too much good all at once.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0031.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The joy of it was almost overwhelming, like too much good all at once.",
+ "raw": {
+ "normalized_reference": "the joy of it was almost overwhelming, like too much good all at once.",
+ "normalized_hypothesis": "the joy of it was almost overwhelming, like too much good all at once.",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the joy of it was almost overwhelming like too much good all at once",
+ "normalized_hypothesis": "the joy of it was almost overwhelming like too much good all at once",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0032",
+ "category": "homographs",
+ "text": "Please close the door while I stand close to the microphone.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0032.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Please close the door while I stand close to the microphone.",
+ "raw": {
+ "normalized_reference": "please close the door while i stand close to the microphone.",
+ "normalized_hypothesis": "please close the door while i stand close to the microphone.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "please close the door while i stand close to the microphone",
+ "normalized_hypothesis": "please close the door while i stand close to the microphone",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0033",
+ "category": "long",
+ "text": "I think the most underrated skill in the world is knowing when to stop explaining yourself and just let people draw their own conclusions.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0033.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I think the most underrated skill in the world is knowing when to stop explaining yourself and just let people draw their own conclusions.",
+ "raw": {
+ "normalized_reference": "i think the most underrated skill in the world is knowing when to stop explaining yourself and just let people draw their own conclusions.",
+ "normalized_hypothesis": "i think the most underrated skill in the world is knowing when to stop explaining yourself and just let people draw their own conclusions.",
+ "errors": 0,
+ "words": 24,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i think the most underrated skill in the world is knowing when to stop explaining yourself and just let people draw their own conclusions",
+ "normalized_hypothesis": "i think the most underrated skill in the world is knowing when to stop explaining yourself and just let people draw their own conclusions",
+ "errors": 0,
+ "words": 24,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0034",
+ "category": "long_range",
+ "text": "When the final bus disappeared around the corner, Nadia remembered that her umbrella, notebook, and lunch were still inside the quiet office.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0034.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "When the final bus disappeared around the corner, Nadia remembered that her umbrella, notebook, and lunch were still inside the quiet office.",
+ "raw": {
+ "normalized_reference": "when the final bus disappeared around the corner, nadia remembered that her umbrella, notebook, and lunch were still inside the quiet office.",
+ "normalized_hypothesis": "when the final bus disappeared around the corner, nadia remembered that her umbrella, notebook, and lunch were still inside the quiet office.",
+ "errors": 0,
+ "words": 22,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "when the final bus disappeared around the corner nadia remembered that her umbrella notebook and lunch were still inside the quiet office",
+ "normalized_hypothesis": "when the final bus disappeared around the corner nadia remembered that her umbrella notebook and lunch were still inside the quiet office",
+ "errors": 0,
+ "words": 22,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0035",
+ "category": "names_places",
+ "text": "Aurelia and Joaquin crossed the Schuylkill River before noon.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0035.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Aurelia and Joaquin crossed the Schalkel River before noon.",
+ "raw": {
+ "normalized_reference": "aurelia and joaquin crossed the schuylkill river before noon.",
+ "normalized_hypothesis": "aurelia and joaquin crossed the schalkel river before noon.",
+ "errors": 1,
+ "words": 9,
+ "wer": 0.1111111111111111
+ },
+ "semantic": {
+ "normalized_reference": "aurelia and joaquin crossed the schuylkill river before noon",
+ "normalized_hypothesis": "aurelia and joaquin crossed the schalkel river before noon",
+ "errors": 1,
+ "words": 9,
+ "wer": 0.1111111111111111
+ }
+ },
+ {
+ "id": "release500_0036",
+ "category": "narrative",
+ "text": "The village had a name, but nobody used it. Everyone just called it home.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0036.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The village had a name, but nobody used it. Everyone just called it home.",
+ "raw": {
+ "normalized_reference": "the village had a name, but nobody used it. everyone just called it home.",
+ "normalized_hypothesis": "the village had a name, but nobody used it. everyone just called it home.",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the village had a name but nobody used it everyone just called it home",
+ "normalized_hypothesis": "the village had a name but nobody used it everyone just called it home",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0037",
+ "category": "numbers",
+ "text": "Increase the gain by three decibels, wait five seconds, then reduce it by one.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0037.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Increase the gain by 3 decibels, wait 5 seconds, then reduce it by 1.",
+ "raw": {
+ "normalized_reference": "increase the gain by three decibels, wait five seconds, then reduce it by one.",
+ "normalized_hypothesis": "increase the gain by 3 decibels, wait 5 seconds, then reduce it by 1.",
+ "errors": 3,
+ "words": 14,
+ "wer": 0.21428571428571427
+ },
+ "semantic": {
+ "normalized_reference": "increase the gain by 3 decibels wait 5 seconds then reduce it by one",
+ "normalized_hypothesis": "increase the gain by 3 decibels wait 5 seconds then reduce it by one",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0038",
+ "category": "punctuation",
+ "text": "The result was unexpected: clear, stable, and surprisingly natural.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0038.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The result was unexpected, clear, stable, and surprisingly natural.",
+ "raw": {
+ "normalized_reference": "the result was unexpected: clear, stable, and surprisingly natural.",
+ "normalized_hypothesis": "the result was unexpected, clear, stable, and surprisingly natural.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the result was unexpected clear stable and surprisingly natural",
+ "normalized_hypothesis": "the result was unexpected clear stable and surprisingly natural",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0039",
+ "category": "question",
+ "text": "Who told you that was the right way to do it?",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0039.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Who told you that was the right way to do it?",
+ "raw": {
+ "normalized_reference": "who told you that was the right way to do it?",
+ "normalized_hypothesis": "who told you that was the right way to do it?",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "who told you that was the right way to do it",
+ "normalized_hypothesis": "who told you that was the right way to do it",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0040",
+ "category": "rare_words",
+ "text": "The meticulous cartographer examined an iridescent archipelago through binoculars.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0040.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The meticulous cartographer examined an iridescent archipelago through binoculars.",
+ "raw": {
+ "normalized_reference": "the meticulous cartographer examined an iridescent archipelago through binoculars.",
+ "normalized_hypothesis": "the meticulous cartographer examined an iridescent archipelago through binoculars.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the meticulous cartographer examined an iridescent archipelago through binoculars",
+ "normalized_hypothesis": "the meticulous cartographer examined an iridescent archipelago through binoculars",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0041",
+ "category": "short",
+ "text": "I mean. Yeah. Kind of.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0041.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I mean, yes, kind of.",
+ "raw": {
+ "normalized_reference": "i mean. yeah. kind of.",
+ "normalized_hypothesis": "i mean, yes, kind of.",
+ "errors": 1,
+ "words": 5,
+ "wer": 0.2
+ },
+ "semantic": {
+ "normalized_reference": "i mean yeah kind of",
+ "normalized_hypothesis": "i mean yes kind of",
+ "errors": 1,
+ "words": 5,
+ "wer": 0.2
+ }
+ },
+ {
+ "id": "release500_0042",
+ "category": "technical",
+ "text": "The logs are your best friend when something behaves unexpectedly.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0042.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The logs are your best friend when something behaves unexpectedly.",
+ "raw": {
+ "normalized_reference": "the logs are your best friend when something behaves unexpectedly.",
+ "normalized_hypothesis": "the logs are your best friend when something behaves unexpectedly.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the logs are your best friend when something behaves unexpectedly",
+ "normalized_hypothesis": "the logs are your best friend when something behaves unexpectedly",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0043",
+ "category": "conversational",
+ "text": "He kept talking and I kept nodding and neither of us was listening.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0043.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He kept talking and I kept nodding and neither of us was listening.",
+ "raw": {
+ "normalized_reference": "he kept talking and i kept nodding and neither of us was listening.",
+ "normalized_hypothesis": "he kept talking and i kept nodding and neither of us was listening.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he kept talking and i kept nodding and neither of us was listening",
+ "normalized_hypothesis": "he kept talking and i kept nodding and neither of us was listening",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0044",
+ "category": "descriptive",
+ "text": "It's one of those places where everyone seems to know each other already.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0044.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It's one of those places where everyone seems to know each other already.",
+ "raw": {
+ "normalized_reference": "it's one of those places where everyone seems to know each other already.",
+ "normalized_hypothesis": "it's one of those places where everyone seems to know each other already.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it is one of those places where everyone seems to know each other already",
+ "normalized_hypothesis": "it is one of those places where everyone seems to know each other already",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0045",
+ "category": "emotional",
+ "text": "He said he was proud of me and nobody had ever said that to me before.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0045.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He said he was proud of me and nobody had ever said that to me before.",
+ "raw": {
+ "normalized_reference": "he said he was proud of me and nobody had ever said that to me before.",
+ "normalized_hypothesis": "he said he was proud of me and nobody had ever said that to me before.",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he said he was proud of me and nobody had ever said that to me before",
+ "normalized_hypothesis": "he said he was proud of me and nobody had ever said that to me before",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0046",
+ "category": "homographs",
+ "text": "The minute detail took a minute to explain clearly.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0046.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The minute detail took a minute to explain clearly.",
+ "raw": {
+ "normalized_reference": "the minute detail took a minute to explain clearly.",
+ "normalized_hypothesis": "the minute detail took a minute to explain clearly.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the minute detail took a minute to explain clearly",
+ "normalized_hypothesis": "the minute detail took a minute to explain clearly",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0047",
+ "category": "long",
+ "text": "The strange thing about really wanting something is that getting it never quite feels the way you imagined it would, and not getting it never quite feels as bad.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0047.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The strange thing about really wanting something is that getting it never quite feels the way you imagined it would, and not getting it never quite feels as bad.",
+ "raw": {
+ "normalized_reference": "the strange thing about really wanting something is that getting it never quite feels the way you imagined it would, and not getting it never quite feels as bad.",
+ "normalized_hypothesis": "the strange thing about really wanting something is that getting it never quite feels the way you imagined it would, and not getting it never quite feels as bad.",
+ "errors": 0,
+ "words": 29,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the strange thing about really wanting something is that getting it never quite feels the way you imagined it would and not getting it never quite feels as bad",
+ "normalized_hypothesis": "the strange thing about really wanting something is that getting it never quite feels the way you imagined it would and not getting it never quite feels as bad",
+ "errors": 0,
+ "words": 29,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0048",
+ "category": "long_range",
+ "text": "Even though the recipe looked simple at first, preparing the sauce required patience, careful timing, and a much larger pan.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0048.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Even though the recipe looked simple at first, preparing the sauce required patience, careful timing, and a much larger pan.",
+ "raw": {
+ "normalized_reference": "even though the recipe looked simple at first, preparing the sauce required patience, careful timing, and a much larger pan.",
+ "normalized_hypothesis": "even though the recipe looked simple at first, preparing the sauce required patience, careful timing, and a much larger pan.",
+ "errors": 0,
+ "words": 20,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "even though the recipe looked simple at 1st preparing the sauce required patience careful timing and a much larger pan",
+ "normalized_hypothesis": "even though the recipe looked simple at 1st preparing the sauce required patience careful timing and a much larger pan",
+ "errors": 0,
+ "words": 20,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0049",
+ "category": "names_places",
+ "text": "Xiomara met Siobhan near the old library in Worcestershire.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0049.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Zamora met Siobhan near the old library in Worcestershire.",
+ "raw": {
+ "normalized_reference": "xiomara met siobhan near the old library in worcestershire.",
+ "normalized_hypothesis": "zamora met siobhan near the old library in worcestershire.",
+ "errors": 1,
+ "words": 9,
+ "wer": 0.1111111111111111
+ },
+ "semantic": {
+ "normalized_reference": "xiomara met siobhan near the old library in worcestershire",
+ "normalized_hypothesis": "zamora met siobhan near the old library in worcestershire",
+ "errors": 1,
+ "words": 9,
+ "wer": 0.1111111111111111
+ }
+ },
+ {
+ "id": "release500_0050",
+ "category": "narrative",
+ "text": "She walked into the room and everyone went quiet, not because she was loud, but because she wasn't.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0050.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "She walked into the room and everyone went quiet, not because she was loud, but because she was.",
+ "raw": {
+ "normalized_reference": "she walked into the room and everyone went quiet, not because she was loud, but because she wasn't.",
+ "normalized_hypothesis": "she walked into the room and everyone went quiet, not because she was loud, but because she was.",
+ "errors": 1,
+ "words": 18,
+ "wer": 0.05555555555555555
+ },
+ "semantic": {
+ "normalized_reference": "she walked into the room and everyone went quiet not because she was loud but because she was not",
+ "normalized_hypothesis": "she walked into the room and everyone went quiet not because she was loud but because she was",
+ "errors": 1,
+ "words": 19,
+ "wer": 0.05263157894736842
+ }
+ },
+ {
+ "id": "release500_0051",
+ "category": "numbers",
+ "text": "At exactly seven forty-five, Morgan placed eighteen dollars and fifty cents beside the notebook.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0051.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "At exactly 7.45, Morgan placed $18.50 beside the notebook.",
+ "raw": {
+ "normalized_reference": "at exactly seven forty-five, morgan placed eighteen dollars and fifty cents beside the notebook.",
+ "normalized_hypothesis": "at exactly 7.45, morgan placed $18.50 beside the notebook.",
+ "errors": 8,
+ "words": 15,
+ "wer": 0.5333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "at exactly 745 morgan placed $18.50 beside the notebook",
+ "normalized_hypothesis": "at exactly 7.45 morgan placed $18.50 beside the notebook",
+ "errors": 2,
+ "words": 10,
+ "wer": 0.2
+ }
+ },
+ {
+ "id": "release500_0052",
+ "category": "punctuation",
+ "text": "Does punctuation work? Yes: commas, semicolons, and periods should create sensible boundaries.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0052.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Does punctuation work? Yes, commas, semicolons, and periods should create sensible boundaries.",
+ "raw": {
+ "normalized_reference": "does punctuation work? yes: commas, semicolons, and periods should create sensible boundaries.",
+ "normalized_hypothesis": "does punctuation work? yes, commas, semicolons, and periods should create sensible boundaries.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "does punctuation work yes commas semicolons and periods should create sensible boundaries",
+ "normalized_hypothesis": "does punctuation work yes commas semicolons and periods should create sensible boundaries",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0053",
+ "category": "question",
+ "text": "Have you ever just stopped and thought about how strange all of this is?",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0053.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Have you ever just stopped and thought about how strange all of this is?",
+ "raw": {
+ "normalized_reference": "have you ever just stopped and thought about how strange all of this is?",
+ "normalized_hypothesis": "have you ever just stopped and thought about how strange all of this is?",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "have you ever just stopped and thought about how strange all of this is",
+ "normalized_hypothesis": "have you ever just stopped and thought about how strange all of this is",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0054",
+ "category": "rare_words",
+ "text": "Her idiosyncratic explanation sounded plausible but remained fundamentally ambiguous.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0054.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Her idiosyncratic explanation sounded plausible, but remained fundamentally ambiguous.",
+ "raw": {
+ "normalized_reference": "her idiosyncratic explanation sounded plausible but remained fundamentally ambiguous.",
+ "normalized_hypothesis": "her idiosyncratic explanation sounded plausible, but remained fundamentally ambiguous.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "her idiosyncratic explanation sounded plausible but remained fundamentally ambiguous",
+ "normalized_hypothesis": "her idiosyncratic explanation sounded plausible but remained fundamentally ambiguous",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0055",
+ "category": "short",
+ "text": "That tracks.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0055.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "That tracks.",
+ "raw": {
+ "normalized_reference": "that tracks.",
+ "normalized_hypothesis": "that tracks.",
+ "errors": 0,
+ "words": 2,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "that tracks",
+ "normalized_hypothesis": "that tracks",
+ "errors": 0,
+ "words": 2,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0056",
+ "category": "technical",
+ "text": "The key thing to remember is that order matters here more than you'd think.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0056.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The key thing to remember is that order matters here more than you'd think.",
+ "raw": {
+ "normalized_reference": "the key thing to remember is that order matters here more than you'd think.",
+ "normalized_hypothesis": "the key thing to remember is that order matters here more than you'd think.",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the key thing to remember is that order matters here more than you would think",
+ "normalized_hypothesis": "the key thing to remember is that order matters here more than you would think",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0057",
+ "category": "conversational",
+ "text": "I would not say that again in this room if I were you.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0057.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I would not say that again in this room if I were you.",
+ "raw": {
+ "normalized_reference": "i would not say that again in this room if i were you.",
+ "normalized_hypothesis": "i would not say that again in this room if i were you.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i would not say that again in this room if i were you",
+ "normalized_hypothesis": "i would not say that again in this room if i were you",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0058",
+ "category": "descriptive",
+ "text": "The room was filled with the quiet that means something is about to change.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0058.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The room was filled with the quiet that means something is about to change.",
+ "raw": {
+ "normalized_reference": "the room was filled with the quiet that means something is about to change.",
+ "normalized_hypothesis": "the room was filled with the quiet that means something is about to change.",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the room was filled with the quiet that means something is about to change",
+ "normalized_hypothesis": "the room was filled with the quiet that means something is about to change",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0059",
+ "category": "emotional",
+ "text": "I didn't expect to feel that way about it, but I did.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0059.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I didn't expect to feel that way about it, but I did.",
+ "raw": {
+ "normalized_reference": "i didn't expect to feel that way about it, but i did.",
+ "normalized_hypothesis": "i didn't expect to feel that way about it, but i did.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i did not expect to feel that way about it but i did",
+ "normalized_hypothesis": "i did not expect to feel that way about it but i did",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0060",
+ "category": "homographs",
+ "text": "I object to the strange object that was left near the entrance.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0060.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I object to the strange object that was left near the entrance.",
+ "raw": {
+ "normalized_reference": "i object to the strange object that was left near the entrance.",
+ "normalized_hypothesis": "i object to the strange object that was left near the entrance.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i object to the strange object that was left near the entrance",
+ "normalized_hypothesis": "i object to the strange object that was left near the entrance",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0061",
+ "category": "long",
+ "text": "I've learned more from the things that went wrong than from the things that worked, which is a frustrating way to have to learn things but apparently the only way that stuck.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0061.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I've learned more from the things that went wrong than from the things that worked, which is a frustrating way to have to learn things, but apparently the only way that stuck.",
+ "raw": {
+ "normalized_reference": "i've learned more from the things that went wrong than from the things that worked, which is a frustrating way to have to learn things but apparently the only way that stuck.",
+ "normalized_hypothesis": "i've learned more from the things that went wrong than from the things that worked, which is a frustrating way to have to learn things, but apparently the only way that stuck.",
+ "errors": 0,
+ "words": 32,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i have learned more from the things that went wrong than from the things that worked which is a frustrating way to have to learn things but apparently the only way that stuck",
+ "normalized_hypothesis": "i have learned more from the things that went wrong than from the things that worked which is a frustrating way to have to learn things but apparently the only way that stuck",
+ "errors": 0,
+ "words": 33,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0062",
+ "category": "long_range",
+ "text": "Because every nearby hotel was full, the exhausted hikers followed the narrow road until they found a farmhouse with one empty room.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0062.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Because every nearby hotel was full, the exhausted hikers followed the narrow road until they found a farmhouse with one empty room.",
+ "raw": {
+ "normalized_reference": "because every nearby hotel was full, the exhausted hikers followed the narrow road until they found a farmhouse with one empty room.",
+ "normalized_hypothesis": "because every nearby hotel was full, the exhausted hikers followed the narrow road until they found a farmhouse with one empty room.",
+ "errors": 0,
+ "words": 22,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "because every nearby hotel was full the exhausted hikers followed the narrow road until they found a farmhouse with one empty room",
+ "normalized_hypothesis": "because every nearby hotel was full the exhausted hikers followed the narrow road until they found a farmhouse with one empty room",
+ "errors": 0,
+ "words": 22,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0063",
+ "category": "names_places",
+ "text": "The itinerary sends Beatrice from Albuquerque to Des Moines and then to Boise.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0063.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The itinerary sends Beatrice from Albuquerque to Des Moines and then to Boise.",
+ "raw": {
+ "normalized_reference": "the itinerary sends beatrice from albuquerque to des moines and then to boise.",
+ "normalized_hypothesis": "the itinerary sends beatrice from albuquerque to des moines and then to boise.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the itinerary sends beatrice from albuquerque to des moines and then to boise",
+ "normalized_hypothesis": "the itinerary sends beatrice from albuquerque to des moines and then to boise",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0064",
+ "category": "narrative",
+ "text": "We had been driving for six hours when we finally saw the sign for the town.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0064.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "We had been driving for six hours when we finally saw the sign for the town.",
+ "raw": {
+ "normalized_reference": "we had been driving for six hours when we finally saw the sign for the town.",
+ "normalized_hypothesis": "we had been driving for six hours when we finally saw the sign for the town.",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "we had been driving for 6 hours when we finally saw the sign for the town",
+ "normalized_hypothesis": "we had been driving for 6 hours when we finally saw the sign for the town",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0065",
+ "category": "numbers",
+ "text": "Call extension zero four nine after eight thirty in the morning.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0065.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Call extension 049 after 8.30 in the morning.",
+ "raw": {
+ "normalized_reference": "call extension zero four nine after eight thirty in the morning.",
+ "normalized_hypothesis": "call extension 049 after 8.30 in the morning.",
+ "errors": 5,
+ "words": 11,
+ "wer": 0.45454545454545453
+ },
+ "semantic": {
+ "normalized_reference": "call extension 049 after 830 in the morning",
+ "normalized_hypothesis": "call extension 49 after 8.30 in the morning",
+ "errors": 3,
+ "words": 8,
+ "wer": 0.375
+ }
+ },
+ {
+ "id": "release500_0066",
+ "category": "punctuation",
+ "text": "Stop - listen carefully - then answer the question without guessing.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0066.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Stop, listen carefully, then answer the question without guessing.",
+ "raw": {
+ "normalized_reference": "stop - listen carefully - then answer the question without guessing.",
+ "normalized_hypothesis": "stop, listen carefully, then answer the question without guessing.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "stop listen carefully then answer the question without guessing",
+ "normalized_hypothesis": "stop listen carefully then answer the question without guessing",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0067",
+ "category": "question",
+ "text": "Does it matter at this point, honestly?",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0067.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Does it matter at this point, honestly?",
+ "raw": {
+ "normalized_reference": "does it matter at this point, honestly?",
+ "normalized_hypothesis": "does it matter at this point, honestly?",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "does it matter at this point honestly",
+ "normalized_hypothesis": "does it matter at this point honestly",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0068",
+ "category": "rare_words",
+ "text": "An entrepreneurial meteorologist described the unprecedented atmospheric phenomenon.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0068.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "An entrepreneurial meteorologist described the unprecedented atmospheric phenomenon.",
+ "raw": {
+ "normalized_reference": "an entrepreneurial meteorologist described the unprecedented atmospheric phenomenon.",
+ "normalized_hypothesis": "an entrepreneurial meteorologist described the unprecedented atmospheric phenomenon.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "an entrepreneurial meteorologist described the unprecedented atmospheric phenomenon",
+ "normalized_hypothesis": "an entrepreneurial meteorologist described the unprecedented atmospheric phenomenon",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0069",
+ "category": "short",
+ "text": "That's not what I expected.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0069.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "That's not what I expected.",
+ "raw": {
+ "normalized_reference": "that's not what i expected.",
+ "normalized_hypothesis": "that's not what i expected.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "that is not what i expected",
+ "normalized_hypothesis": "that is not what i expected",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0070",
+ "category": "technical",
+ "text": "There's a known issue with that version that causes failures under load.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0070.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "There's a known issue with that version that causes failures under load.",
+ "raw": {
+ "normalized_reference": "there's a known issue with that version that causes failures under load.",
+ "normalized_hypothesis": "there's a known issue with that version that causes failures under load.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "there is a known issue with that version that causes failures under load",
+ "normalized_hypothesis": "there is a known issue with that version that causes failures under load",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0071",
+ "category": "conversational",
+ "text": "Don't look now but I think that's him over there.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0071.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Don't look now, but I think that's him over there.",
+ "raw": {
+ "normalized_reference": "don't look now but i think that's him over there.",
+ "normalized_hypothesis": "don't look now, but i think that's him over there.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "do not look now but i think that is him over there",
+ "normalized_hypothesis": "do not look now but i think that is him over there",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0072",
+ "category": "descriptive",
+ "text": "The city at night looks completely different from how it feels during the day.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0072.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The city at night looks completely different from how it feels during the day.",
+ "raw": {
+ "normalized_reference": "the city at night looks completely different from how it feels during the day.",
+ "normalized_hypothesis": "the city at night looks completely different from how it feels during the day.",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the city at night looks completely different from how it feels during the day",
+ "normalized_hypothesis": "the city at night looks completely different from how it feels during the day",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0073",
+ "category": "emotional",
+ "text": "It's the kind of happy that makes you nervous because you know it can't last.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0073.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It's the kind of happy that makes you nervous because you know it can't last.",
+ "raw": {
+ "normalized_reference": "it's the kind of happy that makes you nervous because you know it can't last.",
+ "normalized_hypothesis": "it's the kind of happy that makes you nervous because you know it can't last.",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it is the kind of happy that makes you nervous because you know it can not last",
+ "normalized_hypothesis": "it is the kind of happy that makes you nervous because you know it can not last",
+ "errors": 0,
+ "words": 17,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0074",
+ "category": "homographs",
+ "text": "The dove dove beneath the branch while the wind began to wind the loose cable.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0074.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The dove dove beneath the branch while the wind began to wind the loose cable.",
+ "raw": {
+ "normalized_reference": "the dove dove beneath the branch while the wind began to wind the loose cable.",
+ "normalized_hypothesis": "the dove dove beneath the branch while the wind began to wind the loose cable.",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the dove dove beneath the branch while the wind began to wind the loose cable",
+ "normalized_hypothesis": "the dove dove beneath the branch while the wind began to wind the loose cable",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0075",
+ "category": "long",
+ "text": "I've thought a lot about what I would do differently if I could go back, and the honest answer is probably not as much as I'd like to think.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0075.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I've thought a lot about what I would do differently if I could go back, and the honest answer is probably not as much as I'd like to think.",
+ "raw": {
+ "normalized_reference": "i've thought a lot about what i would do differently if i could go back, and the honest answer is probably not as much as i'd like to think.",
+ "normalized_hypothesis": "i've thought a lot about what i would do differently if i could go back, and the honest answer is probably not as much as i'd like to think.",
+ "errors": 0,
+ "words": 29,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i have thought a lot about what i would do differently if i could go back and the honest answer is probably not as much as i would like to think",
+ "normalized_hypothesis": "i have thought a lot about what i would do differently if i could go back and the honest answer is probably not as much as i would like to think",
+ "errors": 0,
+ "words": 31,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0076",
+ "category": "long_range",
+ "text": "Although the weather changed without warning, the musicians finished packing before anyone opened the enormous glass door.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0076.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Although the weather changed without warning, the musicians finished packing before anyone opened the enormous glass door.",
+ "raw": {
+ "normalized_reference": "although the weather changed without warning, the musicians finished packing before anyone opened the enormous glass door.",
+ "normalized_hypothesis": "although the weather changed without warning, the musicians finished packing before anyone opened the enormous glass door.",
+ "errors": 0,
+ "words": 17,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "although the weather changed without warning the musicians finished packing before anyone opened the enormous glass door",
+ "normalized_hypothesis": "although the weather changed without warning the musicians finished packing before anyone opened the enormous glass door",
+ "errors": 0,
+ "words": 17,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0077",
+ "category": "names_places",
+ "text": "Saskatchewan looked unusually bright beneath the pale February sunrise.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0077.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Saskatchewan looked unusually bright beneath the pale February sunrise.",
+ "raw": {
+ "normalized_reference": "saskatchewan looked unusually bright beneath the pale february sunrise.",
+ "normalized_hypothesis": "saskatchewan looked unusually bright beneath the pale february sunrise.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "saskatchewan looked unusually bright beneath the pale february sunrise",
+ "normalized_hypothesis": "saskatchewan looked unusually bright beneath the pale february sunrise",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0078",
+ "category": "narrative",
+ "text": "She handled it with a grace I genuinely did not expect.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0078.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "She handled it with a grace I genuinely did not expect.",
+ "raw": {
+ "normalized_reference": "she handled it with a grace i genuinely did not expect.",
+ "normalized_hypothesis": "she handled it with a grace i genuinely did not expect.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "she handled it with a grace i genuinely did not expect",
+ "normalized_hypothesis": "she handled it with a grace i genuinely did not expect",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0079",
+ "category": "numbers",
+ "text": "Apartment four hundred seven is across from room twenty-nine on the third floor.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0079.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Apartment 407 is across from room 29 on the third floor.",
+ "raw": {
+ "normalized_reference": "apartment four hundred seven is across from room twenty-nine on the third floor.",
+ "normalized_hypothesis": "apartment 407 is across from room 29 on the third floor.",
+ "errors": 5,
+ "words": 14,
+ "wer": 0.35714285714285715
+ },
+ "semantic": {
+ "normalized_reference": "apartment 407 is across from room 29 on the 3rd floor",
+ "normalized_hypothesis": "apartment 407 is across from room 29 on the 3rd floor",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0080",
+ "category": "punctuation",
+ "text": "\"Not yet,\" Rowan whispered. \"Give me another minute... please.\"",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0080.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Not yet, Rowan whispered. Give me another minute. Please.",
+ "raw": {
+ "normalized_reference": "\"not yet,\" rowan whispered. \"give me another minute... please.\"",
+ "normalized_hypothesis": "not yet, rowan whispered. give me another minute. please.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "not yet rowan whispered give me another minute please",
+ "normalized_hypothesis": "not yet rowan whispered give me another minute please",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0081",
+ "category": "question",
+ "text": "What are we actually trying to accomplish here?",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0081.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "What are we actually trying to accomplish here?",
+ "raw": {
+ "normalized_reference": "what are we actually trying to accomplish here?",
+ "normalized_hypothesis": "what are we actually trying to accomplish here?",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "what are we actually trying to accomplish here",
+ "normalized_hypothesis": "what are we actually trying to accomplish here",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0082",
+ "category": "rare_words",
+ "text": "The physician recommended a noninvasive procedure for the recurring neurological symptoms.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0082.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The physician recommended a non-invasive procedure for the recurring neurological symptoms.",
+ "raw": {
+ "normalized_reference": "the physician recommended a noninvasive procedure for the recurring neurological symptoms.",
+ "normalized_hypothesis": "the physician recommended a non-invasive procedure for the recurring neurological symptoms.",
+ "errors": 2,
+ "words": 11,
+ "wer": 0.18181818181818182
+ },
+ "semantic": {
+ "normalized_reference": "the physician recommended a noninvasive procedure for the recurring neurological symptoms",
+ "normalized_hypothesis": "the physician recommended a non invasive procedure for the recurring neurological symptoms",
+ "errors": 2,
+ "words": 11,
+ "wer": 0.18181818181818182
+ }
+ },
+ {
+ "id": "release500_0083",
+ "category": "short",
+ "text": "Are you serious right now?",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0083.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Are you serious right now?",
+ "raw": {
+ "normalized_reference": "are you serious right now?",
+ "normalized_hypothesis": "are you serious right now?",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "are you serious right now",
+ "normalized_hypothesis": "are you serious right now",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0084",
+ "category": "technical",
+ "text": "The performance improvement is real, but it comes with tradeoffs.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0084.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The performance improvement is real, but it comes with trade-offs.",
+ "raw": {
+ "normalized_reference": "the performance improvement is real, but it comes with tradeoffs.",
+ "normalized_hypothesis": "the performance improvement is real, but it comes with trade-offs.",
+ "errors": 2,
+ "words": 10,
+ "wer": 0.2
+ },
+ "semantic": {
+ "normalized_reference": "the performance improvement is real but it comes with tradeoffs",
+ "normalized_hypothesis": "the performance improvement is real but it comes with trade offs",
+ "errors": 2,
+ "words": 10,
+ "wer": 0.2
+ }
+ },
+ {
+ "id": "release500_0085",
+ "category": "conversational",
+ "text": "I don't know. It's just been one of those weeks, you know?",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0085.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I don't know. It's just been one of those weeks, you know.",
+ "raw": {
+ "normalized_reference": "i don't know. it's just been one of those weeks, you know?",
+ "normalized_hypothesis": "i don't know. it's just been one of those weeks, you know.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i do not know it is just been one of those weeks you know",
+ "normalized_hypothesis": "i do not know it is just been one of those weeks you know",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0086",
+ "category": "descriptive",
+ "text": "The moment you stop expecting perfection, things get a lot more interesting.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0086.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The moment you stop expecting perfection, things get a lot more interesting.",
+ "raw": {
+ "normalized_reference": "the moment you stop expecting perfection, things get a lot more interesting.",
+ "normalized_hypothesis": "the moment you stop expecting perfection, things get a lot more interesting.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the moment you stop expecting perfection things get a lot more interesting",
+ "normalized_hypothesis": "the moment you stop expecting perfection things get a lot more interesting",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0087",
+ "category": "emotional",
+ "text": "It's the kind of thing that stays with you even years later.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0087.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It's the kind of thing that stays with you even years later.",
+ "raw": {
+ "normalized_reference": "it's the kind of thing that stays with you even years later.",
+ "normalized_hypothesis": "it's the kind of thing that stays with you even years later.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it is the kind of thing that stays with you even years later",
+ "normalized_hypothesis": "it is the kind of thing that stays with you even years later",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0088",
+ "category": "long",
+ "text": "She told me once that the most important skill she'd developed was learning to disagree with someone without making them feel they were wrong.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0088.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "She told me once that the most important skill she'd developed was learning to disagree with someone without making them feel they were wrong.",
+ "raw": {
+ "normalized_reference": "she told me once that the most important skill she'd developed was learning to disagree with someone without making them feel they were wrong.",
+ "normalized_hypothesis": "she told me once that the most important skill she'd developed was learning to disagree with someone without making them feel they were wrong.",
+ "errors": 0,
+ "words": 24,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "she told me once that the most important skill she would developed was learning to disagree with someone without making them feel they were wrong",
+ "normalized_hypothesis": "she told me once that the most important skill she would developed was learning to disagree with someone without making them feel they were wrong",
+ "errors": 0,
+ "words": 25,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0089",
+ "category": "narrative",
+ "text": "He stood at the edge of the cliff and wondered, not for the first time, how he'd gotten here.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0089.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He stood at the edge of the cliff and wondered, not for the first time, how he'd gotten here.",
+ "raw": {
+ "normalized_reference": "he stood at the edge of the cliff and wondered, not for the first time, how he'd gotten here.",
+ "normalized_hypothesis": "he stood at the edge of the cliff and wondered, not for the first time, how he'd gotten here.",
+ "errors": 0,
+ "words": 19,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he stood at the edge of the cliff and wondered not for the 1st time how he would gotten here",
+ "normalized_hypothesis": "he stood at the edge of the cliff and wondered not for the 1st time how he would gotten here",
+ "errors": 0,
+ "words": 20,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0090",
+ "category": "question",
+ "text": "Why does asking for help feel so hard when you actually need it?",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0090.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Why does asking for help feel so hard when you actually need it?",
+ "raw": {
+ "normalized_reference": "why does asking for help feel so hard when you actually need it?",
+ "normalized_hypothesis": "why does asking for help feel so hard when you actually need it?",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "why does asking for help feel so hard when you actually need it",
+ "normalized_hypothesis": "why does asking for help feel so hard when you actually need it",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0091",
+ "category": "short",
+ "text": "That's fair.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0091.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "That's fair.",
+ "raw": {
+ "normalized_reference": "that's fair.",
+ "normalized_hypothesis": "that's fair.",
+ "errors": 0,
+ "words": 2,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "that is fair",
+ "normalized_hypothesis": "that is fair",
+ "errors": 0,
+ "words": 3,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0092",
+ "category": "technical",
+ "text": "If you run into that error, the fix is usually just restarting the service.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0092.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "If you run into that error, the fix is usually just restarting the service.",
+ "raw": {
+ "normalized_reference": "if you run into that error, the fix is usually just restarting the service.",
+ "normalized_hypothesis": "if you run into that error, the fix is usually just restarting the service.",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "if you run into that error the fix is usually just restarting the service",
+ "normalized_hypothesis": "if you run into that error the fix is usually just restarting the service",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0093",
+ "category": "conversational",
+ "text": "It was a disaster in the best possible way.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0093.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It was a disaster in the best possible way.",
+ "raw": {
+ "normalized_reference": "it was a disaster in the best possible way.",
+ "normalized_hypothesis": "it was a disaster in the best possible way.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it was a disaster in the best possible way",
+ "normalized_hypothesis": "it was a disaster in the best possible way",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0094",
+ "category": "descriptive",
+ "text": "It had the look of a place that used to mean something to someone.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0094.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It had the look of a place that used to mean something to someone.",
+ "raw": {
+ "normalized_reference": "it had the look of a place that used to mean something to someone.",
+ "normalized_hypothesis": "it had the look of a place that used to mean something to someone.",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it had the look of a place that used to mean something to someone",
+ "normalized_hypothesis": "it had the look of a place that used to mean something to someone",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0095",
+ "category": "emotional",
+ "text": "There's this feeling you get when something finally works after weeks of failure.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0095.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "There's this feeling you get when something finally works after weeks of failure.",
+ "raw": {
+ "normalized_reference": "there's this feeling you get when something finally works after weeks of failure.",
+ "normalized_hypothesis": "there's this feeling you get when something finally works after weeks of failure.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "there is this feeling you get when something finally works after weeks of failure",
+ "normalized_hypothesis": "there is this feeling you get when something finally works after weeks of failure",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0096",
+ "category": "long",
+ "text": "He had a theory that most arguments aren't actually about what they seem to be about, and the longer I live the more I think he was onto something.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0096.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He had a theory that most arguments aren't actually about what they seem to be about, and the longer I lived the more I think he was onto something.",
+ "raw": {
+ "normalized_reference": "he had a theory that most arguments aren't actually about what they seem to be about, and the longer i live the more i think he was onto something.",
+ "normalized_hypothesis": "he had a theory that most arguments aren't actually about what they seem to be about, and the longer i lived the more i think he was onto something.",
+ "errors": 1,
+ "words": 29,
+ "wer": 0.034482758620689655
+ },
+ "semantic": {
+ "normalized_reference": "he had a theory that most arguments are not actually about what they seem to be about and the longer i live the more i think he was onto something",
+ "normalized_hypothesis": "he had a theory that most arguments are not actually about what they seem to be about and the longer i lived the more i think he was onto something",
+ "errors": 1,
+ "words": 30,
+ "wer": 0.03333333333333333
+ }
+ },
+ {
+ "id": "release500_0097",
+ "category": "narrative",
+ "text": "The thing nobody tells you about starting over is how ordinary most of it feels.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0097.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The thing nobody tells you about starting over is how ordinary most of it feels.",
+ "raw": {
+ "normalized_reference": "the thing nobody tells you about starting over is how ordinary most of it feels.",
+ "normalized_hypothesis": "the thing nobody tells you about starting over is how ordinary most of it feels.",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the thing nobody tells you about starting over is how ordinary most of it feels",
+ "normalized_hypothesis": "the thing nobody tells you about starting over is how ordinary most of it feels",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0098",
+ "category": "question",
+ "text": "What exactly are we waiting for at this point?",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0098.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "What exactly are we waiting for at this point?",
+ "raw": {
+ "normalized_reference": "what exactly are we waiting for at this point?",
+ "normalized_hypothesis": "what exactly are we waiting for at this point?",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "what exactly are we waiting for at this point",
+ "normalized_hypothesis": "what exactly are we waiting for at this point",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0099",
+ "category": "short",
+ "text": "I stand by what I said.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0099.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I stand by what I said.",
+ "raw": {
+ "normalized_reference": "i stand by what i said.",
+ "normalized_hypothesis": "i stand by what i said.",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i stand by what i said",
+ "normalized_hypothesis": "i stand by what i said",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0100",
+ "category": "technical",
+ "text": "Double-check that value before you push anything to production.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0100.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Double check that value before you push anything to production.",
+ "raw": {
+ "normalized_reference": "double-check that value before you push anything to production.",
+ "normalized_hypothesis": "double check that value before you push anything to production.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "double check that value before you push anything to production",
+ "normalized_hypothesis": "double check that value before you push anything to production",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0101",
+ "category": "conversational",
+ "text": "I understand your point; I just do not agree with the conclusion.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0101.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I understand your point, I just do not agree with the conclusion.",
+ "raw": {
+ "normalized_reference": "i understand your point; i just do not agree with the conclusion.",
+ "normalized_hypothesis": "i understand your point, i just do not agree with the conclusion.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i understand your point i just do not agree with the conclusion",
+ "normalized_hypothesis": "i understand your point i just do not agree with the conclusion",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0102",
+ "category": "descriptive",
+ "text": "The sky that day was a strange shade of orange I'd never seen before.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0102.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The sky that day was a strange shade of orange I'd never seen before.",
+ "raw": {
+ "normalized_reference": "the sky that day was a strange shade of orange i'd never seen before.",
+ "normalized_hypothesis": "the sky that day was a strange shade of orange i'd never seen before.",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the sky that day was a strange shade of orange i would never seen before",
+ "normalized_hypothesis": "the sky that day was a strange shade of orange i would never seen before",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0103",
+ "category": "emotional",
+ "text": "It hurt more than I expected, and I thought I was prepared.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0103.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It hurt more than I expected, and I thought I was prepared.",
+ "raw": {
+ "normalized_reference": "it hurt more than i expected, and i thought i was prepared.",
+ "normalized_hypothesis": "it hurt more than i expected, and i thought i was prepared.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it hurt more than i expected and i thought i was prepared",
+ "normalized_hypothesis": "it hurt more than i expected and i thought i was prepared",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0104",
+ "category": "long",
+ "text": "The hardest part wasn't making the decision, it was sitting with the uncertainty of not knowing whether it was the right one for months afterward.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0104.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The hardest part was, making the decision, it was sitting with the uncertainty of not knowing whether it was the right one for months afterward.",
+ "raw": {
+ "normalized_reference": "the hardest part wasn't making the decision, it was sitting with the uncertainty of not knowing whether it was the right one for months afterward.",
+ "normalized_hypothesis": "the hardest part was, making the decision, it was sitting with the uncertainty of not knowing whether it was the right one for months afterward.",
+ "errors": 1,
+ "words": 25,
+ "wer": 0.04
+ },
+ "semantic": {
+ "normalized_reference": "the hardest part was not making the decision it was sitting with the uncertainty of not knowing whether it was the right one for months afterward",
+ "normalized_hypothesis": "the hardest part was making the decision it was sitting with the uncertainty of not knowing whether it was the right one for months afterward",
+ "errors": 1,
+ "words": 26,
+ "wer": 0.038461538461538464
+ }
+ },
+ {
+ "id": "release500_0105",
+ "category": "narrative",
+ "text": "There was this one day last summer when everything just clicked into place.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0105.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "There was this one day last summer when everything just clicked into place.",
+ "raw": {
+ "normalized_reference": "there was this one day last summer when everything just clicked into place.",
+ "normalized_hypothesis": "there was this one day last summer when everything just clicked into place.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "there was this one day last summer when everything just clicked into place",
+ "normalized_hypothesis": "there was this one day last summer when everything just clicked into place",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0106",
+ "category": "question",
+ "text": "Have you ever thought about why we keep doing things the same way when they stop working?",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0106.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Have you ever thought about why we keep doing things the same way when they stop working?",
+ "raw": {
+ "normalized_reference": "have you ever thought about why we keep doing things the same way when they stop working?",
+ "normalized_hypothesis": "have you ever thought about why we keep doing things the same way when they stop working?",
+ "errors": 0,
+ "words": 17,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "have you ever thought about why we keep doing things the same way when they stop working",
+ "normalized_hypothesis": "have you ever thought about why we keep doing things the same way when they stop working",
+ "errors": 0,
+ "words": 17,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0107",
+ "category": "short",
+ "text": "I can't believe you just said that.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0107.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I can't believe you just said that.",
+ "raw": {
+ "normalized_reference": "i can't believe you just said that.",
+ "normalized_hypothesis": "i can't believe you just said that.",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i can not believe you just said that",
+ "normalized_hypothesis": "i can not believe you just said that",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0108",
+ "category": "technical",
+ "text": "You'll want to isolate that variable before drawing any conclusions.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0108.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "You'll want to isolate that variable before drawing any conclusions.",
+ "raw": {
+ "normalized_reference": "you'll want to isolate that variable before drawing any conclusions.",
+ "normalized_hypothesis": "you'll want to isolate that variable before drawing any conclusions.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "you will want to isolate that variable before drawing any conclusions",
+ "normalized_hypothesis": "you will want to isolate that variable before drawing any conclusions",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0109",
+ "category": "conversational",
+ "text": "The meeting that could have been an email lasted two hours.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0109.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The meeting that could have been an email lasted two hours.",
+ "raw": {
+ "normalized_reference": "the meeting that could have been an email lasted two hours.",
+ "normalized_hypothesis": "the meeting that could have been an email lasted two hours.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the meeting that could have been an email lasted 2 hours",
+ "normalized_hypothesis": "the meeting that could have been an email lasted 2 hours",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0110",
+ "category": "descriptive",
+ "text": "There's a specific kind of tired that sleep doesn't fix.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0110.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "There's a specific kind of tired that sleep doesn't fix.",
+ "raw": {
+ "normalized_reference": "there's a specific kind of tired that sleep doesn't fix.",
+ "normalized_hypothesis": "there's a specific kind of tired that sleep doesn't fix.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "there is a specific kind of tired that sleep does not fix",
+ "normalized_hypothesis": "there is a specific kind of tired that sleep does not fix",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0111",
+ "category": "emotional",
+ "text": "Something about that moment broke something open in me that I hadn't realized was closed.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0111.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Something about that moment broke something open in me that I hadn't realized was closed.",
+ "raw": {
+ "normalized_reference": "something about that moment broke something open in me that i hadn't realized was closed.",
+ "normalized_hypothesis": "something about that moment broke something open in me that i hadn't realized was closed.",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "something about that moment broke something open in me that i had not realized was closed",
+ "normalized_hypothesis": "something about that moment broke something open in me that i had not realized was closed",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0112",
+ "category": "long",
+ "text": "The thing about making something good is that you don't usually know it's good while you're making it. You just know you're not done yet.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0112.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The thing about making something good is that you don't usually know it's good while you're making it. You just know you're not done yet.",
+ "raw": {
+ "normalized_reference": "the thing about making something good is that you don't usually know it's good while you're making it. you just know you're not done yet.",
+ "normalized_hypothesis": "the thing about making something good is that you don't usually know it's good while you're making it. you just know you're not done yet.",
+ "errors": 0,
+ "words": 25,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the thing about making something good is that you do not usually know it is good while you are making it you just know you are not done yet",
+ "normalized_hypothesis": "the thing about making something good is that you do not usually know it is good while you are making it you just know you are not done yet",
+ "errors": 0,
+ "words": 29,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0113",
+ "category": "narrative",
+ "text": "She opened the door slowly, as if she expected to find something she wasn't ready for.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0113.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "She opened the door slowly, as if she expected to find something she was ready for.",
+ "raw": {
+ "normalized_reference": "she opened the door slowly, as if she expected to find something she wasn't ready for.",
+ "normalized_hypothesis": "she opened the door slowly, as if she expected to find something she was ready for.",
+ "errors": 1,
+ "words": 16,
+ "wer": 0.0625
+ },
+ "semantic": {
+ "normalized_reference": "she opened the door slowly as if she expected to find something she was not ready for",
+ "normalized_hypothesis": "she opened the door slowly as if she expected to find something she was ready for",
+ "errors": 1,
+ "words": 17,
+ "wer": 0.058823529411764705
+ }
+ },
+ {
+ "id": "release500_0114",
+ "category": "question",
+ "text": "What would you do differently if you could go back?",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0114.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "What would you do differently if you could go back?",
+ "raw": {
+ "normalized_reference": "what would you do differently if you could go back?",
+ "normalized_hypothesis": "what would you do differently if you could go back?",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "what would you do differently if you could go back",
+ "normalized_hypothesis": "what would you do differently if you could go back",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0115",
+ "category": "short",
+ "text": "Okay. Okay okay okay.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0115.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Okay. Okay, okay, okay.",
+ "raw": {
+ "normalized_reference": "okay. okay okay okay.",
+ "normalized_hypothesis": "okay. okay, okay, okay.",
+ "errors": 0,
+ "words": 4,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "okay okay okay okay",
+ "normalized_hypothesis": "okay okay okay okay",
+ "errors": 0,
+ "words": 4,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0116",
+ "category": "technical",
+ "text": "The latency goes up significantly at scale, which is expected.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0116.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The latency goes up significantly at scale, which is expected.",
+ "raw": {
+ "normalized_reference": "the latency goes up significantly at scale, which is expected.",
+ "normalized_hypothesis": "the latency goes up significantly at scale, which is expected.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the latency goes up significantly at scale which is expected",
+ "normalized_hypothesis": "the latency goes up significantly at scale which is expected",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0117",
+ "category": "conversational",
+ "text": "We ended up staying until two in the morning talking about nothing.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0117.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "We ended up staying until two in the morning, talking about nothing.",
+ "raw": {
+ "normalized_reference": "we ended up staying until two in the morning talking about nothing.",
+ "normalized_hypothesis": "we ended up staying until two in the morning, talking about nothing.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "we ended up staying until 2 in the morning talking about nothing",
+ "normalized_hypothesis": "we ended up staying until 2 in the morning talking about nothing",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0118",
+ "category": "descriptive",
+ "text": "The rain started slowly, and then all at once, the way it usually does.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0118.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The rain started slowly and then all at once, the way it usually does.",
+ "raw": {
+ "normalized_reference": "the rain started slowly, and then all at once, the way it usually does.",
+ "normalized_hypothesis": "the rain started slowly and then all at once, the way it usually does.",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the rain started slowly and then all at once the way it usually does",
+ "normalized_hypothesis": "the rain started slowly and then all at once the way it usually does",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0119",
+ "category": "emotional",
+ "text": "I honestly did not think I was going to make it through that year.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0119.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I honestly did not think I was going to make it through that year.",
+ "raw": {
+ "normalized_reference": "i honestly did not think i was going to make it through that year.",
+ "normalized_hypothesis": "i honestly did not think i was going to make it through that year.",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i honestly did not think i was going to make it through that year",
+ "normalized_hypothesis": "i honestly did not think i was going to make it through that year",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0120",
+ "category": "long",
+ "text": "There's an assumption that people who seem confident all the time actually feel confident, and in my experience that's almost never true.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0120.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "There's an assumption that people who seem confident all the time actually feel confident, and in my experience that's almost never true.",
+ "raw": {
+ "normalized_reference": "there's an assumption that people who seem confident all the time actually feel confident, and in my experience that's almost never true.",
+ "normalized_hypothesis": "there's an assumption that people who seem confident all the time actually feel confident, and in my experience that's almost never true.",
+ "errors": 0,
+ "words": 22,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "there is an assumption that people who seem confident all the time actually feel confident and in my experience that is almost never true",
+ "normalized_hypothesis": "there is an assumption that people who seem confident all the time actually feel confident and in my experience that is almost never true",
+ "errors": 0,
+ "words": 24,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0121",
+ "category": "narrative",
+ "text": "The moment I opened the door I knew something was different, even if I couldn't say what.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0121.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The moment I opened the door, I knew something was different, even if I couldn't say what.",
+ "raw": {
+ "normalized_reference": "the moment i opened the door i knew something was different, even if i couldn't say what.",
+ "normalized_hypothesis": "the moment i opened the door, i knew something was different, even if i couldn't say what.",
+ "errors": 0,
+ "words": 17,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the moment i opened the door i knew something was different even if i could not say what",
+ "normalized_hypothesis": "the moment i opened the door i knew something was different even if i could not say what",
+ "errors": 0,
+ "words": 18,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0122",
+ "category": "question",
+ "text": "How do you explain something that doesn't make sense even to you?",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0122.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "How do you explain something that doesn't make sense even to you?",
+ "raw": {
+ "normalized_reference": "how do you explain something that doesn't make sense even to you?",
+ "normalized_hypothesis": "how do you explain something that doesn't make sense even to you?",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "how do you explain something that does not make sense even to you",
+ "normalized_hypothesis": "how do you explain something that does not make sense even to you",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0123",
+ "category": "short",
+ "text": "Hold on a second.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0123.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Hold on a second.",
+ "raw": {
+ "normalized_reference": "hold on a second.",
+ "normalized_hypothesis": "hold on a second.",
+ "errors": 0,
+ "words": 4,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "hold on a 2nd",
+ "normalized_hypothesis": "hold on a 2nd",
+ "errors": 0,
+ "words": 4,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0124",
+ "category": "technical",
+ "text": "A deterministic split prevents duplicate normalized text from leaking into evaluation.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0124.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "A deterministic split prevents duplicate normalized text from leaking into evaluation.",
+ "raw": {
+ "normalized_reference": "a deterministic split prevents duplicate normalized text from leaking into evaluation.",
+ "normalized_hypothesis": "a deterministic split prevents duplicate normalized text from leaking into evaluation.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "a deterministic split prevents duplicate normalized text from leaking into evaluation",
+ "normalized_hypothesis": "a deterministic split prevents duplicate normalized text from leaking into evaluation",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0125",
+ "category": "conversational",
+ "text": "Honestly, that was much better than I expected, but the ending still felt rushed.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0125.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Honestly, that was much better than I expected, but the ending still felt rushed.",
+ "raw": {
+ "normalized_reference": "honestly, that was much better than i expected, but the ending still felt rushed.",
+ "normalized_hypothesis": "honestly, that was much better than i expected, but the ending still felt rushed.",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "honestly that was much better than i expected but the ending still felt rushed",
+ "normalized_hypothesis": "honestly that was much better than i expected but the ending still felt rushed",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0126",
+ "category": "descriptive",
+ "text": "She had a habit of tilting her head slightly when thinking hard about something.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0126.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "She had a habit of tilting her head slightly when thinking hard about something.",
+ "raw": {
+ "normalized_reference": "she had a habit of tilting her head slightly when thinking hard about something.",
+ "normalized_hypothesis": "she had a habit of tilting her head slightly when thinking hard about something.",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "she had a habit of tilting her head slightly when thinking hard about something",
+ "normalized_hypothesis": "she had a habit of tilting her head slightly when thinking hard about something",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0127",
+ "category": "emotional",
+ "text": "I cannot stress enough how much I did not want to be right about this.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0127.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I cannot stress enough how much I did not want to be right about this.",
+ "raw": {
+ "normalized_reference": "i cannot stress enough how much i did not want to be right about this.",
+ "normalized_hypothesis": "i cannot stress enough how much i did not want to be right about this.",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i cannot stress enough how much i did not want to be right about this",
+ "normalized_hypothesis": "i cannot stress enough how much i did not want to be right about this",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0128",
+ "category": "long",
+ "text": "I used to think the goal was to get to a point where nothing was hard anymore, and now I think that point doesn't exist and I'm not sure I'd want it to.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0128.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I used to think the goal was to get to a point where nothing was hard anymore, and now I think that point doesn't exist, and I'm not sure I'd want it to.",
+ "raw": {
+ "normalized_reference": "i used to think the goal was to get to a point where nothing was hard anymore, and now i think that point doesn't exist and i'm not sure i'd want it to.",
+ "normalized_hypothesis": "i used to think the goal was to get to a point where nothing was hard anymore, and now i think that point doesn't exist, and i'm not sure i'd want it to.",
+ "errors": 0,
+ "words": 33,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i used to think the goal was to get to a point where nothing was hard anymore and now i think that point does not exist and i am not sure i would want it to",
+ "normalized_hypothesis": "i used to think the goal was to get to a point where nothing was hard anymore and now i think that point does not exist and i am not sure i would want it to",
+ "errors": 0,
+ "words": 36,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0129",
+ "category": "narrative",
+ "text": "So this is the part of the story where everything goes completely sideways.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0129.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "So this is the part of the story where everything goes completely sideways.",
+ "raw": {
+ "normalized_reference": "so this is the part of the story where everything goes completely sideways.",
+ "normalized_hypothesis": "so this is the part of the story where everything goes completely sideways.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "so this is the part of the story where everything goes completely sideways",
+ "normalized_hypothesis": "so this is the part of the story where everything goes completely sideways",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0130",
+ "category": "question",
+ "text": "Is there a version of this where it actually works out okay?",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0130.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Is there a version of this where it actually works out okay?",
+ "raw": {
+ "normalized_reference": "is there a version of this where it actually works out okay?",
+ "normalized_hypothesis": "is there a version of this where it actually works out okay?",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "is there a version of this where it actually works out okay",
+ "normalized_hypothesis": "is there a version of this where it actually works out okay",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0131",
+ "category": "short",
+ "text": "Oh no. Oh no no no.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0131.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Oh, no. Oh, no, no, no.",
+ "raw": {
+ "normalized_reference": "oh no. oh no no no.",
+ "normalized_hypothesis": "oh, no. oh, no, no, no.",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "0 no 0 no no no",
+ "normalized_hypothesis": "0 no 0 no no no",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0132",
+ "category": "technical",
+ "text": "Always test on a copy before modifying the original.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0132.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Always test on a copy before modifying the original.",
+ "raw": {
+ "normalized_reference": "always test on a copy before modifying the original.",
+ "normalized_hypothesis": "always test on a copy before modifying the original.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "always test on a copy before modifying the original",
+ "normalized_hypothesis": "always test on a copy before modifying the original",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0133",
+ "category": "conversational",
+ "text": "I wasn't even supposed to be there. That's the wild part.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0133.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I wasn't even supposed to be there. That's the wild part.",
+ "raw": {
+ "normalized_reference": "i wasn't even supposed to be there. that's the wild part.",
+ "normalized_hypothesis": "i wasn't even supposed to be there. that's the wild part.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i was not even supposed to be there that is the wild part",
+ "normalized_hypothesis": "i was not even supposed to be there that is the wild part",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0134",
+ "category": "descriptive",
+ "text": "It's a small town. Not the kind you leave. The kind you leave and come back to.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0134.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It's a small town. Not the kind you leave. The kind you leave and come back to.",
+ "raw": {
+ "normalized_reference": "it's a small town. not the kind you leave. the kind you leave and come back to.",
+ "normalized_hypothesis": "it's a small town. not the kind you leave. the kind you leave and come back to.",
+ "errors": 0,
+ "words": 17,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it is a small town not the kind you leave the kind you leave and come back to",
+ "normalized_hypothesis": "it is a small town not the kind you leave the kind you leave and come back to",
+ "errors": 0,
+ "words": 18,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0135",
+ "category": "emotional",
+ "text": "She cried. I cried. We all just sort of stood there crying.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0135.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "She cried. I cried. We all just sort of stood there crying.",
+ "raw": {
+ "normalized_reference": "she cried. i cried. we all just sort of stood there crying.",
+ "normalized_hypothesis": "she cried. i cried. we all just sort of stood there crying.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "she cried i cried we all just sort of stood there crying",
+ "normalized_hypothesis": "she cried i cried we all just sort of stood there crying",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0136",
+ "category": "long",
+ "text": "There's a version of being busy that's actually just a very elaborate way of avoiding the one thing you know you should be doing.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0136.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "There's a version of being busy that's actually just a very elaborate way of avoiding the one thing you know you should be doing.",
+ "raw": {
+ "normalized_reference": "there's a version of being busy that's actually just a very elaborate way of avoiding the one thing you know you should be doing.",
+ "normalized_hypothesis": "there's a version of being busy that's actually just a very elaborate way of avoiding the one thing you know you should be doing.",
+ "errors": 0,
+ "words": 24,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "there is a version of being busy that is actually just a very elaborate way of avoiding the one thing you know you should be doing",
+ "normalized_hypothesis": "there is a version of being busy that is actually just a very elaborate way of avoiding the one thing you know you should be doing",
+ "errors": 0,
+ "words": 26,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0137",
+ "category": "narrative",
+ "text": "The letter arrived on a Tuesday, which felt appropriate somehow.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0137.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The letter arrived on a Tuesday, which felt appropriate somehow.",
+ "raw": {
+ "normalized_reference": "the letter arrived on a tuesday, which felt appropriate somehow.",
+ "normalized_hypothesis": "the letter arrived on a tuesday, which felt appropriate somehow.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the letter arrived on a tuesday which felt appropriate somehow",
+ "normalized_hypothesis": "the letter arrived on a tuesday which felt appropriate somehow",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0138",
+ "category": "question",
+ "text": "What exactly did she say when you told her?",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0138.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "What exactly did she say when you told her?",
+ "raw": {
+ "normalized_reference": "what exactly did she say when you told her?",
+ "normalized_hypothesis": "what exactly did she say when you told her?",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "what exactly did she say when you told her",
+ "normalized_hypothesis": "what exactly did she say when you told her",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0139",
+ "category": "short",
+ "text": "No way. Absolutely not.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0139.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "No way. Absolutely not.",
+ "raw": {
+ "normalized_reference": "no way. absolutely not.",
+ "normalized_hypothesis": "no way. absolutely not.",
+ "errors": 0,
+ "words": 4,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "no way absolutely not",
+ "normalized_hypothesis": "no way absolutely not",
+ "errors": 0,
+ "words": 4,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0140",
+ "category": "technical",
+ "text": "The first thing you want to do is make sure everything is backed up.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0140.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The first thing you want to do is make sure everything is backed up.",
+ "raw": {
+ "normalized_reference": "the first thing you want to do is make sure everything is backed up.",
+ "normalized_hypothesis": "the first thing you want to do is make sure everything is backed up.",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the 1st thing you want to do is make sure everything is backed up",
+ "normalized_hypothesis": "the 1st thing you want to do is make sure everything is backed up",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0141",
+ "category": "conversational",
+ "text": "I have the most insane idea and you need to just hear me out.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0141.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I have the most insane idea, and you need to just hear me out.",
+ "raw": {
+ "normalized_reference": "i have the most insane idea and you need to just hear me out.",
+ "normalized_hypothesis": "i have the most insane idea, and you need to just hear me out.",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i have the most insane idea and you need to just hear me out",
+ "normalized_hypothesis": "i have the most insane idea and you need to just hear me out",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0142",
+ "category": "descriptive",
+ "text": "There's a coffee shop on the corner that's been there longer than anyone can remember.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0142.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "There's a coffee shop on the corner that's been there longer than anyone can remember.",
+ "raw": {
+ "normalized_reference": "there's a coffee shop on the corner that's been there longer than anyone can remember.",
+ "normalized_hypothesis": "there's a coffee shop on the corner that's been there longer than anyone can remember.",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "there is a coffee shop on the corner that has been there longer than anyone can remember",
+ "normalized_hypothesis": "there is a coffee shop on the corner that has been there longer than anyone can remember",
+ "errors": 0,
+ "words": 17,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0143",
+ "category": "emotional",
+ "text": "I kept it together until I got to the car, and then I completely fell apart.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0143.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I kept it together until I got to the car, and then I completely fell apart.",
+ "raw": {
+ "normalized_reference": "i kept it together until i got to the car, and then i completely fell apart.",
+ "normalized_hypothesis": "i kept it together until i got to the car, and then i completely fell apart.",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i kept it together until i got to the car and then i completely fell apart",
+ "normalized_hypothesis": "i kept it together until i got to the car and then i completely fell apart",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0144",
+ "category": "long",
+ "text": "The older I get, the more I think that most of what we call wisdom is just pattern recognition built from having been wrong enough times to start noticing.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0144.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The older I get, the more I think that most of what we call wisdom is just pattern recognition built from having been wrong enough times to start noticing.",
+ "raw": {
+ "normalized_reference": "the older i get, the more i think that most of what we call wisdom is just pattern recognition built from having been wrong enough times to start noticing.",
+ "normalized_hypothesis": "the older i get, the more i think that most of what we call wisdom is just pattern recognition built from having been wrong enough times to start noticing.",
+ "errors": 0,
+ "words": 29,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the older i get the more i think that most of what we call wisdom is just pattern recognition built from having been wrong enough times to start noticing",
+ "normalized_hypothesis": "the older i get the more i think that most of what we call wisdom is just pattern recognition built from having been wrong enough times to start noticing",
+ "errors": 0,
+ "words": 29,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0145",
+ "category": "narrative",
+ "text": "I remember the exact moment I realized I had made a huge mistake.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0145.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I remember the exact moment I realized I had made a huge mistake.",
+ "raw": {
+ "normalized_reference": "i remember the exact moment i realized i had made a huge mistake.",
+ "normalized_hypothesis": "i remember the exact moment i realized i had made a huge mistake.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i remember the exact moment i realized i had made a huge mistake",
+ "normalized_hypothesis": "i remember the exact moment i realized i had made a huge mistake",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0146",
+ "category": "question",
+ "text": "Do you think it would have gone differently if we'd left earlier?",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0146.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Do you think it would have gone differently if we'd left earlier?",
+ "raw": {
+ "normalized_reference": "do you think it would have gone differently if we'd left earlier?",
+ "normalized_hypothesis": "do you think it would have gone differently if we'd left earlier?",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "do you think it would have gone differently if we would left earlier",
+ "normalized_hypothesis": "do you think it would have gone differently if we would left earlier",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0147",
+ "category": "short",
+ "text": "You have to be kidding me.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0147.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "You have to be kidding me.",
+ "raw": {
+ "normalized_reference": "you have to be kidding me.",
+ "normalized_hypothesis": "you have to be kidding me.",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "you have to be kidding me",
+ "normalized_hypothesis": "you have to be kidding me",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0148",
+ "category": "technical",
+ "text": "If the checksum fails, restart the worker process without deleting the cached tensors.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0148.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "If the checksum fails, restart the worker process without deleting the cached tensors.",
+ "raw": {
+ "normalized_reference": "if the checksum fails, restart the worker process without deleting the cached tensors.",
+ "normalized_hypothesis": "if the checksum fails, restart the worker process without deleting the cached tensors.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "if the checksum fails restart the worker process without deleting the cached tensors",
+ "normalized_hypothesis": "if the checksum fails restart the worker process without deleting the cached tensors",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0149",
+ "category": "conversational",
+ "text": "I need you to understand how close that was to not working.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0149.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I need you to understand how close that was to not working.",
+ "raw": {
+ "normalized_reference": "i need you to understand how close that was to not working.",
+ "normalized_hypothesis": "i need you to understand how close that was to not working.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i need you to understand how close that was to not working",
+ "normalized_hypothesis": "i need you to understand how close that was to not working",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0150",
+ "category": "descriptive",
+ "text": "It was the sort of morning where the whole world looks rinsed clean.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0150.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It was the sort of morning where the whole world looks rinsed clean.",
+ "raw": {
+ "normalized_reference": "it was the sort of morning where the whole world looks rinsed clean.",
+ "normalized_hypothesis": "it was the sort of morning where the whole world looks rinsed clean.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it was the sort of morning where the whole world looks rinsed clean",
+ "normalized_hypothesis": "it was the sort of morning where the whole world looks rinsed clean",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0151",
+ "category": "emotional",
+ "text": "She started laughing and I started laughing and we couldn't stop for a full minute.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0151.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "She started laughing and I started laughing and we couldn't stop for a full minute.",
+ "raw": {
+ "normalized_reference": "she started laughing and i started laughing and we couldn't stop for a full minute.",
+ "normalized_hypothesis": "she started laughing and i started laughing and we couldn't stop for a full minute.",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "she started laughing and i started laughing and we could not stop for a full minute",
+ "normalized_hypothesis": "she started laughing and i started laughing and we could not stop for a full minute",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0152",
+ "category": "long",
+ "text": "We spent three years building something we were genuinely proud of, and then the market shifted and none of it mattered the way we thought it would.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0152.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "We spent three years building something we were genuinely proud of, and then the market shifted and none of it mattered the way we thought it would.",
+ "raw": {
+ "normalized_reference": "we spent three years building something we were genuinely proud of, and then the market shifted and none of it mattered the way we thought it would.",
+ "normalized_hypothesis": "we spent three years building something we were genuinely proud of, and then the market shifted and none of it mattered the way we thought it would.",
+ "errors": 0,
+ "words": 27,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "we spent 3 years building something we were genuinely proud of and then the market shifted and none of it mattered the way we thought it would",
+ "normalized_hypothesis": "we spent 3 years building something we were genuinely proud of and then the market shifted and none of it mattered the way we thought it would",
+ "errors": 0,
+ "words": 27,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0153",
+ "category": "narrative",
+ "text": "And somehow, despite all of that, it actually turned out okay.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0153.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "And somehow, despite all of that, it actually turned out okay.",
+ "raw": {
+ "normalized_reference": "and somehow, despite all of that, it actually turned out okay.",
+ "normalized_hypothesis": "and somehow, despite all of that, it actually turned out okay.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "and somehow despite all of that it actually turned out okay",
+ "normalized_hypothesis": "and somehow despite all of that it actually turned out okay",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0154",
+ "category": "question",
+ "text": "Is that something you actually want, or something you think you should want?",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0154.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Is that something you actually want or something you think you should want?",
+ "raw": {
+ "normalized_reference": "is that something you actually want, or something you think you should want?",
+ "normalized_hypothesis": "is that something you actually want or something you think you should want?",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "is that something you actually want or something you think you should want",
+ "normalized_hypothesis": "is that something you actually want or something you think you should want",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0155",
+ "category": "short",
+ "text": "This is fine.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0155.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "This is fine.",
+ "raw": {
+ "normalized_reference": "this is fine.",
+ "normalized_hypothesis": "this is fine.",
+ "errors": 0,
+ "words": 3,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "this is fine",
+ "normalized_hypothesis": "this is fine",
+ "errors": 0,
+ "words": 3,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0156",
+ "category": "technical",
+ "text": "The tricky part isn't the setup, it's making sure all dependencies line up.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0156.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The tricky part isn't the setup, it's making sure all dependencies line up.",
+ "raw": {
+ "normalized_reference": "the tricky part isn't the setup, it's making sure all dependencies line up.",
+ "normalized_hypothesis": "the tricky part isn't the setup, it's making sure all dependencies line up.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the tricky part is not the setup it is making sure all dependencies line up",
+ "normalized_hypothesis": "the tricky part is not the setup it is making sure all dependencies line up",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0157",
+ "category": "conversational",
+ "text": "He called it a minor setback. I called it a disaster.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0157.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He called it a minor setback. I called it a disaster.",
+ "raw": {
+ "normalized_reference": "he called it a minor setback. i called it a disaster.",
+ "normalized_hypothesis": "he called it a minor setback. i called it a disaster.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he called it a minor setback i called it a disaster",
+ "normalized_hypothesis": "he called it a minor setback i called it a disaster",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0158",
+ "category": "descriptive",
+ "text": "The whole place smelled like pine and old wood and cold air.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0158.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The whole place smelled like pine and old wood and cold air.",
+ "raw": {
+ "normalized_reference": "the whole place smelled like pine and old wood and cold air.",
+ "normalized_hypothesis": "the whole place smelled like pine and old wood and cold air.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the whole place smelled like pine and old wood and cold air",
+ "normalized_hypothesis": "the whole place smelled like pine and old wood and cold air",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0159",
+ "category": "emotional",
+ "text": "I've never been so relieved in my entire life.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0159.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I've never been so relieved in my entire life.",
+ "raw": {
+ "normalized_reference": "i've never been so relieved in my entire life.",
+ "normalized_hypothesis": "i've never been so relieved in my entire life.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i have never been so relieved in my entire life",
+ "normalized_hypothesis": "i have never been so relieved in my entire life",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0160",
+ "category": "narrative",
+ "text": "I wrote it all down that night so I wouldn't forget, and I still haven't looked at it.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0160.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I wrote it all down that night so I wouldn't forget, and I still haven't looked at it.",
+ "raw": {
+ "normalized_reference": "i wrote it all down that night so i wouldn't forget, and i still haven't looked at it.",
+ "normalized_hypothesis": "i wrote it all down that night so i wouldn't forget, and i still haven't looked at it.",
+ "errors": 0,
+ "words": 18,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i wrote it all down that night so i would not forget and i still have not looked at it",
+ "normalized_hypothesis": "i wrote it all down that night so i would not forget and i still have not looked at it",
+ "errors": 0,
+ "words": 20,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0161",
+ "category": "short",
+ "text": "That's kind of wild actually.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0161.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "That's kind of wild, actually.",
+ "raw": {
+ "normalized_reference": "that's kind of wild actually.",
+ "normalized_hypothesis": "that's kind of wild, actually.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "that is kind of wild actually",
+ "normalized_hypothesis": "that is kind of wild actually",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0162",
+ "category": "technical",
+ "text": "Version pinning will save you a lot of pain later down the line.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0162.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Version pinning will save you a lot of pain later down the line.",
+ "raw": {
+ "normalized_reference": "version pinning will save you a lot of pain later down the line.",
+ "normalized_hypothesis": "version pinning will save you a lot of pain later down the line.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "version pinning will save you a lot of pain later down the line",
+ "normalized_hypothesis": "version pinning will save you a lot of pain later down the line",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0163",
+ "category": "conversational",
+ "text": "At some point you just have to commit and see what happens.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0163.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "At some point you just have to commit and see what happens.",
+ "raw": {
+ "normalized_reference": "at some point you just have to commit and see what happens.",
+ "normalized_hypothesis": "at some point you just have to commit and see what happens.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "at some point you just have to commit and see what happens",
+ "normalized_hypothesis": "at some point you just have to commit and see what happens",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0164",
+ "category": "emotional",
+ "text": "Nobody told me it was going to be this hard, and honestly, I'm grateful.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0164.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Nobody told me it was going to be this hard, and honestly, I'm grateful.",
+ "raw": {
+ "normalized_reference": "nobody told me it was going to be this hard, and honestly, i'm grateful.",
+ "normalized_hypothesis": "nobody told me it was going to be this hard, and honestly, i'm grateful.",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "nobody told me it was going to be this hard and honestly i am grateful",
+ "normalized_hypothesis": "nobody told me it was going to be this hard and honestly i am grateful",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0165",
+ "category": "narrative",
+ "text": "Chapter seven. In which everything that could go wrong did, and a few things that couldn't.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0165.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Chapter 7 In which everything that could go wrong did, and a few things that couldn't.",
+ "raw": {
+ "normalized_reference": "chapter seven. in which everything that could go wrong did, and a few things that couldn't.",
+ "normalized_hypothesis": "chapter 7 in which everything that could go wrong did, and a few things that couldn't.",
+ "errors": 1,
+ "words": 16,
+ "wer": 0.0625
+ },
+ "semantic": {
+ "normalized_reference": "chapter 7 in which everything that could go wrong did and a few things that could not",
+ "normalized_hypothesis": "chapter 7 in which everything that could go wrong did and a few things that could not",
+ "errors": 0,
+ "words": 17,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0166",
+ "category": "short",
+ "text": "Wait, what?",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0166.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Wait, what?",
+ "raw": {
+ "normalized_reference": "wait, what?",
+ "normalized_hypothesis": "wait, what?",
+ "errors": 0,
+ "words": 2,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "wait what",
+ "normalized_hypothesis": "wait what",
+ "errors": 0,
+ "words": 2,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0167",
+ "category": "technical",
+ "text": "The graphics processor remained idle because the data loader exhausted every central processing core.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0167.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The graphics processor remained idle because the data loader exhausted every central processing core.",
+ "raw": {
+ "normalized_reference": "the graphics processor remained idle because the data loader exhausted every central processing core.",
+ "normalized_hypothesis": "the graphics processor remained idle because the data loader exhausted every central processing core.",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the graphics processor remained idle because the data loader exhausted every central processing core",
+ "normalized_hypothesis": "the graphics processor remained idle because the data loader exhausted every central processing core",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0168",
+ "category": "conversational",
+ "text": "Could you speak a little more slowly? I am trying to write every detail down.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0168.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Could you speak a little more slowly? I am trying to write every detail down.",
+ "raw": {
+ "normalized_reference": "could you speak a little more slowly? i am trying to write every detail down.",
+ "normalized_hypothesis": "could you speak a little more slowly? i am trying to write every detail down.",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "could you speak a little more slowly i am trying to write every detail down",
+ "normalized_hypothesis": "could you speak a little more slowly i am trying to write every detail down",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0169",
+ "category": "emotional",
+ "text": "I was so excited I could barely form a complete sentence.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0169.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I was so excited I could barely form a complete sentence.",
+ "raw": {
+ "normalized_reference": "i was so excited i could barely form a complete sentence.",
+ "normalized_hypothesis": "i was so excited i could barely form a complete sentence.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i was so excited i could barely form a complete sentence",
+ "normalized_hypothesis": "i was so excited i could barely form a complete sentence",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0170",
+ "category": "narrative",
+ "text": "He looked at me like I had said something in a language he didn't speak.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0170.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He looked at me like I had said something in a language he didn't speak.",
+ "raw": {
+ "normalized_reference": "he looked at me like i had said something in a language he didn't speak.",
+ "normalized_hypothesis": "he looked at me like i had said something in a language he didn't speak.",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he looked at me like i had said something in a language he did not speak",
+ "normalized_hypothesis": "he looked at me like i had said something in a language he did not speak",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0171",
+ "category": "short",
+ "text": "That's actually hilarious.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0171.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "That's actually hilarious.",
+ "raw": {
+ "normalized_reference": "that's actually hilarious.",
+ "normalized_hypothesis": "that's actually hilarious.",
+ "errors": 0,
+ "words": 3,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "that is actually hilarious",
+ "normalized_hypothesis": "that is actually hilarious",
+ "errors": 0,
+ "words": 4,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0172",
+ "category": "technical",
+ "text": "The encoder maps phonemes, stress, and phrase boundaries into a shared hidden representation.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0172.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The encoder maps phonemes, stress, and phrase boundaries into a shared hidden representation.",
+ "raw": {
+ "normalized_reference": "the encoder maps phonemes, stress, and phrase boundaries into a shared hidden representation.",
+ "normalized_hypothesis": "the encoder maps phonemes, stress, and phrase boundaries into a shared hidden representation.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the encoder maps phonemes stress and phrase boundaries into a shared hidden representation",
+ "normalized_hypothesis": "the encoder maps phonemes stress and phrase boundaries into a shared hidden representation",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0173",
+ "category": "conversational",
+ "text": "She showed up three hours late and acted like nothing had happened.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0173.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "She showed up three hours late and acted like nothing had happened.",
+ "raw": {
+ "normalized_reference": "she showed up three hours late and acted like nothing had happened.",
+ "normalized_hypothesis": "she showed up three hours late and acted like nothing had happened.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "she showed up 3 hours late and acted like nothing had happened",
+ "normalized_hypothesis": "she showed up 3 hours late and acted like nothing had happened",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0174",
+ "category": "emotional",
+ "text": "Oh my god. This is the best thing I've ever tasted.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0174.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Oh my god. This is the best thing I've ever tasted.",
+ "raw": {
+ "normalized_reference": "oh my god. this is the best thing i've ever tasted.",
+ "normalized_hypothesis": "oh my god. this is the best thing i've ever tasted.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "0 my god this is the best thing i have ever tasted",
+ "normalized_hypothesis": "0 my god this is the best thing i have ever tasted",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0175",
+ "category": "narrative",
+ "text": "The note was short, which was somehow more frightening than a long one would have been.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0175.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The note was short, which was somehow more frightening than a long one would have been.",
+ "raw": {
+ "normalized_reference": "the note was short, which was somehow more frightening than a long one would have been.",
+ "normalized_hypothesis": "the note was short, which was somehow more frightening than a long one would have been.",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the note was short which was somehow more frightening than a long one would have been",
+ "normalized_hypothesis": "the note was short which was somehow more frightening than a long one would have been",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0176",
+ "category": "short",
+ "text": "Noted.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0176.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Noted.",
+ "raw": {
+ "normalized_reference": "noted.",
+ "normalized_hypothesis": "noted.",
+ "errors": 0,
+ "words": 1,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "noted",
+ "normalized_hypothesis": "noted",
+ "errors": 0,
+ "words": 1,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0177",
+ "category": "technical",
+ "text": "It's a simple process once you understand the underlying structure.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0177.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It's a simple process once you understand the underlying structure.",
+ "raw": {
+ "normalized_reference": "it's a simple process once you understand the underlying structure.",
+ "normalized_hypothesis": "it's a simple process once you understand the underlying structure.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it is a simple process once you understand the underlying structure",
+ "normalized_hypothesis": "it is a simple process once you understand the underlying structure",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0178",
+ "category": "conversational",
+ "text": "You have to hear this story because it's almost too good to be true.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0178.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "You have to hear this story because it's almost too good to be true.",
+ "raw": {
+ "normalized_reference": "you have to hear this story because it's almost too good to be true.",
+ "normalized_hypothesis": "you have to hear this story because it's almost too good to be true.",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "you have to hear this story because it is almost too good to be true",
+ "normalized_hypothesis": "you have to hear this story because it is almost too good to be true",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0179",
+ "category": "emotional",
+ "text": "I have never felt more seen in my entire life than in that moment.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0179.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I have never felt more seen in my entire life than in that moment.",
+ "raw": {
+ "normalized_reference": "i have never felt more seen in my entire life than in that moment.",
+ "normalized_hypothesis": "i have never felt more seen in my entire life than in that moment.",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i have never felt more seen in my entire life than in that moment",
+ "normalized_hypothesis": "i have never felt more seen in my entire life than in that moment",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0180",
+ "category": "narrative",
+ "text": "She knew. She had always known. She just waited for the right moment.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0180.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "She knew. She had always known. She just waited for the right moment.",
+ "raw": {
+ "normalized_reference": "she knew. she had always known. she just waited for the right moment.",
+ "normalized_hypothesis": "she knew. she had always known. she just waited for the right moment.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "she knew she had always known she just waited for the right moment",
+ "normalized_hypothesis": "she knew she had always known she just waited for the right moment",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0181",
+ "category": "short",
+ "text": "Yeah, no. Hard pass.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0181.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Yes, no. Hard pass.",
+ "raw": {
+ "normalized_reference": "yeah, no. hard pass.",
+ "normalized_hypothesis": "yes, no. hard pass.",
+ "errors": 1,
+ "words": 4,
+ "wer": 0.25
+ },
+ "semantic": {
+ "normalized_reference": "yeah no hard pass",
+ "normalized_hypothesis": "yes no hard pass",
+ "errors": 1,
+ "words": 4,
+ "wer": 0.25
+ }
+ },
+ {
+ "id": "release500_0182",
+ "category": "technical",
+ "text": "Please verify the sample rate, hop length, and mel filter configuration before inference.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0182.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Please verify the sample rate, hop length, and MEL filter configuration before inference.",
+ "raw": {
+ "normalized_reference": "please verify the sample rate, hop length, and mel filter configuration before inference.",
+ "normalized_hypothesis": "please verify the sample rate, hop length, and mel filter configuration before inference.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "please verify the sample rate hop length and mel filter configuration before inference",
+ "normalized_hypothesis": "please verify the sample rate hop length and mel filter configuration before inference",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0183",
+ "category": "conversational",
+ "text": "That's the worst idea I've ever heard and I'm completely in.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0183.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "That's the worst idea I've ever heard, and I'm completely in.",
+ "raw": {
+ "normalized_reference": "that's the worst idea i've ever heard and i'm completely in.",
+ "normalized_hypothesis": "that's the worst idea i've ever heard, and i'm completely in.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "that is the worst idea i have ever heard and i am completely in",
+ "normalized_hypothesis": "that is the worst idea i have ever heard and i am completely in",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0184",
+ "category": "emotional",
+ "text": "I remember thinking: this is it. This is the thing I've been waiting for.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0184.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I remember thinking this is it. This is the thing I've been waiting for.",
+ "raw": {
+ "normalized_reference": "i remember thinking: this is it. this is the thing i've been waiting for.",
+ "normalized_hypothesis": "i remember thinking this is it. this is the thing i've been waiting for.",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i remember thinking this is it this is the thing i have been waiting for",
+ "normalized_hypothesis": "i remember thinking this is it this is the thing i have been waiting for",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0185",
+ "category": "narrative",
+ "text": "We never talked about it again, which I think was the right call.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0185.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "We never talked about it again, which I think was the right call.",
+ "raw": {
+ "normalized_reference": "we never talked about it again, which i think was the right call.",
+ "normalized_hypothesis": "we never talked about it again, which i think was the right call.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "we never talked about it again which i think was the right call",
+ "normalized_hypothesis": "we never talked about it again which i think was the right call",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0186",
+ "category": "short",
+ "text": "In hindsight, obvious.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0186.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "In hindsight, obvious.",
+ "raw": {
+ "normalized_reference": "in hindsight, obvious.",
+ "normalized_hypothesis": "in hindsight, obvious.",
+ "errors": 0,
+ "words": 3,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "in hindsight obvious",
+ "normalized_hypothesis": "in hindsight obvious",
+ "errors": 0,
+ "words": 3,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0187",
+ "category": "technical",
+ "text": "Once it's configured correctly, it basically runs itself.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0187.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Once it's configured correctly, it basically runs itself.",
+ "raw": {
+ "normalized_reference": "once it's configured correctly, it basically runs itself.",
+ "normalized_hypothesis": "once it's configured correctly, it basically runs itself.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "once it is configured correctly it basically runs itself",
+ "normalized_hypothesis": "once it is configured correctly it basically runs itself",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0188",
+ "category": "conversational",
+ "text": "You got engaged? When? How? Tell me everything right now.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0188.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "You got engaged. When? How? Tell me everything right now.",
+ "raw": {
+ "normalized_reference": "you got engaged? when? how? tell me everything right now.",
+ "normalized_hypothesis": "you got engaged. when? how? tell me everything right now.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "you got engaged when how tell me everything right now",
+ "normalized_hypothesis": "you got engaged when how tell me everything right now",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0189",
+ "category": "emotional",
+ "text": "It's hard to explain how much that moment meant to me.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0189.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It's hard to explain how much that moment meant to me.",
+ "raw": {
+ "normalized_reference": "it's hard to explain how much that moment meant to me.",
+ "normalized_hypothesis": "it's hard to explain how much that moment meant to me.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it is hard to explain how much that moment meant to me",
+ "normalized_hypothesis": "it is hard to explain how much that moment meant to me",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0190",
+ "category": "narrative",
+ "text": "He told the story three different times that night and somehow it got better each time.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0190.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He told the story three different times that night, and somehow it got better each time.",
+ "raw": {
+ "normalized_reference": "he told the story three different times that night and somehow it got better each time.",
+ "normalized_hypothesis": "he told the story three different times that night, and somehow it got better each time.",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he told the story 3 different times that night and somehow it got better each time",
+ "normalized_hypothesis": "he told the story 3 different times that night and somehow it got better each time",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0191",
+ "category": "short",
+ "text": "Right. Moving on.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0191.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Right. Moving on.",
+ "raw": {
+ "normalized_reference": "right. moving on.",
+ "normalized_hypothesis": "right. moving on.",
+ "errors": 0,
+ "words": 3,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "right moving on",
+ "normalized_hypothesis": "right moving on",
+ "errors": 0,
+ "words": 3,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0192",
+ "category": "conversational",
+ "text": "No, I did not move your keys; check beside the small ceramic bowl.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0192.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "No, I did not move your keys. Check beside the small ceramic bowl.",
+ "raw": {
+ "normalized_reference": "no, i did not move your keys; check beside the small ceramic bowl.",
+ "normalized_hypothesis": "no, i did not move your keys. check beside the small ceramic bowl.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "no i did not move your keys check beside the small ceramic bowl",
+ "normalized_hypothesis": "no i did not move your keys check beside the small ceramic bowl",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0193",
+ "category": "emotional",
+ "text": "I was devastated. I'm not going to pretend otherwise.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0193.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I was devastated. I'm not going to pretend otherwise.",
+ "raw": {
+ "normalized_reference": "i was devastated. i'm not going to pretend otherwise.",
+ "normalized_hypothesis": "i was devastated. i'm not going to pretend otherwise.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i was devastated i am not going to pretend otherwise",
+ "normalized_hypothesis": "i was devastated i am not going to pretend otherwise",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0194",
+ "category": "narrative",
+ "text": "That was the last summer before everything changed, though we didn't know it yet.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0194.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "That was the last summer before everything changed, though we didn't know it yet.",
+ "raw": {
+ "normalized_reference": "that was the last summer before everything changed, though we didn't know it yet.",
+ "normalized_hypothesis": "that was the last summer before everything changed, though we didn't know it yet.",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "that was the last summer before everything changed though we did not know it yet",
+ "normalized_hypothesis": "that was the last summer before everything changed though we did not know it yet",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0195",
+ "category": "short",
+ "text": "Bold of you to assume.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0195.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Bold of you to assume.",
+ "raw": {
+ "normalized_reference": "bold of you to assume.",
+ "normalized_hypothesis": "bold of you to assume.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "bold of you to assume",
+ "normalized_hypothesis": "bold of you to assume",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0196",
+ "category": "conversational",
+ "text": "It's one of those decisions that makes total sense in the moment.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0196.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It's one of those decisions that makes total sense in the moment.",
+ "raw": {
+ "normalized_reference": "it's one of those decisions that makes total sense in the moment.",
+ "normalized_hypothesis": "it's one of those decisions that makes total sense in the moment.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it is one of those decisions that makes total sense in the moment",
+ "normalized_hypothesis": "it is one of those decisions that makes total sense in the moment",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0197",
+ "category": "emotional",
+ "text": "I was furious. Not the loud kind. The quiet kind that lasts for days.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0197.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "i was furious not the loud kind the quiet kind that lasts for days",
+ "raw": {
+ "normalized_reference": "i was furious. not the loud kind. the quiet kind that lasts for days.",
+ "normalized_hypothesis": "i was furious not the loud kind the quiet kind that lasts for days",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i was furious not the loud kind the quiet kind that lasts for days",
+ "normalized_hypothesis": "i was furious not the loud kind the quiet kind that lasts for days",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0198",
+ "category": "narrative",
+ "text": "That's the part of the story I usually leave out.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0198.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "That's the part of the story I usually leave out.",
+ "raw": {
+ "normalized_reference": "that's the part of the story i usually leave out.",
+ "normalized_hypothesis": "that's the part of the story i usually leave out.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "that is the part of the story i usually leave out",
+ "normalized_hypothesis": "that is the part of the story i usually leave out",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0199",
+ "category": "short",
+ "text": "I've said too much.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0199.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I've said too much.",
+ "raw": {
+ "normalized_reference": "i've said too much.",
+ "normalized_hypothesis": "i've said too much.",
+ "errors": 0,
+ "words": 4,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i have said too much",
+ "normalized_hypothesis": "i have said too much",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0200",
+ "category": "conversational",
+ "text": "I've been faking that I understood this the entire time.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0200.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I've been faking that I understood this the entire time.",
+ "raw": {
+ "normalized_reference": "i've been faking that i understood this the entire time.",
+ "normalized_hypothesis": "i've been faking that i understood this the entire time.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i have been faking that i understood this the entire time",
+ "normalized_hypothesis": "i have been faking that i understood this the entire time",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0201",
+ "category": "phonetic_balance",
+ "text": "Peace be unto you and grace before the Lord.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0201.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Peace be unto you and grace before the Lord.",
+ "raw": {
+ "normalized_reference": "peace be unto you and grace before the lord.",
+ "normalized_hypothesis": "peace be unto you and grace before the lord.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "peace be unto you and grace before the lord",
+ "normalized_hypothesis": "peace be unto you and grace before the lord",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0202",
+ "category": "phonetic_balance",
+ "text": "From now on we're pals.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0202.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "From now on, we're pals.",
+ "raw": {
+ "normalized_reference": "from now on we're pals.",
+ "normalized_hypothesis": "from now on, we're pals.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "from now on we are pals",
+ "normalized_hypothesis": "from now on we are pals",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0203",
+ "category": "phonetic_balance",
+ "text": "Here we allow our solicitors to look after our legal work.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0203.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Here we allow our solicitors to look after our legal work.",
+ "raw": {
+ "normalized_reference": "here we allow our solicitors to look after our legal work.",
+ "normalized_hypothesis": "here we allow our solicitors to look after our legal work.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "here we allow our solicitors to look after our legal work",
+ "normalized_hypothesis": "here we allow our solicitors to look after our legal work",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0204",
+ "category": "phonetic_balance",
+ "text": "Her words sent a strange chill through Philip.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0204.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Her words sent a strange chill through Philip.",
+ "raw": {
+ "normalized_reference": "her words sent a strange chill through philip.",
+ "normalized_hypothesis": "her words sent a strange chill through philip.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "her words sent a strange chill through philip",
+ "normalized_hypothesis": "her words sent a strange chill through philip",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0205",
+ "category": "phonetic_balance",
+ "text": "It was introduced by Representative Dick of Ohio.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0205.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It was introduced by Representative Dick of Ohio.",
+ "raw": {
+ "normalized_reference": "it was introduced by representative dick of ohio.",
+ "normalized_hypothesis": "it was introduced by representative dick of ohio.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it was introduced by representative dick of ohio",
+ "normalized_hypothesis": "it was introduced by representative dick of ohio",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0206",
+ "category": "phonetic_balance",
+ "text": "It's the aurora borealis.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0206.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It's the Aurora Borealis.",
+ "raw": {
+ "normalized_reference": "it's the aurora borealis.",
+ "normalized_hypothesis": "it's the aurora borealis.",
+ "errors": 0,
+ "words": 4,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it is the aurora borealis",
+ "normalized_hypothesis": "it is the aurora borealis",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0207",
+ "category": "phonetic_balance",
+ "text": "The boy grew and prospered.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0207.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The boy grew and prospered.",
+ "raw": {
+ "normalized_reference": "the boy grew and prospered.",
+ "normalized_hypothesis": "the boy grew and prospered.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the boy grew and prospered",
+ "normalized_hypothesis": "the boy grew and prospered",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0208",
+ "category": "phonetic_balance",
+ "text": "His voice was passionately rebellious.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0208.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "His voice was passionately rebellious.",
+ "raw": {
+ "normalized_reference": "his voice was passionately rebellious.",
+ "normalized_hypothesis": "his voice was passionately rebellious.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "his voice was passionately rebellious",
+ "normalized_hypothesis": "his voice was passionately rebellious",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0209",
+ "category": "phonetic_balance",
+ "text": "That's where they cut off the Scottish Chiefs and killed all hands.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0209.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "That's where they cut off the Scottish chiefs and killed all hands.",
+ "raw": {
+ "normalized_reference": "that's where they cut off the scottish chiefs and killed all hands.",
+ "normalized_hypothesis": "that's where they cut off the scottish chiefs and killed all hands.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "that is where they cut off the scottish chiefs and killed all hands",
+ "normalized_hypothesis": "that is where they cut off the scottish chiefs and killed all hands",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0210",
+ "category": "phonetic_balance",
+ "text": "And Tom King patiently endured.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0210.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "And Tom King patiently endured.",
+ "raw": {
+ "normalized_reference": "and tom king patiently endured.",
+ "normalized_hypothesis": "and tom king patiently endured.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "and tom king patiently endured",
+ "normalized_hypothesis": "and tom king patiently endured",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0211",
+ "category": "phonetic_balance",
+ "text": "From that moment his friendship for Belize turns to hatred and jealousy.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0211.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "From that moment his friendship for Belize turns to hatred and jealousy.",
+ "raw": {
+ "normalized_reference": "from that moment his friendship for belize turns to hatred and jealousy.",
+ "normalized_hypothesis": "from that moment his friendship for belize turns to hatred and jealousy.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "from that moment his friendship for belize turns to hatred and jealousy",
+ "normalized_hypothesis": "from that moment his friendship for belize turns to hatred and jealousy",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0212",
+ "category": "phonetic_balance",
+ "text": "Then you don't believe in altruism.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0212.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Then you don't believe in altruism.",
+ "raw": {
+ "normalized_reference": "then you don't believe in altruism.",
+ "normalized_hypothesis": "then you don't believe in altruism.",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "then you do not believe in altruism",
+ "normalized_hypothesis": "then you do not believe in altruism",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0213",
+ "category": "phonetic_balance",
+ "text": "What Pascal glimpsed with the vision of a seer, I have lived.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0213.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "What Pascal glimpsed with the vision of a seer, I have lived.",
+ "raw": {
+ "normalized_reference": "what pascal glimpsed with the vision of a seer, i have lived.",
+ "normalized_hypothesis": "what pascal glimpsed with the vision of a seer, i have lived.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "what pascal glimpsed with the vision of a seer i have lived",
+ "normalized_hypothesis": "what pascal glimpsed with the vision of a seer i have lived",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0214",
+ "category": "phonetic_balance",
+ "text": "You were making them talk shop, Ruth charged him.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0214.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "You were making them talk shop, Ruth charged him.",
+ "raw": {
+ "normalized_reference": "you were making them talk shop, ruth charged him.",
+ "normalized_hypothesis": "you were making them talk shop, ruth charged him.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "you were making them talk shop ruth charged him",
+ "normalized_hypothesis": "you were making them talk shop ruth charged him",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0215",
+ "category": "phonetic_balance",
+ "text": "Philip snatched at the letter which Gregson held out to him.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0215.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Philip snatched at the letter which Gregson held out to him.",
+ "raw": {
+ "normalized_reference": "philip snatched at the letter which gregson held out to him.",
+ "normalized_hypothesis": "philip snatched at the letter which gregson held out to him.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "philip snatched at the letter which gregson held out to him",
+ "normalized_hypothesis": "philip snatched at the letter which gregson held out to him",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0216",
+ "category": "phonetic_balance",
+ "text": "For a time the exciting thrill of his adventure was gone.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0216.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "For a time, the exciting thrill of his adventure was gone.",
+ "raw": {
+ "normalized_reference": "for a time the exciting thrill of his adventure was gone.",
+ "normalized_hypothesis": "for a time, the exciting thrill of his adventure was gone.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "for a time the exciting thrill of his adventure was gone",
+ "normalized_hypothesis": "for a time the exciting thrill of his adventure was gone",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0217",
+ "category": "phonetic_balance",
+ "text": "Obviously, it was a disease that could be contracted by contact.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0217.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Obviously, it was a disease that could be contracted by contact.",
+ "raw": {
+ "normalized_reference": "obviously, it was a disease that could be contracted by contact.",
+ "normalized_hypothesis": "obviously, it was a disease that could be contracted by contact.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "obviously it was a disease that could be contracted by contact",
+ "normalized_hypothesis": "obviously it was a disease that could be contracted by contact",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0218",
+ "category": "phonetic_balance",
+ "text": "Blind with rage, he darted in.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0218.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Blind with rage, he darted.",
+ "raw": {
+ "normalized_reference": "blind with rage, he darted in.",
+ "normalized_hypothesis": "blind with rage, he darted.",
+ "errors": 1,
+ "words": 6,
+ "wer": 0.16666666666666666
+ },
+ "semantic": {
+ "normalized_reference": "blind with rage he darted in",
+ "normalized_hypothesis": "blind with rage he darted",
+ "errors": 1,
+ "words": 6,
+ "wer": 0.16666666666666666
+ }
+ },
+ {
+ "id": "release500_0219",
+ "category": "phonetic_balance",
+ "text": "You live on an income which your father earned.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0219.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "You live on an income which your father earned.",
+ "raw": {
+ "normalized_reference": "you live on an income which your father earned.",
+ "normalized_hypothesis": "you live on an income which your father earned.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "you live on an income which your father earned",
+ "normalized_hypothesis": "you live on an income which your father earned",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0220",
+ "category": "phonetic_balance",
+ "text": "Something about it seemed to fascinate him, to challenge his presence.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0220.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Something about it seemed to fascinate him, to challenge his presence.",
+ "raw": {
+ "normalized_reference": "something about it seemed to fascinate him, to challenge his presence.",
+ "normalized_hypothesis": "something about it seemed to fascinate him, to challenge his presence.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "something about it seemed to fascinate him to challenge his presence",
+ "normalized_hypothesis": "something about it seemed to fascinate him to challenge his presence",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0221",
+ "category": "phonetic_balance",
+ "text": "Her beautiful hair was done up in shining coils.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0221.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Her beautiful hair was done up in shining coils.",
+ "raw": {
+ "normalized_reference": "her beautiful hair was done up in shining coils.",
+ "normalized_hypothesis": "her beautiful hair was done up in shining coils.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "her beautiful hair was done up in shining coils",
+ "normalized_hypothesis": "her beautiful hair was done up in shining coils",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0222",
+ "category": "phonetic_balance",
+ "text": "It does, was her audacious answer.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0222.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It does was her audacious answer.",
+ "raw": {
+ "normalized_reference": "it does, was her audacious answer.",
+ "normalized_hypothesis": "it does was her audacious answer.",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it does was her audacious answer",
+ "normalized_hypothesis": "it does was her audacious answer",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0223",
+ "category": "phonetic_balance",
+ "text": "He caught himself with a jerk.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0223.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He caught himself with a jerk.",
+ "raw": {
+ "normalized_reference": "he caught himself with a jerk.",
+ "normalized_hypothesis": "he caught himself with a jerk.",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he caught himself with a jerk",
+ "normalized_hypothesis": "he caught himself with a jerk",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0224",
+ "category": "phonetic_balance",
+ "text": "He did not believe in the burning of daylight for such a luxury.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0224.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He did not believe in the burning of daylight for such a luxury.",
+ "raw": {
+ "normalized_reference": "he did not believe in the burning of daylight for such a luxury.",
+ "normalized_hypothesis": "he did not believe in the burning of daylight for such a luxury.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he did not believe in the burning of daylight for such a luxury",
+ "normalized_hypothesis": "he did not believe in the burning of daylight for such a luxury",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0225",
+ "category": "phonetic_balance",
+ "text": "He had comparatively no advantages at first.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0225.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He had comparatively no advantages at first.",
+ "raw": {
+ "normalized_reference": "he had comparatively no advantages at first.",
+ "normalized_hypothesis": "he had comparatively no advantages at first.",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he had comparatively no advantages at 1st",
+ "normalized_hypothesis": "he had comparatively no advantages at 1st",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0226",
+ "category": "phonetic_balance",
+ "text": "Such men believe, when they come together.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0226.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Such men believe when they come together.",
+ "raw": {
+ "normalized_reference": "such men believe, when they come together.",
+ "normalized_hypothesis": "such men believe when they come together.",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "such men believe when they come together",
+ "normalized_hypothesis": "such men believe when they come together",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0227",
+ "category": "phonetic_balance",
+ "text": "The ship should be in within a week or ten days.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0227.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The ship should be in within a week or ten days.",
+ "raw": {
+ "normalized_reference": "the ship should be in within a week or ten days.",
+ "normalized_hypothesis": "the ship should be in within a week or ten days.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the ship should be in within a week or 10 days",
+ "normalized_hypothesis": "the ship should be in within a week or 10 days",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0228",
+ "category": "phonetic_balance",
+ "text": "Such a dog the wise driver kills, or turns loose.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0228.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Such a dog the wise driver kills or turns loose.",
+ "raw": {
+ "normalized_reference": "such a dog the wise driver kills, or turns loose.",
+ "normalized_hypothesis": "such a dog the wise driver kills or turns loose.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "such a dog the wise driver kills or turns loose",
+ "normalized_hypothesis": "such a dog the wise driver kills or turns loose",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0229",
+ "category": "phonetic_balance",
+ "text": "You have heard always how he was the lover of the Princess Naomi.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0229.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "You have heard always how he was the lover of the princess Naomi.",
+ "raw": {
+ "normalized_reference": "you have heard always how he was the lover of the princess naomi.",
+ "normalized_hypothesis": "you have heard always how he was the lover of the princess naomi.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "you have heard always how he was the lover of the princess naomi",
+ "normalized_hypothesis": "you have heard always how he was the lover of the princess naomi",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0230",
+ "category": "phonetic_balance",
+ "text": "I want to know how all this is possible.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0230.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I want to know how all this is possible.",
+ "raw": {
+ "normalized_reference": "i want to know how all this is possible.",
+ "normalized_hypothesis": "i want to know how all this is possible.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i want to know how all this is possible",
+ "normalized_hypothesis": "i want to know how all this is possible",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0231",
+ "category": "phonetic_balance",
+ "text": "It was my idea to a tee.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0231.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It was my idea to a T.",
+ "raw": {
+ "normalized_reference": "it was my idea to a tee.",
+ "normalized_hypothesis": "it was my idea to a t.",
+ "errors": 1,
+ "words": 7,
+ "wer": 0.14285714285714285
+ },
+ "semantic": {
+ "normalized_reference": "it was my idea to a tee",
+ "normalized_hypothesis": "it was my idea to a t",
+ "errors": 1,
+ "words": 7,
+ "wer": 0.14285714285714285
+ }
+ },
+ {
+ "id": "release500_0232",
+ "category": "phonetic_balance",
+ "text": "They likewise are disinclined to being eaten.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0232.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "They likewise are disinclined to being eaten.",
+ "raw": {
+ "normalized_reference": "they likewise are disinclined to being eaten.",
+ "normalized_hypothesis": "they likewise are disinclined to being eaten.",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "they likewise are disinclined to being eaten",
+ "normalized_hypothesis": "they likewise are disinclined to being eaten",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0233",
+ "category": "phonetic_balance",
+ "text": "This state of mind comes of an undue prominence of the ego.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0233.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "This state of mind comes of an undue prominence of the ego.",
+ "raw": {
+ "normalized_reference": "this state of mind comes of an undue prominence of the ego.",
+ "normalized_hypothesis": "this state of mind comes of an undue prominence of the ego.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "this state of mind comes of an undue prominence of the ego",
+ "normalized_hypothesis": "this state of mind comes of an undue prominence of the ego",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0234",
+ "category": "phonetic_balance",
+ "text": "Then it is as I said, Womble announced with finality.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0234.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Then it is as I said, Womble announced with finality.",
+ "raw": {
+ "normalized_reference": "then it is as i said, womble announced with finality.",
+ "normalized_hypothesis": "then it is as i said, womble announced with finality.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "then it is as i said womble announced with finality",
+ "normalized_hypothesis": "then it is as i said womble announced with finality",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0235",
+ "category": "phonetic_balance",
+ "text": "I can see that knife now.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0235.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I can see that knife now.",
+ "raw": {
+ "normalized_reference": "i can see that knife now.",
+ "normalized_hypothesis": "i can see that knife now.",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i can see that knife now",
+ "normalized_hypothesis": "i can see that knife now",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0236",
+ "category": "phonetic_balance",
+ "text": "If I meet her again I shall apologize, said Eileen.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0236.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "If I meet her again, I shall apologize, said Eileen.",
+ "raw": {
+ "normalized_reference": "if i meet her again i shall apologize, said eileen.",
+ "normalized_hypothesis": "if i meet her again, i shall apologize, said eileen.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "if i meet her again i shall apologize said eileen",
+ "normalized_hypothesis": "if i meet her again i shall apologize said eileen",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0237",
+ "category": "phonetic_balance",
+ "text": "A wildly exciting time was his during the week preceding Thursday the eighteenth.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0237.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "A wildly exciting time was his during the week preceding Thursday the 18th.",
+ "raw": {
+ "normalized_reference": "a wildly exciting time was his during the week preceding thursday the eighteenth.",
+ "normalized_hypothesis": "a wildly exciting time was his during the week preceding thursday the 18th.",
+ "errors": 1,
+ "words": 13,
+ "wer": 0.07692307692307693
+ },
+ "semantic": {
+ "normalized_reference": "a wildly exciting time was his during the week preceding thursday the 18th",
+ "normalized_hypothesis": "a wildly exciting time was his during the week preceding thursday the 18th",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0238",
+ "category": "phonetic_balance",
+ "text": "In what bucolic school of fence he had been taught was beyond imagining.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0238.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "In what bucolic school of fence he had been taught was beyond imagining.",
+ "raw": {
+ "normalized_reference": "in what bucolic school of fence he had been taught was beyond imagining.",
+ "normalized_hypothesis": "in what bucolic school of fence he had been taught was beyond imagining.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "in what bucolic school of fence he had been taught was beyond imagining",
+ "normalized_hypothesis": "in what bucolic school of fence he had been taught was beyond imagining",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0239",
+ "category": "phonetic_balance",
+ "text": "So she said, the irate skipper dashed on.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0239.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "So she said the irate skipper dashed on.",
+ "raw": {
+ "normalized_reference": "so she said, the irate skipper dashed on.",
+ "normalized_hypothesis": "so she said the irate skipper dashed on.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "so she said the irate skipper dashed on",
+ "normalized_hypothesis": "so she said the irate skipper dashed on",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0240",
+ "category": "phonetic_balance",
+ "text": "It is not an attempt to smash the market.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0240.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It is not an attempt to smash the market.",
+ "raw": {
+ "normalized_reference": "it is not an attempt to smash the market.",
+ "normalized_hypothesis": "it is not an attempt to smash the market.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it is not an attempt to smash the market",
+ "normalized_hypothesis": "it is not an attempt to smash the market",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0241",
+ "category": "phonetic_balance",
+ "text": "He destroyed everything that had belonged to the woman.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0241.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He destroyed everything that had belonged to the woman.",
+ "raw": {
+ "normalized_reference": "he destroyed everything that had belonged to the woman.",
+ "normalized_hypothesis": "he destroyed everything that had belonged to the woman.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he destroyed everything that had belonged to the woman",
+ "normalized_hypothesis": "he destroyed everything that had belonged to the woman",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0242",
+ "category": "phonetic_balance",
+ "text": "Did Thorpe go to see any one in Churchill.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0242.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Did Thorpe go to see anyone in Churchill?",
+ "raw": {
+ "normalized_reference": "did thorpe go to see any one in churchill.",
+ "normalized_hypothesis": "did thorpe go to see anyone in churchill?",
+ "errors": 2,
+ "words": 9,
+ "wer": 0.2222222222222222
+ },
+ "semantic": {
+ "normalized_reference": "did thorpe go to see any one in churchill",
+ "normalized_hypothesis": "did thorpe go to see anyone in churchill",
+ "errors": 2,
+ "words": 9,
+ "wer": 0.2222222222222222
+ }
+ },
+ {
+ "id": "release500_0243",
+ "category": "phonetic_balance",
+ "text": "Many other similar disconcerting omissions will be noticed in the Manuscript.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0243.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Many other similar disconcerting omissions will be noticed in the manuscript.",
+ "raw": {
+ "normalized_reference": "many other similar disconcerting omissions will be noticed in the manuscript.",
+ "normalized_hypothesis": "many other similar disconcerting omissions will be noticed in the manuscript.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "many other similar disconcerting omissions will be noticed in the manuscript",
+ "normalized_hypothesis": "many other similar disconcerting omissions will be noticed in the manuscript",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0244",
+ "category": "phonetic_balance",
+ "text": "The ringing of the big bell aroused him.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0244.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The ringing of the big bell aroused him.",
+ "raw": {
+ "normalized_reference": "the ringing of the big bell aroused him.",
+ "normalized_hypothesis": "the ringing of the big bell aroused him.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the ringing of the big bell aroused him",
+ "normalized_hypothesis": "the ringing of the big bell aroused him",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0245",
+ "category": "phonetic_balance",
+ "text": "A sinewy hand, dripping with water, was clutching the rail.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0245.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "A sinewy hand dripping with water was clutching the rail.",
+ "raw": {
+ "normalized_reference": "a sinewy hand, dripping with water, was clutching the rail.",
+ "normalized_hypothesis": "a sinewy hand dripping with water was clutching the rail.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "a sinewy hand dripping with water was clutching the rail",
+ "normalized_hypothesis": "a sinewy hand dripping with water was clutching the rail",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0246",
+ "category": "phonetic_balance",
+ "text": "Enters now the psychology of the situation.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0246.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Enters now the psychology of the situation.",
+ "raw": {
+ "normalized_reference": "enters now the psychology of the situation.",
+ "normalized_hypothesis": "enters now the psychology of the situation.",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "enters now the psychology of the situation",
+ "normalized_hypothesis": "enters now the psychology of the situation",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0247",
+ "category": "phonetic_balance",
+ "text": "Wash your hands of me.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0247.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Wash your hands of me.",
+ "raw": {
+ "normalized_reference": "wash your hands of me.",
+ "normalized_hypothesis": "wash your hands of me.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "wash your hands of me",
+ "normalized_hypothesis": "wash your hands of me",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0248",
+ "category": "phonetic_balance",
+ "text": "He confessed that the sketch had startled him.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0248.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He confessed that the sketch had startled him.",
+ "raw": {
+ "normalized_reference": "he confessed that the sketch had startled him.",
+ "normalized_hypothesis": "he confessed that the sketch had startled him.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he confessed that the sketch had startled him",
+ "normalized_hypothesis": "he confessed that the sketch had startled him",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0249",
+ "category": "phonetic_balance",
+ "text": "Her face was against his breast.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0249.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Her face was against his breast.",
+ "raw": {
+ "normalized_reference": "her face was against his breast.",
+ "normalized_hypothesis": "her face was against his breast.",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "her face was against his breast",
+ "normalized_hypothesis": "her face was against his breast",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0250",
+ "category": "phonetic_balance",
+ "text": "He was a pariah; a wanderer without a friend or a home.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0250.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He was a pariah, a wanderer without a friend or home.",
+ "raw": {
+ "normalized_reference": "he was a pariah; a wanderer without a friend or a home.",
+ "normalized_hypothesis": "he was a pariah, a wanderer without a friend or home.",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "he was a pariah a wanderer without a friend or a home",
+ "normalized_hypothesis": "he was a pariah a wanderer without a friend or home",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ }
+ },
+ {
+ "id": "release500_0251",
+ "category": "phonetic_balance",
+ "text": "A cry of joy burst from Philip's lips.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0251.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "A cry of joy burst from Philip's lips.",
+ "raw": {
+ "normalized_reference": "a cry of joy burst from philip's lips.",
+ "normalized_hypothesis": "a cry of joy burst from philip's lips.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "a cry of joy burst from philip is lips",
+ "normalized_hypothesis": "a cry of joy burst from philip is lips",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0252",
+ "category": "phonetic_balance",
+ "text": "And the big Persian knew of his existence before he did of hers.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0252.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "And the big Persian knew of his existence before he did of hers.",
+ "raw": {
+ "normalized_reference": "and the big persian knew of his existence before he did of hers.",
+ "normalized_hypothesis": "and the big persian knew of his existence before he did of hers.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "and the big persian knew of his existence before he did of hers",
+ "normalized_hypothesis": "and the big persian knew of his existence before he did of hers",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0253",
+ "category": "phonetic_balance",
+ "text": "The men stared into each other's face.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0253.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The men stared into each other's face.",
+ "raw": {
+ "normalized_reference": "the men stared into each other's face.",
+ "normalized_hypothesis": "the men stared into each other's face.",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the men stared into each other is face",
+ "normalized_hypothesis": "the men stared into each other is face",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0254",
+ "category": "phonetic_balance",
+ "text": "It is the fire, partly, she said.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0254.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It is the fire, partly, she said.",
+ "raw": {
+ "normalized_reference": "it is the fire, partly, she said.",
+ "normalized_hypothesis": "it is the fire, partly, she said.",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it is the fire partly she said",
+ "normalized_hypothesis": "it is the fire partly she said",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0255",
+ "category": "phonetic_balance",
+ "text": "It seemed nearer to him since he had seen and talked with Gregson.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0255.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It seemed nearer to him since he had seen and talked with Gregson.",
+ "raw": {
+ "normalized_reference": "it seemed nearer to him since he had seen and talked with gregson.",
+ "normalized_hypothesis": "it seemed nearer to him since he had seen and talked with gregson.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it seemed nearer to him since he had seen and talked with gregson",
+ "normalized_hypothesis": "it seemed nearer to him since he had seen and talked with gregson",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0256",
+ "category": "phonetic_balance",
+ "text": "Among my minor afflictions, I may mention a new and mysterious one.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0256.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Among my minor afflictions, I may mention a new and mysterious one.",
+ "raw": {
+ "normalized_reference": "among my minor afflictions, i may mention a new and mysterious one.",
+ "normalized_hypothesis": "among my minor afflictions, i may mention a new and mysterious one.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "among my minor afflictions i may mention a new and mysterious one",
+ "normalized_hypothesis": "among my minor afflictions i may mention a new and mysterious one",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0257",
+ "category": "phonetic_balance",
+ "text": "Between him and all domestic animals there must be no hostilities.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0257.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Between him and all domestic animals there must be no hostilities.",
+ "raw": {
+ "normalized_reference": "between him and all domestic animals there must be no hostilities.",
+ "normalized_hypothesis": "between him and all domestic animals there must be no hostilities.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "between him and all domestic animals there must be no hostilities",
+ "normalized_hypothesis": "between him and all domestic animals there must be no hostilities",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0258",
+ "category": "phonetic_balance",
+ "text": "With these arguments he convinced himself that he should go on alone.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0258.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "With these arguments he convinced himself that he should go on alone.",
+ "raw": {
+ "normalized_reference": "with these arguments he convinced himself that he should go on alone.",
+ "normalized_hypothesis": "with these arguments he convinced himself that he should go on alone.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "with these arguments he convinced himself that he should go on alone",
+ "normalized_hypothesis": "with these arguments he convinced himself that he should go on alone",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0259",
+ "category": "phonetic_balance",
+ "text": "We are consumed in our own flesh-pots.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0259.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "We are consumed in our own flesh pots.",
+ "raw": {
+ "normalized_reference": "we are consumed in our own flesh-pots.",
+ "normalized_hypothesis": "we are consumed in our own flesh pots.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "we are consumed in our own flesh pots",
+ "normalized_hypothesis": "we are consumed in our own flesh pots",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0260",
+ "category": "phonetic_balance",
+ "text": "Mercedes screamed, cried, laughed, and manifested the chaotic abandonment of hysteria.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0260.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Mercedes screamed, cried, laughed, and manifested the chaotic abandonment of hysteria.",
+ "raw": {
+ "normalized_reference": "mercedes screamed, cried, laughed, and manifested the chaotic abandonment of hysteria.",
+ "normalized_hypothesis": "mercedes screamed, cried, laughed, and manifested the chaotic abandonment of hysteria.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "mercedes screamed cried laughed and manifested the chaotic abandonment of hysteria",
+ "normalized_hypothesis": "mercedes screamed cried laughed and manifested the chaotic abandonment of hysteria",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0261",
+ "category": "phonetic_balance",
+ "text": "Philip made no effort to follow.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0261.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Philip made no effort to follow.",
+ "raw": {
+ "normalized_reference": "philip made no effort to follow.",
+ "normalized_hypothesis": "philip made no effort to follow.",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "philip made no effort to follow",
+ "normalized_hypothesis": "philip made no effort to follow",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0262",
+ "category": "phonetic_balance",
+ "text": "She said with chattering teeth.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0262.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "She said with chattering teeth.",
+ "raw": {
+ "normalized_reference": "she said with chattering teeth.",
+ "normalized_hypothesis": "she said with chattering teeth.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "she said with chattering teeth",
+ "normalized_hypothesis": "she said with chattering teeth",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0263",
+ "category": "phonetic_balance",
+ "text": "Halfway around the track one donkey got into an argument with its rider.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0263.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Halfway around the track, one donkey got into an argument with its rider.",
+ "raw": {
+ "normalized_reference": "halfway around the track one donkey got into an argument with its rider.",
+ "normalized_hypothesis": "halfway around the track, one donkey got into an argument with its rider.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "halfway around the track one donkey got into an argument with its rider",
+ "normalized_hypothesis": "halfway around the track one donkey got into an argument with its rider",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0264",
+ "category": "phonetic_balance",
+ "text": "Earth and gravel seemed to fill the pan.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0264.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Earth and gravel seemed to fill the pan.",
+ "raw": {
+ "normalized_reference": "earth and gravel seemed to fill the pan.",
+ "normalized_hypothesis": "earth and gravel seemed to fill the pan.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "earth and gravel seemed to fill the pan",
+ "normalized_hypothesis": "earth and gravel seemed to fill the pan",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0265",
+ "category": "phonetic_balance",
+ "text": "Her achievements with cocoanuts were a revelation.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0265.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Her achievements with coconuts were a revelation.",
+ "raw": {
+ "normalized_reference": "her achievements with cocoanuts were a revelation.",
+ "normalized_hypothesis": "her achievements with coconuts were a revelation.",
+ "errors": 1,
+ "words": 7,
+ "wer": 0.14285714285714285
+ },
+ "semantic": {
+ "normalized_reference": "her achievements with cocoanuts were a revelation",
+ "normalized_hypothesis": "her achievements with coconuts were a revelation",
+ "errors": 1,
+ "words": 7,
+ "wer": 0.14285714285714285
+ }
+ },
+ {
+ "id": "release500_0266",
+ "category": "phonetic_balance",
+ "text": "For a full minute he crouched and listened.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0266.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "For a full minute he crouched and listened.",
+ "raw": {
+ "normalized_reference": "for a full minute he crouched and listened.",
+ "normalized_hypothesis": "for a full minute he crouched and listened.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "for a full minute he crouched and listened",
+ "normalized_hypothesis": "for a full minute he crouched and listened",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0267",
+ "category": "phonetic_balance",
+ "text": "It was edged with ice.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0267.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It was edged with ice.",
+ "raw": {
+ "normalized_reference": "it was edged with ice.",
+ "normalized_hypothesis": "it was edged with ice.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it was edged with ice",
+ "normalized_hypothesis": "it was edged with ice",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0268",
+ "category": "phonetic_balance",
+ "text": "The Portuguese boy passed the Hawaiian.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0268.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The Portuguese boy passed the Hawaiian.",
+ "raw": {
+ "normalized_reference": "the portuguese boy passed the hawaiian.",
+ "normalized_hypothesis": "the portuguese boy passed the hawaiian.",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the portuguese boy passed the hawaiian",
+ "normalized_hypothesis": "the portuguese boy passed the hawaiian",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0269",
+ "category": "phonetic_balance",
+ "text": "But he did not broach it, preferring to mature it carefully.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0269.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "But he did not broach it, preferring to mature it carefully.",
+ "raw": {
+ "normalized_reference": "but he did not broach it, preferring to mature it carefully.",
+ "normalized_hypothesis": "but he did not broach it, preferring to mature it carefully.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "but he did not broach it preferring to mature it carefully",
+ "normalized_hypothesis": "but he did not broach it preferring to mature it carefully",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0270",
+ "category": "phonetic_balance",
+ "text": "Come on, Del Mar challenged.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0270.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Come on, Delmar Challenged.",
+ "raw": {
+ "normalized_reference": "come on, del mar challenged.",
+ "normalized_hypothesis": "come on, delmar challenged.",
+ "errors": 2,
+ "words": 5,
+ "wer": 0.4
+ },
+ "semantic": {
+ "normalized_reference": "come on del mar challenged",
+ "normalized_hypothesis": "come on delmar challenged",
+ "errors": 2,
+ "words": 5,
+ "wer": 0.4
+ }
+ },
+ {
+ "id": "release500_0271",
+ "category": "phonetic_balance",
+ "text": "For a full half minute Jeanne looked at him without speaking.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0271.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "For a full half minute, Gene looked at him without speaking.",
+ "raw": {
+ "normalized_reference": "for a full half minute jeanne looked at him without speaking.",
+ "normalized_hypothesis": "for a full half minute, gene looked at him without speaking.",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ },
+ "semantic": {
+ "normalized_reference": "for a full half minute jeanne looked at him without speaking",
+ "normalized_hypothesis": "for a full half minute gene looked at him without speaking",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ }
+ },
+ {
+ "id": "release500_0272",
+ "category": "phonetic_balance",
+ "text": "He bore no grudges and had few enemies.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0272.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He bore no grudges and had few enemies.",
+ "raw": {
+ "normalized_reference": "he bore no grudges and had few enemies.",
+ "normalized_hypothesis": "he bore no grudges and had few enemies.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he bore no grudges and had few enemies",
+ "normalized_hypothesis": "he bore no grudges and had few enemies",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0273",
+ "category": "phonetic_balance",
+ "text": "I'll go over tomorrow afternoon.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0273.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I'll go over tomorrow afternoon.",
+ "raw": {
+ "normalized_reference": "i'll go over tomorrow afternoon.",
+ "normalized_hypothesis": "i'll go over tomorrow afternoon.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i will go over tomorrow afternoon",
+ "normalized_hypothesis": "i will go over tomorrow afternoon",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0274",
+ "category": "phonetic_balance",
+ "text": "Well, there are better men in Hawaii, that's all.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0274.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Well, there are better men in Hawaii, that's all.",
+ "raw": {
+ "normalized_reference": "well, there are better men in hawaii, that's all.",
+ "normalized_hypothesis": "well, there are better men in hawaii, that's all.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "well there are better men in hawaii that is all",
+ "normalized_hypothesis": "well there are better men in hawaii that is all",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0275",
+ "category": "phonetic_balance",
+ "text": "He was an athlete and a giant.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0275.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He was an athlete and a giant.",
+ "raw": {
+ "normalized_reference": "he was an athlete and a giant.",
+ "normalized_hypothesis": "he was an athlete and a giant.",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he was an athlete and a giant",
+ "normalized_hypothesis": "he was an athlete and a giant",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0276",
+ "category": "phonetic_balance",
+ "text": "McVeigh when he returns from a trip to Honolulu.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0276.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "McVeigh when he returns from a trip to Honolulu.",
+ "raw": {
+ "normalized_reference": "mcveigh when he returns from a trip to honolulu.",
+ "normalized_hypothesis": "mcveigh when he returns from a trip to honolulu.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "mcveigh when he returns from a trip to honolulu",
+ "normalized_hypothesis": "mcveigh when he returns from a trip to honolulu",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0277",
+ "category": "phonetic_balance",
+ "text": "You're a devil for fighting, and will surely win.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0277.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "You're a devil for fighting and will surely win.",
+ "raw": {
+ "normalized_reference": "you're a devil for fighting, and will surely win.",
+ "normalized_hypothesis": "you're a devil for fighting and will surely win.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "you are a devil for fighting and will surely win",
+ "normalized_hypothesis": "you are a devil for fighting and will surely win",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0278",
+ "category": "phonetic_balance",
+ "text": "All an appearance can know is mirage.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0278.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "All an appearance can know is Mirage.",
+ "raw": {
+ "normalized_reference": "all an appearance can know is mirage.",
+ "normalized_hypothesis": "all an appearance can know is mirage.",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "all an appearance can know is mirage",
+ "normalized_hypothesis": "all an appearance can know is mirage",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0279",
+ "category": "phonetic_balance",
+ "text": "You fired me out of your house, in short.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0279.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "You fired me out of your house in short.",
+ "raw": {
+ "normalized_reference": "you fired me out of your house, in short.",
+ "normalized_hypothesis": "you fired me out of your house in short.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "you fired me out of your house in short",
+ "normalized_hypothesis": "you fired me out of your house in short",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0280",
+ "category": "phonetic_balance",
+ "text": "He made sure that the magazine was loaded, and resumed his paddling.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0280.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He made sure that the magazine was loaded and resumed his paddling.",
+ "raw": {
+ "normalized_reference": "he made sure that the magazine was loaded, and resumed his paddling.",
+ "normalized_hypothesis": "he made sure that the magazine was loaded and resumed his paddling.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he made sure that the magazine was loaded and resumed his paddling",
+ "normalized_hypothesis": "he made sure that the magazine was loaded and resumed his paddling",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0281",
+ "category": "phonetic_balance",
+ "text": "Change chairs, Daylight commanded.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0281.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Change chairs. Daylight commanded.",
+ "raw": {
+ "normalized_reference": "change chairs, daylight commanded.",
+ "normalized_hypothesis": "change chairs. daylight commanded.",
+ "errors": 0,
+ "words": 4,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "change chairs daylight commanded",
+ "normalized_hypothesis": "change chairs daylight commanded",
+ "errors": 0,
+ "words": 4,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0282",
+ "category": "phonetic_balance",
+ "text": "By that answer my professional medical prestige stood or fell.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0282.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "By that answer, my professional medical prestige stood or fell.",
+ "raw": {
+ "normalized_reference": "by that answer my professional medical prestige stood or fell.",
+ "normalized_hypothesis": "by that answer, my professional medical prestige stood or fell.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "by that answer my professional medical prestige stood or fell",
+ "normalized_hypothesis": "by that answer my professional medical prestige stood or fell",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0283",
+ "category": "phonetic_balance",
+ "text": "The moon had already begun its westward decline.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0283.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The moon had already begun its westward decline.",
+ "raw": {
+ "normalized_reference": "the moon had already begun its westward decline.",
+ "normalized_hypothesis": "the moon had already begun its westward decline.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the moon had already begun its westward decline",
+ "normalized_hypothesis": "the moon had already begun its westward decline",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0284",
+ "category": "phonetic_balance",
+ "text": "At sea, Monday, March 16, 1908.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0284.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "At Sea Monday, March 16, 1988",
+ "raw": {
+ "normalized_reference": "at sea, monday, march 16, 1908.",
+ "normalized_hypothesis": "at sea monday, march 16, 1988",
+ "errors": 1,
+ "words": 6,
+ "wer": 0.16666666666666666
+ },
+ "semantic": {
+ "normalized_reference": "at sea monday march 16 1908",
+ "normalized_hypothesis": "at sea monday march 16 1988",
+ "errors": 1,
+ "words": 6,
+ "wer": 0.16666666666666666
+ }
+ },
+ {
+ "id": "release500_0285",
+ "category": "phonetic_balance",
+ "text": "And wherever I ranged, the way lay along alcohol-drenched roads.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0285.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "And wherever I ranged, the way lay along alcohol-drenched roads.",
+ "raw": {
+ "normalized_reference": "and wherever i ranged, the way lay along alcohol-drenched roads.",
+ "normalized_hypothesis": "and wherever i ranged, the way lay along alcohol-drenched roads.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "and wherever i ranged the way lay along alcohol drenched roads",
+ "normalized_hypothesis": "and wherever i ranged the way lay along alcohol drenched roads",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0286",
+ "category": "phonetic_balance",
+ "text": "The date was nearly eighteen years old.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0286.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The date was nearly 18 years old.",
+ "raw": {
+ "normalized_reference": "the date was nearly eighteen years old.",
+ "normalized_hypothesis": "the date was nearly 18 years old.",
+ "errors": 1,
+ "words": 7,
+ "wer": 0.14285714285714285
+ },
+ "semantic": {
+ "normalized_reference": "the date was nearly 18 years old",
+ "normalized_hypothesis": "the date was nearly 18 years old",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0287",
+ "category": "phonetic_balance",
+ "text": "Very early in my life, I separated from my mother.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0287.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Very early in my life, I separated from my mother.",
+ "raw": {
+ "normalized_reference": "very early in my life, i separated from my mother.",
+ "normalized_hypothesis": "very early in my life, i separated from my mother.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "very early in my life i separated from my mother",
+ "normalized_hypothesis": "very early in my life i separated from my mother",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0288",
+ "category": "phonetic_balance",
+ "text": "There was pride and strength, the ring of triumph in his voice.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0288.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "There was pride and strength, the ring of triumph in his voice.",
+ "raw": {
+ "normalized_reference": "there was pride and strength, the ring of triumph in his voice.",
+ "normalized_hypothesis": "there was pride and strength, the ring of triumph in his voice.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "there was pride and strength the ring of triumph in his voice",
+ "normalized_hypothesis": "there was pride and strength the ring of triumph in his voice",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0289",
+ "category": "phonetic_balance",
+ "text": "From my earliest recollection my sleep was a period of terror.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0289.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "From my earliest recollection, my sleep was a period of terror.",
+ "raw": {
+ "normalized_reference": "from my earliest recollection my sleep was a period of terror.",
+ "normalized_hypothesis": "from my earliest recollection, my sleep was a period of terror.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "from my earliest recollection my sleep was a period of terror",
+ "normalized_hypothesis": "from my earliest recollection my sleep was a period of terror",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0290",
+ "category": "phonetic_balance",
+ "text": "We'll have to watch our chances.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0290.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "We'll have to watch our chances.",
+ "raw": {
+ "normalized_reference": "we'll have to watch our chances.",
+ "normalized_hypothesis": "we'll have to watch our chances.",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "we will have to watch our chances",
+ "normalized_hypothesis": "we will have to watch our chances",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0291",
+ "category": "phonetic_balance",
+ "text": "Close beside him gleamed the white fangs of the wolf-dog.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0291.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Close beside him gleamed the white fangs of the wolf-dog.",
+ "raw": {
+ "normalized_reference": "close beside him gleamed the white fangs of the wolf-dog.",
+ "normalized_hypothesis": "close beside him gleamed the white fangs of the wolf-dog.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "close beside him gleamed the white fangs of the wolf dog",
+ "normalized_hypothesis": "close beside him gleamed the white fangs of the wolf dog",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0292",
+ "category": "phonetic_balance",
+ "text": "Would you be satisfied with that one hundredth part of me.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0292.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Would you be satisfied with that one hundredth part of me?",
+ "raw": {
+ "normalized_reference": "would you be satisfied with that one hundredth part of me.",
+ "normalized_hypothesis": "would you be satisfied with that one hundredth part of me?",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "would you be satisfied with that 100th part of me",
+ "normalized_hypothesis": "would you be satisfied with that 100th part of me",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0293",
+ "category": "phonetic_balance",
+ "text": "Two gigantic owls were tearing at the carcass.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0293.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Two gigantic owls were tearing at the carcass.",
+ "raw": {
+ "normalized_reference": "two gigantic owls were tearing at the carcass.",
+ "normalized_hypothesis": "two gigantic owls were tearing at the carcass.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "2 gigantic owls were tearing at the carcass",
+ "normalized_hypothesis": "2 gigantic owls were tearing at the carcass",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0294",
+ "category": "phonetic_balance",
+ "text": "The lush vegetation of that sheltered spot make a natural shield.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0294.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The lush vegetation of that sheltered spot make a natural shield.",
+ "raw": {
+ "normalized_reference": "the lush vegetation of that sheltered spot make a natural shield.",
+ "normalized_hypothesis": "the lush vegetation of that sheltered spot make a natural shield.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the lush vegetation of that sheltered spot make a natural shield",
+ "normalized_hypothesis": "the lush vegetation of that sheltered spot make a natural shield",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0295",
+ "category": "phonetic_balance",
+ "text": "Below him the shadow was broken into a pool of rippling starlight.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0295.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Below him, the shadow was broken into a pool of rippling starlight.",
+ "raw": {
+ "normalized_reference": "below him the shadow was broken into a pool of rippling starlight.",
+ "normalized_hypothesis": "below him, the shadow was broken into a pool of rippling starlight.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "below him the shadow was broken into a pool of rippling starlight",
+ "normalized_hypothesis": "below him the shadow was broken into a pool of rippling starlight",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0296",
+ "category": "phonetic_balance",
+ "text": "Lots of men take women buggy riding.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0296.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Lots of men take women buggy riding.",
+ "raw": {
+ "normalized_reference": "lots of men take women buggy riding.",
+ "normalized_hypothesis": "lots of men take women buggy riding.",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "lots of men take women buggy riding",
+ "normalized_hypothesis": "lots of men take women buggy riding",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0297",
+ "category": "phonetic_balance",
+ "text": "But it contributed to the smash.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0297.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "But it contributed to the smash.",
+ "raw": {
+ "normalized_reference": "but it contributed to the smash.",
+ "normalized_hypothesis": "but it contributed to the smash.",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "but it contributed to the smash",
+ "normalized_hypothesis": "but it contributed to the smash",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0298",
+ "category": "phonetic_balance",
+ "text": "No man ate of the seal meat or the oil.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0298.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "No man ate of the seal meat or the oil.",
+ "raw": {
+ "normalized_reference": "no man ate of the seal meat or the oil.",
+ "normalized_hypothesis": "no man ate of the seal meat or the oil.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "no man ate of the seal meat or the oil",
+ "normalized_hypothesis": "no man ate of the seal meat or the oil",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0299",
+ "category": "phonetic_balance",
+ "text": "You're going in for grab sharing.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0299.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "You're going in for grab sharing.",
+ "raw": {
+ "normalized_reference": "you're going in for grab sharing.",
+ "normalized_hypothesis": "you're going in for grab sharing.",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "you are going in for grab sharing",
+ "normalized_hypothesis": "you are going in for grab sharing",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0300",
+ "category": "phonetic_balance",
+ "text": "He it was that lived to found the family of the Patino.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0300.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He it was that lived to found the family of the patino.",
+ "raw": {
+ "normalized_reference": "he it was that lived to found the family of the patino.",
+ "normalized_hypothesis": "he it was that lived to found the family of the patino.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he it was that lived to found the family of the patino",
+ "normalized_hypothesis": "he it was that lived to found the family of the patino",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0301",
+ "category": "phonetic_balance",
+ "text": "For such countries nothing remained but reorganization.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0301.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "For such countries, nothing remained but reorganization.",
+ "raw": {
+ "normalized_reference": "for such countries nothing remained but reorganization.",
+ "normalized_hypothesis": "for such countries, nothing remained but reorganization.",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "for such countries nothing remained but reorganization",
+ "normalized_hypothesis": "for such countries nothing remained but reorganization",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0302",
+ "category": "phonetic_balance",
+ "text": "Stand off butcher and baker and all the rest.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0302.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Standoff Butcher and Baker and all the rest.",
+ "raw": {
+ "normalized_reference": "stand off butcher and baker and all the rest.",
+ "normalized_hypothesis": "standoff butcher and baker and all the rest.",
+ "errors": 2,
+ "words": 9,
+ "wer": 0.2222222222222222
+ },
+ "semantic": {
+ "normalized_reference": "stand off butcher and baker and all the rest",
+ "normalized_hypothesis": "standoff butcher and baker and all the rest",
+ "errors": 2,
+ "words": 9,
+ "wer": 0.2222222222222222
+ }
+ },
+ {
+ "id": "release500_0303",
+ "category": "phonetic_balance",
+ "text": "All eyes, however, were staring at him in certitude of expectancy.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0303.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "All eyes, however, were staring at him in certitude of expectancy.",
+ "raw": {
+ "normalized_reference": "all eyes, however, were staring at him in certitude of expectancy.",
+ "normalized_hypothesis": "all eyes, however, were staring at him in certitude of expectancy.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "all eyes however were staring at him in certitude of expectancy",
+ "normalized_hypothesis": "all eyes however were staring at him in certitude of expectancy",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0304",
+ "category": "phonetic_balance",
+ "text": "Two days ago Jeanne learned where her father's men were hiding.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0304.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Two days ago, Jean learned where her father's men were hiding.",
+ "raw": {
+ "normalized_reference": "two days ago jeanne learned where her father's men were hiding.",
+ "normalized_hypothesis": "two days ago, jean learned where her father's men were hiding.",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ },
+ "semantic": {
+ "normalized_reference": "2 days ago jeanne learned where her father is men were hiding",
+ "normalized_hypothesis": "2 days ago jean learned where her father is men were hiding",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ }
+ },
+ {
+ "id": "release500_0305",
+ "category": "phonetic_balance",
+ "text": "His previous wives had never lived long enough to bear him children.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0305.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "His previous wives had never lived long enough to bear him children.",
+ "raw": {
+ "normalized_reference": "his previous wives had never lived long enough to bear him children.",
+ "normalized_hypothesis": "his previous wives had never lived long enough to bear him children.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "his previous wives had never lived long enough to bear him children",
+ "normalized_hypothesis": "his previous wives had never lived long enough to bear him children",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0306",
+ "category": "phonetic_balance",
+ "text": "But this little defect did not worry him.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0306.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "But this little defect did not worry him.",
+ "raw": {
+ "normalized_reference": "but this little defect did not worry him.",
+ "normalized_hypothesis": "but this little defect did not worry him.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "but this little defect did not worry him",
+ "normalized_hypothesis": "but this little defect did not worry him",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0307",
+ "category": "phonetic_balance",
+ "text": "Eli Harding asked, as Shunk started to follow.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0307.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Illy Harding asked as Shunk started to follow.",
+ "raw": {
+ "normalized_reference": "eli harding asked, as shunk started to follow.",
+ "normalized_hypothesis": "illy harding asked as shunk started to follow.",
+ "errors": 1,
+ "words": 8,
+ "wer": 0.125
+ },
+ "semantic": {
+ "normalized_reference": "eli harding asked as shunk started to follow",
+ "normalized_hypothesis": "illy harding asked as shunk started to follow",
+ "errors": 1,
+ "words": 8,
+ "wer": 0.125
+ }
+ },
+ {
+ "id": "release500_0308",
+ "category": "phonetic_balance",
+ "text": "And this was their sole conversation throughout the meal.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0308.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "And this was their sole conversation throughout the meal.",
+ "raw": {
+ "normalized_reference": "and this was their sole conversation throughout the meal.",
+ "normalized_hypothesis": "and this was their sole conversation throughout the meal.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "and this was their sole conversation throughout the meal",
+ "normalized_hypothesis": "and this was their sole conversation throughout the meal",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0309",
+ "category": "phonetic_balance",
+ "text": "Then, again, Tudor had such an irritating way about him.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0309.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Then again, Tudor had such an irritating way about him.",
+ "raw": {
+ "normalized_reference": "then, again, tudor had such an irritating way about him.",
+ "normalized_hypothesis": "then again, tudor had such an irritating way about him.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "then again tudor had such an irritating way about him",
+ "normalized_hypothesis": "then again tudor had such an irritating way about him",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0310",
+ "category": "phonetic_balance",
+ "text": "In it there was something that was almost tragedy.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0310.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "In it there was something that was almost tragedy.",
+ "raw": {
+ "normalized_reference": "in it there was something that was almost tragedy.",
+ "normalized_hypothesis": "in it there was something that was almost tragedy.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "in it there was something that was almost tragedy",
+ "normalized_hypothesis": "in it there was something that was almost tragedy",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0311",
+ "category": "phonetic_balance",
+ "text": "Thirty pounds, said the captain with finality.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0311.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Thirty pounds, said the captain with finality.",
+ "raw": {
+ "normalized_reference": "thirty pounds, said the captain with finality.",
+ "normalized_hypothesis": "thirty pounds, said the captain with finality.",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "\u00a330 said the captain with finality",
+ "normalized_hypothesis": "\u00a330 said the captain with finality",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0312",
+ "category": "phonetic_balance",
+ "text": "It was over when he made his way through the ring of spectators.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0312.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It was over when he made his way through the ring of spectators.",
+ "raw": {
+ "normalized_reference": "it was over when he made his way through the ring of spectators.",
+ "normalized_hypothesis": "it was over when he made his way through the ring of spectators.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it was over when he made his way through the ring of spectators",
+ "normalized_hypothesis": "it was over when he made his way through the ring of spectators",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0313",
+ "category": "phonetic_balance",
+ "text": "She had died from cold and starvation.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0313.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "She had died from cold and starvation.",
+ "raw": {
+ "normalized_reference": "she had died from cold and starvation.",
+ "normalized_hypothesis": "she had died from cold and starvation.",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "she had died from cold and starvation",
+ "normalized_hypothesis": "she had died from cold and starvation",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0314",
+ "category": "phonetic_balance",
+ "text": "We fished sharks on Niihau together.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0314.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "We fished sharks on Niihau together.",
+ "raw": {
+ "normalized_reference": "we fished sharks on niihau together.",
+ "normalized_hypothesis": "we fished sharks on niihau together.",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "we fished sharks on niihau together",
+ "normalized_hypothesis": "we fished sharks on niihau together",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0315",
+ "category": "phonetic_balance",
+ "text": "His slim hands gripped the edges of the table.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0315.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "His slim hands gripped the edges of the table.",
+ "raw": {
+ "normalized_reference": "his slim hands gripped the edges of the table.",
+ "normalized_hypothesis": "his slim hands gripped the edges of the table.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "his slim hands gripped the edges of the table",
+ "normalized_hypothesis": "his slim hands gripped the edges of the table",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0316",
+ "category": "phonetic_balance",
+ "text": "We had been chased by them ourselves, more than once.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0316.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "We had been chased by them ourselves more than once.",
+ "raw": {
+ "normalized_reference": "we had been chased by them ourselves, more than once.",
+ "normalized_hypothesis": "we had been chased by them ourselves more than once.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "we had been chased by them ourselves more than once",
+ "normalized_hypothesis": "we had been chased by them ourselves more than once",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0317",
+ "category": "phonetic_balance",
+ "text": "Or have they already devised one.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0317.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Or have they already devised one?",
+ "raw": {
+ "normalized_reference": "or have they already devised one.",
+ "normalized_hypothesis": "or have they already devised one?",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "or have they already devised one",
+ "normalized_hypothesis": "or have they already devised one",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0318",
+ "category": "phonetic_balance",
+ "text": "I cannot follow you, she said.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0318.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I cannot follow you, she said.",
+ "raw": {
+ "normalized_reference": "i cannot follow you, she said.",
+ "normalized_hypothesis": "i cannot follow you, she said.",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i cannot follow you she said",
+ "normalized_hypothesis": "i cannot follow you she said",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0319",
+ "category": "phonetic_balance",
+ "text": "He went down in midstream, searching the shadows of both shores.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0319.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He went down in midstream, searching the shadows of both shores.",
+ "raw": {
+ "normalized_reference": "he went down in midstream, searching the shadows of both shores.",
+ "normalized_hypothesis": "he went down in midstream, searching the shadows of both shores.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he went down in midstream searching the shadows of both shores",
+ "normalized_hypothesis": "he went down in midstream searching the shadows of both shores",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0320",
+ "category": "phonetic_balance",
+ "text": "I was sick once -- typhoid.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0320.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I was sick once typhoid.",
+ "raw": {
+ "normalized_reference": "i was sick once -- typhoid.",
+ "normalized_hypothesis": "i was sick once typhoid.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i was sick once typhoid",
+ "normalized_hypothesis": "i was sick once typhoid",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0321",
+ "category": "phonetic_balance",
+ "text": "Why not like any railroad station or ferry depot.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0321.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Why not like any railroad station or ferry depot?",
+ "raw": {
+ "normalized_reference": "why not like any railroad station or ferry depot.",
+ "normalized_hypothesis": "why not like any railroad station or ferry depot?",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "why not like any railroad station or ferry depot",
+ "normalized_hypothesis": "why not like any railroad station or ferry depot",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0322",
+ "category": "phonetic_balance",
+ "text": "Red-Eye swung back and forth on the branch farther down.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0322.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Red-Eye swung back and forth on the branch farther down.",
+ "raw": {
+ "normalized_reference": "red-eye swung back and forth on the branch farther down.",
+ "normalized_hypothesis": "red-eye swung back and forth on the branch farther down.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "red eye swung back and forth on the branch farther down",
+ "normalized_hypothesis": "red eye swung back and forth on the branch farther down",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0323",
+ "category": "phonetic_balance",
+ "text": "Within himself he called it no longer his own.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0323.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Within himself he called it no longer his own.",
+ "raw": {
+ "normalized_reference": "within himself he called it no longer his own.",
+ "normalized_hypothesis": "within himself he called it no longer his own.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "within himself he called it no longer his own",
+ "normalized_hypothesis": "within himself he called it no longer his own",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0324",
+ "category": "phonetic_balance",
+ "text": "It lasted as a deterrent for two days.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0324.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It lasted as a deterrent for two days.",
+ "raw": {
+ "normalized_reference": "it lasted as a deterrent for two days.",
+ "normalized_hypothesis": "it lasted as a deterrent for two days.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it lasted as a deterrent for 2 days",
+ "normalized_hypothesis": "it lasted as a deterrent for 2 days",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0325",
+ "category": "phonetic_balance",
+ "text": "He loved to play Chinese lottery.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0325.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He loved to play Chinese lottery.",
+ "raw": {
+ "normalized_reference": "he loved to play chinese lottery.",
+ "normalized_hypothesis": "he loved to play chinese lottery.",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he loved to play chinese lottery",
+ "normalized_hypothesis": "he loved to play chinese lottery",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0326",
+ "category": "phonetic_balance",
+ "text": "Lord, but I'm glad to see you again, Phil.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0326.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Lord, but I'm glad to see you again, Phil.",
+ "raw": {
+ "normalized_reference": "lord, but i'm glad to see you again, phil.",
+ "normalized_hypothesis": "lord, but i'm glad to see you again, phil.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "lord but i am glad to see you again phil",
+ "normalized_hypothesis": "lord but i am glad to see you again phil",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0327",
+ "category": "phonetic_balance",
+ "text": "There is not an iota of truth in it, certainly not.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0327.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "There is not an iota of truth in it, certainly not.",
+ "raw": {
+ "normalized_reference": "there is not an iota of truth in it, certainly not.",
+ "normalized_hypothesis": "there is not an iota of truth in it, certainly not.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "there is not an iota of truth in it certainly not",
+ "normalized_hypothesis": "there is not an iota of truth in it certainly not",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0328",
+ "category": "phonetic_balance",
+ "text": "She added, with genuine sympathy in her face and voice.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0328.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "She added, with genuine sympathy in her face and voice.",
+ "raw": {
+ "normalized_reference": "she added, with genuine sympathy in her face and voice.",
+ "normalized_hypothesis": "she added, with genuine sympathy in her face and voice.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "she added with genuine sympathy in her face and voice",
+ "normalized_hypothesis": "she added with genuine sympathy in her face and voice",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0329",
+ "category": "phonetic_balance",
+ "text": "We have plenty of capital ourselves, and yet we want more.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0329.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "We have plenty of capital ourselves, and yet we want more.",
+ "raw": {
+ "normalized_reference": "we have plenty of capital ourselves, and yet we want more.",
+ "normalized_hypothesis": "we have plenty of capital ourselves, and yet we want more.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "we have plenty of capital ourselves and yet we want more",
+ "normalized_hypothesis": "we have plenty of capital ourselves and yet we want more",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0330",
+ "category": "phonetic_balance",
+ "text": "And now put yourself in my place for a moment.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0330.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "And now put yourself in my place for a moment.",
+ "raw": {
+ "normalized_reference": "and now put yourself in my place for a moment.",
+ "normalized_hypothesis": "and now put yourself in my place for a moment.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "and now put yourself in my place for a moment",
+ "normalized_hypothesis": "and now put yourself in my place for a moment",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0331",
+ "category": "phonetic_balance",
+ "text": "Fast, but endure.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0331.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Fast, but endure.",
+ "raw": {
+ "normalized_reference": "fast, but endure.",
+ "normalized_hypothesis": "fast, but endure.",
+ "errors": 0,
+ "words": 3,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "fast but endure",
+ "normalized_hypothesis": "fast but endure",
+ "errors": 0,
+ "words": 3,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0332",
+ "category": "phonetic_balance",
+ "text": "You were looking squeamish this afternoon, he began.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0332.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "You were looking squeamish this afternoon, he began.",
+ "raw": {
+ "normalized_reference": "you were looking squeamish this afternoon, he began.",
+ "normalized_hypothesis": "you were looking squeamish this afternoon, he began.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "you were looking squeamish this afternoon he began",
+ "normalized_hypothesis": "you were looking squeamish this afternoon he began",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0333",
+ "category": "phonetic_balance",
+ "text": "It is the nearest refuge.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0333.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It is the nearest refuge.",
+ "raw": {
+ "normalized_reference": "it is the nearest refuge.",
+ "normalized_hypothesis": "it is the nearest refuge.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it is the nearest refuge",
+ "normalized_hypothesis": "it is the nearest refuge",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0334",
+ "category": "phonetic_balance",
+ "text": "I want to die in it.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0334.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I want to die in it.",
+ "raw": {
+ "normalized_reference": "i want to die in it.",
+ "normalized_hypothesis": "i want to die in it.",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i want to die in it",
+ "normalized_hypothesis": "i want to die in it",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0335",
+ "category": "phonetic_balance",
+ "text": "He stopped, and Philip nodded at the horrified question in his eyes.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0335.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He stopped, and Philip nodded at the horrified question in his eyes.",
+ "raw": {
+ "normalized_reference": "he stopped, and philip nodded at the horrified question in his eyes.",
+ "normalized_hypothesis": "he stopped, and philip nodded at the horrified question in his eyes.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he stopped and philip nodded at the horrified question in his eyes",
+ "normalized_hypothesis": "he stopped and philip nodded at the horrified question in his eyes",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0336",
+ "category": "phonetic_balance",
+ "text": "She turned, fearing that Jacques might see what was in her face.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0336.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "She turned, fearing that Jack might see what was in her face.",
+ "raw": {
+ "normalized_reference": "she turned, fearing that jacques might see what was in her face.",
+ "normalized_hypothesis": "she turned, fearing that jack might see what was in her face.",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "she turned fearing that jacques might see what was in her face",
+ "normalized_hypothesis": "she turned fearing that jack might see what was in her face",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ }
+ },
+ {
+ "id": "release500_0337",
+ "category": "phonetic_balance",
+ "text": "It was a curious coincidence.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0337.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It was a curious coincidence.",
+ "raw": {
+ "normalized_reference": "it was a curious coincidence.",
+ "normalized_hypothesis": "it was a curious coincidence.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it was a curious coincidence",
+ "normalized_hypothesis": "it was a curious coincidence",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0338",
+ "category": "phonetic_balance",
+ "text": "Daylight made no answer, and the door closed behind him.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0338.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Daylight made no answer, and the door closed behind him.",
+ "raw": {
+ "normalized_reference": "daylight made no answer, and the door closed behind him.",
+ "normalized_hypothesis": "daylight made no answer, and the door closed behind him.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "daylight made no answer and the door closed behind him",
+ "normalized_hypothesis": "daylight made no answer and the door closed behind him",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0339",
+ "category": "phonetic_balance",
+ "text": "And he did hurt my arm.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0339.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "And he did hurt my arm.",
+ "raw": {
+ "normalized_reference": "and he did hurt my arm.",
+ "normalized_hypothesis": "and he did hurt my arm.",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "and he did hurt my arm",
+ "normalized_hypothesis": "and he did hurt my arm",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0340",
+ "category": "phonetic_balance",
+ "text": "Dunham, can your boy go along with Jesse.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0340.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Dunham, can your boy go along with Jesse?",
+ "raw": {
+ "normalized_reference": "dunham, can your boy go along with jesse.",
+ "normalized_hypothesis": "dunham, can your boy go along with jesse?",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "dunham can your boy go along with jesse",
+ "normalized_hypothesis": "dunham can your boy go along with jesse",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0341",
+ "category": "phonetic_balance",
+ "text": "The creative joy, I murmured.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0341.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The creative joy, I murmured.",
+ "raw": {
+ "normalized_reference": "the creative joy, i murmured.",
+ "normalized_hypothesis": "the creative joy, i murmured.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the creative joy i murmured",
+ "normalized_hypothesis": "the creative joy i murmured",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0342",
+ "category": "phonetic_balance",
+ "text": "He pulled, and the log crashed down to break his back.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0342.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He pulled and the log crashed down to break his back.",
+ "raw": {
+ "normalized_reference": "he pulled, and the log crashed down to break his back.",
+ "normalized_hypothesis": "he pulled and the log crashed down to break his back.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he pulled and the log crashed down to break his back",
+ "normalized_hypothesis": "he pulled and the log crashed down to break his back",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0343",
+ "category": "phonetic_balance",
+ "text": "By golly, the boy wins.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0343.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "By golly, the boy wins.",
+ "raw": {
+ "normalized_reference": "by golly, the boy wins.",
+ "normalized_hypothesis": "by golly, the boy wins.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "by golly the boy wins",
+ "normalized_hypothesis": "by golly the boy wins",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0344",
+ "category": "phonetic_balance",
+ "text": "The boy threw back his head with pride.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0344.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The boy threw back his head with pride.",
+ "raw": {
+ "normalized_reference": "the boy threw back his head with pride.",
+ "normalized_hypothesis": "the boy threw back his head with pride.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the boy threw back his head with pride",
+ "normalized_hypothesis": "the boy threw back his head with pride",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0345",
+ "category": "phonetic_balance",
+ "text": "The last one I knew was an overseer.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0345.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The last one I knew was an overseer.",
+ "raw": {
+ "normalized_reference": "the last one i knew was an overseer.",
+ "normalized_hypothesis": "the last one i knew was an overseer.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the last one i knew was an overseer",
+ "normalized_hypothesis": "the last one i knew was an overseer",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0346",
+ "category": "phonetic_balance",
+ "text": "Two years ago I gave up civilization for this.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0346.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Two years ago, I gave up civilization for this.",
+ "raw": {
+ "normalized_reference": "two years ago i gave up civilization for this.",
+ "normalized_hypothesis": "two years ago, i gave up civilization for this.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "2 years ago i gave up civilization for this",
+ "normalized_hypothesis": "2 years ago i gave up civilization for this",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0347",
+ "category": "phonetic_balance",
+ "text": "The river bared its bosom, and snorting steamboats challenged the wilderness.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0347.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The river bared its bosom, and snorting steamboats challenged the wilderness.",
+ "raw": {
+ "normalized_reference": "the river bared its bosom, and snorting steamboats challenged the wilderness.",
+ "normalized_hypothesis": "the river bared its bosom, and snorting steamboats challenged the wilderness.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the river bared its bosom and snorting steamboats challenged the wilderness",
+ "normalized_hypothesis": "the river bared its bosom and snorting steamboats challenged the wilderness",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0348",
+ "category": "phonetic_balance",
+ "text": "Mab, she said.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0348.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Mab, she said.",
+ "raw": {
+ "normalized_reference": "mab, she said.",
+ "normalized_hypothesis": "mab, she said.",
+ "errors": 0,
+ "words": 3,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "mab she said",
+ "normalized_hypothesis": "mab she said",
+ "errors": 0,
+ "words": 3,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0349",
+ "category": "phonetic_balance",
+ "text": "The regret in his voice was provocative of a second burst of laughter.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0349.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The regret in his voice was provocative of a second burst of laughter.",
+ "raw": {
+ "normalized_reference": "the regret in his voice was provocative of a second burst of laughter.",
+ "normalized_hypothesis": "the regret in his voice was provocative of a second burst of laughter.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the regret in his voice was provocative of a 2nd burst of laughter",
+ "normalized_hypothesis": "the regret in his voice was provocative of a 2nd burst of laughter",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0350",
+ "category": "phonetic_balance",
+ "text": "And each year something happened, and I did not go.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0350.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "And each year something happened and I did not go.",
+ "raw": {
+ "normalized_reference": "and each year something happened, and i did not go.",
+ "normalized_hypothesis": "and each year something happened and i did not go.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "and each year something happened and i did not go",
+ "normalized_hypothesis": "and each year something happened and i did not go",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0351",
+ "category": "phonetic_balance",
+ "text": "Gad, your letter came just in time.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0351.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Gad, your letter came just in time.",
+ "raw": {
+ "normalized_reference": "gad, your letter came just in time.",
+ "normalized_hypothesis": "gad, your letter came just in time.",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "gad your letter came just in time",
+ "normalized_hypothesis": "gad your letter came just in time",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0352",
+ "category": "phonetic_balance",
+ "text": "But if Pierre did not return, until tomorrow.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0352.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "But if Pierre did not return until tomorrow...",
+ "raw": {
+ "normalized_reference": "but if pierre did not return, until tomorrow.",
+ "normalized_hypothesis": "but if pierre did not return until tomorrow...",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "but if pierre did not return until tomorrow",
+ "normalized_hypothesis": "but if pierre did not return until tomorrow .",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0353",
+ "category": "phonetic_balance",
+ "text": "Billinger may arrive in time.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0353.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Billinger may arrive in time.",
+ "raw": {
+ "normalized_reference": "billinger may arrive in time.",
+ "normalized_hypothesis": "billinger may arrive in time.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "billinger may arrive in time",
+ "normalized_hypothesis": "billinger may arrive in time",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0354",
+ "category": "phonetic_balance",
+ "text": "In such a tumbling of values was no time to sell.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0354.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "In such a tumbling of values was no time to sell.",
+ "raw": {
+ "normalized_reference": "in such a tumbling of values was no time to sell.",
+ "normalized_hypothesis": "in such a tumbling of values was no time to sell.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "in such a tumbling of values was no time to sell",
+ "normalized_hypothesis": "in such a tumbling of values was no time to sell",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0355",
+ "category": "phonetic_balance",
+ "text": "There's not an iota of truth in it.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0355.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "There's not an iota of truth in it.",
+ "raw": {
+ "normalized_reference": "there's not an iota of truth in it.",
+ "normalized_hypothesis": "there's not an iota of truth in it.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "there is not an iota of truth in it",
+ "normalized_hypothesis": "there is not an iota of truth in it",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0356",
+ "category": "phonetic_balance",
+ "text": "It is dog eat dog, and you ate them up.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0356.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It is dog eat dog and you ate them up.",
+ "raw": {
+ "normalized_reference": "it is dog eat dog, and you ate them up.",
+ "normalized_hypothesis": "it is dog eat dog and you ate them up.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it is dog eat dog and you ate them up",
+ "normalized_hypothesis": "it is dog eat dog and you ate them up",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0357",
+ "category": "phonetic_balance",
+ "text": "You can take a vacation on pay.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0357.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "You can take a vacation on pay.",
+ "raw": {
+ "normalized_reference": "you can take a vacation on pay.",
+ "normalized_hypothesis": "you can take a vacation on pay.",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "you can take a vacation on pay",
+ "normalized_hypothesis": "you can take a vacation on pay",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0358",
+ "category": "phonetic_balance",
+ "text": "I have seen myself that one man contemplated by Pascal's philosophic eye.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0358.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I have seen myself that one man contemplated by Pascal's philosophic eye.",
+ "raw": {
+ "normalized_reference": "i have seen myself that one man contemplated by pascal's philosophic eye.",
+ "normalized_hypothesis": "i have seen myself that one man contemplated by pascal's philosophic eye.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i have seen myself that one man contemplated by pascal is philosophic eye",
+ "normalized_hypothesis": "i have seen myself that one man contemplated by pascal is philosophic eye",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0359",
+ "category": "phonetic_balance",
+ "text": "Ah, I had forgotten, he exclaimed.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0359.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "All I had forgotten, he exclaimed.",
+ "raw": {
+ "normalized_reference": "ah, i had forgotten, he exclaimed.",
+ "normalized_hypothesis": "all i had forgotten, he exclaimed.",
+ "errors": 1,
+ "words": 6,
+ "wer": 0.16666666666666666
+ },
+ "semantic": {
+ "normalized_reference": "ah i had forgotten he exclaimed",
+ "normalized_hypothesis": "all i had forgotten he exclaimed",
+ "errors": 1,
+ "words": 6,
+ "wer": 0.16666666666666666
+ }
+ },
+ {
+ "id": "release500_0360",
+ "category": "phonetic_balance",
+ "text": "Providence had delivered him through the maelstrom.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0360.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Providence had delivered him through the maelstrom.",
+ "raw": {
+ "normalized_reference": "providence had delivered him through the maelstrom.",
+ "normalized_hypothesis": "providence had delivered him through the maelstrom.",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "providence had delivered him through the maelstrom",
+ "normalized_hypothesis": "providence had delivered him through the maelstrom",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0361",
+ "category": "phonetic_balance",
+ "text": "She was his now, forever.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0361.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "She was his now forever.",
+ "raw": {
+ "normalized_reference": "she was his now, forever.",
+ "normalized_hypothesis": "she was his now forever.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "she was his now forever",
+ "normalized_hypothesis": "she was his now forever",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0362",
+ "category": "phonetic_balance",
+ "text": "It was unobtrusive, yet it was there.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0362.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It was unobtrusive, yet it was there.",
+ "raw": {
+ "normalized_reference": "it was unobtrusive, yet it was there.",
+ "normalized_hypothesis": "it was unobtrusive, yet it was there.",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it was unobtrusive yet it was there",
+ "normalized_hypothesis": "it was unobtrusive yet it was there",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0363",
+ "category": "phonetic_balance",
+ "text": "Men like Joe Goose dated existence from drunk to drunk.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0363.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Men like Joe Goose dated existence from drunk to drunk.",
+ "raw": {
+ "normalized_reference": "men like joe goose dated existence from drunk to drunk.",
+ "normalized_hypothesis": "men like joe goose dated existence from drunk to drunk.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "men like joe goose dated existence from drunk to drunk",
+ "normalized_hypothesis": "men like joe goose dated existence from drunk to drunk",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0364",
+ "category": "phonetic_balance",
+ "text": "I was completely lost in my work.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0364.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I was completely lost in my work.",
+ "raw": {
+ "normalized_reference": "i was completely lost in my work.",
+ "normalized_hypothesis": "i was completely lost in my work.",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i was completely lost in my work",
+ "normalized_hypothesis": "i was completely lost in my work",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0365",
+ "category": "phonetic_balance",
+ "text": "The very thing, Ernest agreed.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0365.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The very thing Ernest agreed.",
+ "raw": {
+ "normalized_reference": "the very thing, ernest agreed.",
+ "normalized_hypothesis": "the very thing ernest agreed.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the very thing ernest agreed",
+ "normalized_hypothesis": "the very thing ernest agreed",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0366",
+ "category": "phonetic_balance",
+ "text": "He was the leader, and Tudor was his lieutenant.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0366.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He was the leader and Tudor was his lieutenant.",
+ "raw": {
+ "normalized_reference": "he was the leader, and tudor was his lieutenant.",
+ "normalized_hypothesis": "he was the leader and tudor was his lieutenant.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he was the leader and tudor was his lieutenant",
+ "normalized_hypothesis": "he was the leader and tudor was his lieutenant",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0367",
+ "category": "phonetic_balance",
+ "text": "The Fire-Men wore animal skins around their waists and across their shoulders.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0367.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The firemen wore animal skins around their waists and across their shoulders.",
+ "raw": {
+ "normalized_reference": "the fire-men wore animal skins around their waists and across their shoulders.",
+ "normalized_hypothesis": "the firemen wore animal skins around their waists and across their shoulders.",
+ "errors": 2,
+ "words": 13,
+ "wer": 0.15384615384615385
+ },
+ "semantic": {
+ "normalized_reference": "the fire men wore animal skins around their waists and across their shoulders",
+ "normalized_hypothesis": "the firemen wore animal skins around their waists and across their shoulders",
+ "errors": 2,
+ "words": 13,
+ "wer": 0.15384615384615385
+ }
+ },
+ {
+ "id": "release500_0368",
+ "category": "phonetic_balance",
+ "text": "We could throw stones with our feet.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0368.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "We could throw stones with our feet.",
+ "raw": {
+ "normalized_reference": "we could throw stones with our feet.",
+ "normalized_hypothesis": "we could throw stones with our feet.",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "we could throw stones with our feet",
+ "normalized_hypothesis": "we could throw stones with our feet",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0369",
+ "category": "phonetic_balance",
+ "text": "He turned sharply, and faced Gregson across the table.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0369.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He turned sharply and faced Gregson across the table.",
+ "raw": {
+ "normalized_reference": "he turned sharply, and faced gregson across the table.",
+ "normalized_hypothesis": "he turned sharply and faced gregson across the table.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he turned sharply and faced gregson across the table",
+ "normalized_hypothesis": "he turned sharply and faced gregson across the table",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0370",
+ "category": "phonetic_balance",
+ "text": "They are to attack your camp tomorrow night.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0370.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "They are to attack your camp tomorrow night.",
+ "raw": {
+ "normalized_reference": "they are to attack your camp tomorrow night.",
+ "normalized_hypothesis": "they are to attack your camp tomorrow night.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "they are to attack your camp tomorrow night",
+ "normalized_hypothesis": "they are to attack your camp tomorrow night",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0371",
+ "category": "phonetic_balance",
+ "text": "And watch out for wet feet, was his parting advice.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0371.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "And watch out for wet feet was his parting advice.",
+ "raw": {
+ "normalized_reference": "and watch out for wet feet, was his parting advice.",
+ "normalized_hypothesis": "and watch out for wet feet was his parting advice.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "and watch out for wet feet was his parting advice",
+ "normalized_hypothesis": "and watch out for wet feet was his parting advice",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0372",
+ "category": "phonetic_balance",
+ "text": "Two weeks passed, and in that time Thorpe left camp three times.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0372.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Two weeks passed, and in that time Thorpe left camp three times.",
+ "raw": {
+ "normalized_reference": "two weeks passed, and in that time thorpe left camp three times.",
+ "normalized_hypothesis": "two weeks passed, and in that time thorpe left camp three times.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "2 weeks passed and in that time thorpe left camp 3 times",
+ "normalized_hypothesis": "2 weeks passed and in that time thorpe left camp 3 times",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0373",
+ "category": "phonetic_balance",
+ "text": "But ever his gaze returned to that Crouched Venus on the piano.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0373.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "But ever his gaze returned to that crouched Venus on the piano.",
+ "raw": {
+ "normalized_reference": "but ever his gaze returned to that crouched venus on the piano.",
+ "normalized_hypothesis": "but ever his gaze returned to that crouched venus on the piano.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "but ever his gaze returned to that crouched venus on the piano",
+ "normalized_hypothesis": "but ever his gaze returned to that crouched venus on the piano",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0374",
+ "category": "phonetic_balance",
+ "text": "But she had become an automaton.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0374.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "But she had become an automaton.",
+ "raw": {
+ "normalized_reference": "but she had become an automaton.",
+ "normalized_hypothesis": "but she had become an automaton.",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "but she had become an automaton",
+ "normalized_hypothesis": "but she had become an automaton",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0375",
+ "category": "phonetic_balance",
+ "text": "I do not blame you for anything; remember that.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0375.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I do not blame you for anything. Remember that.",
+ "raw": {
+ "normalized_reference": "i do not blame you for anything; remember that.",
+ "normalized_hypothesis": "i do not blame you for anything. remember that.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i do not blame you for anything remember that",
+ "normalized_hypothesis": "i do not blame you for anything remember that",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0376",
+ "category": "phonetic_balance",
+ "text": "The next thing to watch out for is bed sores.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0376.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The next thing to watch out for is bed sores.",
+ "raw": {
+ "normalized_reference": "the next thing to watch out for is bed sores.",
+ "normalized_hypothesis": "the next thing to watch out for is bed sores.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the next thing to watch out for is bed sores",
+ "normalized_hypothesis": "the next thing to watch out for is bed sores",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0377",
+ "category": "phonetic_balance",
+ "text": "He will follow us soon.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0377.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He will follow us soon.",
+ "raw": {
+ "normalized_reference": "he will follow us soon.",
+ "normalized_hypothesis": "he will follow us soon.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he will follow us soon",
+ "normalized_hypothesis": "he will follow us soon",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0378",
+ "category": "phonetic_balance",
+ "text": "His beady black eyes saw bargains where other men saw bankruptcy.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0378.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "His beady black eyes saw bargains where other men saw bankruptcy.",
+ "raw": {
+ "normalized_reference": "his beady black eyes saw bargains where other men saw bankruptcy.",
+ "normalized_hypothesis": "his beady black eyes saw bargains where other men saw bankruptcy.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "his beady black eyes saw bargains where other men saw bankruptcy",
+ "normalized_hypothesis": "his beady black eyes saw bargains where other men saw bankruptcy",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0379",
+ "category": "phonetic_balance",
+ "text": "They robbed me a few years later.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0379.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "They robbed me a few years later.",
+ "raw": {
+ "normalized_reference": "they robbed me a few years later.",
+ "normalized_hypothesis": "they robbed me a few years later.",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "they robbed me a few years later",
+ "normalized_hypothesis": "they robbed me a few years later",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0380",
+ "category": "phonetic_balance",
+ "text": "Gregson held a lighted match until it burnt his fingertips.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0380.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Gregson held a lighted match until it burnt his fingertips.",
+ "raw": {
+ "normalized_reference": "gregson held a lighted match until it burnt his fingertips.",
+ "normalized_hypothesis": "gregson held a lighted match until it burnt his fingertips.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "gregson held a lighted match until it burnt his fingertips",
+ "normalized_hypothesis": "gregson held a lighted match until it burnt his fingertips",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0381",
+ "category": "phonetic_balance",
+ "text": "A burst of laughter was his reward.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0381.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "A burst of laughter was his reward.",
+ "raw": {
+ "normalized_reference": "a burst of laughter was his reward.",
+ "normalized_hypothesis": "a burst of laughter was his reward.",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "a burst of laughter was his reward",
+ "normalized_hypothesis": "a burst of laughter was his reward",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0382",
+ "category": "phonetic_balance",
+ "text": "He cried, and swung the club wildly.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0382.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He cried and swung the club wildly.",
+ "raw": {
+ "normalized_reference": "he cried, and swung the club wildly.",
+ "normalized_hypothesis": "he cried and swung the club wildly.",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he cried and swung the club wildly",
+ "normalized_hypothesis": "he cried and swung the club wildly",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0383",
+ "category": "phonetic_balance",
+ "text": "Sandel would never become a world champion.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0383.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Sandal would never become a world champion.",
+ "raw": {
+ "normalized_reference": "sandel would never become a world champion.",
+ "normalized_hypothesis": "sandal would never become a world champion.",
+ "errors": 1,
+ "words": 7,
+ "wer": 0.14285714285714285
+ },
+ "semantic": {
+ "normalized_reference": "sandel would never become a world champion",
+ "normalized_hypothesis": "sandal would never become a world champion",
+ "errors": 1,
+ "words": 7,
+ "wer": 0.14285714285714285
+ }
+ },
+ {
+ "id": "release500_0384",
+ "category": "phonetic_balance",
+ "text": "And after the bath a shave would not be bad.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0384.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "And after the bath, a shave would not be bad.",
+ "raw": {
+ "normalized_reference": "and after the bath a shave would not be bad.",
+ "normalized_hypothesis": "and after the bath, a shave would not be bad.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "and after the bath a shave would not be bad",
+ "normalized_hypothesis": "and after the bath a shave would not be bad",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0385",
+ "category": "phonetic_balance",
+ "text": "They saw each other for the first time in Boston.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0385.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "They saw each other for the first time in Boston.",
+ "raw": {
+ "normalized_reference": "they saw each other for the first time in boston.",
+ "normalized_hypothesis": "they saw each other for the first time in boston.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "they saw each other for the 1st time in boston",
+ "normalized_hypothesis": "they saw each other for the 1st time in boston",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0386",
+ "category": "phonetic_balance",
+ "text": "I had been born with no organic, chemical predisposition toward alcohol.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0386.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I had been born with no organic, chemical predisposition toward alcohol.",
+ "raw": {
+ "normalized_reference": "i had been born with no organic, chemical predisposition toward alcohol.",
+ "normalized_hypothesis": "i had been born with no organic, chemical predisposition toward alcohol.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i had been born with no organic chemical predisposition toward alcohol",
+ "normalized_hypothesis": "i had been born with no organic chemical predisposition toward alcohol",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0387",
+ "category": "phonetic_balance",
+ "text": "They had no fixed values, to be altered by adjectives and adverbs.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0387.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "They had no fixed values to be altered by adjectives and adverbs.",
+ "raw": {
+ "normalized_reference": "they had no fixed values, to be altered by adjectives and adverbs.",
+ "normalized_hypothesis": "they had no fixed values to be altered by adjectives and adverbs.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "they had no fixed values to be altered by adjectives and adverbs",
+ "normalized_hypothesis": "they had no fixed values to be altered by adjectives and adverbs",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0388",
+ "category": "phonetic_balance",
+ "text": "Nor was Elam Harnish an exception.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0388.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Nor was a lamb-harnish an exception.",
+ "raw": {
+ "normalized_reference": "nor was elam harnish an exception.",
+ "normalized_hypothesis": "nor was a lamb-harnish an exception.",
+ "errors": 2,
+ "words": 6,
+ "wer": 0.3333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "nor was elam harnish an exception",
+ "normalized_hypothesis": "nor was a lamb harnish an exception",
+ "errors": 2,
+ "words": 6,
+ "wer": 0.3333333333333333
+ }
+ },
+ {
+ "id": "release500_0389",
+ "category": "phonetic_balance",
+ "text": "He was a wise hyena.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0389.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He was a wise hyena.",
+ "raw": {
+ "normalized_reference": "he was a wise hyena.",
+ "normalized_hypothesis": "he was a wise hyena.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he was a wise hyena",
+ "normalized_hypothesis": "he was a wise hyena",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0390",
+ "category": "phonetic_balance",
+ "text": "And now behold the perversity of things.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0390.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "And now behold the perversity of things.",
+ "raw": {
+ "normalized_reference": "and now behold the perversity of things.",
+ "normalized_hypothesis": "and now behold the perversity of things.",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "and now behold the perversity of things",
+ "normalized_hypothesis": "and now behold the perversity of things",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0391",
+ "category": "phonetic_balance",
+ "text": "The Fire People, like ourselves, lived in caves.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0391.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The fire people, like ourselves, lived in caves.",
+ "raw": {
+ "normalized_reference": "the fire people, like ourselves, lived in caves.",
+ "normalized_hypothesis": "the fire people, like ourselves, lived in caves.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the fire people like ourselves lived in caves",
+ "normalized_hypothesis": "the fire people like ourselves lived in caves",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0392",
+ "category": "phonetic_balance",
+ "text": "To her the bridge was tambo, which is the native for taboo.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0392.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "To her the bridge was Tambo, which is the native for Taboo.",
+ "raw": {
+ "normalized_reference": "to her the bridge was tambo, which is the native for taboo.",
+ "normalized_hypothesis": "to her the bridge was tambo, which is the native for taboo.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "to her the bridge was tambo which is the native for taboo",
+ "normalized_hypothesis": "to her the bridge was tambo which is the native for taboo",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0393",
+ "category": "phonetic_balance",
+ "text": "The voyage was our idea of a good time.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0393.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The voyage was our idea of a good time.",
+ "raw": {
+ "normalized_reference": "the voyage was our idea of a good time.",
+ "normalized_hypothesis": "the voyage was our idea of a good time.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the voyage was our idea of a good time",
+ "normalized_hypothesis": "the voyage was our idea of a good time",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0394",
+ "category": "phonetic_balance",
+ "text": "Of course much grumbling went on, and little outbursts were continually occurring.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0394.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Of course much grumbling went on, and little outbursts were continually occurring.",
+ "raw": {
+ "normalized_reference": "of course much grumbling went on, and little outbursts were continually occurring.",
+ "normalized_hypothesis": "of course much grumbling went on, and little outbursts were continually occurring.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "of course much grumbling went on and little outbursts were continually occurring",
+ "normalized_hypothesis": "of course much grumbling went on and little outbursts were continually occurring",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0395",
+ "category": "phonetic_balance",
+ "text": "Get down and dig in.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0395.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Get down and dig in.",
+ "raw": {
+ "normalized_reference": "get down and dig in.",
+ "normalized_hypothesis": "get down and dig in.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "get down and dig in",
+ "normalized_hypothesis": "get down and dig in",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0396",
+ "category": "phonetic_balance",
+ "text": "It was put together in a casual, helter-skelter sort of way.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0396.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It was put together in a casual, helter-skelter sort of way.",
+ "raw": {
+ "normalized_reference": "it was put together in a casual, helter-skelter sort of way.",
+ "normalized_hypothesis": "it was put together in a casual, helter-skelter sort of way.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it was put together in a casual helter skelter sort of way",
+ "normalized_hypothesis": "it was put together in a casual helter skelter sort of way",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0397",
+ "category": "phonetic_balance",
+ "text": "He deluged me, overwhelmed me with argument.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0397.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He deluged me, overwhelmed me with argument.",
+ "raw": {
+ "normalized_reference": "he deluged me, overwhelmed me with argument.",
+ "normalized_hypothesis": "he deluged me, overwhelmed me with argument.",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he deluged me overwhelmed me with argument",
+ "normalized_hypothesis": "he deluged me overwhelmed me with argument",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0398",
+ "category": "phonetic_balance",
+ "text": "The girl faced him, her eyes shining with sudden fear.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0398.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The girl faced him, her eyes shining with sudden fear.",
+ "raw": {
+ "normalized_reference": "the girl faced him, her eyes shining with sudden fear.",
+ "normalized_hypothesis": "the girl faced him, her eyes shining with sudden fear.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the girl faced him her eyes shining with sudden fear",
+ "normalized_hypothesis": "the girl faced him her eyes shining with sudden fear",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0399",
+ "category": "phonetic_balance",
+ "text": "It will break our hearts and our backs to hoist anchor by hand.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0399.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It will break our hearts and our backs to hoist anchor by hand.",
+ "raw": {
+ "normalized_reference": "it will break our hearts and our backs to hoist anchor by hand.",
+ "normalized_hypothesis": "it will break our hearts and our backs to hoist anchor by hand.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it will break our hearts and our backs to hoist anchor by hand",
+ "normalized_hypothesis": "it will break our hearts and our backs to hoist anchor by hand",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0400",
+ "category": "phonetic_balance",
+ "text": "He made no reply as he waited for Whittemore to continue.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0400.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He made no reply as he waited for Whittemore to continue.",
+ "raw": {
+ "normalized_reference": "he made no reply as he waited for whittemore to continue.",
+ "normalized_hypothesis": "he made no reply as he waited for whittemore to continue.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he made no reply as he waited for whittemore to continue",
+ "normalized_hypothesis": "he made no reply as he waited for whittemore to continue",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0401",
+ "category": "phonetic_balance",
+ "text": "Who the devil gave it to you to be judge and jury.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0401.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Who the devil gave it to you to be judge and jury?",
+ "raw": {
+ "normalized_reference": "who the devil gave it to you to be judge and jury.",
+ "normalized_hypothesis": "who the devil gave it to you to be judge and jury?",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "who the devil gave it to you to be judge and jury",
+ "normalized_hypothesis": "who the devil gave it to you to be judge and jury",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0402",
+ "category": "phonetic_balance",
+ "text": "Daylight found himself charmed and made curious by the little man.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0402.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Daylight found himself charmed and made curious by the little man.",
+ "raw": {
+ "normalized_reference": "daylight found himself charmed and made curious by the little man.",
+ "normalized_hypothesis": "daylight found himself charmed and made curious by the little man.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "daylight found himself charmed and made curious by the little man",
+ "normalized_hypothesis": "daylight found himself charmed and made curious by the little man",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0403",
+ "category": "phonetic_balance",
+ "text": "It's a Yankee, Joan cried.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0403.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It's a Yankee, Joan cried.",
+ "raw": {
+ "normalized_reference": "it's a yankee, joan cried.",
+ "normalized_hypothesis": "it's a yankee, joan cried.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it is a yankee joan cried",
+ "normalized_hypothesis": "it is a yankee joan cried",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0404",
+ "category": "phonetic_balance",
+ "text": "But what they want with your toothbrush is more than I can imagine.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0404.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "But what they want with your toothbrush is more than I can imagine.",
+ "raw": {
+ "normalized_reference": "but what they want with your toothbrush is more than i can imagine.",
+ "normalized_hypothesis": "but what they want with your toothbrush is more than i can imagine.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "but what they want with your toothbrush is more than i can imagine",
+ "normalized_hypothesis": "but what they want with your toothbrush is more than i can imagine",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0405",
+ "category": "phonetic_balance",
+ "text": "They were babbling and chattering all together.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0405.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "They were babbling and chattering all together.",
+ "raw": {
+ "normalized_reference": "they were babbling and chattering all together.",
+ "normalized_hypothesis": "they were babbling and chattering all together.",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "they were babbling and chattering all together",
+ "normalized_hypothesis": "they were babbling and chattering all together",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0406",
+ "category": "phonetic_balance",
+ "text": "Men of Selden's stamp don't stop at women and children.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0406.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Men of Selden's stamp don't stop at women and children.",
+ "raw": {
+ "normalized_reference": "men of selden's stamp don't stop at women and children.",
+ "normalized_hypothesis": "men of selden's stamp don't stop at women and children.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "men of selden is stamp do not stop at women and children",
+ "normalized_hypothesis": "men of selden is stamp do not stop at women and children",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0407",
+ "category": "phonetic_balance",
+ "text": "Cameron looked at his hands with their long, sinewy fingers.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0407.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Cameron looked at his hands with their long, sinewy fingers.",
+ "raw": {
+ "normalized_reference": "cameron looked at his hands with their long, sinewy fingers.",
+ "normalized_hypothesis": "cameron looked at his hands with their long, sinewy fingers.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "cameron looked at his hands with their long sinewy fingers",
+ "normalized_hypothesis": "cameron looked at his hands with their long sinewy fingers",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0408",
+ "category": "phonetic_balance",
+ "text": "Saxon waited, for she knew a fresh idea had struck Billy.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0408.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Saxon waited, for she knew a fresh idea had struck Billy.",
+ "raw": {
+ "normalized_reference": "saxon waited, for she knew a fresh idea had struck billy.",
+ "normalized_hypothesis": "saxon waited, for she knew a fresh idea had struck billy.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "saxon waited for she knew a fresh idea had struck billy",
+ "normalized_hypothesis": "saxon waited for she knew a fresh idea had struck billy",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0409",
+ "category": "phonetic_balance",
+ "text": "He saw only the effect in a general, sketchy way.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0409.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He saw only the effect in a general, sketchy way.",
+ "raw": {
+ "normalized_reference": "he saw only the effect in a general, sketchy way.",
+ "normalized_hypothesis": "he saw only the effect in a general, sketchy way.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he saw only the effect in a general sketchy way",
+ "normalized_hypothesis": "he saw only the effect in a general sketchy way",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0410",
+ "category": "phonetic_balance",
+ "text": "He began to follow the footprints of the dog.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0410.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He began to follow the footprints of the dog.",
+ "raw": {
+ "normalized_reference": "he began to follow the footprints of the dog.",
+ "normalized_hypothesis": "he began to follow the footprints of the dog.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he began to follow the footprints of the dog",
+ "normalized_hypothesis": "he began to follow the footprints of the dog",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0411",
+ "category": "phonetic_balance",
+ "text": "Do you value your hide.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0411.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Do you value your hide?",
+ "raw": {
+ "normalized_reference": "do you value your hide.",
+ "normalized_hypothesis": "do you value your hide?",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "do you value your hide",
+ "normalized_hypothesis": "do you value your hide",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0412",
+ "category": "phonetic_balance",
+ "text": "A rising tide of fat had submerged them.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0412.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "A rising tide of fat had submerged them.",
+ "raw": {
+ "normalized_reference": "a rising tide of fat had submerged them.",
+ "normalized_hypothesis": "a rising tide of fat had submerged them.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "a rising tide of fat had submerged them",
+ "normalized_hypothesis": "a rising tide of fat had submerged them",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0413",
+ "category": "phonetic_balance",
+ "text": "There was nothing on the rock.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0413.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "There was nothing on the rock.",
+ "raw": {
+ "normalized_reference": "there was nothing on the rock.",
+ "normalized_hypothesis": "there was nothing on the rock.",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "there was nothing on the rock",
+ "normalized_hypothesis": "there was nothing on the rock",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0414",
+ "category": "phonetic_balance",
+ "text": "Whatever he guessed he locked away in the taboo room of Naomi.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0414.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Whatever he guessed, he locked away in the taboo room of Naomi.",
+ "raw": {
+ "normalized_reference": "whatever he guessed he locked away in the taboo room of naomi.",
+ "normalized_hypothesis": "whatever he guessed, he locked away in the taboo room of naomi.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "whatever he guessed he locked away in the taboo room of naomi",
+ "normalized_hypothesis": "whatever he guessed he locked away in the taboo room of naomi",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0415",
+ "category": "phonetic_balance",
+ "text": "Will we ever forget it.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0415.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Will we ever forget it?",
+ "raw": {
+ "normalized_reference": "will we ever forget it.",
+ "normalized_hypothesis": "will we ever forget it?",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "will we ever forget it",
+ "normalized_hypothesis": "will we ever forget it",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0416",
+ "category": "phonetic_balance",
+ "text": "Jeanne and Pierre both gazed toward the great rock.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0416.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Jean and Pierre both gazed toward the great rock.",
+ "raw": {
+ "normalized_reference": "jeanne and pierre both gazed toward the great rock.",
+ "normalized_hypothesis": "jean and pierre both gazed toward the great rock.",
+ "errors": 1,
+ "words": 9,
+ "wer": 0.1111111111111111
+ },
+ "semantic": {
+ "normalized_reference": "jeanne and pierre both gazed toward the great rock",
+ "normalized_hypothesis": "jean and pierre both gazed toward the great rock",
+ "errors": 1,
+ "words": 9,
+ "wer": 0.1111111111111111
+ }
+ },
+ {
+ "id": "release500_0417",
+ "category": "phonetic_balance",
+ "text": "Does the old boy often go off at half-cock that way.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0417.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Does the old boy often go off at half-cock that way?",
+ "raw": {
+ "normalized_reference": "does the old boy often go off at half-cock that way.",
+ "normalized_hypothesis": "does the old boy often go off at half-cock that way?",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "does the old boy often go off at half cock that way",
+ "normalized_hypothesis": "does the old boy often go off at half cock that way",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0418",
+ "category": "phonetic_balance",
+ "text": "The temperature dropped to fifty below zero and remained there the whole trip.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0418.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The temperature dropped to 50 below zero and remained there the whole trip.",
+ "raw": {
+ "normalized_reference": "the temperature dropped to fifty below zero and remained there the whole trip.",
+ "normalized_hypothesis": "the temperature dropped to 50 below zero and remained there the whole trip.",
+ "errors": 1,
+ "words": 13,
+ "wer": 0.07692307692307693
+ },
+ "semantic": {
+ "normalized_reference": "the temperature dropped to 50 below 0 and remained there the whole trip",
+ "normalized_hypothesis": "the temperature dropped to 50 below 0 and remained there the whole trip",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0419",
+ "category": "phonetic_balance",
+ "text": "It's worth eight dollars.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0419.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It's worth eight dollars.",
+ "raw": {
+ "normalized_reference": "it's worth eight dollars.",
+ "normalized_hypothesis": "it's worth eight dollars.",
+ "errors": 0,
+ "words": 4,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it is worth $8",
+ "normalized_hypothesis": "it is worth $8",
+ "errors": 0,
+ "words": 4,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0420",
+ "category": "phonetic_balance",
+ "text": "But I did not enjoy it long.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0420.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "But I did not enjoy it long.",
+ "raw": {
+ "normalized_reference": "but i did not enjoy it long.",
+ "normalized_hypothesis": "but i did not enjoy it long.",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "but i did not enjoy it long",
+ "normalized_hypothesis": "but i did not enjoy it long",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0421",
+ "category": "phonetic_balance",
+ "text": "Shorty turned to their employers.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0421.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Shorty turned to their employers.",
+ "raw": {
+ "normalized_reference": "shorty turned to their employers.",
+ "normalized_hypothesis": "shorty turned to their employers.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "shorty turned to their employers",
+ "normalized_hypothesis": "shorty turned to their employers",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0422",
+ "category": "phonetic_balance",
+ "text": "It was the third or fourth time that Philip had heard MacDougall swear.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0422.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It was the third or fourth time that Philip had heard MacDougall swear.",
+ "raw": {
+ "normalized_reference": "it was the third or fourth time that philip had heard macdougall swear.",
+ "normalized_hypothesis": "it was the third or fourth time that philip had heard macdougall swear.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it was the 3rd or 4th time that philip had heard macdougall swear",
+ "normalized_hypothesis": "it was the 3rd or 4th time that philip had heard macdougall swear",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0423",
+ "category": "phonetic_balance",
+ "text": "Ernest saw in the affair the most sinister import.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0423.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Ernest saw in the affair the most sinister import.",
+ "raw": {
+ "normalized_reference": "ernest saw in the affair the most sinister import.",
+ "normalized_hypothesis": "ernest saw in the affair the most sinister import.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "ernest saw in the affair the most sinister import",
+ "normalized_hypothesis": "ernest saw in the affair the most sinister import",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0424",
+ "category": "phonetic_balance",
+ "text": "Also, at regular intervals, he would mutter.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0424.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Also at regular intervals he would mutter,",
+ "raw": {
+ "normalized_reference": "also, at regular intervals, he would mutter.",
+ "normalized_hypothesis": "also at regular intervals he would mutter,",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "also at regular intervals he would mutter",
+ "normalized_hypothesis": "also at regular intervals he would mutter",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0425",
+ "category": "phonetic_balance",
+ "text": "He may anticipate the day of his death.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0425.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He may anticipate the day of his death.",
+ "raw": {
+ "normalized_reference": "he may anticipate the day of his death.",
+ "normalized_hypothesis": "he may anticipate the day of his death.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he may anticipate the day of his death",
+ "normalized_hypothesis": "he may anticipate the day of his death",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0426",
+ "category": "phonetic_balance",
+ "text": "I defy any man to get a Solomon Island sore in California.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0426.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I defy any man to get a Solomon Island sore in California.",
+ "raw": {
+ "normalized_reference": "i defy any man to get a solomon island sore in california.",
+ "normalized_hypothesis": "i defy any man to get a solomon island sore in california.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i defy any man to get a solomon island sore in california",
+ "normalized_hypothesis": "i defy any man to get a solomon island sore in california",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0427",
+ "category": "phonetic_balance",
+ "text": "Now our figuring was all right.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0427.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Now our figuring was all right.",
+ "raw": {
+ "normalized_reference": "now our figuring was all right.",
+ "normalized_hypothesis": "now our figuring was all right.",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "now our figuring was all right",
+ "normalized_hypothesis": "now our figuring was all right",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0428",
+ "category": "phonetic_balance",
+ "text": "MacDougall tapped his forehead suspiciously with a stubby forefinger.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0428.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "MacDougall tapped his forehead suspiciously with a stubby forefinger.",
+ "raw": {
+ "normalized_reference": "macdougall tapped his forehead suspiciously with a stubby forefinger.",
+ "normalized_hypothesis": "macdougall tapped his forehead suspiciously with a stubby forefinger.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "macdougall tapped his forehead suspiciously with a stubby forefinger",
+ "normalized_hypothesis": "macdougall tapped his forehead suspiciously with a stubby forefinger",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0429",
+ "category": "phonetic_balance",
+ "text": "O'Brien took off his coat and bared his right arm.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0429.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "O'Brien took off his coat and bared his right arm.",
+ "raw": {
+ "normalized_reference": "o'brien took off his coat and bared his right arm.",
+ "normalized_hypothesis": "o'brien took off his coat and bared his right arm.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "0 brien took off his coat and bared his right arm",
+ "normalized_hypothesis": "0 brien took off his coat and bared his right arm",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0430",
+ "category": "phonetic_balance",
+ "text": "But it won't continue, she said with easy confidence.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0430.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "But it won't continue, she said with easy confidence.",
+ "raw": {
+ "normalized_reference": "but it won't continue, she said with easy confidence.",
+ "normalized_hypothesis": "but it won't continue, she said with easy confidence.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "but it will not continue she said with easy confidence",
+ "normalized_hypothesis": "but it will not continue she said with easy confidence",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0431",
+ "category": "phonetic_balance",
+ "text": "I was still weak from my prolonged immersion.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0431.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I was still weak from my prolonged immersion.",
+ "raw": {
+ "normalized_reference": "i was still weak from my prolonged immersion.",
+ "normalized_hypothesis": "i was still weak from my prolonged immersion.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i was still weak from my prolonged immersion",
+ "normalized_hypothesis": "i was still weak from my prolonged immersion",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0432",
+ "category": "phonetic_balance",
+ "text": "Oolong was two hundred and fifty miles from the nearest land.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0432.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Oolong was 250 miles from the nearest land.",
+ "raw": {
+ "normalized_reference": "oolong was two hundred and fifty miles from the nearest land.",
+ "normalized_hypothesis": "oolong was 250 miles from the nearest land.",
+ "errors": 4,
+ "words": 11,
+ "wer": 0.36363636363636365
+ },
+ "semantic": {
+ "normalized_reference": "oolong was 250 miles from the nearest land",
+ "normalized_hypothesis": "oolong was 250 miles from the nearest land",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0433",
+ "category": "phonetic_balance",
+ "text": "I just do appreciate it without being able to express my feelings.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0433.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I just do appreciate it without being able to express my feelings.",
+ "raw": {
+ "normalized_reference": "i just do appreciate it without being able to express my feelings.",
+ "normalized_hypothesis": "i just do appreciate it without being able to express my feelings.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i just do appreciate it without being able to express my feelings",
+ "normalized_hypothesis": "i just do appreciate it without being able to express my feelings",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0434",
+ "category": "phonetic_balance",
+ "text": "Now just what do you want to know.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0434.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Now just what do you want to know?",
+ "raw": {
+ "normalized_reference": "now just what do you want to know.",
+ "normalized_hypothesis": "now just what do you want to know?",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "now just what do you want to know",
+ "normalized_hypothesis": "now just what do you want to know",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0435",
+ "category": "phonetic_balance",
+ "text": "But she swung obediently on her heel into the wind.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0435.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "But she swung obediently on her heel into the wind.",
+ "raw": {
+ "normalized_reference": "but she swung obediently on her heel into the wind.",
+ "normalized_hypothesis": "but she swung obediently on her heel into the wind.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "but she swung obediently on her heel into the wind",
+ "normalized_hypothesis": "but she swung obediently on her heel into the wind",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0436",
+ "category": "phonetic_balance",
+ "text": "Ask him, I laughed, then turned to Pasquini.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0436.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Ask him I laughed, then turned to Pasquini.",
+ "raw": {
+ "normalized_reference": "ask him, i laughed, then turned to pasquini.",
+ "normalized_hypothesis": "ask him i laughed, then turned to pasquini.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "ask him i laughed then turned to pasquini",
+ "normalized_hypothesis": "ask him i laughed then turned to pasquini",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0437",
+ "category": "phonetic_balance",
+ "text": "He did not rush in.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0437.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He did not rush in.",
+ "raw": {
+ "normalized_reference": "he did not rush in.",
+ "normalized_hypothesis": "he did not rush in.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he did not rush in",
+ "normalized_hypothesis": "he did not rush in",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0438",
+ "category": "phonetic_balance",
+ "text": "They obeyed him, and went here and there at his commands.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0438.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "They obeyed him and went here and there at his commands.",
+ "raw": {
+ "normalized_reference": "they obeyed him, and went here and there at his commands.",
+ "normalized_hypothesis": "they obeyed him and went here and there at his commands.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "they obeyed him and went here and there at his commands",
+ "normalized_hypothesis": "they obeyed him and went here and there at his commands",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0439",
+ "category": "phonetic_balance",
+ "text": "Tom Spink has a harpoon.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0439.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Tom Spink has a harpoon.",
+ "raw": {
+ "normalized_reference": "tom spink has a harpoon.",
+ "normalized_hypothesis": "tom spink has a harpoon.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "tom spink has a harpoon",
+ "normalized_hypothesis": "tom spink has a harpoon",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0440",
+ "category": "phonetic_balance",
+ "text": "It was my reports from the north which chiefly induced people to buy.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0440.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It was my reports from the north which chiefly induced people to buy.",
+ "raw": {
+ "normalized_reference": "it was my reports from the north which chiefly induced people to buy.",
+ "normalized_hypothesis": "it was my reports from the north which chiefly induced people to buy.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it was my reports from the north which chiefly induced people to buy",
+ "normalized_hypothesis": "it was my reports from the north which chiefly induced people to buy",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0441",
+ "category": "phonetic_balance",
+ "text": "Cherokee identified himself with his instinct.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0441.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Cherokee identified himself with his instinct.",
+ "raw": {
+ "normalized_reference": "cherokee identified himself with his instinct.",
+ "normalized_hypothesis": "cherokee identified himself with his instinct.",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "cherokee identified himself with his instinct",
+ "normalized_hypothesis": "cherokee identified himself with his instinct",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0442",
+ "category": "phonetic_balance",
+ "text": "I graduated last of my class.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0442.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I graduated last of my class.",
+ "raw": {
+ "normalized_reference": "i graduated last of my class.",
+ "normalized_hypothesis": "i graduated last of my class.",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i graduated last of my class",
+ "normalized_hypothesis": "i graduated last of my class",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0443",
+ "category": "phonetic_balance",
+ "text": "It was a gigantic inadequacy.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0443.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It was a gigantic inadequacy.",
+ "raw": {
+ "normalized_reference": "it was a gigantic inadequacy.",
+ "normalized_hypothesis": "it was a gigantic inadequacy.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it was a gigantic inadequacy",
+ "normalized_hypothesis": "it was a gigantic inadequacy",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0444",
+ "category": "phonetic_balance",
+ "text": "Did I possess too much vitality.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0444.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Did I possess too much vitality?",
+ "raw": {
+ "normalized_reference": "did i possess too much vitality.",
+ "normalized_hypothesis": "did i possess too much vitality?",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "did i possess too much vitality",
+ "normalized_hypothesis": "did i possess too much vitality",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0445",
+ "category": "phonetic_balance",
+ "text": "Therefore, hurrah for the game.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0445.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Therefore, hurrah for the game.",
+ "raw": {
+ "normalized_reference": "therefore, hurrah for the game.",
+ "normalized_hypothesis": "therefore, hurrah for the game.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "therefore hurrah for the game",
+ "normalized_hypothesis": "therefore hurrah for the game",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0446",
+ "category": "phonetic_balance",
+ "text": "He had fulfilled his duty and paid properly.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0446.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He had fulfilled his duty and paid properly.",
+ "raw": {
+ "normalized_reference": "he had fulfilled his duty and paid properly.",
+ "normalized_hypothesis": "he had fulfilled his duty and paid properly.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he had fulfilled his duty and paid properly",
+ "normalized_hypothesis": "he had fulfilled his duty and paid properly",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0447",
+ "category": "phonetic_balance",
+ "text": "How can you manage all alone, Mr Young.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0447.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "How can you manage all alone, Mr. Young?",
+ "raw": {
+ "normalized_reference": "how can you manage all alone, mr young.",
+ "normalized_hypothesis": "how can you manage all alone, mr. young?",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "how can you manage all alone mister young",
+ "normalized_hypothesis": "how can you manage all alone mister young",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0448",
+ "category": "phonetic_balance",
+ "text": "Red-Eye never committed a more outrageous deed.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0448.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Red-Eye never committed a more outrageous deed.",
+ "raw": {
+ "normalized_reference": "red-eye never committed a more outrageous deed.",
+ "normalized_hypothesis": "red-eye never committed a more outrageous deed.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "red eye never committed a more outrageous deed",
+ "normalized_hypothesis": "red eye never committed a more outrageous deed",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0449",
+ "category": "phonetic_balance",
+ "text": "He would keep his faith with Gregson for the promised day or two.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0449.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He would keep his faith with Gregson for the promised day or two.",
+ "raw": {
+ "normalized_reference": "he would keep his faith with gregson for the promised day or two.",
+ "normalized_hypothesis": "he would keep his faith with gregson for the promised day or two.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he would keep his faith with gregson for the promised day or 2",
+ "normalized_hypothesis": "he would keep his faith with gregson for the promised day or 2",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0450",
+ "category": "phonetic_balance",
+ "text": "There was one chance, and only one, of saving Jeanne.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0450.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "There was one chance and only one of saving Gene.",
+ "raw": {
+ "normalized_reference": "there was one chance, and only one, of saving jeanne.",
+ "normalized_hypothesis": "there was one chance and only one of saving gene.",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ },
+ "semantic": {
+ "normalized_reference": "there was one chance and only one of saving jeanne",
+ "normalized_hypothesis": "there was one chance and only one of saving gene",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ }
+ },
+ {
+ "id": "release500_0451",
+ "category": "phonetic_balance",
+ "text": "I was beneath the water, suffocating and drowning.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0451.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I was beneath the water suffocating and drowning.",
+ "raw": {
+ "normalized_reference": "i was beneath the water, suffocating and drowning.",
+ "normalized_hypothesis": "i was beneath the water suffocating and drowning.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i was beneath the water suffocating and drowning",
+ "normalized_hypothesis": "i was beneath the water suffocating and drowning",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0452",
+ "category": "phonetic_balance",
+ "text": "Give them their choice between a fine or an official whipping.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0452.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Give them their choice between a fine or an official whipping.",
+ "raw": {
+ "normalized_reference": "give them their choice between a fine or an official whipping.",
+ "normalized_hypothesis": "give them their choice between a fine or an official whipping.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "give them their choice between a fine or an official whipping",
+ "normalized_hypothesis": "give them their choice between a fine or an official whipping",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0453",
+ "category": "phonetic_balance",
+ "text": "They must have been swept away by the chaotic currents.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0453.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "They must have been swept away by the chaotic currents.",
+ "raw": {
+ "normalized_reference": "they must have been swept away by the chaotic currents.",
+ "normalized_hypothesis": "they must have been swept away by the chaotic currents.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "they must have been swept away by the chaotic currents",
+ "normalized_hypothesis": "they must have been swept away by the chaotic currents",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0454",
+ "category": "phonetic_balance",
+ "text": "You yellow giant thing of the frost.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0454.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "You yellow giant thing of the frost.",
+ "raw": {
+ "normalized_reference": "you yellow giant thing of the frost.",
+ "normalized_hypothesis": "you yellow giant thing of the frost.",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "you yellow giant thing of the frost",
+ "normalized_hypothesis": "you yellow giant thing of the frost",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0455",
+ "category": "phonetic_balance",
+ "text": "He had peeled off his shirt and was wildly waving it.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0455.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He had peeled off his shirt and was wildly waving it.",
+ "raw": {
+ "normalized_reference": "he had peeled off his shirt and was wildly waving it.",
+ "normalized_hypothesis": "he had peeled off his shirt and was wildly waving it.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he had peeled off his shirt and was wildly waving it",
+ "normalized_hypothesis": "he had peeled off his shirt and was wildly waving it",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0456",
+ "category": "phonetic_balance",
+ "text": "To these he gave castor oil.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0456.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "To these he gave castor oil.",
+ "raw": {
+ "normalized_reference": "to these he gave castor oil.",
+ "normalized_hypothesis": "to these he gave castor oil.",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "to these he gave castor oil",
+ "normalized_hypothesis": "to these he gave castor oil",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0457",
+ "category": "phonetic_balance",
+ "text": "Tudor surveyed him with withering disgust.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0457.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Tudor surveyed him with withering disgust.",
+ "raw": {
+ "normalized_reference": "tudor surveyed him with withering disgust.",
+ "normalized_hypothesis": "tudor surveyed him with withering disgust.",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "tudor surveyed him with withering disgust",
+ "normalized_hypothesis": "tudor surveyed him with withering disgust",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0458",
+ "category": "phonetic_balance",
+ "text": "After all, it was simply a mistake in judgment.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0458.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "After all, it was simply a mistake in judgment.",
+ "raw": {
+ "normalized_reference": "after all, it was simply a mistake in judgment.",
+ "normalized_hypothesis": "after all, it was simply a mistake in judgment.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "after all it was simply a mistake in judgment",
+ "normalized_hypothesis": "after all it was simply a mistake in judgment",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0459",
+ "category": "phonetic_balance",
+ "text": "That is the strange part of it.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0459.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "That is the strange part of it.",
+ "raw": {
+ "normalized_reference": "that is the strange part of it.",
+ "normalized_hypothesis": "that is the strange part of it.",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "that is the strange part of it",
+ "normalized_hypothesis": "that is the strange part of it",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0460",
+ "category": "phonetic_balance",
+ "text": "Not a wheel moved in his empire.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0460.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Not a wheel moved in his empire.",
+ "raw": {
+ "normalized_reference": "not a wheel moved in his empire.",
+ "normalized_hypothesis": "not a wheel moved in his empire.",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "not a wheel moved in his empire",
+ "normalized_hypothesis": "not a wheel moved in his empire",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0461",
+ "category": "phonetic_balance",
+ "text": "Its diameter was not more than two hundred yards.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0461.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Its diameter was not more than 200 yards.",
+ "raw": {
+ "normalized_reference": "its diameter was not more than two hundred yards.",
+ "normalized_hypothesis": "its diameter was not more than 200 yards.",
+ "errors": 2,
+ "words": 9,
+ "wer": 0.2222222222222222
+ },
+ "semantic": {
+ "normalized_reference": "its diameter was not more than 200 yards",
+ "normalized_hypothesis": "its diameter was not more than 200 yards",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0462",
+ "category": "phonetic_balance",
+ "text": "Nimrod replied, with a slight manifestation of sensitiveness.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0462.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Nimrod replied with a slight manifestation of sensitiveness.",
+ "raw": {
+ "normalized_reference": "nimrod replied, with a slight manifestation of sensitiveness.",
+ "normalized_hypothesis": "nimrod replied with a slight manifestation of sensitiveness.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "nimrod replied with a slight manifestation of sensitiveness",
+ "normalized_hypothesis": "nimrod replied with a slight manifestation of sensitiveness",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0463",
+ "category": "phonetic_balance",
+ "text": "I tried to read George Moore last night, and was dreadfully bored.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0463.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I tried to read George Moore last night and was dreadfully bored.",
+ "raw": {
+ "normalized_reference": "i tried to read george moore last night, and was dreadfully bored.",
+ "normalized_hypothesis": "i tried to read george moore last night and was dreadfully bored.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i tried to read george moore last night and was dreadfully bored",
+ "normalized_hypothesis": "i tried to read george moore last night and was dreadfully bored",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0464",
+ "category": "phonetic_balance",
+ "text": "Yes, it was a man who asked, a stranger.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0464.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Yes, it was a man who asked a stranger.",
+ "raw": {
+ "normalized_reference": "yes, it was a man who asked, a stranger.",
+ "normalized_hypothesis": "yes, it was a man who asked a stranger.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "yes it was a man who asked a stranger",
+ "normalized_hypothesis": "yes it was a man who asked a stranger",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0465",
+ "category": "phonetic_balance",
+ "text": "We were now good friends.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0465.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "We were now good friends.",
+ "raw": {
+ "normalized_reference": "we were now good friends.",
+ "normalized_hypothesis": "we were now good friends.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "we were now good friends",
+ "normalized_hypothesis": "we were now good friends",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0466",
+ "category": "phonetic_balance",
+ "text": "Yet, in accordance with Ernest's test of truth, it worked.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0466.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Yet in accordance with Ernest's test of truth, it worked.",
+ "raw": {
+ "normalized_reference": "yet, in accordance with ernest's test of truth, it worked.",
+ "normalized_hypothesis": "yet in accordance with ernest's test of truth, it worked.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "yet in accordance with ernest is test of truth it worked",
+ "normalized_hypothesis": "yet in accordance with ernest is test of truth it worked",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0467",
+ "category": "phonetic_balance",
+ "text": "It's the strap hangers that'll keep us from going under.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0467.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It's the strap hangers that'll keep us from going under.",
+ "raw": {
+ "normalized_reference": "it's the strap hangers that'll keep us from going under.",
+ "normalized_hypothesis": "it's the strap hangers that'll keep us from going under.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it is the strap hangers that will keep us from going under",
+ "normalized_hypothesis": "it is the strap hangers that will keep us from going under",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0468",
+ "category": "phonetic_balance",
+ "text": "I never allow what can't be changed to annoy me.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0468.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I never allow what can't be changed to annoy me.",
+ "raw": {
+ "normalized_reference": "i never allow what can't be changed to annoy me.",
+ "normalized_hypothesis": "i never allow what can't be changed to annoy me.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i never allow what can not be changed to annoy me",
+ "normalized_hypothesis": "i never allow what can not be changed to annoy me",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0469",
+ "category": "phonetic_balance",
+ "text": "After all superfluous flesh is gone what is left is stringy and resistant.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0469.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "After all superfluous flesh is gone, what is left is stringy and resistant.",
+ "raw": {
+ "normalized_reference": "after all superfluous flesh is gone what is left is stringy and resistant.",
+ "normalized_hypothesis": "after all superfluous flesh is gone, what is left is stringy and resistant.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "after all superfluous flesh is gone what is left is stringy and resistant",
+ "normalized_hypothesis": "after all superfluous flesh is gone what is left is stringy and resistant",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0470",
+ "category": "phonetic_balance",
+ "text": "Then he shouted, Shut up.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0470.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Then he shouted, shut up.",
+ "raw": {
+ "normalized_reference": "then he shouted, shut up.",
+ "normalized_hypothesis": "then he shouted, shut up.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "then he shouted shut up",
+ "normalized_hypothesis": "then he shouted shut up",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0471",
+ "category": "phonetic_balance",
+ "text": "He had worshiped her, as Dante might have worshiped Beatrice.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0471.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He had worshipped her as Dante might have worshipped Beatrice.",
+ "raw": {
+ "normalized_reference": "he had worshiped her, as dante might have worshiped beatrice.",
+ "normalized_hypothesis": "he had worshipped her as dante might have worshipped beatrice.",
+ "errors": 2,
+ "words": 10,
+ "wer": 0.2
+ },
+ "semantic": {
+ "normalized_reference": "he had worshiped her as dante might have worshiped beatrice",
+ "normalized_hypothesis": "he had worshipped her as dante might have worshipped beatrice",
+ "errors": 2,
+ "words": 10,
+ "wer": 0.2
+ }
+ },
+ {
+ "id": "release500_0472",
+ "category": "phonetic_balance",
+ "text": "He looked like one who had passed through an uncomfortable hour or two.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0472.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He looked like one who had passed through an uncomfortable hour or two.",
+ "raw": {
+ "normalized_reference": "he looked like one who had passed through an uncomfortable hour or two.",
+ "normalized_hypothesis": "he looked like one who had passed through an uncomfortable hour or two.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he looked like one who had passed through an uncomfortable hour or 2",
+ "normalized_hypothesis": "he looked like one who had passed through an uncomfortable hour or 2",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0473",
+ "category": "phonetic_balance",
+ "text": "The history of our westward-faring race is written in it.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0473.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The history of our westward-faring race is written in it.",
+ "raw": {
+ "normalized_reference": "the history of our westward-faring race is written in it.",
+ "normalized_hypothesis": "the history of our westward-faring race is written in it.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the history of our westward faring race is written in it",
+ "normalized_hypothesis": "the history of our westward faring race is written in it",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0474",
+ "category": "phonetic_balance",
+ "text": "He may desire to escape pain, or to enjoy pleasure.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0474.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He may desire to escape pain or to enjoy pleasure.",
+ "raw": {
+ "normalized_reference": "he may desire to escape pain, or to enjoy pleasure.",
+ "normalized_hypothesis": "he may desire to escape pain or to enjoy pleasure.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he may desire to escape pain or to enjoy pleasure",
+ "normalized_hypothesis": "he may desire to escape pain or to enjoy pleasure",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0475",
+ "category": "phonetic_balance",
+ "text": "And the air was growing chilly.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0475.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "And the air was growing chilly.",
+ "raw": {
+ "normalized_reference": "and the air was growing chilly.",
+ "normalized_hypothesis": "and the air was growing chilly.",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "and the air was growing chilly",
+ "normalized_hypothesis": "and the air was growing chilly",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0476",
+ "category": "phonetic_balance",
+ "text": "Dig in; you're sure good, was Daylight's answer.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0476.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Dig in, you're sure good, was Daylight's answer.",
+ "raw": {
+ "normalized_reference": "dig in; you're sure good, was daylight's answer.",
+ "normalized_hypothesis": "dig in, you're sure good, was daylight's answer.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "dig in you are sure good was daylight is answer",
+ "normalized_hypothesis": "dig in you are sure good was daylight is answer",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0477",
+ "category": "phonetic_balance",
+ "text": "My idea was that he would have more influence over the natives.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0477.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "My idea was that he would have more influence over the natives.",
+ "raw": {
+ "normalized_reference": "my idea was that he would have more influence over the natives.",
+ "normalized_hypothesis": "my idea was that he would have more influence over the natives.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "my idea was that he would have more influence over the natives",
+ "normalized_hypothesis": "my idea was that he would have more influence over the natives",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0478",
+ "category": "phonetic_balance",
+ "text": "Now run along, and tell them to hurry.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0478.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Now run along and tell them to hurry.",
+ "raw": {
+ "normalized_reference": "now run along, and tell them to hurry.",
+ "normalized_hypothesis": "now run along and tell them to hurry.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "now run along and tell them to hurry",
+ "normalized_hypothesis": "now run along and tell them to hurry",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0479",
+ "category": "phonetic_balance",
+ "text": "I have no idea, replied Philip.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0479.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I have no idea, replied Philip.",
+ "raw": {
+ "normalized_reference": "i have no idea, replied philip.",
+ "normalized_hypothesis": "i have no idea, replied philip.",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i have no idea replied philip",
+ "normalized_hypothesis": "i have no idea replied philip",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0480",
+ "category": "phonetic_balance",
+ "text": "For the rest, he was a mere automaton.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0480.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "For the rest, he was a mere automaton.",
+ "raw": {
+ "normalized_reference": "for the rest, he was a mere automaton.",
+ "normalized_hypothesis": "for the rest, he was a mere automaton.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "for the rest he was a mere automaton",
+ "normalized_hypothesis": "for the rest he was a mere automaton",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0481",
+ "category": "phonetic_balance",
+ "text": "Why, the average review is more nauseating than cod liver oil.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0481.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Why the average review is more nauseating than cod liver oil.",
+ "raw": {
+ "normalized_reference": "why, the average review is more nauseating than cod liver oil.",
+ "normalized_hypothesis": "why the average review is more nauseating than cod liver oil.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "why the average review is more nauseating than cod liver oil",
+ "normalized_hypothesis": "why the average review is more nauseating than cod liver oil",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0482",
+ "category": "phonetic_balance",
+ "text": "I have been robbed, sir, I amended.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0482.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I have been robbed, sir, I amend it.",
+ "raw": {
+ "normalized_reference": "i have been robbed, sir, i amended.",
+ "normalized_hypothesis": "i have been robbed, sir, i amend it.",
+ "errors": 2,
+ "words": 7,
+ "wer": 0.2857142857142857
+ },
+ "semantic": {
+ "normalized_reference": "i have been robbed sir i amended",
+ "normalized_hypothesis": "i have been robbed sir i amend it",
+ "errors": 2,
+ "words": 7,
+ "wer": 0.2857142857142857
+ }
+ },
+ {
+ "id": "release500_0483",
+ "category": "phonetic_balance",
+ "text": "He moved away as quietly as he had come.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0483.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He moved away as quietly as he had come.",
+ "raw": {
+ "normalized_reference": "he moved away as quietly as he had come.",
+ "normalized_hypothesis": "he moved away as quietly as he had come.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he moved away as quietly as he had come",
+ "normalized_hypothesis": "he moved away as quietly as he had come",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0484",
+ "category": "phonetic_balance",
+ "text": "A trickle of fresh blood ran over his face.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0484.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "A trickle of fresh blood ran over his face.",
+ "raw": {
+ "normalized_reference": "a trickle of fresh blood ran over his face.",
+ "normalized_hypothesis": "a trickle of fresh blood ran over his face.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "a trickle of fresh blood ran over his face",
+ "normalized_hypothesis": "a trickle of fresh blood ran over his face",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0485",
+ "category": "phonetic_balance",
+ "text": "Suppose you saw me at work through the window.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0485.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Suppose you saw me at work through the window.",
+ "raw": {
+ "normalized_reference": "suppose you saw me at work through the window.",
+ "normalized_hypothesis": "suppose you saw me at work through the window.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "suppose you saw me at work through the window",
+ "normalized_hypothesis": "suppose you saw me at work through the window",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0486",
+ "category": "phonetic_balance",
+ "text": "But who was Eileen's double.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0486.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "But who was Eileen's double?",
+ "raw": {
+ "normalized_reference": "but who was eileen's double.",
+ "normalized_hypothesis": "but who was eileen's double?",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "but who was eileen is double",
+ "normalized_hypothesis": "but who was eileen is double",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0487",
+ "category": "phonetic_balance",
+ "text": "Darkness hid him from Jeanne.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0487.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Darkness hid him from Gene.",
+ "raw": {
+ "normalized_reference": "darkness hid him from jeanne.",
+ "normalized_hypothesis": "darkness hid him from gene.",
+ "errors": 1,
+ "words": 5,
+ "wer": 0.2
+ },
+ "semantic": {
+ "normalized_reference": "darkness hid him from jeanne",
+ "normalized_hypothesis": "darkness hid him from gene",
+ "errors": 1,
+ "words": 5,
+ "wer": 0.2
+ }
+ },
+ {
+ "id": "release500_0488",
+ "category": "phonetic_balance",
+ "text": "By this time Charley was as enraged as the Greek.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0488.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "By this time Charlie was as enraged as the Greek.",
+ "raw": {
+ "normalized_reference": "by this time charley was as enraged as the greek.",
+ "normalized_hypothesis": "by this time charlie was as enraged as the greek.",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ },
+ "semantic": {
+ "normalized_reference": "by this time charley was as enraged as the greek",
+ "normalized_hypothesis": "by this time charlie was as enraged as the greek",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ }
+ },
+ {
+ "id": "release500_0489",
+ "category": "phonetic_balance",
+ "text": "I am writing these lines in Honolulu, Hawaii.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0489.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I am writing these lines in Honolulu, Hawaii.",
+ "raw": {
+ "normalized_reference": "i am writing these lines in honolulu, hawaii.",
+ "normalized_hypothesis": "i am writing these lines in honolulu, hawaii.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i am writing these lines in honolulu hawaii",
+ "normalized_hypothesis": "i am writing these lines in honolulu hawaii",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0490",
+ "category": "phonetic_balance",
+ "text": "Nobody knows how the natives got them.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0490.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Nobody knows how the natives got them.",
+ "raw": {
+ "normalized_reference": "nobody knows how the natives got them.",
+ "normalized_hypothesis": "nobody knows how the natives got them.",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "nobody knows how the natives got them",
+ "normalized_hypothesis": "nobody knows how the natives got them",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0491",
+ "category": "phonetic_balance",
+ "text": "And how would we ever find ourselves.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0491.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "And how would we ever find ourselves?",
+ "raw": {
+ "normalized_reference": "and how would we ever find ourselves.",
+ "normalized_hypothesis": "and how would we ever find ourselves?",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "and how would we ever find ourselves",
+ "normalized_hypothesis": "and how would we ever find ourselves",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0492",
+ "category": "phonetic_balance",
+ "text": "They are coming ashore, whoever they are.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0492.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "They are coming ashore, whoever they are.",
+ "raw": {
+ "normalized_reference": "they are coming ashore, whoever they are.",
+ "normalized_hypothesis": "they are coming ashore, whoever they are.",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "they are coming ashore whoever they are",
+ "normalized_hypothesis": "they are coming ashore whoever they are",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0493",
+ "category": "phonetic_balance",
+ "text": "This is my fifth voyage.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0493.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "This is my fifth voyage.",
+ "raw": {
+ "normalized_reference": "this is my fifth voyage.",
+ "normalized_hypothesis": "this is my fifth voyage.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "this is my 5th voyage",
+ "normalized_hypothesis": "this is my 5th voyage",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0494",
+ "category": "phonetic_balance",
+ "text": "And yet if she came he had no words to say.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0494.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "And yet if she came he had no words to say.",
+ "raw": {
+ "normalized_reference": "and yet if she came he had no words to say.",
+ "normalized_hypothesis": "and yet if she came he had no words to say.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "and yet if she came he had no words to say",
+ "normalized_hypothesis": "and yet if she came he had no words to say",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0495",
+ "category": "phonetic_balance",
+ "text": "You must sleep, he urged.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0495.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "You must sleep, he urged.",
+ "raw": {
+ "normalized_reference": "you must sleep, he urged.",
+ "normalized_hypothesis": "you must sleep, he urged.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "you must sleep he urged",
+ "normalized_hypothesis": "you must sleep he urged",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0496",
+ "category": "phonetic_balance",
+ "text": "Could the incident have anything to do with Jeanne and Pierre.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0496.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Could the incident have anything to do with Jean and Pierre?",
+ "raw": {
+ "normalized_reference": "could the incident have anything to do with jeanne and pierre.",
+ "normalized_hypothesis": "could the incident have anything to do with jean and pierre?",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ },
+ "semantic": {
+ "normalized_reference": "could the incident have anything to do with jeanne and pierre",
+ "normalized_hypothesis": "could the incident have anything to do with jean and pierre",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ }
+ },
+ {
+ "id": "release500_0497",
+ "category": "phonetic_balance",
+ "text": "Each day she became a more vital part of him.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0497.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Each day she became a more vital part of him.",
+ "raw": {
+ "normalized_reference": "each day she became a more vital part of him.",
+ "normalized_hypothesis": "each day she became a more vital part of him.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "each day she became a more vital part of him",
+ "normalized_hypothesis": "each day she became a more vital part of him",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0498",
+ "category": "phonetic_balance",
+ "text": "But why continue the tirade, for tirade it was.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0498.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "But why continue the tirade for tirade it was?",
+ "raw": {
+ "normalized_reference": "but why continue the tirade, for tirade it was.",
+ "normalized_hypothesis": "but why continue the tirade for tirade it was?",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "but why continue the tirade for tirade it was",
+ "normalized_hypothesis": "but why continue the tirade for tirade it was",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0499",
+ "category": "phonetic_balance",
+ "text": "That is what distinguishes all of us from the lower animals.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0499.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "That is what distinguishes all of us from the lower animals.",
+ "raw": {
+ "normalized_reference": "that is what distinguishes all of us from the lower animals.",
+ "normalized_hypothesis": "that is what distinguishes all of us from the lower animals.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "that is what distinguishes all of us from the lower animals",
+ "normalized_hypothesis": "that is what distinguishes all of us from the lower animals",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0500",
+ "category": "phonetic_balance",
+ "text": "There was something pathetic in the girl's attitude now.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0500.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "There was something pathetic in the girl's attitude now.",
+ "raw": {
+ "normalized_reference": "there was something pathetic in the girl's attitude now.",
+ "normalized_hypothesis": "there was something pathetic in the girl's attitude now.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "there was something pathetic in the girl is attitude now",
+ "normalized_hypothesis": "there was something pathetic in the girl is attitude now",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0001",
+ "category": "conversational",
+ "text": "It wasn't until later that I realized what had actually happened.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0001.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It wasn't until later that I realized what had actually happened.",
+ "raw": {
+ "normalized_reference": "it wasn't until later that i realized what had actually happened.",
+ "normalized_hypothesis": "it wasn't until later that i realized what had actually happened.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it was not until later that i realized what had actually happened",
+ "normalized_hypothesis": "it was not until later that i realized what had actually happened",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0002",
+ "category": "descriptive",
+ "text": "The light at that time of afternoon is perfect for about twenty minutes.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0002.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The light at that time of afternoon is perfect for about 20 minutes.",
+ "raw": {
+ "normalized_reference": "the light at that time of afternoon is perfect for about twenty minutes.",
+ "normalized_hypothesis": "the light at that time of afternoon is perfect for about 20 minutes.",
+ "errors": 1,
+ "words": 13,
+ "wer": 0.07692307692307693
+ },
+ "semantic": {
+ "normalized_reference": "the light at that time of afternoon is perfect for about 20 minutes",
+ "normalized_hypothesis": "the light at that time of afternoon is perfect for about 20 minutes",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0003",
+ "category": "emotional",
+ "text": "That was the scariest thing I've ever done, and also maybe the best.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0003.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "That was the scariest thing I've ever done, and also maybe the best.",
+ "raw": {
+ "normalized_reference": "that was the scariest thing i've ever done, and also maybe the best.",
+ "normalized_hypothesis": "that was the scariest thing i've ever done, and also maybe the best.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "that was the scariest thing i have ever done and also maybe the best",
+ "normalized_hypothesis": "that was the scariest thing i have ever done and also maybe the best",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0004",
+ "category": "homographs",
+ "text": "After I record the record, please present the present to the bass player.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0004.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "After I record the record, please present the present to the bass player.",
+ "raw": {
+ "normalized_reference": "after i record the record, please present the present to the bass player.",
+ "normalized_hypothesis": "after i record the record, please present the present to the bass player.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "after i record the record please present the present to the bass player",
+ "normalized_hypothesis": "after i record the record please present the present to the bass player",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0005",
+ "category": "long",
+ "text": "If I'm being completely honest, I had no idea what I was doing for most of it, and I think that might be the most normal thing about the whole experience.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0005.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "If I'm being completely honest, I had no idea what I was doing for most of it, and I think that might be the most normal thing about the whole experience.",
+ "raw": {
+ "normalized_reference": "if i'm being completely honest, i had no idea what i was doing for most of it, and i think that might be the most normal thing about the whole experience.",
+ "normalized_hypothesis": "if i'm being completely honest, i had no idea what i was doing for most of it, and i think that might be the most normal thing about the whole experience.",
+ "errors": 0,
+ "words": 31,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "if i am being completely honest i had no idea what i was doing for most of it and i think that might be the most normal thing about the whole experience",
+ "normalized_hypothesis": "if i am being completely honest i had no idea what i was doing for most of it and i think that might be the most normal thing about the whole experience",
+ "errors": 0,
+ "words": 32,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0006",
+ "category": "long_range",
+ "text": "The committee postponed its decision after discovering that three pages of the report had been printed in the wrong order.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0006.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The committee postponed its decision after discovering that three pages of the report had been printed in the wrong order.",
+ "raw": {
+ "normalized_reference": "the committee postponed its decision after discovering that three pages of the report had been printed in the wrong order.",
+ "normalized_hypothesis": "the committee postponed its decision after discovering that three pages of the report had been printed in the wrong order.",
+ "errors": 0,
+ "words": 20,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the committee postponed its decision after discovering that 3 pages of the report had been printed in the wrong order",
+ "normalized_hypothesis": "the committee postponed its decision after discovering that 3 pages of the report had been printed in the wrong order",
+ "errors": 0,
+ "words": 20,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0007",
+ "category": "names_places",
+ "text": "Gwendolyn photographed the eucalyptus trees outside Ljubljana.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0007.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Gwendolyn photographed the eucalyptus trees outside Ljubljana.",
+ "raw": {
+ "normalized_reference": "gwendolyn photographed the eucalyptus trees outside ljubljana.",
+ "normalized_hypothesis": "gwendolyn photographed the eucalyptus trees outside ljubljana.",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "gwendolyn photographed the eucalyptus trees outside ljubljana",
+ "normalized_hypothesis": "gwendolyn photographed the eucalyptus trees outside ljubljana",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0008",
+ "category": "narrative",
+ "text": "He had this theory, and the more time passes, the more I think he was right.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0008.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He had this theory and the more time passes, the more I think he was right.",
+ "raw": {
+ "normalized_reference": "he had this theory, and the more time passes, the more i think he was right.",
+ "normalized_hypothesis": "he had this theory and the more time passes, the more i think he was right.",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he had this theory and the more time passes the more i think he was right",
+ "normalized_hypothesis": "he had this theory and the more time passes the more i think he was right",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0009",
+ "category": "numbers",
+ "text": "The package weighs twelve point six kilograms and arrived on July twenty-first.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0009.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The package weighs 12.6 kilograms and arrived on July 21st.",
+ "raw": {
+ "normalized_reference": "the package weighs twelve point six kilograms and arrived on july twenty-first.",
+ "normalized_hypothesis": "the package weighs 12.6 kilograms and arrived on july 21st.",
+ "errors": 5,
+ "words": 13,
+ "wer": 0.38461538461538464
+ },
+ "semantic": {
+ "normalized_reference": "the package weighs 12.6 kilograms and arrived on july 21st",
+ "normalized_hypothesis": "the package weighs 12.6 kilograms and arrived on july 21st",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0010",
+ "category": "punctuation",
+ "text": "First, close the window; second, turn off the lamp; finally, lock the door.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0010.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "First, close the window. Second, turn off the lamp. Finally, lock the door.",
+ "raw": {
+ "normalized_reference": "first, close the window; second, turn off the lamp; finally, lock the door.",
+ "normalized_hypothesis": "first, close the window. second, turn off the lamp. finally, lock the door.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "1st close the window 2nd turn off the lamp finally lock the door",
+ "normalized_hypothesis": "1st close the window 2nd turn off the lamp finally lock the door",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0011",
+ "category": "question",
+ "text": "When did you first realize something was wrong?",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0011.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "When did you first realize something was wrong?",
+ "raw": {
+ "normalized_reference": "when did you first realize something was wrong?",
+ "normalized_hypothesis": "when did you first realize something was wrong?",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "when did you 1st realize something was wrong",
+ "normalized_hypothesis": "when did you 1st realize something was wrong",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0012",
+ "category": "rare_words",
+ "text": "The archaeologist catalogued a Byzantine reliquary and several anachronistic fragments.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0012.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The archaeologist catalogued a Byzantine reliquary and several anachronistic fragments.",
+ "raw": {
+ "normalized_reference": "the archaeologist catalogued a byzantine reliquary and several anachronistic fragments.",
+ "normalized_hypothesis": "the archaeologist catalogued a byzantine reliquary and several anachronistic fragments.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the archaeologist catalogued a byzantine reliquary and several anachronistic fragments",
+ "normalized_hypothesis": "the archaeologist catalogued a byzantine reliquary and several anachronistic fragments",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0013",
+ "category": "short",
+ "text": "I have no idea.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0013.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I have no idea.",
+ "raw": {
+ "normalized_reference": "i have no idea.",
+ "normalized_hypothesis": "i have no idea.",
+ "errors": 0,
+ "words": 4,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i have no idea",
+ "normalized_hypothesis": "i have no idea",
+ "errors": 0,
+ "words": 4,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0014",
+ "category": "technical",
+ "text": "The system runs on three core components that all have to stay in sync.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0014.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The system runs on three core components that all have to stay in sync.",
+ "raw": {
+ "normalized_reference": "the system runs on three core components that all have to stay in sync.",
+ "normalized_hypothesis": "the system runs on three core components that all have to stay in sync.",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the system runs on 3 core components that all have to stay in sync",
+ "normalized_hypothesis": "the system runs on 3 core components that all have to stay in sync",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0015",
+ "category": "conversational",
+ "text": "We improvised the entire second half and somehow nobody noticed.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0015.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "We improvised the entire second half and somehow nobody noticed.",
+ "raw": {
+ "normalized_reference": "we improvised the entire second half and somehow nobody noticed.",
+ "normalized_hypothesis": "we improvised the entire second half and somehow nobody noticed.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "we improvised the entire 2nd half and somehow nobody noticed",
+ "normalized_hypothesis": "we improvised the entire 2nd half and somehow nobody noticed",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0016",
+ "category": "descriptive",
+ "text": "There's a smell in old buildings that's impossible to describe but immediately familiar.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0016.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "There's a smell in old buildings that's impossible to describe but immediately familiar.",
+ "raw": {
+ "normalized_reference": "there's a smell in old buildings that's impossible to describe but immediately familiar.",
+ "normalized_hypothesis": "there's a smell in old buildings that's impossible to describe but immediately familiar.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "there is a smell in old buildings that is impossible to describe but immediately familiar",
+ "normalized_hypothesis": "there is a smell in old buildings that is impossible to describe but immediately familiar",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0017",
+ "category": "emotional",
+ "text": "I specifically asked for one thing. One thing. And it still didn't happen.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0017.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I specifically asked for one thing. One thing, and it still didn't happen.",
+ "raw": {
+ "normalized_reference": "i specifically asked for one thing. one thing. and it still didn't happen.",
+ "normalized_hypothesis": "i specifically asked for one thing. one thing, and it still didn't happen.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i specifically asked for one thing one thing and it still did not happen",
+ "normalized_hypothesis": "i specifically asked for one thing one thing and it still did not happen",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0018",
+ "category": "homographs",
+ "text": "They refuse to place the refuse beside the produce aisle.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0018.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "They refused to place the refuse beside the produce aisle.",
+ "raw": {
+ "normalized_reference": "they refuse to place the refuse beside the produce aisle.",
+ "normalized_hypothesis": "they refused to place the refuse beside the produce aisle.",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ },
+ "semantic": {
+ "normalized_reference": "they refuse to place the refuse beside the produce aisle",
+ "normalized_hypothesis": "they refused to place the refuse beside the produce aisle",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ }
+ },
+ {
+ "id": "release500_0019",
+ "category": "long",
+ "text": "There were days last year when I genuinely didn't know how I was going to get through it, and somehow I did, and I still don't fully understand how.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0019.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "There were days last year when I genuinely didn't know how I was going to get through it, and somehow I did, and I still don't fully understand how.",
+ "raw": {
+ "normalized_reference": "there were days last year when i genuinely didn't know how i was going to get through it, and somehow i did, and i still don't fully understand how.",
+ "normalized_hypothesis": "there were days last year when i genuinely didn't know how i was going to get through it, and somehow i did, and i still don't fully understand how.",
+ "errors": 0,
+ "words": 29,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "there were days last year when i genuinely did not know how i was going to get through it and somehow i did and i still do not fully understand how",
+ "normalized_hypothesis": "there were days last year when i genuinely did not know how i was going to get through it and somehow i did and i still do not fully understand how",
+ "errors": 0,
+ "words": 31,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0020",
+ "category": "long_range",
+ "text": "Before announcing the winner, the judge thanked the volunteers who had worked through the storm to repair the damaged stage.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0020.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Before announcing the winner, the judge thanked the volunteers who had worked through the storm to repair the damaged stage.",
+ "raw": {
+ "normalized_reference": "before announcing the winner, the judge thanked the volunteers who had worked through the storm to repair the damaged stage.",
+ "normalized_hypothesis": "before announcing the winner, the judge thanked the volunteers who had worked through the storm to repair the damaged stage.",
+ "errors": 0,
+ "words": 20,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "before announcing the winner the judge thanked the volunteers who had worked through the storm to repair the damaged stage",
+ "normalized_hypothesis": "before announcing the winner the judge thanked the volunteers who had worked through the storm to repair the damaged stage",
+ "errors": 0,
+ "words": 20,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0021",
+ "category": "names_places",
+ "text": "Professor Nguyen mailed the fluorescent poster to Reykjavik on Thursday.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0021.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Professor Nguyen mailed the fluorescent poster to Reykjavik on Thursday.",
+ "raw": {
+ "normalized_reference": "professor nguyen mailed the fluorescent poster to reykjavik on thursday.",
+ "normalized_hypothesis": "professor nguyen mailed the fluorescent poster to reykjavik on thursday.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "professor nguyen mailed the fluorescent poster to reykjavik on thursday",
+ "normalized_hypothesis": "professor nguyen mailed the fluorescent poster to reykjavik on thursday",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0022",
+ "category": "narrative",
+ "text": "The plan was simple. The execution was anything but.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0022.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The plan was simple. The execution was anything but.",
+ "raw": {
+ "normalized_reference": "the plan was simple. the execution was anything but.",
+ "normalized_hypothesis": "the plan was simple. the execution was anything but.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the plan was simple the execution was anything but",
+ "normalized_hypothesis": "the plan was simple the execution was anything but",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0023",
+ "category": "numbers",
+ "text": "Her score improved from eighty-two percent to ninety-seven percent in two weeks.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0023.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Her score improved from 82% to 97% in two weeks.",
+ "raw": {
+ "normalized_reference": "her score improved from eighty-two percent to ninety-seven percent in two weeks.",
+ "normalized_hypothesis": "her score improved from 82% to 97% in two weeks.",
+ "errors": 6,
+ "words": 14,
+ "wer": 0.42857142857142855
+ },
+ "semantic": {
+ "normalized_reference": "her score improved from 82% to 97% in 2 weeks",
+ "normalized_hypothesis": "her score improved from 82% to 97% in 2 weeks",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0024",
+ "category": "punctuation",
+ "text": "Why would anyone pack spoons, batteries, and a violin for one short trip?",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0024.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Why would anyone pack spoons, batteries, and a violin for one short trip?",
+ "raw": {
+ "normalized_reference": "why would anyone pack spoons, batteries, and a violin for one short trip?",
+ "normalized_hypothesis": "why would anyone pack spoons, batteries, and a violin for one short trip?",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "why would anyone pack spoons batteries and a violin for one short trip",
+ "normalized_hypothesis": "why would anyone pack spoons batteries and a violin for one short trip",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0025",
+ "category": "question",
+ "text": "How long have you been sitting on this?",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0025.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "How long have you been sitting on this?",
+ "raw": {
+ "normalized_reference": "how long have you been sitting on this?",
+ "normalized_hypothesis": "how long have you been sitting on this?",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "how long have you been sitting on this",
+ "normalized_hypothesis": "how long have you been sitting on this",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0026",
+ "category": "rare_words",
+ "text": "A mischievous chameleon rested beside the chrysanthemum and the bougainvillea.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0026.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "A mischievous chameleon rested beside the Chrysanthemum and the Balgainvillia.",
+ "raw": {
+ "normalized_reference": "a mischievous chameleon rested beside the chrysanthemum and the bougainvillea.",
+ "normalized_hypothesis": "a mischievous chameleon rested beside the chrysanthemum and the balgainvillia.",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ },
+ "semantic": {
+ "normalized_reference": "a mischievous chameleon rested beside the chrysanthemum and the bougainvillea",
+ "normalized_hypothesis": "a mischievous chameleon rested beside the chrysanthemum and the balgainvillia",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ }
+ },
+ {
+ "id": "release500_0027",
+ "category": "short",
+ "text": "We don't talk about that.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0027.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "We don't talk about that.",
+ "raw": {
+ "normalized_reference": "we don't talk about that.",
+ "normalized_hypothesis": "we don't talk about that.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "we do not talk about that",
+ "normalized_hypothesis": "we do not talk about that",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0028",
+ "category": "technical",
+ "text": "Hash the UTF eight payload with S H A two fifty-six before opening the checkpoint.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0028.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Hash the UTF-8 payload with SH-256 before opening the checkpoint.",
+ "raw": {
+ "normalized_reference": "hash the utf eight payload with s h a two fifty-six before opening the checkpoint.",
+ "normalized_hypothesis": "hash the utf-8 payload with sh-256 before opening the checkpoint.",
+ "errors": 7,
+ "words": 16,
+ "wer": 0.4375
+ },
+ "semantic": {
+ "normalized_reference": "hash the utf 8 payload with s h a 256 before opening the checkpoint",
+ "normalized_hypothesis": "hash the utf 8 payload with sh 256 before opening the checkpoint",
+ "errors": 3,
+ "words": 14,
+ "wer": 0.21428571428571427
+ }
+ },
+ {
+ "id": "release500_0029",
+ "category": "conversational",
+ "text": "I thought I had it all figured out, and then this happened.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0029.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I thought I had it all figured out and then this happened.",
+ "raw": {
+ "normalized_reference": "i thought i had it all figured out, and then this happened.",
+ "normalized_hypothesis": "i thought i had it all figured out and then this happened.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i thought i had it all figured out and then this happened",
+ "normalized_hypothesis": "i thought i had it all figured out and then this happened",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0030",
+ "category": "descriptive",
+ "text": "The kind of quiet you only get at three in the morning in a city.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0030.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The kind of quiet you only get at three in the morning in a city.",
+ "raw": {
+ "normalized_reference": "the kind of quiet you only get at three in the morning in a city.",
+ "normalized_hypothesis": "the kind of quiet you only get at three in the morning in a city.",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the kind of quiet you only get at 3 in the morning in a city",
+ "normalized_hypothesis": "the kind of quiet you only get at 3 in the morning in a city",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0031",
+ "category": "emotional",
+ "text": "The joy of it was almost overwhelming, like too much good all at once.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0031.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The joy of it was almost overwhelming, like too much good all at once.",
+ "raw": {
+ "normalized_reference": "the joy of it was almost overwhelming, like too much good all at once.",
+ "normalized_hypothesis": "the joy of it was almost overwhelming, like too much good all at once.",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the joy of it was almost overwhelming like too much good all at once",
+ "normalized_hypothesis": "the joy of it was almost overwhelming like too much good all at once",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0032",
+ "category": "homographs",
+ "text": "Please close the door while I stand close to the microphone.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0032.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Please close the door while I stand close to the microphone.",
+ "raw": {
+ "normalized_reference": "please close the door while i stand close to the microphone.",
+ "normalized_hypothesis": "please close the door while i stand close to the microphone.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "please close the door while i stand close to the microphone",
+ "normalized_hypothesis": "please close the door while i stand close to the microphone",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0033",
+ "category": "long",
+ "text": "I think the most underrated skill in the world is knowing when to stop explaining yourself and just let people draw their own conclusions.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0033.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I think the most underrated skill in the world is knowing when to stop explaining yourself and just let people draw their own conclusions.",
+ "raw": {
+ "normalized_reference": "i think the most underrated skill in the world is knowing when to stop explaining yourself and just let people draw their own conclusions.",
+ "normalized_hypothesis": "i think the most underrated skill in the world is knowing when to stop explaining yourself and just let people draw their own conclusions.",
+ "errors": 0,
+ "words": 24,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i think the most underrated skill in the world is knowing when to stop explaining yourself and just let people draw their own conclusions",
+ "normalized_hypothesis": "i think the most underrated skill in the world is knowing when to stop explaining yourself and just let people draw their own conclusions",
+ "errors": 0,
+ "words": 24,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0034",
+ "category": "long_range",
+ "text": "When the final bus disappeared around the corner, Nadia remembered that her umbrella, notebook, and lunch were still inside the quiet office.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0034.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "When the final bus disappeared around the corner, Nadia remembered that her umbrella, notebook, and lunch were still inside the quiet office.",
+ "raw": {
+ "normalized_reference": "when the final bus disappeared around the corner, nadia remembered that her umbrella, notebook, and lunch were still inside the quiet office.",
+ "normalized_hypothesis": "when the final bus disappeared around the corner, nadia remembered that her umbrella, notebook, and lunch were still inside the quiet office.",
+ "errors": 0,
+ "words": 22,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "when the final bus disappeared around the corner nadia remembered that her umbrella notebook and lunch were still inside the quiet office",
+ "normalized_hypothesis": "when the final bus disappeared around the corner nadia remembered that her umbrella notebook and lunch were still inside the quiet office",
+ "errors": 0,
+ "words": 22,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0035",
+ "category": "names_places",
+ "text": "Aurelia and Joaquin crossed the Schuylkill River before noon.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0035.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Aurelia and Joaquin crossed the Schalkel River before noon.",
+ "raw": {
+ "normalized_reference": "aurelia and joaquin crossed the schuylkill river before noon.",
+ "normalized_hypothesis": "aurelia and joaquin crossed the schalkel river before noon.",
+ "errors": 1,
+ "words": 9,
+ "wer": 0.1111111111111111
+ },
+ "semantic": {
+ "normalized_reference": "aurelia and joaquin crossed the schuylkill river before noon",
+ "normalized_hypothesis": "aurelia and joaquin crossed the schalkel river before noon",
+ "errors": 1,
+ "words": 9,
+ "wer": 0.1111111111111111
+ }
+ },
+ {
+ "id": "release500_0036",
+ "category": "narrative",
+ "text": "The village had a name, but nobody used it. Everyone just called it home.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0036.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The village had a name, but nobody used it. Everyone just called it home.",
+ "raw": {
+ "normalized_reference": "the village had a name, but nobody used it. everyone just called it home.",
+ "normalized_hypothesis": "the village had a name, but nobody used it. everyone just called it home.",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the village had a name but nobody used it everyone just called it home",
+ "normalized_hypothesis": "the village had a name but nobody used it everyone just called it home",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0037",
+ "category": "numbers",
+ "text": "Increase the gain by three decibels, wait five seconds, then reduce it by one.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0037.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Increase the gain by 3 decibels, wait 5 seconds, then reduce it by 1.",
+ "raw": {
+ "normalized_reference": "increase the gain by three decibels, wait five seconds, then reduce it by one.",
+ "normalized_hypothesis": "increase the gain by 3 decibels, wait 5 seconds, then reduce it by 1.",
+ "errors": 3,
+ "words": 14,
+ "wer": 0.21428571428571427
+ },
+ "semantic": {
+ "normalized_reference": "increase the gain by 3 decibels wait 5 seconds then reduce it by one",
+ "normalized_hypothesis": "increase the gain by 3 decibels wait 5 seconds then reduce it by one",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0038",
+ "category": "punctuation",
+ "text": "The result was unexpected: clear, stable, and surprisingly natural.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0038.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The result was unexpected, clear, stable, and surprisingly natural.",
+ "raw": {
+ "normalized_reference": "the result was unexpected: clear, stable, and surprisingly natural.",
+ "normalized_hypothesis": "the result was unexpected, clear, stable, and surprisingly natural.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the result was unexpected clear stable and surprisingly natural",
+ "normalized_hypothesis": "the result was unexpected clear stable and surprisingly natural",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0039",
+ "category": "question",
+ "text": "Who told you that was the right way to do it?",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0039.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Who told you that was the right way to do it?",
+ "raw": {
+ "normalized_reference": "who told you that was the right way to do it?",
+ "normalized_hypothesis": "who told you that was the right way to do it?",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "who told you that was the right way to do it",
+ "normalized_hypothesis": "who told you that was the right way to do it",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0040",
+ "category": "rare_words",
+ "text": "The meticulous cartographer examined an iridescent archipelago through binoculars.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0040.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The meticulous cartographer examined an iridescent archipelago through binoculars.",
+ "raw": {
+ "normalized_reference": "the meticulous cartographer examined an iridescent archipelago through binoculars.",
+ "normalized_hypothesis": "the meticulous cartographer examined an iridescent archipelago through binoculars.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the meticulous cartographer examined an iridescent archipelago through binoculars",
+ "normalized_hypothesis": "the meticulous cartographer examined an iridescent archipelago through binoculars",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0041",
+ "category": "short",
+ "text": "I mean. Yeah. Kind of.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0041.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I mean, yes, kind of.",
+ "raw": {
+ "normalized_reference": "i mean. yeah. kind of.",
+ "normalized_hypothesis": "i mean, yes, kind of.",
+ "errors": 1,
+ "words": 5,
+ "wer": 0.2
+ },
+ "semantic": {
+ "normalized_reference": "i mean yeah kind of",
+ "normalized_hypothesis": "i mean yes kind of",
+ "errors": 1,
+ "words": 5,
+ "wer": 0.2
+ }
+ },
+ {
+ "id": "release500_0042",
+ "category": "technical",
+ "text": "The logs are your best friend when something behaves unexpectedly.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0042.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The logs are your best friend when something behaves unexpectedly.",
+ "raw": {
+ "normalized_reference": "the logs are your best friend when something behaves unexpectedly.",
+ "normalized_hypothesis": "the logs are your best friend when something behaves unexpectedly.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the logs are your best friend when something behaves unexpectedly",
+ "normalized_hypothesis": "the logs are your best friend when something behaves unexpectedly",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0043",
+ "category": "conversational",
+ "text": "He kept talking and I kept nodding and neither of us was listening.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0043.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He kept talking and I kept nodding and neither of us was listening.",
+ "raw": {
+ "normalized_reference": "he kept talking and i kept nodding and neither of us was listening.",
+ "normalized_hypothesis": "he kept talking and i kept nodding and neither of us was listening.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he kept talking and i kept nodding and neither of us was listening",
+ "normalized_hypothesis": "he kept talking and i kept nodding and neither of us was listening",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0044",
+ "category": "descriptive",
+ "text": "It's one of those places where everyone seems to know each other already.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0044.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It's one of those places where everyone seems to know each other already.",
+ "raw": {
+ "normalized_reference": "it's one of those places where everyone seems to know each other already.",
+ "normalized_hypothesis": "it's one of those places where everyone seems to know each other already.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it is one of those places where everyone seems to know each other already",
+ "normalized_hypothesis": "it is one of those places where everyone seems to know each other already",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0045",
+ "category": "emotional",
+ "text": "He said he was proud of me and nobody had ever said that to me before.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0045.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He said he was proud of me and nobody had ever said that to me before.",
+ "raw": {
+ "normalized_reference": "he said he was proud of me and nobody had ever said that to me before.",
+ "normalized_hypothesis": "he said he was proud of me and nobody had ever said that to me before.",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he said he was proud of me and nobody had ever said that to me before",
+ "normalized_hypothesis": "he said he was proud of me and nobody had ever said that to me before",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0046",
+ "category": "homographs",
+ "text": "The minute detail took a minute to explain clearly.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0046.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The minute detail took a minute to explain clearly.",
+ "raw": {
+ "normalized_reference": "the minute detail took a minute to explain clearly.",
+ "normalized_hypothesis": "the minute detail took a minute to explain clearly.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the minute detail took a minute to explain clearly",
+ "normalized_hypothesis": "the minute detail took a minute to explain clearly",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0047",
+ "category": "long",
+ "text": "The strange thing about really wanting something is that getting it never quite feels the way you imagined it would, and not getting it never quite feels as bad.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0047.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The strange thing about really wanting something is that getting it never quite feels the way you imagined it would, and not getting it never quite feels as bad.",
+ "raw": {
+ "normalized_reference": "the strange thing about really wanting something is that getting it never quite feels the way you imagined it would, and not getting it never quite feels as bad.",
+ "normalized_hypothesis": "the strange thing about really wanting something is that getting it never quite feels the way you imagined it would, and not getting it never quite feels as bad.",
+ "errors": 0,
+ "words": 29,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the strange thing about really wanting something is that getting it never quite feels the way you imagined it would and not getting it never quite feels as bad",
+ "normalized_hypothesis": "the strange thing about really wanting something is that getting it never quite feels the way you imagined it would and not getting it never quite feels as bad",
+ "errors": 0,
+ "words": 29,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0048",
+ "category": "long_range",
+ "text": "Even though the recipe looked simple at first, preparing the sauce required patience, careful timing, and a much larger pan.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0048.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Even though the recipe looked simple at first, preparing the sauce required patience, careful timing, and a much larger pan.",
+ "raw": {
+ "normalized_reference": "even though the recipe looked simple at first, preparing the sauce required patience, careful timing, and a much larger pan.",
+ "normalized_hypothesis": "even though the recipe looked simple at first, preparing the sauce required patience, careful timing, and a much larger pan.",
+ "errors": 0,
+ "words": 20,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "even though the recipe looked simple at 1st preparing the sauce required patience careful timing and a much larger pan",
+ "normalized_hypothesis": "even though the recipe looked simple at 1st preparing the sauce required patience careful timing and a much larger pan",
+ "errors": 0,
+ "words": 20,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0049",
+ "category": "names_places",
+ "text": "Xiomara met Siobhan near the old library in Worcestershire.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0049.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Ziamara met Siobhan near the old library in Worcestershire.",
+ "raw": {
+ "normalized_reference": "xiomara met siobhan near the old library in worcestershire.",
+ "normalized_hypothesis": "ziamara met siobhan near the old library in worcestershire.",
+ "errors": 1,
+ "words": 9,
+ "wer": 0.1111111111111111
+ },
+ "semantic": {
+ "normalized_reference": "xiomara met siobhan near the old library in worcestershire",
+ "normalized_hypothesis": "ziamara met siobhan near the old library in worcestershire",
+ "errors": 1,
+ "words": 9,
+ "wer": 0.1111111111111111
+ }
+ },
+ {
+ "id": "release500_0050",
+ "category": "narrative",
+ "text": "She walked into the room and everyone went quiet, not because she was loud, but because she wasn't.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0050.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "She walked into the room and everyone went quiet, not because she was loud, but because she was.",
+ "raw": {
+ "normalized_reference": "she walked into the room and everyone went quiet, not because she was loud, but because she wasn't.",
+ "normalized_hypothesis": "she walked into the room and everyone went quiet, not because she was loud, but because she was.",
+ "errors": 1,
+ "words": 18,
+ "wer": 0.05555555555555555
+ },
+ "semantic": {
+ "normalized_reference": "she walked into the room and everyone went quiet not because she was loud but because she was not",
+ "normalized_hypothesis": "she walked into the room and everyone went quiet not because she was loud but because she was",
+ "errors": 1,
+ "words": 19,
+ "wer": 0.05263157894736842
+ }
+ },
+ {
+ "id": "release500_0051",
+ "category": "numbers",
+ "text": "At exactly seven forty-five, Morgan placed eighteen dollars and fifty cents beside the notebook.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0051.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "At exactly 7.45, Morgan placed $18.50 beside the notebook.",
+ "raw": {
+ "normalized_reference": "at exactly seven forty-five, morgan placed eighteen dollars and fifty cents beside the notebook.",
+ "normalized_hypothesis": "at exactly 7.45, morgan placed $18.50 beside the notebook.",
+ "errors": 8,
+ "words": 15,
+ "wer": 0.5333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "at exactly 745 morgan placed $18.50 beside the notebook",
+ "normalized_hypothesis": "at exactly 7.45 morgan placed $18.50 beside the notebook",
+ "errors": 2,
+ "words": 10,
+ "wer": 0.2
+ }
+ },
+ {
+ "id": "release500_0052",
+ "category": "punctuation",
+ "text": "Does punctuation work? Yes: commas, semicolons, and periods should create sensible boundaries.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0052.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Does punctuation work? Yes, commas, semicolons, and periods should create sensible boundaries.",
+ "raw": {
+ "normalized_reference": "does punctuation work? yes: commas, semicolons, and periods should create sensible boundaries.",
+ "normalized_hypothesis": "does punctuation work? yes, commas, semicolons, and periods should create sensible boundaries.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "does punctuation work yes commas semicolons and periods should create sensible boundaries",
+ "normalized_hypothesis": "does punctuation work yes commas semicolons and periods should create sensible boundaries",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0053",
+ "category": "question",
+ "text": "Have you ever just stopped and thought about how strange all of this is?",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0053.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Have you ever just stopped and thought about how strange all of this is?",
+ "raw": {
+ "normalized_reference": "have you ever just stopped and thought about how strange all of this is?",
+ "normalized_hypothesis": "have you ever just stopped and thought about how strange all of this is?",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "have you ever just stopped and thought about how strange all of this is",
+ "normalized_hypothesis": "have you ever just stopped and thought about how strange all of this is",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0054",
+ "category": "rare_words",
+ "text": "Her idiosyncratic explanation sounded plausible but remained fundamentally ambiguous.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0054.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Her idiosyncratic explanation sounded plausible, but remained fundamentally ambiguous.",
+ "raw": {
+ "normalized_reference": "her idiosyncratic explanation sounded plausible but remained fundamentally ambiguous.",
+ "normalized_hypothesis": "her idiosyncratic explanation sounded plausible, but remained fundamentally ambiguous.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "her idiosyncratic explanation sounded plausible but remained fundamentally ambiguous",
+ "normalized_hypothesis": "her idiosyncratic explanation sounded plausible but remained fundamentally ambiguous",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0055",
+ "category": "short",
+ "text": "That tracks.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0055.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Get Trax.",
+ "raw": {
+ "normalized_reference": "that tracks.",
+ "normalized_hypothesis": "get trax.",
+ "errors": 2,
+ "words": 2,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "that tracks",
+ "normalized_hypothesis": "get trax",
+ "errors": 2,
+ "words": 2,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0056",
+ "category": "technical",
+ "text": "The key thing to remember is that order matters here more than you'd think.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0056.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The key thing to remember is that order matters here more than you'd think.",
+ "raw": {
+ "normalized_reference": "the key thing to remember is that order matters here more than you'd think.",
+ "normalized_hypothesis": "the key thing to remember is that order matters here more than you'd think.",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the key thing to remember is that order matters here more than you would think",
+ "normalized_hypothesis": "the key thing to remember is that order matters here more than you would think",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0057",
+ "category": "conversational",
+ "text": "I would not say that again in this room if I were you.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0057.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I would not say that again in this room if I were you.",
+ "raw": {
+ "normalized_reference": "i would not say that again in this room if i were you.",
+ "normalized_hypothesis": "i would not say that again in this room if i were you.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i would not say that again in this room if i were you",
+ "normalized_hypothesis": "i would not say that again in this room if i were you",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0058",
+ "category": "descriptive",
+ "text": "The room was filled with the quiet that means something is about to change.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0058.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The room was filled with the quiet that means something is about to change.",
+ "raw": {
+ "normalized_reference": "the room was filled with the quiet that means something is about to change.",
+ "normalized_hypothesis": "the room was filled with the quiet that means something is about to change.",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the room was filled with the quiet that means something is about to change",
+ "normalized_hypothesis": "the room was filled with the quiet that means something is about to change",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0059",
+ "category": "emotional",
+ "text": "I didn't expect to feel that way about it, but I did.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0059.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I didn't expect to feel that way about it, but I did.",
+ "raw": {
+ "normalized_reference": "i didn't expect to feel that way about it, but i did.",
+ "normalized_hypothesis": "i didn't expect to feel that way about it, but i did.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i did not expect to feel that way about it but i did",
+ "normalized_hypothesis": "i did not expect to feel that way about it but i did",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0060",
+ "category": "homographs",
+ "text": "I object to the strange object that was left near the entrance.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0060.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I object to the strange object that was left near the entrance.",
+ "raw": {
+ "normalized_reference": "i object to the strange object that was left near the entrance.",
+ "normalized_hypothesis": "i object to the strange object that was left near the entrance.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i object to the strange object that was left near the entrance",
+ "normalized_hypothesis": "i object to the strange object that was left near the entrance",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0061",
+ "category": "long",
+ "text": "I've learned more from the things that went wrong than from the things that worked, which is a frustrating way to have to learn things but apparently the only way that stuck.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0061.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I've learned more from the things that went wrong than from the things that worked, which is a frustrating way to have to learn things, but apparently the only way that stuck.",
+ "raw": {
+ "normalized_reference": "i've learned more from the things that went wrong than from the things that worked, which is a frustrating way to have to learn things but apparently the only way that stuck.",
+ "normalized_hypothesis": "i've learned more from the things that went wrong than from the things that worked, which is a frustrating way to have to learn things, but apparently the only way that stuck.",
+ "errors": 0,
+ "words": 32,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i have learned more from the things that went wrong than from the things that worked which is a frustrating way to have to learn things but apparently the only way that stuck",
+ "normalized_hypothesis": "i have learned more from the things that went wrong than from the things that worked which is a frustrating way to have to learn things but apparently the only way that stuck",
+ "errors": 0,
+ "words": 33,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0062",
+ "category": "long_range",
+ "text": "Because every nearby hotel was full, the exhausted hikers followed the narrow road until they found a farmhouse with one empty room.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0062.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Because every nearby hotel was full, the exhausted hikers followed the narrow road until they found a farmhouse with one empty room.",
+ "raw": {
+ "normalized_reference": "because every nearby hotel was full, the exhausted hikers followed the narrow road until they found a farmhouse with one empty room.",
+ "normalized_hypothesis": "because every nearby hotel was full, the exhausted hikers followed the narrow road until they found a farmhouse with one empty room.",
+ "errors": 0,
+ "words": 22,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "because every nearby hotel was full the exhausted hikers followed the narrow road until they found a farmhouse with one empty room",
+ "normalized_hypothesis": "because every nearby hotel was full the exhausted hikers followed the narrow road until they found a farmhouse with one empty room",
+ "errors": 0,
+ "words": 22,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0063",
+ "category": "names_places",
+ "text": "The itinerary sends Beatrice from Albuquerque to Des Moines and then to Boise.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0063.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The itinerary sends Beatrice from Albuquerque to Des Moines and then to Boise.",
+ "raw": {
+ "normalized_reference": "the itinerary sends beatrice from albuquerque to des moines and then to boise.",
+ "normalized_hypothesis": "the itinerary sends beatrice from albuquerque to des moines and then to boise.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the itinerary sends beatrice from albuquerque to des moines and then to boise",
+ "normalized_hypothesis": "the itinerary sends beatrice from albuquerque to des moines and then to boise",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0064",
+ "category": "narrative",
+ "text": "We had been driving for six hours when we finally saw the sign for the town.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0064.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "We had been driving for six hours when we finally saw the sign for the town.",
+ "raw": {
+ "normalized_reference": "we had been driving for six hours when we finally saw the sign for the town.",
+ "normalized_hypothesis": "we had been driving for six hours when we finally saw the sign for the town.",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "we had been driving for 6 hours when we finally saw the sign for the town",
+ "normalized_hypothesis": "we had been driving for 6 hours when we finally saw the sign for the town",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0065",
+ "category": "numbers",
+ "text": "Call extension zero four nine after eight thirty in the morning.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0065.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Call extension 049 after 8.30 in the morning.",
+ "raw": {
+ "normalized_reference": "call extension zero four nine after eight thirty in the morning.",
+ "normalized_hypothesis": "call extension 049 after 8.30 in the morning.",
+ "errors": 5,
+ "words": 11,
+ "wer": 0.45454545454545453
+ },
+ "semantic": {
+ "normalized_reference": "call extension 049 after 830 in the morning",
+ "normalized_hypothesis": "call extension 49 after 8.30 in the morning",
+ "errors": 3,
+ "words": 8,
+ "wer": 0.375
+ }
+ },
+ {
+ "id": "release500_0066",
+ "category": "punctuation",
+ "text": "Stop - listen carefully - then answer the question without guessing.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0066.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Stop, listen carefully, then answer the question without guessing.",
+ "raw": {
+ "normalized_reference": "stop - listen carefully - then answer the question without guessing.",
+ "normalized_hypothesis": "stop, listen carefully, then answer the question without guessing.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "stop listen carefully then answer the question without guessing",
+ "normalized_hypothesis": "stop listen carefully then answer the question without guessing",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0067",
+ "category": "question",
+ "text": "Does it matter at this point, honestly?",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0067.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Does it matter at this point, honestly?",
+ "raw": {
+ "normalized_reference": "does it matter at this point, honestly?",
+ "normalized_hypothesis": "does it matter at this point, honestly?",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "does it matter at this point honestly",
+ "normalized_hypothesis": "does it matter at this point honestly",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0068",
+ "category": "rare_words",
+ "text": "An entrepreneurial meteorologist described the unprecedented atmospheric phenomenon.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0068.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "An entrepreneurial meteorologist described the unprecedented atmospheric phenomenon.",
+ "raw": {
+ "normalized_reference": "an entrepreneurial meteorologist described the unprecedented atmospheric phenomenon.",
+ "normalized_hypothesis": "an entrepreneurial meteorologist described the unprecedented atmospheric phenomenon.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "an entrepreneurial meteorologist described the unprecedented atmospheric phenomenon",
+ "normalized_hypothesis": "an entrepreneurial meteorologist described the unprecedented atmospheric phenomenon",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0069",
+ "category": "short",
+ "text": "That's not what I expected.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0069.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "That's not what I expected.",
+ "raw": {
+ "normalized_reference": "that's not what i expected.",
+ "normalized_hypothesis": "that's not what i expected.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "that is not what i expected",
+ "normalized_hypothesis": "that is not what i expected",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0070",
+ "category": "technical",
+ "text": "There's a known issue with that version that causes failures under load.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0070.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "There is a known issue with that version that causes failures under load.",
+ "raw": {
+ "normalized_reference": "there's a known issue with that version that causes failures under load.",
+ "normalized_hypothesis": "there is a known issue with that version that causes failures under load.",
+ "errors": 2,
+ "words": 12,
+ "wer": 0.16666666666666666
+ },
+ "semantic": {
+ "normalized_reference": "there is a known issue with that version that causes failures under load",
+ "normalized_hypothesis": "there is a known issue with that version that causes failures under load",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0071",
+ "category": "conversational",
+ "text": "Don't look now but I think that's him over there.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0071.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Don't look now, but I think that's him over there.",
+ "raw": {
+ "normalized_reference": "don't look now but i think that's him over there.",
+ "normalized_hypothesis": "don't look now, but i think that's him over there.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "do not look now but i think that is him over there",
+ "normalized_hypothesis": "do not look now but i think that is him over there",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0072",
+ "category": "descriptive",
+ "text": "The city at night looks completely different from how it feels during the day.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0072.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The city at night looks completely different from how it feels during the day.",
+ "raw": {
+ "normalized_reference": "the city at night looks completely different from how it feels during the day.",
+ "normalized_hypothesis": "the city at night looks completely different from how it feels during the day.",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the city at night looks completely different from how it feels during the day",
+ "normalized_hypothesis": "the city at night looks completely different from how it feels during the day",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0073",
+ "category": "emotional",
+ "text": "It's the kind of happy that makes you nervous because you know it can't last.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0073.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It's the kind of happy that makes you nervous because you know it can't last.",
+ "raw": {
+ "normalized_reference": "it's the kind of happy that makes you nervous because you know it can't last.",
+ "normalized_hypothesis": "it's the kind of happy that makes you nervous because you know it can't last.",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it is the kind of happy that makes you nervous because you know it can not last",
+ "normalized_hypothesis": "it is the kind of happy that makes you nervous because you know it can not last",
+ "errors": 0,
+ "words": 17,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0074",
+ "category": "homographs",
+ "text": "The dove dove beneath the branch while the wind began to wind the loose cable.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0074.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The dove dove beneath the branch while the wind began to wind the loose cable.",
+ "raw": {
+ "normalized_reference": "the dove dove beneath the branch while the wind began to wind the loose cable.",
+ "normalized_hypothesis": "the dove dove beneath the branch while the wind began to wind the loose cable.",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the dove dove beneath the branch while the wind began to wind the loose cable",
+ "normalized_hypothesis": "the dove dove beneath the branch while the wind began to wind the loose cable",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0075",
+ "category": "long",
+ "text": "I've thought a lot about what I would do differently if I could go back, and the honest answer is probably not as much as I'd like to think.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0075.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I thought a lot about what I would do differently if I could go back, and the honest answer is probably not as much as I'd like to think.",
+ "raw": {
+ "normalized_reference": "i've thought a lot about what i would do differently if i could go back, and the honest answer is probably not as much as i'd like to think.",
+ "normalized_hypothesis": "i thought a lot about what i would do differently if i could go back, and the honest answer is probably not as much as i'd like to think.",
+ "errors": 1,
+ "words": 29,
+ "wer": 0.034482758620689655
+ },
+ "semantic": {
+ "normalized_reference": "i have thought a lot about what i would do differently if i could go back and the honest answer is probably not as much as i would like to think",
+ "normalized_hypothesis": "i thought a lot about what i would do differently if i could go back and the honest answer is probably not as much as i would like to think",
+ "errors": 1,
+ "words": 31,
+ "wer": 0.03225806451612903
+ }
+ },
+ {
+ "id": "release500_0076",
+ "category": "long_range",
+ "text": "Although the weather changed without warning, the musicians finished packing before anyone opened the enormous glass door.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0076.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Although the weather changed without warning, the musicians finished packing before anyone opened the enormous glass door.",
+ "raw": {
+ "normalized_reference": "although the weather changed without warning, the musicians finished packing before anyone opened the enormous glass door.",
+ "normalized_hypothesis": "although the weather changed without warning, the musicians finished packing before anyone opened the enormous glass door.",
+ "errors": 0,
+ "words": 17,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "although the weather changed without warning the musicians finished packing before anyone opened the enormous glass door",
+ "normalized_hypothesis": "although the weather changed without warning the musicians finished packing before anyone opened the enormous glass door",
+ "errors": 0,
+ "words": 17,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0077",
+ "category": "names_places",
+ "text": "Saskatchewan looked unusually bright beneath the pale February sunrise.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0077.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Saskatchewan looked unusually bright beneath the pale February sunrise.",
+ "raw": {
+ "normalized_reference": "saskatchewan looked unusually bright beneath the pale february sunrise.",
+ "normalized_hypothesis": "saskatchewan looked unusually bright beneath the pale february sunrise.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "saskatchewan looked unusually bright beneath the pale february sunrise",
+ "normalized_hypothesis": "saskatchewan looked unusually bright beneath the pale february sunrise",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0078",
+ "category": "narrative",
+ "text": "She handled it with a grace I genuinely did not expect.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0078.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "She handled it with a grace I genuinely did not expect.",
+ "raw": {
+ "normalized_reference": "she handled it with a grace i genuinely did not expect.",
+ "normalized_hypothesis": "she handled it with a grace i genuinely did not expect.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "she handled it with a grace i genuinely did not expect",
+ "normalized_hypothesis": "she handled it with a grace i genuinely did not expect",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0079",
+ "category": "numbers",
+ "text": "Apartment four hundred seven is across from room twenty-nine on the third floor.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0079.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Apartment 407 is across from room 29 on the third floor.",
+ "raw": {
+ "normalized_reference": "apartment four hundred seven is across from room twenty-nine on the third floor.",
+ "normalized_hypothesis": "apartment 407 is across from room 29 on the third floor.",
+ "errors": 5,
+ "words": 14,
+ "wer": 0.35714285714285715
+ },
+ "semantic": {
+ "normalized_reference": "apartment 407 is across from room 29 on the 3rd floor",
+ "normalized_hypothesis": "apartment 407 is across from room 29 on the 3rd floor",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0080",
+ "category": "punctuation",
+ "text": "\"Not yet,\" Rowan whispered. \"Give me another minute... please.\"",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0080.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Not yet, Rowan whispered. Give me another minute. Please.",
+ "raw": {
+ "normalized_reference": "\"not yet,\" rowan whispered. \"give me another minute... please.\"",
+ "normalized_hypothesis": "not yet, rowan whispered. give me another minute. please.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "not yet rowan whispered give me another minute please",
+ "normalized_hypothesis": "not yet rowan whispered give me another minute please",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0081",
+ "category": "question",
+ "text": "What are we actually trying to accomplish here?",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0081.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "What are we actually trying to accomplish here?",
+ "raw": {
+ "normalized_reference": "what are we actually trying to accomplish here?",
+ "normalized_hypothesis": "what are we actually trying to accomplish here?",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "what are we actually trying to accomplish here",
+ "normalized_hypothesis": "what are we actually trying to accomplish here",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0082",
+ "category": "rare_words",
+ "text": "The physician recommended a noninvasive procedure for the recurring neurological symptoms.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0082.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The physician recommended a non-invasive procedure for the recurring neurological symptoms.",
+ "raw": {
+ "normalized_reference": "the physician recommended a noninvasive procedure for the recurring neurological symptoms.",
+ "normalized_hypothesis": "the physician recommended a non-invasive procedure for the recurring neurological symptoms.",
+ "errors": 2,
+ "words": 11,
+ "wer": 0.18181818181818182
+ },
+ "semantic": {
+ "normalized_reference": "the physician recommended a noninvasive procedure for the recurring neurological symptoms",
+ "normalized_hypothesis": "the physician recommended a non invasive procedure for the recurring neurological symptoms",
+ "errors": 2,
+ "words": 11,
+ "wer": 0.18181818181818182
+ }
+ },
+ {
+ "id": "release500_0083",
+ "category": "short",
+ "text": "Are you serious right now?",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0083.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Are you serious right now?",
+ "raw": {
+ "normalized_reference": "are you serious right now?",
+ "normalized_hypothesis": "are you serious right now?",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "are you serious right now",
+ "normalized_hypothesis": "are you serious right now",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0084",
+ "category": "technical",
+ "text": "The performance improvement is real, but it comes with tradeoffs.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0084.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The performance improvement is real, but it comes with trade-offs.",
+ "raw": {
+ "normalized_reference": "the performance improvement is real, but it comes with tradeoffs.",
+ "normalized_hypothesis": "the performance improvement is real, but it comes with trade-offs.",
+ "errors": 2,
+ "words": 10,
+ "wer": 0.2
+ },
+ "semantic": {
+ "normalized_reference": "the performance improvement is real but it comes with tradeoffs",
+ "normalized_hypothesis": "the performance improvement is real but it comes with trade offs",
+ "errors": 2,
+ "words": 10,
+ "wer": 0.2
+ }
+ },
+ {
+ "id": "release500_0085",
+ "category": "conversational",
+ "text": "I don't know. It's just been one of those weeks, you know?",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0085.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I don't know. It's just been one of those weeks, you know.",
+ "raw": {
+ "normalized_reference": "i don't know. it's just been one of those weeks, you know?",
+ "normalized_hypothesis": "i don't know. it's just been one of those weeks, you know.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i do not know it is just been one of those weeks you know",
+ "normalized_hypothesis": "i do not know it is just been one of those weeks you know",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0086",
+ "category": "descriptive",
+ "text": "The moment you stop expecting perfection, things get a lot more interesting.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0086.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The moment you stop expecting perfection, things get a lot more interesting.",
+ "raw": {
+ "normalized_reference": "the moment you stop expecting perfection, things get a lot more interesting.",
+ "normalized_hypothesis": "the moment you stop expecting perfection, things get a lot more interesting.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the moment you stop expecting perfection things get a lot more interesting",
+ "normalized_hypothesis": "the moment you stop expecting perfection things get a lot more interesting",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0087",
+ "category": "emotional",
+ "text": "It's the kind of thing that stays with you even years later.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0087.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It's the kind of thing that stays with you even years later.",
+ "raw": {
+ "normalized_reference": "it's the kind of thing that stays with you even years later.",
+ "normalized_hypothesis": "it's the kind of thing that stays with you even years later.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it is the kind of thing that stays with you even years later",
+ "normalized_hypothesis": "it is the kind of thing that stays with you even years later",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0088",
+ "category": "long",
+ "text": "She told me once that the most important skill she'd developed was learning to disagree with someone without making them feel they were wrong.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0088.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "She told me once that the most important skill she'd developed was learning to disagree with someone without making them feel they were wrong.",
+ "raw": {
+ "normalized_reference": "she told me once that the most important skill she'd developed was learning to disagree with someone without making them feel they were wrong.",
+ "normalized_hypothesis": "she told me once that the most important skill she'd developed was learning to disagree with someone without making them feel they were wrong.",
+ "errors": 0,
+ "words": 24,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "she told me once that the most important skill she would developed was learning to disagree with someone without making them feel they were wrong",
+ "normalized_hypothesis": "she told me once that the most important skill she would developed was learning to disagree with someone without making them feel they were wrong",
+ "errors": 0,
+ "words": 25,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0089",
+ "category": "narrative",
+ "text": "He stood at the edge of the cliff and wondered, not for the first time, how he'd gotten here.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0089.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He stood at the edge of the cliff and wondered, not for the first time, how he'd got here.",
+ "raw": {
+ "normalized_reference": "he stood at the edge of the cliff and wondered, not for the first time, how he'd gotten here.",
+ "normalized_hypothesis": "he stood at the edge of the cliff and wondered, not for the first time, how he'd got here.",
+ "errors": 1,
+ "words": 19,
+ "wer": 0.05263157894736842
+ },
+ "semantic": {
+ "normalized_reference": "he stood at the edge of the cliff and wondered not for the 1st time how he would gotten here",
+ "normalized_hypothesis": "he stood at the edge of the cliff and wondered not for the 1st time how he would got here",
+ "errors": 1,
+ "words": 20,
+ "wer": 0.05
+ }
+ },
+ {
+ "id": "release500_0090",
+ "category": "question",
+ "text": "Why does asking for help feel so hard when you actually need it?",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0090.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Why does asking for help feel so hard when you actually need it?",
+ "raw": {
+ "normalized_reference": "why does asking for help feel so hard when you actually need it?",
+ "normalized_hypothesis": "why does asking for help feel so hard when you actually need it?",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "why does asking for help feel so hard when you actually need it",
+ "normalized_hypothesis": "why does asking for help feel so hard when you actually need it",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0091",
+ "category": "short",
+ "text": "That's fair.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0091.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "That's fair.",
+ "raw": {
+ "normalized_reference": "that's fair.",
+ "normalized_hypothesis": "that's fair.",
+ "errors": 0,
+ "words": 2,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "that is fair",
+ "normalized_hypothesis": "that is fair",
+ "errors": 0,
+ "words": 3,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0092",
+ "category": "technical",
+ "text": "If you run into that error, the fix is usually just restarting the service.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0092.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "If you run into that error, the fix is usually just restarting the service.",
+ "raw": {
+ "normalized_reference": "if you run into that error, the fix is usually just restarting the service.",
+ "normalized_hypothesis": "if you run into that error, the fix is usually just restarting the service.",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "if you run into that error the fix is usually just restarting the service",
+ "normalized_hypothesis": "if you run into that error the fix is usually just restarting the service",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0093",
+ "category": "conversational",
+ "text": "It was a disaster in the best possible way.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0093.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It was a disaster in the best possible way.",
+ "raw": {
+ "normalized_reference": "it was a disaster in the best possible way.",
+ "normalized_hypothesis": "it was a disaster in the best possible way.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it was a disaster in the best possible way",
+ "normalized_hypothesis": "it was a disaster in the best possible way",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0094",
+ "category": "descriptive",
+ "text": "It had the look of a place that used to mean something to someone.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0094.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It had the look of a place that used to mean something to someone.",
+ "raw": {
+ "normalized_reference": "it had the look of a place that used to mean something to someone.",
+ "normalized_hypothesis": "it had the look of a place that used to mean something to someone.",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it had the look of a place that used to mean something to someone",
+ "normalized_hypothesis": "it had the look of a place that used to mean something to someone",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0095",
+ "category": "emotional",
+ "text": "There's this feeling you get when something finally works after weeks of failure.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0095.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "There's this feeling you get when something finally works after weeks of failure.",
+ "raw": {
+ "normalized_reference": "there's this feeling you get when something finally works after weeks of failure.",
+ "normalized_hypothesis": "there's this feeling you get when something finally works after weeks of failure.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "there is this feeling you get when something finally works after weeks of failure",
+ "normalized_hypothesis": "there is this feeling you get when something finally works after weeks of failure",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0096",
+ "category": "long",
+ "text": "He had a theory that most arguments aren't actually about what they seem to be about, and the longer I live the more I think he was onto something.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0096.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He had a theory that most arguments aren't actually about what they seem to be about, and the longer I lived the more I think he was onto something.",
+ "raw": {
+ "normalized_reference": "he had a theory that most arguments aren't actually about what they seem to be about, and the longer i live the more i think he was onto something.",
+ "normalized_hypothesis": "he had a theory that most arguments aren't actually about what they seem to be about, and the longer i lived the more i think he was onto something.",
+ "errors": 1,
+ "words": 29,
+ "wer": 0.034482758620689655
+ },
+ "semantic": {
+ "normalized_reference": "he had a theory that most arguments are not actually about what they seem to be about and the longer i live the more i think he was onto something",
+ "normalized_hypothesis": "he had a theory that most arguments are not actually about what they seem to be about and the longer i lived the more i think he was onto something",
+ "errors": 1,
+ "words": 30,
+ "wer": 0.03333333333333333
+ }
+ },
+ {
+ "id": "release500_0097",
+ "category": "narrative",
+ "text": "The thing nobody tells you about starting over is how ordinary most of it feels.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0097.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The thing nobody tells you about starting over is how ordinary most of it feels.",
+ "raw": {
+ "normalized_reference": "the thing nobody tells you about starting over is how ordinary most of it feels.",
+ "normalized_hypothesis": "the thing nobody tells you about starting over is how ordinary most of it feels.",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the thing nobody tells you about starting over is how ordinary most of it feels",
+ "normalized_hypothesis": "the thing nobody tells you about starting over is how ordinary most of it feels",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0098",
+ "category": "question",
+ "text": "What exactly are we waiting for at this point?",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0098.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "What exactly are we waiting for at this point?",
+ "raw": {
+ "normalized_reference": "what exactly are we waiting for at this point?",
+ "normalized_hypothesis": "what exactly are we waiting for at this point?",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "what exactly are we waiting for at this point",
+ "normalized_hypothesis": "what exactly are we waiting for at this point",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0099",
+ "category": "short",
+ "text": "I stand by what I said.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0099.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I stand by what I said.",
+ "raw": {
+ "normalized_reference": "i stand by what i said.",
+ "normalized_hypothesis": "i stand by what i said.",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i stand by what i said",
+ "normalized_hypothesis": "i stand by what i said",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0100",
+ "category": "technical",
+ "text": "Double-check that value before you push anything to production.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0100.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Double check that value before you push anything to production.",
+ "raw": {
+ "normalized_reference": "double-check that value before you push anything to production.",
+ "normalized_hypothesis": "double check that value before you push anything to production.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "double check that value before you push anything to production",
+ "normalized_hypothesis": "double check that value before you push anything to production",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0101",
+ "category": "conversational",
+ "text": "I understand your point; I just do not agree with the conclusion.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0101.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I understand your point. I just do not agree with the conclusion.",
+ "raw": {
+ "normalized_reference": "i understand your point; i just do not agree with the conclusion.",
+ "normalized_hypothesis": "i understand your point. i just do not agree with the conclusion.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i understand your point i just do not agree with the conclusion",
+ "normalized_hypothesis": "i understand your point i just do not agree with the conclusion",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0102",
+ "category": "descriptive",
+ "text": "The sky that day was a strange shade of orange I'd never seen before.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0102.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The sky that day was a strange shade of orange I'd never seen before.",
+ "raw": {
+ "normalized_reference": "the sky that day was a strange shade of orange i'd never seen before.",
+ "normalized_hypothesis": "the sky that day was a strange shade of orange i'd never seen before.",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the sky that day was a strange shade of orange i would never seen before",
+ "normalized_hypothesis": "the sky that day was a strange shade of orange i would never seen before",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0103",
+ "category": "emotional",
+ "text": "It hurt more than I expected, and I thought I was prepared.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0103.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It hurt more than I expected, and I thought I was prepared.",
+ "raw": {
+ "normalized_reference": "it hurt more than i expected, and i thought i was prepared.",
+ "normalized_hypothesis": "it hurt more than i expected, and i thought i was prepared.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it hurt more than i expected and i thought i was prepared",
+ "normalized_hypothesis": "it hurt more than i expected and i thought i was prepared",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0104",
+ "category": "long",
+ "text": "The hardest part wasn't making the decision, it was sitting with the uncertainty of not knowing whether it was the right one for months afterward.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0104.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The hardest part was, making the decision, it was sitting with the uncertainty of not knowing whether it was the right one for months afterward.",
+ "raw": {
+ "normalized_reference": "the hardest part wasn't making the decision, it was sitting with the uncertainty of not knowing whether it was the right one for months afterward.",
+ "normalized_hypothesis": "the hardest part was, making the decision, it was sitting with the uncertainty of not knowing whether it was the right one for months afterward.",
+ "errors": 1,
+ "words": 25,
+ "wer": 0.04
+ },
+ "semantic": {
+ "normalized_reference": "the hardest part was not making the decision it was sitting with the uncertainty of not knowing whether it was the right one for months afterward",
+ "normalized_hypothesis": "the hardest part was making the decision it was sitting with the uncertainty of not knowing whether it was the right one for months afterward",
+ "errors": 1,
+ "words": 26,
+ "wer": 0.038461538461538464
+ }
+ },
+ {
+ "id": "release500_0105",
+ "category": "narrative",
+ "text": "There was this one day last summer when everything just clicked into place.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0105.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "There was this one day last summer when everything just clicked into place.",
+ "raw": {
+ "normalized_reference": "there was this one day last summer when everything just clicked into place.",
+ "normalized_hypothesis": "there was this one day last summer when everything just clicked into place.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "there was this one day last summer when everything just clicked into place",
+ "normalized_hypothesis": "there was this one day last summer when everything just clicked into place",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0106",
+ "category": "question",
+ "text": "Have you ever thought about why we keep doing things the same way when they stop working?",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0106.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Have you ever thought about why we keep doing things the same way when they stop working?",
+ "raw": {
+ "normalized_reference": "have you ever thought about why we keep doing things the same way when they stop working?",
+ "normalized_hypothesis": "have you ever thought about why we keep doing things the same way when they stop working?",
+ "errors": 0,
+ "words": 17,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "have you ever thought about why we keep doing things the same way when they stop working",
+ "normalized_hypothesis": "have you ever thought about why we keep doing things the same way when they stop working",
+ "errors": 0,
+ "words": 17,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0107",
+ "category": "short",
+ "text": "I can't believe you just said that.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0107.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I can't believe you just said that.",
+ "raw": {
+ "normalized_reference": "i can't believe you just said that.",
+ "normalized_hypothesis": "i can't believe you just said that.",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i can not believe you just said that",
+ "normalized_hypothesis": "i can not believe you just said that",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0108",
+ "category": "technical",
+ "text": "You'll want to isolate that variable before drawing any conclusions.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0108.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "You'll want to isolate that variable before drawing any conclusions.",
+ "raw": {
+ "normalized_reference": "you'll want to isolate that variable before drawing any conclusions.",
+ "normalized_hypothesis": "you'll want to isolate that variable before drawing any conclusions.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "you will want to isolate that variable before drawing any conclusions",
+ "normalized_hypothesis": "you will want to isolate that variable before drawing any conclusions",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0109",
+ "category": "conversational",
+ "text": "The meeting that could have been an email lasted two hours.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0109.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The meeting that could have been an email lasted two hours.",
+ "raw": {
+ "normalized_reference": "the meeting that could have been an email lasted two hours.",
+ "normalized_hypothesis": "the meeting that could have been an email lasted two hours.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the meeting that could have been an email lasted 2 hours",
+ "normalized_hypothesis": "the meeting that could have been an email lasted 2 hours",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0110",
+ "category": "descriptive",
+ "text": "There's a specific kind of tired that sleep doesn't fix.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0110.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "There's a specific kind of tired that sleep doesn't fix.",
+ "raw": {
+ "normalized_reference": "there's a specific kind of tired that sleep doesn't fix.",
+ "normalized_hypothesis": "there's a specific kind of tired that sleep doesn't fix.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "there is a specific kind of tired that sleep does not fix",
+ "normalized_hypothesis": "there is a specific kind of tired that sleep does not fix",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0111",
+ "category": "emotional",
+ "text": "Something about that moment broke something open in me that I hadn't realized was closed.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0111.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Something about that moment broke something open in me that I hadn't realized was closed.",
+ "raw": {
+ "normalized_reference": "something about that moment broke something open in me that i hadn't realized was closed.",
+ "normalized_hypothesis": "something about that moment broke something open in me that i hadn't realized was closed.",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "something about that moment broke something open in me that i had not realized was closed",
+ "normalized_hypothesis": "something about that moment broke something open in me that i had not realized was closed",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0112",
+ "category": "long",
+ "text": "The thing about making something good is that you don't usually know it's good while you're making it. You just know you're not done yet.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0112.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The thing about making something good is that you don't usually know it's good while you're making it. You just know you're not done yet.",
+ "raw": {
+ "normalized_reference": "the thing about making something good is that you don't usually know it's good while you're making it. you just know you're not done yet.",
+ "normalized_hypothesis": "the thing about making something good is that you don't usually know it's good while you're making it. you just know you're not done yet.",
+ "errors": 0,
+ "words": 25,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the thing about making something good is that you do not usually know it is good while you are making it you just know you are not done yet",
+ "normalized_hypothesis": "the thing about making something good is that you do not usually know it is good while you are making it you just know you are not done yet",
+ "errors": 0,
+ "words": 29,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0113",
+ "category": "narrative",
+ "text": "She opened the door slowly, as if she expected to find something she wasn't ready for.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0113.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "She opened the door slowly as if she expected to find something she was ready for.",
+ "raw": {
+ "normalized_reference": "she opened the door slowly, as if she expected to find something she wasn't ready for.",
+ "normalized_hypothesis": "she opened the door slowly as if she expected to find something she was ready for.",
+ "errors": 1,
+ "words": 16,
+ "wer": 0.0625
+ },
+ "semantic": {
+ "normalized_reference": "she opened the door slowly as if she expected to find something she was not ready for",
+ "normalized_hypothesis": "she opened the door slowly as if she expected to find something she was ready for",
+ "errors": 1,
+ "words": 17,
+ "wer": 0.058823529411764705
+ }
+ },
+ {
+ "id": "release500_0114",
+ "category": "question",
+ "text": "What would you do differently if you could go back?",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0114.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "What would you do differently if you could go back?",
+ "raw": {
+ "normalized_reference": "what would you do differently if you could go back?",
+ "normalized_hypothesis": "what would you do differently if you could go back?",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "what would you do differently if you could go back",
+ "normalized_hypothesis": "what would you do differently if you could go back",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0115",
+ "category": "short",
+ "text": "Okay. Okay okay okay.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0115.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Okay. Okay, okay, okay.",
+ "raw": {
+ "normalized_reference": "okay. okay okay okay.",
+ "normalized_hypothesis": "okay. okay, okay, okay.",
+ "errors": 0,
+ "words": 4,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "okay okay okay okay",
+ "normalized_hypothesis": "okay okay okay okay",
+ "errors": 0,
+ "words": 4,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0116",
+ "category": "technical",
+ "text": "The latency goes up significantly at scale, which is expected.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0116.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The latency goes up significantly at scale, which is expected.",
+ "raw": {
+ "normalized_reference": "the latency goes up significantly at scale, which is expected.",
+ "normalized_hypothesis": "the latency goes up significantly at scale, which is expected.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the latency goes up significantly at scale which is expected",
+ "normalized_hypothesis": "the latency goes up significantly at scale which is expected",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0117",
+ "category": "conversational",
+ "text": "We ended up staying until two in the morning talking about nothing.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0117.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "We ended up staying until two in the morning talking about nothing.",
+ "raw": {
+ "normalized_reference": "we ended up staying until two in the morning talking about nothing.",
+ "normalized_hypothesis": "we ended up staying until two in the morning talking about nothing.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "we ended up staying until 2 in the morning talking about nothing",
+ "normalized_hypothesis": "we ended up staying until 2 in the morning talking about nothing",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0118",
+ "category": "descriptive",
+ "text": "The rain started slowly, and then all at once, the way it usually does.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0118.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The rain started slowly and then all at once, the way it usually does.",
+ "raw": {
+ "normalized_reference": "the rain started slowly, and then all at once, the way it usually does.",
+ "normalized_hypothesis": "the rain started slowly and then all at once, the way it usually does.",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the rain started slowly and then all at once the way it usually does",
+ "normalized_hypothesis": "the rain started slowly and then all at once the way it usually does",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0119",
+ "category": "emotional",
+ "text": "I honestly did not think I was going to make it through that year.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0119.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I honestly did not think I was going to make it through that year.",
+ "raw": {
+ "normalized_reference": "i honestly did not think i was going to make it through that year.",
+ "normalized_hypothesis": "i honestly did not think i was going to make it through that year.",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i honestly did not think i was going to make it through that year",
+ "normalized_hypothesis": "i honestly did not think i was going to make it through that year",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0120",
+ "category": "long",
+ "text": "There's an assumption that people who seem confident all the time actually feel confident, and in my experience that's almost never true.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0120.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "There's an assumption that people who seem confident all the time actually feel confident, and in my experience that's almost never true.",
+ "raw": {
+ "normalized_reference": "there's an assumption that people who seem confident all the time actually feel confident, and in my experience that's almost never true.",
+ "normalized_hypothesis": "there's an assumption that people who seem confident all the time actually feel confident, and in my experience that's almost never true.",
+ "errors": 0,
+ "words": 22,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "there is an assumption that people who seem confident all the time actually feel confident and in my experience that is almost never true",
+ "normalized_hypothesis": "there is an assumption that people who seem confident all the time actually feel confident and in my experience that is almost never true",
+ "errors": 0,
+ "words": 24,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0121",
+ "category": "narrative",
+ "text": "The moment I opened the door I knew something was different, even if I couldn't say what.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0121.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The moment I opened the door, I knew something was different, even if I couldn't say what.",
+ "raw": {
+ "normalized_reference": "the moment i opened the door i knew something was different, even if i couldn't say what.",
+ "normalized_hypothesis": "the moment i opened the door, i knew something was different, even if i couldn't say what.",
+ "errors": 0,
+ "words": 17,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the moment i opened the door i knew something was different even if i could not say what",
+ "normalized_hypothesis": "the moment i opened the door i knew something was different even if i could not say what",
+ "errors": 0,
+ "words": 18,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0122",
+ "category": "question",
+ "text": "How do you explain something that doesn't make sense even to you?",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0122.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "How do you explain something that doesn't make sense even to you?",
+ "raw": {
+ "normalized_reference": "how do you explain something that doesn't make sense even to you?",
+ "normalized_hypothesis": "how do you explain something that doesn't make sense even to you?",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "how do you explain something that does not make sense even to you",
+ "normalized_hypothesis": "how do you explain something that does not make sense even to you",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0123",
+ "category": "short",
+ "text": "Hold on a second.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0123.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Hold on a second.",
+ "raw": {
+ "normalized_reference": "hold on a second.",
+ "normalized_hypothesis": "hold on a second.",
+ "errors": 0,
+ "words": 4,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "hold on a 2nd",
+ "normalized_hypothesis": "hold on a 2nd",
+ "errors": 0,
+ "words": 4,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0124",
+ "category": "technical",
+ "text": "A deterministic split prevents duplicate normalized text from leaking into evaluation.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0124.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "A deterministic split prevents duplicate normalized text from leaking into evaluation.",
+ "raw": {
+ "normalized_reference": "a deterministic split prevents duplicate normalized text from leaking into evaluation.",
+ "normalized_hypothesis": "a deterministic split prevents duplicate normalized text from leaking into evaluation.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "a deterministic split prevents duplicate normalized text from leaking into evaluation",
+ "normalized_hypothesis": "a deterministic split prevents duplicate normalized text from leaking into evaluation",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0125",
+ "category": "conversational",
+ "text": "Honestly, that was much better than I expected, but the ending still felt rushed.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0125.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Honestly, that was much better than expected, but the ending still felt rushed.",
+ "raw": {
+ "normalized_reference": "honestly, that was much better than i expected, but the ending still felt rushed.",
+ "normalized_hypothesis": "honestly, that was much better than expected, but the ending still felt rushed.",
+ "errors": 1,
+ "words": 14,
+ "wer": 0.07142857142857142
+ },
+ "semantic": {
+ "normalized_reference": "honestly that was much better than i expected but the ending still felt rushed",
+ "normalized_hypothesis": "honestly that was much better than expected but the ending still felt rushed",
+ "errors": 1,
+ "words": 14,
+ "wer": 0.07142857142857142
+ }
+ },
+ {
+ "id": "release500_0126",
+ "category": "descriptive",
+ "text": "She had a habit of tilting her head slightly when thinking hard about something.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0126.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "She had a habit of tilting her head slightly when thinking hard about something.",
+ "raw": {
+ "normalized_reference": "she had a habit of tilting her head slightly when thinking hard about something.",
+ "normalized_hypothesis": "she had a habit of tilting her head slightly when thinking hard about something.",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "she had a habit of tilting her head slightly when thinking hard about something",
+ "normalized_hypothesis": "she had a habit of tilting her head slightly when thinking hard about something",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0127",
+ "category": "emotional",
+ "text": "I cannot stress enough how much I did not want to be right about this.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0127.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I cannot stress enough how much I did not want to be right about this.",
+ "raw": {
+ "normalized_reference": "i cannot stress enough how much i did not want to be right about this.",
+ "normalized_hypothesis": "i cannot stress enough how much i did not want to be right about this.",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i cannot stress enough how much i did not want to be right about this",
+ "normalized_hypothesis": "i cannot stress enough how much i did not want to be right about this",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0128",
+ "category": "long",
+ "text": "I used to think the goal was to get to a point where nothing was hard anymore, and now I think that point doesn't exist and I'm not sure I'd want it to.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0128.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I used to think the goal was to get to a point where nothing was hard anymore, and now I think that point doesn't exist, and I'm not sure I'd want it to.",
+ "raw": {
+ "normalized_reference": "i used to think the goal was to get to a point where nothing was hard anymore, and now i think that point doesn't exist and i'm not sure i'd want it to.",
+ "normalized_hypothesis": "i used to think the goal was to get to a point where nothing was hard anymore, and now i think that point doesn't exist, and i'm not sure i'd want it to.",
+ "errors": 0,
+ "words": 33,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i used to think the goal was to get to a point where nothing was hard anymore and now i think that point does not exist and i am not sure i would want it to",
+ "normalized_hypothesis": "i used to think the goal was to get to a point where nothing was hard anymore and now i think that point does not exist and i am not sure i would want it to",
+ "errors": 0,
+ "words": 36,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0129",
+ "category": "narrative",
+ "text": "So this is the part of the story where everything goes completely sideways.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0129.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "So this is the part of the story where everything goes completely sideways.",
+ "raw": {
+ "normalized_reference": "so this is the part of the story where everything goes completely sideways.",
+ "normalized_hypothesis": "so this is the part of the story where everything goes completely sideways.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "so this is the part of the story where everything goes completely sideways",
+ "normalized_hypothesis": "so this is the part of the story where everything goes completely sideways",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0130",
+ "category": "question",
+ "text": "Is there a version of this where it actually works out okay?",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0130.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Is there a version of this where it actually works out okay?",
+ "raw": {
+ "normalized_reference": "is there a version of this where it actually works out okay?",
+ "normalized_hypothesis": "is there a version of this where it actually works out okay?",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "is there a version of this where it actually works out okay",
+ "normalized_hypothesis": "is there a version of this where it actually works out okay",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0131",
+ "category": "short",
+ "text": "Oh no. Oh no no no.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0131.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Oh, no. Oh, no, no, no.",
+ "raw": {
+ "normalized_reference": "oh no. oh no no no.",
+ "normalized_hypothesis": "oh, no. oh, no, no, no.",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "0 no 0 no no no",
+ "normalized_hypothesis": "0 no 0 no no no",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0132",
+ "category": "technical",
+ "text": "Always test on a copy before modifying the original.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0132.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Always test on a copy before modifying the original.",
+ "raw": {
+ "normalized_reference": "always test on a copy before modifying the original.",
+ "normalized_hypothesis": "always test on a copy before modifying the original.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "always test on a copy before modifying the original",
+ "normalized_hypothesis": "always test on a copy before modifying the original",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0133",
+ "category": "conversational",
+ "text": "I wasn't even supposed to be there. That's the wild part.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0133.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I wasn't even supposed to be there. That's the wild part.",
+ "raw": {
+ "normalized_reference": "i wasn't even supposed to be there. that's the wild part.",
+ "normalized_hypothesis": "i wasn't even supposed to be there. that's the wild part.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i was not even supposed to be there that is the wild part",
+ "normalized_hypothesis": "i was not even supposed to be there that is the wild part",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0134",
+ "category": "descriptive",
+ "text": "It's a small town. Not the kind you leave. The kind you leave and come back to.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0134.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It's a small town. Not the kind you leave. The kind you leave and come back to.",
+ "raw": {
+ "normalized_reference": "it's a small town. not the kind you leave. the kind you leave and come back to.",
+ "normalized_hypothesis": "it's a small town. not the kind you leave. the kind you leave and come back to.",
+ "errors": 0,
+ "words": 17,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it is a small town not the kind you leave the kind you leave and come back to",
+ "normalized_hypothesis": "it is a small town not the kind you leave the kind you leave and come back to",
+ "errors": 0,
+ "words": 18,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0135",
+ "category": "emotional",
+ "text": "She cried. I cried. We all just sort of stood there crying.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0135.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "She cried. I cried. We all just sort of stood there crying.",
+ "raw": {
+ "normalized_reference": "she cried. i cried. we all just sort of stood there crying.",
+ "normalized_hypothesis": "she cried. i cried. we all just sort of stood there crying.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "she cried i cried we all just sort of stood there crying",
+ "normalized_hypothesis": "she cried i cried we all just sort of stood there crying",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0136",
+ "category": "long",
+ "text": "There's a version of being busy that's actually just a very elaborate way of avoiding the one thing you know you should be doing.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0136.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "There's a version of being busy that's actually just a very elaborate way of avoiding the one thing you know you should be doing.",
+ "raw": {
+ "normalized_reference": "there's a version of being busy that's actually just a very elaborate way of avoiding the one thing you know you should be doing.",
+ "normalized_hypothesis": "there's a version of being busy that's actually just a very elaborate way of avoiding the one thing you know you should be doing.",
+ "errors": 0,
+ "words": 24,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "there is a version of being busy that is actually just a very elaborate way of avoiding the one thing you know you should be doing",
+ "normalized_hypothesis": "there is a version of being busy that is actually just a very elaborate way of avoiding the one thing you know you should be doing",
+ "errors": 0,
+ "words": 26,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0137",
+ "category": "narrative",
+ "text": "The letter arrived on a Tuesday, which felt appropriate somehow.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0137.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The letter arrived on a Tuesday which felt appropriate somehow.",
+ "raw": {
+ "normalized_reference": "the letter arrived on a tuesday, which felt appropriate somehow.",
+ "normalized_hypothesis": "the letter arrived on a tuesday which felt appropriate somehow.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the letter arrived on a tuesday which felt appropriate somehow",
+ "normalized_hypothesis": "the letter arrived on a tuesday which felt appropriate somehow",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0138",
+ "category": "question",
+ "text": "What exactly did she say when you told her?",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0138.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "What exactly did she say when you told her?",
+ "raw": {
+ "normalized_reference": "what exactly did she say when you told her?",
+ "normalized_hypothesis": "what exactly did she say when you told her?",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "what exactly did she say when you told her",
+ "normalized_hypothesis": "what exactly did she say when you told her",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0139",
+ "category": "short",
+ "text": "No way. Absolutely not.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0139.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "No way. Absolutely not.",
+ "raw": {
+ "normalized_reference": "no way. absolutely not.",
+ "normalized_hypothesis": "no way. absolutely not.",
+ "errors": 0,
+ "words": 4,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "no way absolutely not",
+ "normalized_hypothesis": "no way absolutely not",
+ "errors": 0,
+ "words": 4,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0140",
+ "category": "technical",
+ "text": "The first thing you want to do is make sure everything is backed up.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0140.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The first thing you want to do is make sure everything is backed up.",
+ "raw": {
+ "normalized_reference": "the first thing you want to do is make sure everything is backed up.",
+ "normalized_hypothesis": "the first thing you want to do is make sure everything is backed up.",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the 1st thing you want to do is make sure everything is backed up",
+ "normalized_hypothesis": "the 1st thing you want to do is make sure everything is backed up",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0141",
+ "category": "conversational",
+ "text": "I have the most insane idea and you need to just hear me out.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0141.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I have the most insane idea and you need to just hear me out.",
+ "raw": {
+ "normalized_reference": "i have the most insane idea and you need to just hear me out.",
+ "normalized_hypothesis": "i have the most insane idea and you need to just hear me out.",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i have the most insane idea and you need to just hear me out",
+ "normalized_hypothesis": "i have the most insane idea and you need to just hear me out",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0142",
+ "category": "descriptive",
+ "text": "There's a coffee shop on the corner that's been there longer than anyone can remember.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0142.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "There's a coffee shop on the corner that's been there longer than anyone can remember.",
+ "raw": {
+ "normalized_reference": "there's a coffee shop on the corner that's been there longer than anyone can remember.",
+ "normalized_hypothesis": "there's a coffee shop on the corner that's been there longer than anyone can remember.",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "there is a coffee shop on the corner that has been there longer than anyone can remember",
+ "normalized_hypothesis": "there is a coffee shop on the corner that has been there longer than anyone can remember",
+ "errors": 0,
+ "words": 17,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0143",
+ "category": "emotional",
+ "text": "I kept it together until I got to the car, and then I completely fell apart.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0143.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I kept it together until I got to the car, and then I completely fell apart.",
+ "raw": {
+ "normalized_reference": "i kept it together until i got to the car, and then i completely fell apart.",
+ "normalized_hypothesis": "i kept it together until i got to the car, and then i completely fell apart.",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i kept it together until i got to the car and then i completely fell apart",
+ "normalized_hypothesis": "i kept it together until i got to the car and then i completely fell apart",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0144",
+ "category": "long",
+ "text": "The older I get, the more I think that most of what we call wisdom is just pattern recognition built from having been wrong enough times to start noticing.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0144.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The older I get, the more I think that most of what we call wisdom is just pattern recognition built from having been wrong enough times to start noticing.",
+ "raw": {
+ "normalized_reference": "the older i get, the more i think that most of what we call wisdom is just pattern recognition built from having been wrong enough times to start noticing.",
+ "normalized_hypothesis": "the older i get, the more i think that most of what we call wisdom is just pattern recognition built from having been wrong enough times to start noticing.",
+ "errors": 0,
+ "words": 29,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the older i get the more i think that most of what we call wisdom is just pattern recognition built from having been wrong enough times to start noticing",
+ "normalized_hypothesis": "the older i get the more i think that most of what we call wisdom is just pattern recognition built from having been wrong enough times to start noticing",
+ "errors": 0,
+ "words": 29,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0145",
+ "category": "narrative",
+ "text": "I remember the exact moment I realized I had made a huge mistake.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0145.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I remember the exact moment I realized I had made a huge mistake.",
+ "raw": {
+ "normalized_reference": "i remember the exact moment i realized i had made a huge mistake.",
+ "normalized_hypothesis": "i remember the exact moment i realized i had made a huge mistake.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i remember the exact moment i realized i had made a huge mistake",
+ "normalized_hypothesis": "i remember the exact moment i realized i had made a huge mistake",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0146",
+ "category": "question",
+ "text": "Do you think it would have gone differently if we'd left earlier?",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0146.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Do you think it would have gone differently if we'd left earlier?",
+ "raw": {
+ "normalized_reference": "do you think it would have gone differently if we'd left earlier?",
+ "normalized_hypothesis": "do you think it would have gone differently if we'd left earlier?",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "do you think it would have gone differently if we would left earlier",
+ "normalized_hypothesis": "do you think it would have gone differently if we would left earlier",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0147",
+ "category": "short",
+ "text": "You have to be kidding me.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0147.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "You have to be kidding me.",
+ "raw": {
+ "normalized_reference": "you have to be kidding me.",
+ "normalized_hypothesis": "you have to be kidding me.",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "you have to be kidding me",
+ "normalized_hypothesis": "you have to be kidding me",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0148",
+ "category": "technical",
+ "text": "If the checksum fails, restart the worker process without deleting the cached tensors.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0148.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "If the checksum fails, restart the worker process without deleting the cache tensors.",
+ "raw": {
+ "normalized_reference": "if the checksum fails, restart the worker process without deleting the cached tensors.",
+ "normalized_hypothesis": "if the checksum fails, restart the worker process without deleting the cache tensors.",
+ "errors": 1,
+ "words": 13,
+ "wer": 0.07692307692307693
+ },
+ "semantic": {
+ "normalized_reference": "if the checksum fails restart the worker process without deleting the cached tensors",
+ "normalized_hypothesis": "if the checksum fails restart the worker process without deleting the cache tensors",
+ "errors": 1,
+ "words": 13,
+ "wer": 0.07692307692307693
+ }
+ },
+ {
+ "id": "release500_0149",
+ "category": "conversational",
+ "text": "I need you to understand how close that was to not working.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0149.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I need you to understand how close that was to not working.",
+ "raw": {
+ "normalized_reference": "i need you to understand how close that was to not working.",
+ "normalized_hypothesis": "i need you to understand how close that was to not working.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i need you to understand how close that was to not working",
+ "normalized_hypothesis": "i need you to understand how close that was to not working",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0150",
+ "category": "descriptive",
+ "text": "It was the sort of morning where the whole world looks rinsed clean.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0150.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It was the sort of morning where the whole world looks rinsed clean.",
+ "raw": {
+ "normalized_reference": "it was the sort of morning where the whole world looks rinsed clean.",
+ "normalized_hypothesis": "it was the sort of morning where the whole world looks rinsed clean.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it was the sort of morning where the whole world looks rinsed clean",
+ "normalized_hypothesis": "it was the sort of morning where the whole world looks rinsed clean",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0151",
+ "category": "emotional",
+ "text": "She started laughing and I started laughing and we couldn't stop for a full minute.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0151.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "She started laughing and I started laughing and we couldn't stop for a full minute.",
+ "raw": {
+ "normalized_reference": "she started laughing and i started laughing and we couldn't stop for a full minute.",
+ "normalized_hypothesis": "she started laughing and i started laughing and we couldn't stop for a full minute.",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "she started laughing and i started laughing and we could not stop for a full minute",
+ "normalized_hypothesis": "she started laughing and i started laughing and we could not stop for a full minute",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0152",
+ "category": "long",
+ "text": "We spent three years building something we were genuinely proud of, and then the market shifted and none of it mattered the way we thought it would.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0152.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "We spent three years building something we were genuinely proud of, and then the market shifted and none of it mattered the way we thought it would.",
+ "raw": {
+ "normalized_reference": "we spent three years building something we were genuinely proud of, and then the market shifted and none of it mattered the way we thought it would.",
+ "normalized_hypothesis": "we spent three years building something we were genuinely proud of, and then the market shifted and none of it mattered the way we thought it would.",
+ "errors": 0,
+ "words": 27,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "we spent 3 years building something we were genuinely proud of and then the market shifted and none of it mattered the way we thought it would",
+ "normalized_hypothesis": "we spent 3 years building something we were genuinely proud of and then the market shifted and none of it mattered the way we thought it would",
+ "errors": 0,
+ "words": 27,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0153",
+ "category": "narrative",
+ "text": "And somehow, despite all of that, it actually turned out okay.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0153.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "And somehow, despite all of that, it actually turned out okay.",
+ "raw": {
+ "normalized_reference": "and somehow, despite all of that, it actually turned out okay.",
+ "normalized_hypothesis": "and somehow, despite all of that, it actually turned out okay.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "and somehow despite all of that it actually turned out okay",
+ "normalized_hypothesis": "and somehow despite all of that it actually turned out okay",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0154",
+ "category": "question",
+ "text": "Is that something you actually want, or something you think you should want?",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0154.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Is that something you actually want or something you think you should want?",
+ "raw": {
+ "normalized_reference": "is that something you actually want, or something you think you should want?",
+ "normalized_hypothesis": "is that something you actually want or something you think you should want?",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "is that something you actually want or something you think you should want",
+ "normalized_hypothesis": "is that something you actually want or something you think you should want",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0155",
+ "category": "short",
+ "text": "This is fine.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0155.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "This is fine.",
+ "raw": {
+ "normalized_reference": "this is fine.",
+ "normalized_hypothesis": "this is fine.",
+ "errors": 0,
+ "words": 3,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "this is fine",
+ "normalized_hypothesis": "this is fine",
+ "errors": 0,
+ "words": 3,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0156",
+ "category": "technical",
+ "text": "The tricky part isn't the setup, it's making sure all dependencies line up.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0156.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The tricky part isn't the setup, it's making sure all dependencies line up.",
+ "raw": {
+ "normalized_reference": "the tricky part isn't the setup, it's making sure all dependencies line up.",
+ "normalized_hypothesis": "the tricky part isn't the setup, it's making sure all dependencies line up.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the tricky part is not the setup it is making sure all dependencies line up",
+ "normalized_hypothesis": "the tricky part is not the setup it is making sure all dependencies line up",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0157",
+ "category": "conversational",
+ "text": "He called it a minor setback. I called it a disaster.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0157.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He called it a minor setback. I called it a disaster.",
+ "raw": {
+ "normalized_reference": "he called it a minor setback. i called it a disaster.",
+ "normalized_hypothesis": "he called it a minor setback. i called it a disaster.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he called it a minor setback i called it a disaster",
+ "normalized_hypothesis": "he called it a minor setback i called it a disaster",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0158",
+ "category": "descriptive",
+ "text": "The whole place smelled like pine and old wood and cold air.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0158.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The whole place smelled like pine and old wood and cold air.",
+ "raw": {
+ "normalized_reference": "the whole place smelled like pine and old wood and cold air.",
+ "normalized_hypothesis": "the whole place smelled like pine and old wood and cold air.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the whole place smelled like pine and old wood and cold air",
+ "normalized_hypothesis": "the whole place smelled like pine and old wood and cold air",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0159",
+ "category": "emotional",
+ "text": "I've never been so relieved in my entire life.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0159.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I've never been so relieved in my entire life.",
+ "raw": {
+ "normalized_reference": "i've never been so relieved in my entire life.",
+ "normalized_hypothesis": "i've never been so relieved in my entire life.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i have never been so relieved in my entire life",
+ "normalized_hypothesis": "i have never been so relieved in my entire life",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0160",
+ "category": "narrative",
+ "text": "I wrote it all down that night so I wouldn't forget, and I still haven't looked at it.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0160.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I wrote it all down that night so I wouldn't forget, and I still haven't looked at it.",
+ "raw": {
+ "normalized_reference": "i wrote it all down that night so i wouldn't forget, and i still haven't looked at it.",
+ "normalized_hypothesis": "i wrote it all down that night so i wouldn't forget, and i still haven't looked at it.",
+ "errors": 0,
+ "words": 18,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i wrote it all down that night so i would not forget and i still have not looked at it",
+ "normalized_hypothesis": "i wrote it all down that night so i would not forget and i still have not looked at it",
+ "errors": 0,
+ "words": 20,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0161",
+ "category": "short",
+ "text": "That's kind of wild actually.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0161.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "That's kind of wild, actually.",
+ "raw": {
+ "normalized_reference": "that's kind of wild actually.",
+ "normalized_hypothesis": "that's kind of wild, actually.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "that is kind of wild actually",
+ "normalized_hypothesis": "that is kind of wild actually",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0162",
+ "category": "technical",
+ "text": "Version pinning will save you a lot of pain later down the line.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0162.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Version pinning will save you a lot of pain later down the line.",
+ "raw": {
+ "normalized_reference": "version pinning will save you a lot of pain later down the line.",
+ "normalized_hypothesis": "version pinning will save you a lot of pain later down the line.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "version pinning will save you a lot of pain later down the line",
+ "normalized_hypothesis": "version pinning will save you a lot of pain later down the line",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0163",
+ "category": "conversational",
+ "text": "At some point you just have to commit and see what happens.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0163.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "At some point you just have to commit and see what happens.",
+ "raw": {
+ "normalized_reference": "at some point you just have to commit and see what happens.",
+ "normalized_hypothesis": "at some point you just have to commit and see what happens.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "at some point you just have to commit and see what happens",
+ "normalized_hypothesis": "at some point you just have to commit and see what happens",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0164",
+ "category": "emotional",
+ "text": "Nobody told me it was going to be this hard, and honestly, I'm grateful.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0164.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Nobody told me it was going to be this hard, and honestly, I'm grateful.",
+ "raw": {
+ "normalized_reference": "nobody told me it was going to be this hard, and honestly, i'm grateful.",
+ "normalized_hypothesis": "nobody told me it was going to be this hard, and honestly, i'm grateful.",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "nobody told me it was going to be this hard and honestly i am grateful",
+ "normalized_hypothesis": "nobody told me it was going to be this hard and honestly i am grateful",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0165",
+ "category": "narrative",
+ "text": "Chapter seven. In which everything that could go wrong did, and a few things that couldn't.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0165.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Chapter 7 In which everything that could go wrong did, and a few things that couldn't.",
+ "raw": {
+ "normalized_reference": "chapter seven. in which everything that could go wrong did, and a few things that couldn't.",
+ "normalized_hypothesis": "chapter 7 in which everything that could go wrong did, and a few things that couldn't.",
+ "errors": 1,
+ "words": 16,
+ "wer": 0.0625
+ },
+ "semantic": {
+ "normalized_reference": "chapter 7 in which everything that could go wrong did and a few things that could not",
+ "normalized_hypothesis": "chapter 7 in which everything that could go wrong did and a few things that could not",
+ "errors": 0,
+ "words": 17,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0166",
+ "category": "short",
+ "text": "Wait, what?",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0166.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Wait what?",
+ "raw": {
+ "normalized_reference": "wait, what?",
+ "normalized_hypothesis": "wait what?",
+ "errors": 0,
+ "words": 2,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "wait what",
+ "normalized_hypothesis": "wait what",
+ "errors": 0,
+ "words": 2,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0167",
+ "category": "technical",
+ "text": "The graphics processor remained idle because the data loader exhausted every central processing core.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0167.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The graphics processor remained idle because the data loader exhausted every central processing core.",
+ "raw": {
+ "normalized_reference": "the graphics processor remained idle because the data loader exhausted every central processing core.",
+ "normalized_hypothesis": "the graphics processor remained idle because the data loader exhausted every central processing core.",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the graphics processor remained idle because the data loader exhausted every central processing core",
+ "normalized_hypothesis": "the graphics processor remained idle because the data loader exhausted every central processing core",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0168",
+ "category": "conversational",
+ "text": "Could you speak a little more slowly? I am trying to write every detail down.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0168.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Could you speak a little more slowly? I am trying to write every detail down.",
+ "raw": {
+ "normalized_reference": "could you speak a little more slowly? i am trying to write every detail down.",
+ "normalized_hypothesis": "could you speak a little more slowly? i am trying to write every detail down.",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "could you speak a little more slowly i am trying to write every detail down",
+ "normalized_hypothesis": "could you speak a little more slowly i am trying to write every detail down",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0169",
+ "category": "emotional",
+ "text": "I was so excited I could barely form a complete sentence.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0169.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I was so excited I could barely form a complete sentence.",
+ "raw": {
+ "normalized_reference": "i was so excited i could barely form a complete sentence.",
+ "normalized_hypothesis": "i was so excited i could barely form a complete sentence.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i was so excited i could barely form a complete sentence",
+ "normalized_hypothesis": "i was so excited i could barely form a complete sentence",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0170",
+ "category": "narrative",
+ "text": "He looked at me like I had said something in a language he didn't speak.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0170.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He looked at me like I had said something in a language he didn't speak.",
+ "raw": {
+ "normalized_reference": "he looked at me like i had said something in a language he didn't speak.",
+ "normalized_hypothesis": "he looked at me like i had said something in a language he didn't speak.",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he looked at me like i had said something in a language he did not speak",
+ "normalized_hypothesis": "he looked at me like i had said something in a language he did not speak",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0171",
+ "category": "short",
+ "text": "That's actually hilarious.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0171.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "That's actually hilarious.",
+ "raw": {
+ "normalized_reference": "that's actually hilarious.",
+ "normalized_hypothesis": "that's actually hilarious.",
+ "errors": 0,
+ "words": 3,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "that is actually hilarious",
+ "normalized_hypothesis": "that is actually hilarious",
+ "errors": 0,
+ "words": 4,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0172",
+ "category": "technical",
+ "text": "The encoder maps phonemes, stress, and phrase boundaries into a shared hidden representation.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0172.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The encoder maps phonemes, stress, and phrase boundaries into a shared hidden representation.",
+ "raw": {
+ "normalized_reference": "the encoder maps phonemes, stress, and phrase boundaries into a shared hidden representation.",
+ "normalized_hypothesis": "the encoder maps phonemes, stress, and phrase boundaries into a shared hidden representation.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the encoder maps phonemes stress and phrase boundaries into a shared hidden representation",
+ "normalized_hypothesis": "the encoder maps phonemes stress and phrase boundaries into a shared hidden representation",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0173",
+ "category": "conversational",
+ "text": "She showed up three hours late and acted like nothing had happened.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0173.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "She showed up three hours late and acted like nothing had happened.",
+ "raw": {
+ "normalized_reference": "she showed up three hours late and acted like nothing had happened.",
+ "normalized_hypothesis": "she showed up three hours late and acted like nothing had happened.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "she showed up 3 hours late and acted like nothing had happened",
+ "normalized_hypothesis": "she showed up 3 hours late and acted like nothing had happened",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0174",
+ "category": "emotional",
+ "text": "Oh my god. This is the best thing I've ever tasted.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0174.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Oh my God, this is the best thing I've ever tasted.",
+ "raw": {
+ "normalized_reference": "oh my god. this is the best thing i've ever tasted.",
+ "normalized_hypothesis": "oh my god, this is the best thing i've ever tasted.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "0 my god this is the best thing i have ever tasted",
+ "normalized_hypothesis": "0 my god this is the best thing i have ever tasted",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0175",
+ "category": "narrative",
+ "text": "The note was short, which was somehow more frightening than a long one would have been.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0175.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The note was short, which was somehow more fright-ing than a long one would have been.",
+ "raw": {
+ "normalized_reference": "the note was short, which was somehow more frightening than a long one would have been.",
+ "normalized_hypothesis": "the note was short, which was somehow more fright-ing than a long one would have been.",
+ "errors": 2,
+ "words": 16,
+ "wer": 0.125
+ },
+ "semantic": {
+ "normalized_reference": "the note was short which was somehow more frightening than a long one would have been",
+ "normalized_hypothesis": "the note was short which was somehow more fright ing than a long one would have been",
+ "errors": 2,
+ "words": 16,
+ "wer": 0.125
+ }
+ },
+ {
+ "id": "release500_0176",
+ "category": "short",
+ "text": "Noted.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0176.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Noted.",
+ "raw": {
+ "normalized_reference": "noted.",
+ "normalized_hypothesis": "noted.",
+ "errors": 0,
+ "words": 1,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "noted",
+ "normalized_hypothesis": "noted",
+ "errors": 0,
+ "words": 1,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0177",
+ "category": "technical",
+ "text": "It's a simple process once you understand the underlying structure.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0177.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It's a simple process once you understand the underlying structure.",
+ "raw": {
+ "normalized_reference": "it's a simple process once you understand the underlying structure.",
+ "normalized_hypothesis": "it's a simple process once you understand the underlying structure.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it is a simple process once you understand the underlying structure",
+ "normalized_hypothesis": "it is a simple process once you understand the underlying structure",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0178",
+ "category": "conversational",
+ "text": "You have to hear this story because it's almost too good to be true.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0178.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "You have to hear this story because it's almost too good to be true.",
+ "raw": {
+ "normalized_reference": "you have to hear this story because it's almost too good to be true.",
+ "normalized_hypothesis": "you have to hear this story because it's almost too good to be true.",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "you have to hear this story because it is almost too good to be true",
+ "normalized_hypothesis": "you have to hear this story because it is almost too good to be true",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0179",
+ "category": "emotional",
+ "text": "I have never felt more seen in my entire life than in that moment.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0179.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I have never felt more seen in my entire life than in that moment.",
+ "raw": {
+ "normalized_reference": "i have never felt more seen in my entire life than in that moment.",
+ "normalized_hypothesis": "i have never felt more seen in my entire life than in that moment.",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i have never felt more seen in my entire life than in that moment",
+ "normalized_hypothesis": "i have never felt more seen in my entire life than in that moment",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0180",
+ "category": "narrative",
+ "text": "She knew. She had always known. She just waited for the right moment.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0180.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "She knew. She had always known. She just waited for the right moment.",
+ "raw": {
+ "normalized_reference": "she knew. she had always known. she just waited for the right moment.",
+ "normalized_hypothesis": "she knew. she had always known. she just waited for the right moment.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "she knew she had always known she just waited for the right moment",
+ "normalized_hypothesis": "she knew she had always known she just waited for the right moment",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0181",
+ "category": "short",
+ "text": "Yeah, no. Hard pass.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0181.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Yes, no. Hard pass.",
+ "raw": {
+ "normalized_reference": "yeah, no. hard pass.",
+ "normalized_hypothesis": "yes, no. hard pass.",
+ "errors": 1,
+ "words": 4,
+ "wer": 0.25
+ },
+ "semantic": {
+ "normalized_reference": "yeah no hard pass",
+ "normalized_hypothesis": "yes no hard pass",
+ "errors": 1,
+ "words": 4,
+ "wer": 0.25
+ }
+ },
+ {
+ "id": "release500_0182",
+ "category": "technical",
+ "text": "Please verify the sample rate, hop length, and mel filter configuration before inference.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0182.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Please verify the sample rate, hop length, and MEL filter configuration before inference.",
+ "raw": {
+ "normalized_reference": "please verify the sample rate, hop length, and mel filter configuration before inference.",
+ "normalized_hypothesis": "please verify the sample rate, hop length, and mel filter configuration before inference.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "please verify the sample rate hop length and mel filter configuration before inference",
+ "normalized_hypothesis": "please verify the sample rate hop length and mel filter configuration before inference",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0183",
+ "category": "conversational",
+ "text": "That's the worst idea I've ever heard and I'm completely in.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0183.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "That's the worst idea I've ever heard and I'm completely in.",
+ "raw": {
+ "normalized_reference": "that's the worst idea i've ever heard and i'm completely in.",
+ "normalized_hypothesis": "that's the worst idea i've ever heard and i'm completely in.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "that is the worst idea i have ever heard and i am completely in",
+ "normalized_hypothesis": "that is the worst idea i have ever heard and i am completely in",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0184",
+ "category": "emotional",
+ "text": "I remember thinking: this is it. This is the thing I've been waiting for.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0184.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I remember thinking, this is it. This is the thing I've been waiting for.",
+ "raw": {
+ "normalized_reference": "i remember thinking: this is it. this is the thing i've been waiting for.",
+ "normalized_hypothesis": "i remember thinking, this is it. this is the thing i've been waiting for.",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i remember thinking this is it this is the thing i have been waiting for",
+ "normalized_hypothesis": "i remember thinking this is it this is the thing i have been waiting for",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0185",
+ "category": "narrative",
+ "text": "We never talked about it again, which I think was the right call.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0185.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "We never talked about it again, which I think was the right call.",
+ "raw": {
+ "normalized_reference": "we never talked about it again, which i think was the right call.",
+ "normalized_hypothesis": "we never talked about it again, which i think was the right call.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "we never talked about it again which i think was the right call",
+ "normalized_hypothesis": "we never talked about it again which i think was the right call",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0186",
+ "category": "short",
+ "text": "In hindsight, obvious.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0186.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "In hindsight, obvious.",
+ "raw": {
+ "normalized_reference": "in hindsight, obvious.",
+ "normalized_hypothesis": "in hindsight, obvious.",
+ "errors": 0,
+ "words": 3,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "in hindsight obvious",
+ "normalized_hypothesis": "in hindsight obvious",
+ "errors": 0,
+ "words": 3,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0187",
+ "category": "technical",
+ "text": "Once it's configured correctly, it basically runs itself.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0187.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Once it's configured correctly, it basically runs itself.",
+ "raw": {
+ "normalized_reference": "once it's configured correctly, it basically runs itself.",
+ "normalized_hypothesis": "once it's configured correctly, it basically runs itself.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "once it is configured correctly it basically runs itself",
+ "normalized_hypothesis": "once it is configured correctly it basically runs itself",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0188",
+ "category": "conversational",
+ "text": "You got engaged? When? How? Tell me everything right now.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0188.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "You got engaged. When? How? Tell me everything right now.",
+ "raw": {
+ "normalized_reference": "you got engaged? when? how? tell me everything right now.",
+ "normalized_hypothesis": "you got engaged. when? how? tell me everything right now.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "you got engaged when how tell me everything right now",
+ "normalized_hypothesis": "you got engaged when how tell me everything right now",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0189",
+ "category": "emotional",
+ "text": "It's hard to explain how much that moment meant to me.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0189.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It's hard to explain how much that moment meant to me.",
+ "raw": {
+ "normalized_reference": "it's hard to explain how much that moment meant to me.",
+ "normalized_hypothesis": "it's hard to explain how much that moment meant to me.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it is hard to explain how much that moment meant to me",
+ "normalized_hypothesis": "it is hard to explain how much that moment meant to me",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0190",
+ "category": "narrative",
+ "text": "He told the story three different times that night and somehow it got better each time.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0190.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He told the story three different times that night and somehow it got better each time.",
+ "raw": {
+ "normalized_reference": "he told the story three different times that night and somehow it got better each time.",
+ "normalized_hypothesis": "he told the story three different times that night and somehow it got better each time.",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he told the story 3 different times that night and somehow it got better each time",
+ "normalized_hypothesis": "he told the story 3 different times that night and somehow it got better each time",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0191",
+ "category": "short",
+ "text": "Right. Moving on.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0191.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Right. Moving on.",
+ "raw": {
+ "normalized_reference": "right. moving on.",
+ "normalized_hypothesis": "right. moving on.",
+ "errors": 0,
+ "words": 3,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "right moving on",
+ "normalized_hypothesis": "right moving on",
+ "errors": 0,
+ "words": 3,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0192",
+ "category": "conversational",
+ "text": "No, I did not move your keys; check beside the small ceramic bowl.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0192.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "No, I did not move your keys. Check beside the small ceramic bowl.",
+ "raw": {
+ "normalized_reference": "no, i did not move your keys; check beside the small ceramic bowl.",
+ "normalized_hypothesis": "no, i did not move your keys. check beside the small ceramic bowl.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "no i did not move your keys check beside the small ceramic bowl",
+ "normalized_hypothesis": "no i did not move your keys check beside the small ceramic bowl",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0193",
+ "category": "emotional",
+ "text": "I was devastated. I'm not going to pretend otherwise.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0193.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I was devastated. I'm not going to pretend otherwise.",
+ "raw": {
+ "normalized_reference": "i was devastated. i'm not going to pretend otherwise.",
+ "normalized_hypothesis": "i was devastated. i'm not going to pretend otherwise.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i was devastated i am not going to pretend otherwise",
+ "normalized_hypothesis": "i was devastated i am not going to pretend otherwise",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0194",
+ "category": "narrative",
+ "text": "That was the last summer before everything changed, though we didn't know it yet.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0194.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "That was the last summer before everything changed, though we didn't know it yet.",
+ "raw": {
+ "normalized_reference": "that was the last summer before everything changed, though we didn't know it yet.",
+ "normalized_hypothesis": "that was the last summer before everything changed, though we didn't know it yet.",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "that was the last summer before everything changed though we did not know it yet",
+ "normalized_hypothesis": "that was the last summer before everything changed though we did not know it yet",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0195",
+ "category": "short",
+ "text": "Bold of you to assume.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0195.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Bold of you to assume.",
+ "raw": {
+ "normalized_reference": "bold of you to assume.",
+ "normalized_hypothesis": "bold of you to assume.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "bold of you to assume",
+ "normalized_hypothesis": "bold of you to assume",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0196",
+ "category": "conversational",
+ "text": "It's one of those decisions that makes total sense in the moment.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0196.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It's one of those decisions that makes total sense in the moment.",
+ "raw": {
+ "normalized_reference": "it's one of those decisions that makes total sense in the moment.",
+ "normalized_hypothesis": "it's one of those decisions that makes total sense in the moment.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it is one of those decisions that makes total sense in the moment",
+ "normalized_hypothesis": "it is one of those decisions that makes total sense in the moment",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0197",
+ "category": "emotional",
+ "text": "I was furious. Not the loud kind. The quiet kind that lasts for days.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0197.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "i was furious not the loud kind the quiet kind that lasts for days",
+ "raw": {
+ "normalized_reference": "i was furious. not the loud kind. the quiet kind that lasts for days.",
+ "normalized_hypothesis": "i was furious not the loud kind the quiet kind that lasts for days",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i was furious not the loud kind the quiet kind that lasts for days",
+ "normalized_hypothesis": "i was furious not the loud kind the quiet kind that lasts for days",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0198",
+ "category": "narrative",
+ "text": "That's the part of the story I usually leave out.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0198.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "That's the part of the story I usually leave out.",
+ "raw": {
+ "normalized_reference": "that's the part of the story i usually leave out.",
+ "normalized_hypothesis": "that's the part of the story i usually leave out.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "that is the part of the story i usually leave out",
+ "normalized_hypothesis": "that is the part of the story i usually leave out",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0199",
+ "category": "short",
+ "text": "I've said too much.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0199.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I've said too much.",
+ "raw": {
+ "normalized_reference": "i've said too much.",
+ "normalized_hypothesis": "i've said too much.",
+ "errors": 0,
+ "words": 4,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i have said too much",
+ "normalized_hypothesis": "i have said too much",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0200",
+ "category": "conversational",
+ "text": "I've been faking that I understood this the entire time.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0200.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I've been faking that I understood this the entire time.",
+ "raw": {
+ "normalized_reference": "i've been faking that i understood this the entire time.",
+ "normalized_hypothesis": "i've been faking that i understood this the entire time.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i have been faking that i understood this the entire time",
+ "normalized_hypothesis": "i have been faking that i understood this the entire time",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0201",
+ "category": "phonetic_balance",
+ "text": "Peace be unto you and grace before the Lord.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0201.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Peace be unto you and grace before the Lord.",
+ "raw": {
+ "normalized_reference": "peace be unto you and grace before the lord.",
+ "normalized_hypothesis": "peace be unto you and grace before the lord.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "peace be unto you and grace before the lord",
+ "normalized_hypothesis": "peace be unto you and grace before the lord",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0202",
+ "category": "phonetic_balance",
+ "text": "From now on we're pals.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0202.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "From now on, we're pals.",
+ "raw": {
+ "normalized_reference": "from now on we're pals.",
+ "normalized_hypothesis": "from now on, we're pals.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "from now on we are pals",
+ "normalized_hypothesis": "from now on we are pals",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0203",
+ "category": "phonetic_balance",
+ "text": "Here we allow our solicitors to look after our legal work.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0203.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Here we allow our solicitors to look after our legal work.",
+ "raw": {
+ "normalized_reference": "here we allow our solicitors to look after our legal work.",
+ "normalized_hypothesis": "here we allow our solicitors to look after our legal work.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "here we allow our solicitors to look after our legal work",
+ "normalized_hypothesis": "here we allow our solicitors to look after our legal work",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0204",
+ "category": "phonetic_balance",
+ "text": "Her words sent a strange chill through Philip.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0204.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Her words sent a strange chill through Philip.",
+ "raw": {
+ "normalized_reference": "her words sent a strange chill through philip.",
+ "normalized_hypothesis": "her words sent a strange chill through philip.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "her words sent a strange chill through philip",
+ "normalized_hypothesis": "her words sent a strange chill through philip",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0205",
+ "category": "phonetic_balance",
+ "text": "It was introduced by Representative Dick of Ohio.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0205.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It was introduced by Representative Dick of Ohio.",
+ "raw": {
+ "normalized_reference": "it was introduced by representative dick of ohio.",
+ "normalized_hypothesis": "it was introduced by representative dick of ohio.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it was introduced by representative dick of ohio",
+ "normalized_hypothesis": "it was introduced by representative dick of ohio",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0206",
+ "category": "phonetic_balance",
+ "text": "It's the aurora borealis.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0206.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It's the Aurora Borealis.",
+ "raw": {
+ "normalized_reference": "it's the aurora borealis.",
+ "normalized_hypothesis": "it's the aurora borealis.",
+ "errors": 0,
+ "words": 4,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it is the aurora borealis",
+ "normalized_hypothesis": "it is the aurora borealis",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0207",
+ "category": "phonetic_balance",
+ "text": "The boy grew and prospered.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0207.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The boy grew and prospered.",
+ "raw": {
+ "normalized_reference": "the boy grew and prospered.",
+ "normalized_hypothesis": "the boy grew and prospered.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the boy grew and prospered",
+ "normalized_hypothesis": "the boy grew and prospered",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0208",
+ "category": "phonetic_balance",
+ "text": "His voice was passionately rebellious.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0208.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "His voice was passionately rebellious.",
+ "raw": {
+ "normalized_reference": "his voice was passionately rebellious.",
+ "normalized_hypothesis": "his voice was passionately rebellious.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "his voice was passionately rebellious",
+ "normalized_hypothesis": "his voice was passionately rebellious",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0209",
+ "category": "phonetic_balance",
+ "text": "That's where they cut off the Scottish Chiefs and killed all hands.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0209.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "That's where they cut off the Scottish chiefs and killed all hands.",
+ "raw": {
+ "normalized_reference": "that's where they cut off the scottish chiefs and killed all hands.",
+ "normalized_hypothesis": "that's where they cut off the scottish chiefs and killed all hands.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "that is where they cut off the scottish chiefs and killed all hands",
+ "normalized_hypothesis": "that is where they cut off the scottish chiefs and killed all hands",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0210",
+ "category": "phonetic_balance",
+ "text": "And Tom King patiently endured.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0210.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "And Tom King patiently endured.",
+ "raw": {
+ "normalized_reference": "and tom king patiently endured.",
+ "normalized_hypothesis": "and tom king patiently endured.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "and tom king patiently endured",
+ "normalized_hypothesis": "and tom king patiently endured",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0211",
+ "category": "phonetic_balance",
+ "text": "From that moment his friendship for Belize turns to hatred and jealousy.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0211.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "From that moment, his friendship for Belize turns to hatred and jealousy.",
+ "raw": {
+ "normalized_reference": "from that moment his friendship for belize turns to hatred and jealousy.",
+ "normalized_hypothesis": "from that moment, his friendship for belize turns to hatred and jealousy.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "from that moment his friendship for belize turns to hatred and jealousy",
+ "normalized_hypothesis": "from that moment his friendship for belize turns to hatred and jealousy",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0212",
+ "category": "phonetic_balance",
+ "text": "Then you don't believe in altruism.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0212.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Then you don't believe in altruism.",
+ "raw": {
+ "normalized_reference": "then you don't believe in altruism.",
+ "normalized_hypothesis": "then you don't believe in altruism.",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "then you do not believe in altruism",
+ "normalized_hypothesis": "then you do not believe in altruism",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0213",
+ "category": "phonetic_balance",
+ "text": "What Pascal glimpsed with the vision of a seer, I have lived.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0213.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "What Pascal glimpsed with the vision of a seer I have lived.",
+ "raw": {
+ "normalized_reference": "what pascal glimpsed with the vision of a seer, i have lived.",
+ "normalized_hypothesis": "what pascal glimpsed with the vision of a seer i have lived.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "what pascal glimpsed with the vision of a seer i have lived",
+ "normalized_hypothesis": "what pascal glimpsed with the vision of a seer i have lived",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0214",
+ "category": "phonetic_balance",
+ "text": "You were making them talk shop, Ruth charged him.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0214.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "You were making them talk shop, Ruth charged him.",
+ "raw": {
+ "normalized_reference": "you were making them talk shop, ruth charged him.",
+ "normalized_hypothesis": "you were making them talk shop, ruth charged him.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "you were making them talk shop ruth charged him",
+ "normalized_hypothesis": "you were making them talk shop ruth charged him",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0215",
+ "category": "phonetic_balance",
+ "text": "Philip snatched at the letter which Gregson held out to him.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0215.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Philip snatched at the letter which Gregson held out to him.",
+ "raw": {
+ "normalized_reference": "philip snatched at the letter which gregson held out to him.",
+ "normalized_hypothesis": "philip snatched at the letter which gregson held out to him.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "philip snatched at the letter which gregson held out to him",
+ "normalized_hypothesis": "philip snatched at the letter which gregson held out to him",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0216",
+ "category": "phonetic_balance",
+ "text": "For a time the exciting thrill of his adventure was gone.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0216.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "For a time, the exciting thrill of his adventure was gone.",
+ "raw": {
+ "normalized_reference": "for a time the exciting thrill of his adventure was gone.",
+ "normalized_hypothesis": "for a time, the exciting thrill of his adventure was gone.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "for a time the exciting thrill of his adventure was gone",
+ "normalized_hypothesis": "for a time the exciting thrill of his adventure was gone",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0217",
+ "category": "phonetic_balance",
+ "text": "Obviously, it was a disease that could be contracted by contact.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0217.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Obviously, it was a disease that could be contracted by contact.",
+ "raw": {
+ "normalized_reference": "obviously, it was a disease that could be contracted by contact.",
+ "normalized_hypothesis": "obviously, it was a disease that could be contracted by contact.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "obviously it was a disease that could be contracted by contact",
+ "normalized_hypothesis": "obviously it was a disease that could be contracted by contact",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0218",
+ "category": "phonetic_balance",
+ "text": "Blind with rage, he darted in.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0218.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Blind with rage he darted in.",
+ "raw": {
+ "normalized_reference": "blind with rage, he darted in.",
+ "normalized_hypothesis": "blind with rage he darted in.",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "blind with rage he darted in",
+ "normalized_hypothesis": "blind with rage he darted in",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0219",
+ "category": "phonetic_balance",
+ "text": "You live on an income which your father earned.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0219.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "You live on an income which your father earned.",
+ "raw": {
+ "normalized_reference": "you live on an income which your father earned.",
+ "normalized_hypothesis": "you live on an income which your father earned.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "you live on an income which your father earned",
+ "normalized_hypothesis": "you live on an income which your father earned",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0220",
+ "category": "phonetic_balance",
+ "text": "Something about it seemed to fascinate him, to challenge his presence.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0220.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Something about it seemed to fascinate him to challenge his presence.",
+ "raw": {
+ "normalized_reference": "something about it seemed to fascinate him, to challenge his presence.",
+ "normalized_hypothesis": "something about it seemed to fascinate him to challenge his presence.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "something about it seemed to fascinate him to challenge his presence",
+ "normalized_hypothesis": "something about it seemed to fascinate him to challenge his presence",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0221",
+ "category": "phonetic_balance",
+ "text": "Her beautiful hair was done up in shining coils.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0221.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Her beautiful hair was done up in shining coils.",
+ "raw": {
+ "normalized_reference": "her beautiful hair was done up in shining coils.",
+ "normalized_hypothesis": "her beautiful hair was done up in shining coils.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "her beautiful hair was done up in shining coils",
+ "normalized_hypothesis": "her beautiful hair was done up in shining coils",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0222",
+ "category": "phonetic_balance",
+ "text": "It does, was her audacious answer.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0222.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It does was her audacious answer.",
+ "raw": {
+ "normalized_reference": "it does, was her audacious answer.",
+ "normalized_hypothesis": "it does was her audacious answer.",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it does was her audacious answer",
+ "normalized_hypothesis": "it does was her audacious answer",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0223",
+ "category": "phonetic_balance",
+ "text": "He caught himself with a jerk.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0223.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He caught himself with a jerk.",
+ "raw": {
+ "normalized_reference": "he caught himself with a jerk.",
+ "normalized_hypothesis": "he caught himself with a jerk.",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he caught himself with a jerk",
+ "normalized_hypothesis": "he caught himself with a jerk",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0224",
+ "category": "phonetic_balance",
+ "text": "He did not believe in the burning of daylight for such a luxury.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0224.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He did not believe in the burning of daylight for such a luxury.",
+ "raw": {
+ "normalized_reference": "he did not believe in the burning of daylight for such a luxury.",
+ "normalized_hypothesis": "he did not believe in the burning of daylight for such a luxury.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he did not believe in the burning of daylight for such a luxury",
+ "normalized_hypothesis": "he did not believe in the burning of daylight for such a luxury",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0225",
+ "category": "phonetic_balance",
+ "text": "He had comparatively no advantages at first.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0225.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He had comparatively no advantages at first.",
+ "raw": {
+ "normalized_reference": "he had comparatively no advantages at first.",
+ "normalized_hypothesis": "he had comparatively no advantages at first.",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he had comparatively no advantages at 1st",
+ "normalized_hypothesis": "he had comparatively no advantages at 1st",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0226",
+ "category": "phonetic_balance",
+ "text": "Such men believe, when they come together.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0226.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Such men believe when they come together.",
+ "raw": {
+ "normalized_reference": "such men believe, when they come together.",
+ "normalized_hypothesis": "such men believe when they come together.",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "such men believe when they come together",
+ "normalized_hypothesis": "such men believe when they come together",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0227",
+ "category": "phonetic_balance",
+ "text": "The ship should be in within a week or ten days.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0227.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The ship should be in within a week or ten days.",
+ "raw": {
+ "normalized_reference": "the ship should be in within a week or ten days.",
+ "normalized_hypothesis": "the ship should be in within a week or ten days.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the ship should be in within a week or 10 days",
+ "normalized_hypothesis": "the ship should be in within a week or 10 days",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0228",
+ "category": "phonetic_balance",
+ "text": "Such a dog the wise driver kills, or turns loose.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0228.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Such a dog the wise driver kills or turns loose.",
+ "raw": {
+ "normalized_reference": "such a dog the wise driver kills, or turns loose.",
+ "normalized_hypothesis": "such a dog the wise driver kills or turns loose.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "such a dog the wise driver kills or turns loose",
+ "normalized_hypothesis": "such a dog the wise driver kills or turns loose",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0229",
+ "category": "phonetic_balance",
+ "text": "You have heard always how he was the lover of the Princess Naomi.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0229.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "You have heard always how he was the lover of the Princess Naomi.",
+ "raw": {
+ "normalized_reference": "you have heard always how he was the lover of the princess naomi.",
+ "normalized_hypothesis": "you have heard always how he was the lover of the princess naomi.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "you have heard always how he was the lover of the princess naomi",
+ "normalized_hypothesis": "you have heard always how he was the lover of the princess naomi",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0230",
+ "category": "phonetic_balance",
+ "text": "I want to know how all this is possible.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0230.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I want to know how all this is possible.",
+ "raw": {
+ "normalized_reference": "i want to know how all this is possible.",
+ "normalized_hypothesis": "i want to know how all this is possible.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i want to know how all this is possible",
+ "normalized_hypothesis": "i want to know how all this is possible",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0231",
+ "category": "phonetic_balance",
+ "text": "It was my idea to a tee.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0231.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It was my idea to a T.",
+ "raw": {
+ "normalized_reference": "it was my idea to a tee.",
+ "normalized_hypothesis": "it was my idea to a t.",
+ "errors": 1,
+ "words": 7,
+ "wer": 0.14285714285714285
+ },
+ "semantic": {
+ "normalized_reference": "it was my idea to a tee",
+ "normalized_hypothesis": "it was my idea to a t",
+ "errors": 1,
+ "words": 7,
+ "wer": 0.14285714285714285
+ }
+ },
+ {
+ "id": "release500_0232",
+ "category": "phonetic_balance",
+ "text": "They likewise are disinclined to being eaten.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0232.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "They likewise are disinclined to being eaten.",
+ "raw": {
+ "normalized_reference": "they likewise are disinclined to being eaten.",
+ "normalized_hypothesis": "they likewise are disinclined to being eaten.",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "they likewise are disinclined to being eaten",
+ "normalized_hypothesis": "they likewise are disinclined to being eaten",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0233",
+ "category": "phonetic_balance",
+ "text": "This state of mind comes of an undue prominence of the ego.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0233.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "This state of mind comes of an undue prominence of the ego.",
+ "raw": {
+ "normalized_reference": "this state of mind comes of an undue prominence of the ego.",
+ "normalized_hypothesis": "this state of mind comes of an undue prominence of the ego.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "this state of mind comes of an undue prominence of the ego",
+ "normalized_hypothesis": "this state of mind comes of an undue prominence of the ego",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0234",
+ "category": "phonetic_balance",
+ "text": "Then it is as I said, Womble announced with finality.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0234.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Then it is as I said, Womble announced with finality.",
+ "raw": {
+ "normalized_reference": "then it is as i said, womble announced with finality.",
+ "normalized_hypothesis": "then it is as i said, womble announced with finality.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "then it is as i said womble announced with finality",
+ "normalized_hypothesis": "then it is as i said womble announced with finality",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0235",
+ "category": "phonetic_balance",
+ "text": "I can see that knife now.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0235.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I can see that knife now.",
+ "raw": {
+ "normalized_reference": "i can see that knife now.",
+ "normalized_hypothesis": "i can see that knife now.",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i can see that knife now",
+ "normalized_hypothesis": "i can see that knife now",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0236",
+ "category": "phonetic_balance",
+ "text": "If I meet her again I shall apologize, said Eileen.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0236.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "If I meet her again, I shall apologize, said Eileen.",
+ "raw": {
+ "normalized_reference": "if i meet her again i shall apologize, said eileen.",
+ "normalized_hypothesis": "if i meet her again, i shall apologize, said eileen.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "if i meet her again i shall apologize said eileen",
+ "normalized_hypothesis": "if i meet her again i shall apologize said eileen",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0237",
+ "category": "phonetic_balance",
+ "text": "A wildly exciting time was his during the week preceding Thursday the eighteenth.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0237.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "A wildly exciting time was his during the week preceding Thursday the 18th.",
+ "raw": {
+ "normalized_reference": "a wildly exciting time was his during the week preceding thursday the eighteenth.",
+ "normalized_hypothesis": "a wildly exciting time was his during the week preceding thursday the 18th.",
+ "errors": 1,
+ "words": 13,
+ "wer": 0.07692307692307693
+ },
+ "semantic": {
+ "normalized_reference": "a wildly exciting time was his during the week preceding thursday the 18th",
+ "normalized_hypothesis": "a wildly exciting time was his during the week preceding thursday the 18th",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0238",
+ "category": "phonetic_balance",
+ "text": "In what bucolic school of fence he had been taught was beyond imagining.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0238.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "In what bucolic school of fence he had been taught was beyond imagining.",
+ "raw": {
+ "normalized_reference": "in what bucolic school of fence he had been taught was beyond imagining.",
+ "normalized_hypothesis": "in what bucolic school of fence he had been taught was beyond imagining.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "in what bucolic school of fence he had been taught was beyond imagining",
+ "normalized_hypothesis": "in what bucolic school of fence he had been taught was beyond imagining",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0239",
+ "category": "phonetic_balance",
+ "text": "So she said, the irate skipper dashed on.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0239.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "So she said, the irate skipper dashed on.",
+ "raw": {
+ "normalized_reference": "so she said, the irate skipper dashed on.",
+ "normalized_hypothesis": "so she said, the irate skipper dashed on.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "so she said the irate skipper dashed on",
+ "normalized_hypothesis": "so she said the irate skipper dashed on",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0240",
+ "category": "phonetic_balance",
+ "text": "It is not an attempt to smash the market.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0240.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It is not an attempt to smash the market.",
+ "raw": {
+ "normalized_reference": "it is not an attempt to smash the market.",
+ "normalized_hypothesis": "it is not an attempt to smash the market.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it is not an attempt to smash the market",
+ "normalized_hypothesis": "it is not an attempt to smash the market",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0241",
+ "category": "phonetic_balance",
+ "text": "He destroyed everything that had belonged to the woman.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0241.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He destroyed everything that had belonged to the woman.",
+ "raw": {
+ "normalized_reference": "he destroyed everything that had belonged to the woman.",
+ "normalized_hypothesis": "he destroyed everything that had belonged to the woman.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he destroyed everything that had belonged to the woman",
+ "normalized_hypothesis": "he destroyed everything that had belonged to the woman",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0242",
+ "category": "phonetic_balance",
+ "text": "Did Thorpe go to see any one in Churchill.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0242.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Did Thorpe go to see anyone in Churchill?",
+ "raw": {
+ "normalized_reference": "did thorpe go to see any one in churchill.",
+ "normalized_hypothesis": "did thorpe go to see anyone in churchill?",
+ "errors": 2,
+ "words": 9,
+ "wer": 0.2222222222222222
+ },
+ "semantic": {
+ "normalized_reference": "did thorpe go to see any one in churchill",
+ "normalized_hypothesis": "did thorpe go to see anyone in churchill",
+ "errors": 2,
+ "words": 9,
+ "wer": 0.2222222222222222
+ }
+ },
+ {
+ "id": "release500_0243",
+ "category": "phonetic_balance",
+ "text": "Many other similar disconcerting omissions will be noticed in the Manuscript.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0243.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Many other similar disconcerting omissions will be noticed in the manuscript.",
+ "raw": {
+ "normalized_reference": "many other similar disconcerting omissions will be noticed in the manuscript.",
+ "normalized_hypothesis": "many other similar disconcerting omissions will be noticed in the manuscript.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "many other similar disconcerting omissions will be noticed in the manuscript",
+ "normalized_hypothesis": "many other similar disconcerting omissions will be noticed in the manuscript",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0244",
+ "category": "phonetic_balance",
+ "text": "The ringing of the big bell aroused him.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0244.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The ringing of the big bell aroused him.",
+ "raw": {
+ "normalized_reference": "the ringing of the big bell aroused him.",
+ "normalized_hypothesis": "the ringing of the big bell aroused him.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the ringing of the big bell aroused him",
+ "normalized_hypothesis": "the ringing of the big bell aroused him",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0245",
+ "category": "phonetic_balance",
+ "text": "A sinewy hand, dripping with water, was clutching the rail.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0245.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "A sinewy hand dripping with water was clutching the rail.",
+ "raw": {
+ "normalized_reference": "a sinewy hand, dripping with water, was clutching the rail.",
+ "normalized_hypothesis": "a sinewy hand dripping with water was clutching the rail.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "a sinewy hand dripping with water was clutching the rail",
+ "normalized_hypothesis": "a sinewy hand dripping with water was clutching the rail",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0246",
+ "category": "phonetic_balance",
+ "text": "Enters now the psychology of the situation.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0246.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Enters now the psychology of the situation.",
+ "raw": {
+ "normalized_reference": "enters now the psychology of the situation.",
+ "normalized_hypothesis": "enters now the psychology of the situation.",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "enters now the psychology of the situation",
+ "normalized_hypothesis": "enters now the psychology of the situation",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0247",
+ "category": "phonetic_balance",
+ "text": "Wash your hands of me.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0247.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Wash your hands of me.",
+ "raw": {
+ "normalized_reference": "wash your hands of me.",
+ "normalized_hypothesis": "wash your hands of me.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "wash your hands of me",
+ "normalized_hypothesis": "wash your hands of me",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0248",
+ "category": "phonetic_balance",
+ "text": "He confessed that the sketch had startled him.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0248.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He confessed that the sketch had startled him.",
+ "raw": {
+ "normalized_reference": "he confessed that the sketch had startled him.",
+ "normalized_hypothesis": "he confessed that the sketch had startled him.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he confessed that the sketch had startled him",
+ "normalized_hypothesis": "he confessed that the sketch had startled him",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0249",
+ "category": "phonetic_balance",
+ "text": "Her face was against his breast.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0249.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Her face was against his breast.",
+ "raw": {
+ "normalized_reference": "her face was against his breast.",
+ "normalized_hypothesis": "her face was against his breast.",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "her face was against his breast",
+ "normalized_hypothesis": "her face was against his breast",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0250",
+ "category": "phonetic_balance",
+ "text": "He was a pariah; a wanderer without a friend or a home.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0250.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He was a pariah, a wanderer without a friend or a home.",
+ "raw": {
+ "normalized_reference": "he was a pariah; a wanderer without a friend or a home.",
+ "normalized_hypothesis": "he was a pariah, a wanderer without a friend or a home.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he was a pariah a wanderer without a friend or a home",
+ "normalized_hypothesis": "he was a pariah a wanderer without a friend or a home",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0251",
+ "category": "phonetic_balance",
+ "text": "A cry of joy burst from Philip's lips.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0251.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "A cry of joy burst from Philip's lips.",
+ "raw": {
+ "normalized_reference": "a cry of joy burst from philip's lips.",
+ "normalized_hypothesis": "a cry of joy burst from philip's lips.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "a cry of joy burst from philip is lips",
+ "normalized_hypothesis": "a cry of joy burst from philip is lips",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0252",
+ "category": "phonetic_balance",
+ "text": "And the big Persian knew of his existence before he did of hers.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0252.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "And the big Persian knew of his existence before he did of hers.",
+ "raw": {
+ "normalized_reference": "and the big persian knew of his existence before he did of hers.",
+ "normalized_hypothesis": "and the big persian knew of his existence before he did of hers.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "and the big persian knew of his existence before he did of hers",
+ "normalized_hypothesis": "and the big persian knew of his existence before he did of hers",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0253",
+ "category": "phonetic_balance",
+ "text": "The men stared into each other's face.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0253.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The men stared into each other's face.",
+ "raw": {
+ "normalized_reference": "the men stared into each other's face.",
+ "normalized_hypothesis": "the men stared into each other's face.",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the men stared into each other is face",
+ "normalized_hypothesis": "the men stared into each other is face",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0254",
+ "category": "phonetic_balance",
+ "text": "It is the fire, partly, she said.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0254.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It is the fire, partly, she said.",
+ "raw": {
+ "normalized_reference": "it is the fire, partly, she said.",
+ "normalized_hypothesis": "it is the fire, partly, she said.",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it is the fire partly she said",
+ "normalized_hypothesis": "it is the fire partly she said",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0255",
+ "category": "phonetic_balance",
+ "text": "It seemed nearer to him since he had seen and talked with Gregson.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0255.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It seemed nearer to him since he had seen and talked with Gregson.",
+ "raw": {
+ "normalized_reference": "it seemed nearer to him since he had seen and talked with gregson.",
+ "normalized_hypothesis": "it seemed nearer to him since he had seen and talked with gregson.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it seemed nearer to him since he had seen and talked with gregson",
+ "normalized_hypothesis": "it seemed nearer to him since he had seen and talked with gregson",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0256",
+ "category": "phonetic_balance",
+ "text": "Among my minor afflictions, I may mention a new and mysterious one.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0256.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Among my minor afflictions I may mention a new and mysterious one.",
+ "raw": {
+ "normalized_reference": "among my minor afflictions, i may mention a new and mysterious one.",
+ "normalized_hypothesis": "among my minor afflictions i may mention a new and mysterious one.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "among my minor afflictions i may mention a new and mysterious one",
+ "normalized_hypothesis": "among my minor afflictions i may mention a new and mysterious one",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0257",
+ "category": "phonetic_balance",
+ "text": "Between him and all domestic animals there must be no hostilities.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0257.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Between him and all domestic animals there must be no hostilities.",
+ "raw": {
+ "normalized_reference": "between him and all domestic animals there must be no hostilities.",
+ "normalized_hypothesis": "between him and all domestic animals there must be no hostilities.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "between him and all domestic animals there must be no hostilities",
+ "normalized_hypothesis": "between him and all domestic animals there must be no hostilities",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0258",
+ "category": "phonetic_balance",
+ "text": "With these arguments he convinced himself that he should go on alone.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0258.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "With these arguments he convinced himself that he should go on alone.",
+ "raw": {
+ "normalized_reference": "with these arguments he convinced himself that he should go on alone.",
+ "normalized_hypothesis": "with these arguments he convinced himself that he should go on alone.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "with these arguments he convinced himself that he should go on alone",
+ "normalized_hypothesis": "with these arguments he convinced himself that he should go on alone",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0259",
+ "category": "phonetic_balance",
+ "text": "We are consumed in our own flesh-pots.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0259.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "We are consumed in our own flesh pots.",
+ "raw": {
+ "normalized_reference": "we are consumed in our own flesh-pots.",
+ "normalized_hypothesis": "we are consumed in our own flesh pots.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "we are consumed in our own flesh pots",
+ "normalized_hypothesis": "we are consumed in our own flesh pots",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0260",
+ "category": "phonetic_balance",
+ "text": "Mercedes screamed, cried, laughed, and manifested the chaotic abandonment of hysteria.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0260.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Mercedes screamed, cried, laughed, and manifested the chaotic abandonment of hysteria.",
+ "raw": {
+ "normalized_reference": "mercedes screamed, cried, laughed, and manifested the chaotic abandonment of hysteria.",
+ "normalized_hypothesis": "mercedes screamed, cried, laughed, and manifested the chaotic abandonment of hysteria.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "mercedes screamed cried laughed and manifested the chaotic abandonment of hysteria",
+ "normalized_hypothesis": "mercedes screamed cried laughed and manifested the chaotic abandonment of hysteria",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0261",
+ "category": "phonetic_balance",
+ "text": "Philip made no effort to follow.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0261.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Philip made no effort to follow.",
+ "raw": {
+ "normalized_reference": "philip made no effort to follow.",
+ "normalized_hypothesis": "philip made no effort to follow.",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "philip made no effort to follow",
+ "normalized_hypothesis": "philip made no effort to follow",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0262",
+ "category": "phonetic_balance",
+ "text": "She said with chattering teeth.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0262.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "She said with chattering teeth.",
+ "raw": {
+ "normalized_reference": "she said with chattering teeth.",
+ "normalized_hypothesis": "she said with chattering teeth.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "she said with chattering teeth",
+ "normalized_hypothesis": "she said with chattering teeth",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0263",
+ "category": "phonetic_balance",
+ "text": "Halfway around the track one donkey got into an argument with its rider.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0263.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Halfway around the track, one donkey got into an argument with its rider.",
+ "raw": {
+ "normalized_reference": "halfway around the track one donkey got into an argument with its rider.",
+ "normalized_hypothesis": "halfway around the track, one donkey got into an argument with its rider.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "halfway around the track one donkey got into an argument with its rider",
+ "normalized_hypothesis": "halfway around the track one donkey got into an argument with its rider",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0264",
+ "category": "phonetic_balance",
+ "text": "Earth and gravel seemed to fill the pan.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0264.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Earth and gravel seemed to fill the pan.",
+ "raw": {
+ "normalized_reference": "earth and gravel seemed to fill the pan.",
+ "normalized_hypothesis": "earth and gravel seemed to fill the pan.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "earth and gravel seemed to fill the pan",
+ "normalized_hypothesis": "earth and gravel seemed to fill the pan",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0265",
+ "category": "phonetic_balance",
+ "text": "Her achievements with cocoanuts were a revelation.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0265.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Her achievements with coconuts were a revelation.",
+ "raw": {
+ "normalized_reference": "her achievements with cocoanuts were a revelation.",
+ "normalized_hypothesis": "her achievements with coconuts were a revelation.",
+ "errors": 1,
+ "words": 7,
+ "wer": 0.14285714285714285
+ },
+ "semantic": {
+ "normalized_reference": "her achievements with cocoanuts were a revelation",
+ "normalized_hypothesis": "her achievements with coconuts were a revelation",
+ "errors": 1,
+ "words": 7,
+ "wer": 0.14285714285714285
+ }
+ },
+ {
+ "id": "release500_0266",
+ "category": "phonetic_balance",
+ "text": "For a full minute he crouched and listened.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0266.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "For a full minute he crouched and listened.",
+ "raw": {
+ "normalized_reference": "for a full minute he crouched and listened.",
+ "normalized_hypothesis": "for a full minute he crouched and listened.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "for a full minute he crouched and listened",
+ "normalized_hypothesis": "for a full minute he crouched and listened",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0267",
+ "category": "phonetic_balance",
+ "text": "It was edged with ice.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0267.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It was edged with ice.",
+ "raw": {
+ "normalized_reference": "it was edged with ice.",
+ "normalized_hypothesis": "it was edged with ice.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it was edged with ice",
+ "normalized_hypothesis": "it was edged with ice",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0268",
+ "category": "phonetic_balance",
+ "text": "The Portuguese boy passed the Hawaiian.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0268.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The Portuguese boy passed the Hawaiian.",
+ "raw": {
+ "normalized_reference": "the portuguese boy passed the hawaiian.",
+ "normalized_hypothesis": "the portuguese boy passed the hawaiian.",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the portuguese boy passed the hawaiian",
+ "normalized_hypothesis": "the portuguese boy passed the hawaiian",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0269",
+ "category": "phonetic_balance",
+ "text": "But he did not broach it, preferring to mature it carefully.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0269.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "But he did not broach it, preferring to mature it carefully.",
+ "raw": {
+ "normalized_reference": "but he did not broach it, preferring to mature it carefully.",
+ "normalized_hypothesis": "but he did not broach it, preferring to mature it carefully.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "but he did not broach it preferring to mature it carefully",
+ "normalized_hypothesis": "but he did not broach it preferring to mature it carefully",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0270",
+ "category": "phonetic_balance",
+ "text": "Come on, Del Mar challenged.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0270.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Come on, Delmar Challenged.",
+ "raw": {
+ "normalized_reference": "come on, del mar challenged.",
+ "normalized_hypothesis": "come on, delmar challenged.",
+ "errors": 2,
+ "words": 5,
+ "wer": 0.4
+ },
+ "semantic": {
+ "normalized_reference": "come on del mar challenged",
+ "normalized_hypothesis": "come on delmar challenged",
+ "errors": 2,
+ "words": 5,
+ "wer": 0.4
+ }
+ },
+ {
+ "id": "release500_0271",
+ "category": "phonetic_balance",
+ "text": "For a full half minute Jeanne looked at him without speaking.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0271.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "For a full half minute, Gene looked at him without speaking.",
+ "raw": {
+ "normalized_reference": "for a full half minute jeanne looked at him without speaking.",
+ "normalized_hypothesis": "for a full half minute, gene looked at him without speaking.",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ },
+ "semantic": {
+ "normalized_reference": "for a full half minute jeanne looked at him without speaking",
+ "normalized_hypothesis": "for a full half minute gene looked at him without speaking",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ }
+ },
+ {
+ "id": "release500_0272",
+ "category": "phonetic_balance",
+ "text": "He bore no grudges and had few enemies.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0272.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He bore no grudges and had few enemies.",
+ "raw": {
+ "normalized_reference": "he bore no grudges and had few enemies.",
+ "normalized_hypothesis": "he bore no grudges and had few enemies.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he bore no grudges and had few enemies",
+ "normalized_hypothesis": "he bore no grudges and had few enemies",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0273",
+ "category": "phonetic_balance",
+ "text": "I'll go over tomorrow afternoon.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0273.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I'll go over tomorrow afternoon.",
+ "raw": {
+ "normalized_reference": "i'll go over tomorrow afternoon.",
+ "normalized_hypothesis": "i'll go over tomorrow afternoon.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i will go over tomorrow afternoon",
+ "normalized_hypothesis": "i will go over tomorrow afternoon",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0274",
+ "category": "phonetic_balance",
+ "text": "Well, there are better men in Hawaii, that's all.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0274.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Well, there are better men in Hawaii, that's all.",
+ "raw": {
+ "normalized_reference": "well, there are better men in hawaii, that's all.",
+ "normalized_hypothesis": "well, there are better men in hawaii, that's all.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "well there are better men in hawaii that is all",
+ "normalized_hypothesis": "well there are better men in hawaii that is all",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0275",
+ "category": "phonetic_balance",
+ "text": "He was an athlete and a giant.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0275.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He was an athlete and a giant.",
+ "raw": {
+ "normalized_reference": "he was an athlete and a giant.",
+ "normalized_hypothesis": "he was an athlete and a giant.",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he was an athlete and a giant",
+ "normalized_hypothesis": "he was an athlete and a giant",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0276",
+ "category": "phonetic_balance",
+ "text": "McVeigh when he returns from a trip to Honolulu.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0276.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "McVeigh when he returns from a trip to Honolulu.",
+ "raw": {
+ "normalized_reference": "mcveigh when he returns from a trip to honolulu.",
+ "normalized_hypothesis": "mcveigh when he returns from a trip to honolulu.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "mcveigh when he returns from a trip to honolulu",
+ "normalized_hypothesis": "mcveigh when he returns from a trip to honolulu",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0277",
+ "category": "phonetic_balance",
+ "text": "You're a devil for fighting, and will surely win.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0277.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "You're a devil for fighting and will surely win.",
+ "raw": {
+ "normalized_reference": "you're a devil for fighting, and will surely win.",
+ "normalized_hypothesis": "you're a devil for fighting and will surely win.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "you are a devil for fighting and will surely win",
+ "normalized_hypothesis": "you are a devil for fighting and will surely win",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0278",
+ "category": "phonetic_balance",
+ "text": "All an appearance can know is mirage.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0278.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "All an appearance can know is mirage.",
+ "raw": {
+ "normalized_reference": "all an appearance can know is mirage.",
+ "normalized_hypothesis": "all an appearance can know is mirage.",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "all an appearance can know is mirage",
+ "normalized_hypothesis": "all an appearance can know is mirage",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0279",
+ "category": "phonetic_balance",
+ "text": "You fired me out of your house, in short.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0279.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "You fired me out of your house in short.",
+ "raw": {
+ "normalized_reference": "you fired me out of your house, in short.",
+ "normalized_hypothesis": "you fired me out of your house in short.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "you fired me out of your house in short",
+ "normalized_hypothesis": "you fired me out of your house in short",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0280",
+ "category": "phonetic_balance",
+ "text": "He made sure that the magazine was loaded, and resumed his paddling.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0280.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He made sure that the magazine was loaded and resumed his paddling.",
+ "raw": {
+ "normalized_reference": "he made sure that the magazine was loaded, and resumed his paddling.",
+ "normalized_hypothesis": "he made sure that the magazine was loaded and resumed his paddling.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he made sure that the magazine was loaded and resumed his paddling",
+ "normalized_hypothesis": "he made sure that the magazine was loaded and resumed his paddling",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0281",
+ "category": "phonetic_balance",
+ "text": "Change chairs, Daylight commanded.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0281.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Change chairs, daylight commanded.",
+ "raw": {
+ "normalized_reference": "change chairs, daylight commanded.",
+ "normalized_hypothesis": "change chairs, daylight commanded.",
+ "errors": 0,
+ "words": 4,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "change chairs daylight commanded",
+ "normalized_hypothesis": "change chairs daylight commanded",
+ "errors": 0,
+ "words": 4,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0282",
+ "category": "phonetic_balance",
+ "text": "By that answer my professional medical prestige stood or fell.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0282.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "By that answer, my professional medical prestige stood or fell.",
+ "raw": {
+ "normalized_reference": "by that answer my professional medical prestige stood or fell.",
+ "normalized_hypothesis": "by that answer, my professional medical prestige stood or fell.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "by that answer my professional medical prestige stood or fell",
+ "normalized_hypothesis": "by that answer my professional medical prestige stood or fell",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0283",
+ "category": "phonetic_balance",
+ "text": "The moon had already begun its westward decline.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0283.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The moon had already begun its westward decline.",
+ "raw": {
+ "normalized_reference": "the moon had already begun its westward decline.",
+ "normalized_hypothesis": "the moon had already begun its westward decline.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the moon had already begun its westward decline",
+ "normalized_hypothesis": "the moon had already begun its westward decline",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0284",
+ "category": "phonetic_balance",
+ "text": "At sea, Monday, March 16, 1908.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0284.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "At sea Monday, March 16, 1908.",
+ "raw": {
+ "normalized_reference": "at sea, monday, march 16, 1908.",
+ "normalized_hypothesis": "at sea monday, march 16, 1908.",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "at sea monday march 16 1908",
+ "normalized_hypothesis": "at sea monday march 16 1908",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0285",
+ "category": "phonetic_balance",
+ "text": "And wherever I ranged, the way lay along alcohol-drenched roads.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0285.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "And wherever I ranged, the way lay along alcohol-drenched roads.",
+ "raw": {
+ "normalized_reference": "and wherever i ranged, the way lay along alcohol-drenched roads.",
+ "normalized_hypothesis": "and wherever i ranged, the way lay along alcohol-drenched roads.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "and wherever i ranged the way lay along alcohol drenched roads",
+ "normalized_hypothesis": "and wherever i ranged the way lay along alcohol drenched roads",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0286",
+ "category": "phonetic_balance",
+ "text": "The date was nearly eighteen years old.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0286.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The date was nearly 18 years old.",
+ "raw": {
+ "normalized_reference": "the date was nearly eighteen years old.",
+ "normalized_hypothesis": "the date was nearly 18 years old.",
+ "errors": 1,
+ "words": 7,
+ "wer": 0.14285714285714285
+ },
+ "semantic": {
+ "normalized_reference": "the date was nearly 18 years old",
+ "normalized_hypothesis": "the date was nearly 18 years old",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0287",
+ "category": "phonetic_balance",
+ "text": "Very early in my life, I separated from my mother.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0287.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Very early in my life, I separated from my mother.",
+ "raw": {
+ "normalized_reference": "very early in my life, i separated from my mother.",
+ "normalized_hypothesis": "very early in my life, i separated from my mother.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "very early in my life i separated from my mother",
+ "normalized_hypothesis": "very early in my life i separated from my mother",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0288",
+ "category": "phonetic_balance",
+ "text": "There was pride and strength, the ring of triumph in his voice.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0288.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "There was pride and strength, the ring of triumph in his voice.",
+ "raw": {
+ "normalized_reference": "there was pride and strength, the ring of triumph in his voice.",
+ "normalized_hypothesis": "there was pride and strength, the ring of triumph in his voice.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "there was pride and strength the ring of triumph in his voice",
+ "normalized_hypothesis": "there was pride and strength the ring of triumph in his voice",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0289",
+ "category": "phonetic_balance",
+ "text": "From my earliest recollection my sleep was a period of terror.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0289.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "From my earliest recollection, my sleep was a period of terror.",
+ "raw": {
+ "normalized_reference": "from my earliest recollection my sleep was a period of terror.",
+ "normalized_hypothesis": "from my earliest recollection, my sleep was a period of terror.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "from my earliest recollection my sleep was a period of terror",
+ "normalized_hypothesis": "from my earliest recollection my sleep was a period of terror",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0290",
+ "category": "phonetic_balance",
+ "text": "We'll have to watch our chances.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0290.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "We'll have to watch our chances.",
+ "raw": {
+ "normalized_reference": "we'll have to watch our chances.",
+ "normalized_hypothesis": "we'll have to watch our chances.",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "we will have to watch our chances",
+ "normalized_hypothesis": "we will have to watch our chances",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0291",
+ "category": "phonetic_balance",
+ "text": "Close beside him gleamed the white fangs of the wolf-dog.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0291.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Close beside him gleamed the white fangs of the wolf-dog.",
+ "raw": {
+ "normalized_reference": "close beside him gleamed the white fangs of the wolf-dog.",
+ "normalized_hypothesis": "close beside him gleamed the white fangs of the wolf-dog.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "close beside him gleamed the white fangs of the wolf dog",
+ "normalized_hypothesis": "close beside him gleamed the white fangs of the wolf dog",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0292",
+ "category": "phonetic_balance",
+ "text": "Would you be satisfied with that one hundredth part of me.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0292.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Would you be satisfied with that one hundredth part of me?",
+ "raw": {
+ "normalized_reference": "would you be satisfied with that one hundredth part of me.",
+ "normalized_hypothesis": "would you be satisfied with that one hundredth part of me?",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "would you be satisfied with that 100th part of me",
+ "normalized_hypothesis": "would you be satisfied with that 100th part of me",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0293",
+ "category": "phonetic_balance",
+ "text": "Two gigantic owls were tearing at the carcass.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0293.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Two gigantic owls were tearing at the carcass.",
+ "raw": {
+ "normalized_reference": "two gigantic owls were tearing at the carcass.",
+ "normalized_hypothesis": "two gigantic owls were tearing at the carcass.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "2 gigantic owls were tearing at the carcass",
+ "normalized_hypothesis": "2 gigantic owls were tearing at the carcass",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0294",
+ "category": "phonetic_balance",
+ "text": "The lush vegetation of that sheltered spot make a natural shield.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0294.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The lush vegetation of that sheltered spot make a natural shield.",
+ "raw": {
+ "normalized_reference": "the lush vegetation of that sheltered spot make a natural shield.",
+ "normalized_hypothesis": "the lush vegetation of that sheltered spot make a natural shield.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the lush vegetation of that sheltered spot make a natural shield",
+ "normalized_hypothesis": "the lush vegetation of that sheltered spot make a natural shield",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0295",
+ "category": "phonetic_balance",
+ "text": "Below him the shadow was broken into a pool of rippling starlight.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0295.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Below him the shadow was broken into a pool of rippling starlight.",
+ "raw": {
+ "normalized_reference": "below him the shadow was broken into a pool of rippling starlight.",
+ "normalized_hypothesis": "below him the shadow was broken into a pool of rippling starlight.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "below him the shadow was broken into a pool of rippling starlight",
+ "normalized_hypothesis": "below him the shadow was broken into a pool of rippling starlight",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0296",
+ "category": "phonetic_balance",
+ "text": "Lots of men take women buggy riding.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0296.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Lots of men take women buggy riding.",
+ "raw": {
+ "normalized_reference": "lots of men take women buggy riding.",
+ "normalized_hypothesis": "lots of men take women buggy riding.",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "lots of men take women buggy riding",
+ "normalized_hypothesis": "lots of men take women buggy riding",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0297",
+ "category": "phonetic_balance",
+ "text": "But it contributed to the smash.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0297.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "But it contributed to the smash.",
+ "raw": {
+ "normalized_reference": "but it contributed to the smash.",
+ "normalized_hypothesis": "but it contributed to the smash.",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "but it contributed to the smash",
+ "normalized_hypothesis": "but it contributed to the smash",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0298",
+ "category": "phonetic_balance",
+ "text": "No man ate of the seal meat or the oil.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0298.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "No man ate of the seal meat or the oil.",
+ "raw": {
+ "normalized_reference": "no man ate of the seal meat or the oil.",
+ "normalized_hypothesis": "no man ate of the seal meat or the oil.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "no man ate of the seal meat or the oil",
+ "normalized_hypothesis": "no man ate of the seal meat or the oil",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0299",
+ "category": "phonetic_balance",
+ "text": "You're going in for grab sharing.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0299.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "You're going in for grab sharing.",
+ "raw": {
+ "normalized_reference": "you're going in for grab sharing.",
+ "normalized_hypothesis": "you're going in for grab sharing.",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "you are going in for grab sharing",
+ "normalized_hypothesis": "you are going in for grab sharing",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0300",
+ "category": "phonetic_balance",
+ "text": "He it was that lived to found the family of the Patino.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0300.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He it was that lived to found the family of the patino.",
+ "raw": {
+ "normalized_reference": "he it was that lived to found the family of the patino.",
+ "normalized_hypothesis": "he it was that lived to found the family of the patino.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he it was that lived to found the family of the patino",
+ "normalized_hypothesis": "he it was that lived to found the family of the patino",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0301",
+ "category": "phonetic_balance",
+ "text": "For such countries nothing remained but reorganization.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0301.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "For such countries, nothing remained but reorganization.",
+ "raw": {
+ "normalized_reference": "for such countries nothing remained but reorganization.",
+ "normalized_hypothesis": "for such countries, nothing remained but reorganization.",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "for such countries nothing remained but reorganization",
+ "normalized_hypothesis": "for such countries nothing remained but reorganization",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0302",
+ "category": "phonetic_balance",
+ "text": "Stand off butcher and baker and all the rest.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0302.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Stand off Butcher and Baker and all the rest.",
+ "raw": {
+ "normalized_reference": "stand off butcher and baker and all the rest.",
+ "normalized_hypothesis": "stand off butcher and baker and all the rest.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "stand off butcher and baker and all the rest",
+ "normalized_hypothesis": "stand off butcher and baker and all the rest",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0303",
+ "category": "phonetic_balance",
+ "text": "All eyes, however, were staring at him in certitude of expectancy.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0303.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "All eyes, however, were staring at him in certitude of expectancy.",
+ "raw": {
+ "normalized_reference": "all eyes, however, were staring at him in certitude of expectancy.",
+ "normalized_hypothesis": "all eyes, however, were staring at him in certitude of expectancy.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "all eyes however were staring at him in certitude of expectancy",
+ "normalized_hypothesis": "all eyes however were staring at him in certitude of expectancy",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0304",
+ "category": "phonetic_balance",
+ "text": "Two days ago Jeanne learned where her father's men were hiding.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0304.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Two days ago, Jean learned where her father's men were hiding.",
+ "raw": {
+ "normalized_reference": "two days ago jeanne learned where her father's men were hiding.",
+ "normalized_hypothesis": "two days ago, jean learned where her father's men were hiding.",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ },
+ "semantic": {
+ "normalized_reference": "2 days ago jeanne learned where her father is men were hiding",
+ "normalized_hypothesis": "2 days ago jean learned where her father is men were hiding",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ }
+ },
+ {
+ "id": "release500_0305",
+ "category": "phonetic_balance",
+ "text": "His previous wives had never lived long enough to bear him children.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0305.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "His previous wives had never lived long enough to bear him children.",
+ "raw": {
+ "normalized_reference": "his previous wives had never lived long enough to bear him children.",
+ "normalized_hypothesis": "his previous wives had never lived long enough to bear him children.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "his previous wives had never lived long enough to bear him children",
+ "normalized_hypothesis": "his previous wives had never lived long enough to bear him children",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0306",
+ "category": "phonetic_balance",
+ "text": "But this little defect did not worry him.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0306.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "But this little defect did not worry him.",
+ "raw": {
+ "normalized_reference": "but this little defect did not worry him.",
+ "normalized_hypothesis": "but this little defect did not worry him.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "but this little defect did not worry him",
+ "normalized_hypothesis": "but this little defect did not worry him",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0307",
+ "category": "phonetic_balance",
+ "text": "Eli Harding asked, as Shunk started to follow.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0307.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Illy Harding asked as Shunk started to follow.",
+ "raw": {
+ "normalized_reference": "eli harding asked, as shunk started to follow.",
+ "normalized_hypothesis": "illy harding asked as shunk started to follow.",
+ "errors": 1,
+ "words": 8,
+ "wer": 0.125
+ },
+ "semantic": {
+ "normalized_reference": "eli harding asked as shunk started to follow",
+ "normalized_hypothesis": "illy harding asked as shunk started to follow",
+ "errors": 1,
+ "words": 8,
+ "wer": 0.125
+ }
+ },
+ {
+ "id": "release500_0308",
+ "category": "phonetic_balance",
+ "text": "And this was their sole conversation throughout the meal.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0308.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "And this was their sole conversation throughout the meal.",
+ "raw": {
+ "normalized_reference": "and this was their sole conversation throughout the meal.",
+ "normalized_hypothesis": "and this was their sole conversation throughout the meal.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "and this was their sole conversation throughout the meal",
+ "normalized_hypothesis": "and this was their sole conversation throughout the meal",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0309",
+ "category": "phonetic_balance",
+ "text": "Then, again, Tudor had such an irritating way about him.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0309.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Then again, Tudor had such an irritating way about him.",
+ "raw": {
+ "normalized_reference": "then, again, tudor had such an irritating way about him.",
+ "normalized_hypothesis": "then again, tudor had such an irritating way about him.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "then again tudor had such an irritating way about him",
+ "normalized_hypothesis": "then again tudor had such an irritating way about him",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0310",
+ "category": "phonetic_balance",
+ "text": "In it there was something that was almost tragedy.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0310.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "In it, there was something that was almost tragedy.",
+ "raw": {
+ "normalized_reference": "in it there was something that was almost tragedy.",
+ "normalized_hypothesis": "in it, there was something that was almost tragedy.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "in it there was something that was almost tragedy",
+ "normalized_hypothesis": "in it there was something that was almost tragedy",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0311",
+ "category": "phonetic_balance",
+ "text": "Thirty pounds, said the captain with finality.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0311.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Thirty pounds, said the captain with finality.",
+ "raw": {
+ "normalized_reference": "thirty pounds, said the captain with finality.",
+ "normalized_hypothesis": "thirty pounds, said the captain with finality.",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "\u00a330 said the captain with finality",
+ "normalized_hypothesis": "\u00a330 said the captain with finality",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0312",
+ "category": "phonetic_balance",
+ "text": "It was over when he made his way through the ring of spectators.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0312.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It was over when he made his way through the ring of spectators.",
+ "raw": {
+ "normalized_reference": "it was over when he made his way through the ring of spectators.",
+ "normalized_hypothesis": "it was over when he made his way through the ring of spectators.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it was over when he made his way through the ring of spectators",
+ "normalized_hypothesis": "it was over when he made his way through the ring of spectators",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0313",
+ "category": "phonetic_balance",
+ "text": "She had died from cold and starvation.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0313.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "She had died from cold and starvation.",
+ "raw": {
+ "normalized_reference": "she had died from cold and starvation.",
+ "normalized_hypothesis": "she had died from cold and starvation.",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "she had died from cold and starvation",
+ "normalized_hypothesis": "she had died from cold and starvation",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0314",
+ "category": "phonetic_balance",
+ "text": "We fished sharks on Niihau together.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0314.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "We fished sharks on Ni Hao together.",
+ "raw": {
+ "normalized_reference": "we fished sharks on niihau together.",
+ "normalized_hypothesis": "we fished sharks on ni hao together.",
+ "errors": 2,
+ "words": 6,
+ "wer": 0.3333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "we fished sharks on niihau together",
+ "normalized_hypothesis": "we fished sharks on ni hao together",
+ "errors": 2,
+ "words": 6,
+ "wer": 0.3333333333333333
+ }
+ },
+ {
+ "id": "release500_0315",
+ "category": "phonetic_balance",
+ "text": "His slim hands gripped the edges of the table.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0315.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "His slim hands gripped the edges of the table.",
+ "raw": {
+ "normalized_reference": "his slim hands gripped the edges of the table.",
+ "normalized_hypothesis": "his slim hands gripped the edges of the table.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "his slim hands gripped the edges of the table",
+ "normalized_hypothesis": "his slim hands gripped the edges of the table",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0316",
+ "category": "phonetic_balance",
+ "text": "We had been chased by them ourselves, more than once.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0316.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "We had been chased by them ourselves more than once.",
+ "raw": {
+ "normalized_reference": "we had been chased by them ourselves, more than once.",
+ "normalized_hypothesis": "we had been chased by them ourselves more than once.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "we had been chased by them ourselves more than once",
+ "normalized_hypothesis": "we had been chased by them ourselves more than once",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0317",
+ "category": "phonetic_balance",
+ "text": "Or have they already devised one.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0317.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Or have they already devised one?",
+ "raw": {
+ "normalized_reference": "or have they already devised one.",
+ "normalized_hypothesis": "or have they already devised one?",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "or have they already devised one",
+ "normalized_hypothesis": "or have they already devised one",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0318",
+ "category": "phonetic_balance",
+ "text": "I cannot follow you, she said.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0318.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I cannot follow you, she said.",
+ "raw": {
+ "normalized_reference": "i cannot follow you, she said.",
+ "normalized_hypothesis": "i cannot follow you, she said.",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i cannot follow you she said",
+ "normalized_hypothesis": "i cannot follow you she said",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0319",
+ "category": "phonetic_balance",
+ "text": "He went down in midstream, searching the shadows of both shores.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0319.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He went down in midstream searching the shadows of both shores.",
+ "raw": {
+ "normalized_reference": "he went down in midstream, searching the shadows of both shores.",
+ "normalized_hypothesis": "he went down in midstream searching the shadows of both shores.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he went down in midstream searching the shadows of both shores",
+ "normalized_hypothesis": "he went down in midstream searching the shadows of both shores",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0320",
+ "category": "phonetic_balance",
+ "text": "I was sick once -- typhoid.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0320.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I was sick once typhoid.",
+ "raw": {
+ "normalized_reference": "i was sick once -- typhoid.",
+ "normalized_hypothesis": "i was sick once typhoid.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i was sick once typhoid",
+ "normalized_hypothesis": "i was sick once typhoid",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0321",
+ "category": "phonetic_balance",
+ "text": "Why not like any railroad station or ferry depot.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0321.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Why not like any railroad station or ferry depot?",
+ "raw": {
+ "normalized_reference": "why not like any railroad station or ferry depot.",
+ "normalized_hypothesis": "why not like any railroad station or ferry depot?",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "why not like any railroad station or ferry depot",
+ "normalized_hypothesis": "why not like any railroad station or ferry depot",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0322",
+ "category": "phonetic_balance",
+ "text": "Red-Eye swung back and forth on the branch farther down.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0322.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Red-Eye swung back and forth on the branch farther down.",
+ "raw": {
+ "normalized_reference": "red-eye swung back and forth on the branch farther down.",
+ "normalized_hypothesis": "red-eye swung back and forth on the branch farther down.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "red eye swung back and forth on the branch farther down",
+ "normalized_hypothesis": "red eye swung back and forth on the branch farther down",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0323",
+ "category": "phonetic_balance",
+ "text": "Within himself he called it no longer his own.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0323.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Within himself he called it no longer his own.",
+ "raw": {
+ "normalized_reference": "within himself he called it no longer his own.",
+ "normalized_hypothesis": "within himself he called it no longer his own.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "within himself he called it no longer his own",
+ "normalized_hypothesis": "within himself he called it no longer his own",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0324",
+ "category": "phonetic_balance",
+ "text": "It lasted as a deterrent for two days.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0324.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It lasted as a deterrent for two days.",
+ "raw": {
+ "normalized_reference": "it lasted as a deterrent for two days.",
+ "normalized_hypothesis": "it lasted as a deterrent for two days.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it lasted as a deterrent for 2 days",
+ "normalized_hypothesis": "it lasted as a deterrent for 2 days",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0325",
+ "category": "phonetic_balance",
+ "text": "He loved to play Chinese lottery.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0325.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He loved to play Chinese lottery.",
+ "raw": {
+ "normalized_reference": "he loved to play chinese lottery.",
+ "normalized_hypothesis": "he loved to play chinese lottery.",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he loved to play chinese lottery",
+ "normalized_hypothesis": "he loved to play chinese lottery",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0326",
+ "category": "phonetic_balance",
+ "text": "Lord, but I'm glad to see you again, Phil.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0326.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Lord, but I'm glad to see you again, Phil.",
+ "raw": {
+ "normalized_reference": "lord, but i'm glad to see you again, phil.",
+ "normalized_hypothesis": "lord, but i'm glad to see you again, phil.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "lord but i am glad to see you again phil",
+ "normalized_hypothesis": "lord but i am glad to see you again phil",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0327",
+ "category": "phonetic_balance",
+ "text": "There is not an iota of truth in it, certainly not.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0327.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "There is not an iota of truth in it, certainly not.",
+ "raw": {
+ "normalized_reference": "there is not an iota of truth in it, certainly not.",
+ "normalized_hypothesis": "there is not an iota of truth in it, certainly not.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "there is not an iota of truth in it certainly not",
+ "normalized_hypothesis": "there is not an iota of truth in it certainly not",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0328",
+ "category": "phonetic_balance",
+ "text": "She added, with genuine sympathy in her face and voice.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0328.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "She added with genuine sympathy in her face and voice.",
+ "raw": {
+ "normalized_reference": "she added, with genuine sympathy in her face and voice.",
+ "normalized_hypothesis": "she added with genuine sympathy in her face and voice.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "she added with genuine sympathy in her face and voice",
+ "normalized_hypothesis": "she added with genuine sympathy in her face and voice",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0329",
+ "category": "phonetic_balance",
+ "text": "We have plenty of capital ourselves, and yet we want more.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0329.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "We have plenty of capital ourselves and yet we want more.",
+ "raw": {
+ "normalized_reference": "we have plenty of capital ourselves, and yet we want more.",
+ "normalized_hypothesis": "we have plenty of capital ourselves and yet we want more.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "we have plenty of capital ourselves and yet we want more",
+ "normalized_hypothesis": "we have plenty of capital ourselves and yet we want more",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0330",
+ "category": "phonetic_balance",
+ "text": "And now put yourself in my place for a moment.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0330.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "And now put yourself in my place for a moment.",
+ "raw": {
+ "normalized_reference": "and now put yourself in my place for a moment.",
+ "normalized_hypothesis": "and now put yourself in my place for a moment.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "and now put yourself in my place for a moment",
+ "normalized_hypothesis": "and now put yourself in my place for a moment",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0331",
+ "category": "phonetic_balance",
+ "text": "Fast, but endure.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0331.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Fast but endure.",
+ "raw": {
+ "normalized_reference": "fast, but endure.",
+ "normalized_hypothesis": "fast but endure.",
+ "errors": 0,
+ "words": 3,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "fast but endure",
+ "normalized_hypothesis": "fast but endure",
+ "errors": 0,
+ "words": 3,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0332",
+ "category": "phonetic_balance",
+ "text": "You were looking squeamish this afternoon, he began.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0332.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "You were looking squeamish this afternoon, he began.",
+ "raw": {
+ "normalized_reference": "you were looking squeamish this afternoon, he began.",
+ "normalized_hypothesis": "you were looking squeamish this afternoon, he began.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "you were looking squeamish this afternoon he began",
+ "normalized_hypothesis": "you were looking squeamish this afternoon he began",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0333",
+ "category": "phonetic_balance",
+ "text": "It is the nearest refuge.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0333.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It is the nearest refuge.",
+ "raw": {
+ "normalized_reference": "it is the nearest refuge.",
+ "normalized_hypothesis": "it is the nearest refuge.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it is the nearest refuge",
+ "normalized_hypothesis": "it is the nearest refuge",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0334",
+ "category": "phonetic_balance",
+ "text": "I want to die in it.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0334.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I want to die in it.",
+ "raw": {
+ "normalized_reference": "i want to die in it.",
+ "normalized_hypothesis": "i want to die in it.",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i want to die in it",
+ "normalized_hypothesis": "i want to die in it",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0335",
+ "category": "phonetic_balance",
+ "text": "He stopped, and Philip nodded at the horrified question in his eyes.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0335.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He stopped, and Philip nodded at the horrified question in his eyes.",
+ "raw": {
+ "normalized_reference": "he stopped, and philip nodded at the horrified question in his eyes.",
+ "normalized_hypothesis": "he stopped, and philip nodded at the horrified question in his eyes.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he stopped and philip nodded at the horrified question in his eyes",
+ "normalized_hypothesis": "he stopped and philip nodded at the horrified question in his eyes",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0336",
+ "category": "phonetic_balance",
+ "text": "She turned, fearing that Jacques might see what was in her face.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0336.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "She turned, fearing that Jack might see what was in her face.",
+ "raw": {
+ "normalized_reference": "she turned, fearing that jacques might see what was in her face.",
+ "normalized_hypothesis": "she turned, fearing that jack might see what was in her face.",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "she turned fearing that jacques might see what was in her face",
+ "normalized_hypothesis": "she turned fearing that jack might see what was in her face",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ }
+ },
+ {
+ "id": "release500_0337",
+ "category": "phonetic_balance",
+ "text": "It was a curious coincidence.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0337.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It was a curious coincidence.",
+ "raw": {
+ "normalized_reference": "it was a curious coincidence.",
+ "normalized_hypothesis": "it was a curious coincidence.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it was a curious coincidence",
+ "normalized_hypothesis": "it was a curious coincidence",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0338",
+ "category": "phonetic_balance",
+ "text": "Daylight made no answer, and the door closed behind him.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0338.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Daylight made no answer and the door closed behind him.",
+ "raw": {
+ "normalized_reference": "daylight made no answer, and the door closed behind him.",
+ "normalized_hypothesis": "daylight made no answer and the door closed behind him.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "daylight made no answer and the door closed behind him",
+ "normalized_hypothesis": "daylight made no answer and the door closed behind him",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0339",
+ "category": "phonetic_balance",
+ "text": "And he did hurt my arm.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0339.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "And he did hurt my arm.",
+ "raw": {
+ "normalized_reference": "and he did hurt my arm.",
+ "normalized_hypothesis": "and he did hurt my arm.",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "and he did hurt my arm",
+ "normalized_hypothesis": "and he did hurt my arm",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0340",
+ "category": "phonetic_balance",
+ "text": "Dunham, can your boy go along with Jesse.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0340.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Dunham, can your boy go along with Jesse?",
+ "raw": {
+ "normalized_reference": "dunham, can your boy go along with jesse.",
+ "normalized_hypothesis": "dunham, can your boy go along with jesse?",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "dunham can your boy go along with jesse",
+ "normalized_hypothesis": "dunham can your boy go along with jesse",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0341",
+ "category": "phonetic_balance",
+ "text": "The creative joy, I murmured.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0341.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The creative joy I murmured.",
+ "raw": {
+ "normalized_reference": "the creative joy, i murmured.",
+ "normalized_hypothesis": "the creative joy i murmured.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the creative joy i murmured",
+ "normalized_hypothesis": "the creative joy i murmured",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0342",
+ "category": "phonetic_balance",
+ "text": "He pulled, and the log crashed down to break his back.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0342.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He pulled and the log crashed down to break his back.",
+ "raw": {
+ "normalized_reference": "he pulled, and the log crashed down to break his back.",
+ "normalized_hypothesis": "he pulled and the log crashed down to break his back.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he pulled and the log crashed down to break his back",
+ "normalized_hypothesis": "he pulled and the log crashed down to break his back",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0343",
+ "category": "phonetic_balance",
+ "text": "By golly, the boy wins.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0343.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "By golly, the boy wins.",
+ "raw": {
+ "normalized_reference": "by golly, the boy wins.",
+ "normalized_hypothesis": "by golly, the boy wins.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "by golly the boy wins",
+ "normalized_hypothesis": "by golly the boy wins",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0344",
+ "category": "phonetic_balance",
+ "text": "The boy threw back his head with pride.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0344.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The boy threw back his head with pride.",
+ "raw": {
+ "normalized_reference": "the boy threw back his head with pride.",
+ "normalized_hypothesis": "the boy threw back his head with pride.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the boy threw back his head with pride",
+ "normalized_hypothesis": "the boy threw back his head with pride",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0345",
+ "category": "phonetic_balance",
+ "text": "The last one I knew was an overseer.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0345.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The last one I knew was an overseer.",
+ "raw": {
+ "normalized_reference": "the last one i knew was an overseer.",
+ "normalized_hypothesis": "the last one i knew was an overseer.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the last one i knew was an overseer",
+ "normalized_hypothesis": "the last one i knew was an overseer",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0346",
+ "category": "phonetic_balance",
+ "text": "Two years ago I gave up civilization for this.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0346.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Two years ago, I gave up civilization for this.",
+ "raw": {
+ "normalized_reference": "two years ago i gave up civilization for this.",
+ "normalized_hypothesis": "two years ago, i gave up civilization for this.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "2 years ago i gave up civilization for this",
+ "normalized_hypothesis": "2 years ago i gave up civilization for this",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0347",
+ "category": "phonetic_balance",
+ "text": "The river bared its bosom, and snorting steamboats challenged the wilderness.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0347.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The river bared its bosom, and snorting steamboats challenged the wilderness.",
+ "raw": {
+ "normalized_reference": "the river bared its bosom, and snorting steamboats challenged the wilderness.",
+ "normalized_hypothesis": "the river bared its bosom, and snorting steamboats challenged the wilderness.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the river bared its bosom and snorting steamboats challenged the wilderness",
+ "normalized_hypothesis": "the river bared its bosom and snorting steamboats challenged the wilderness",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0348",
+ "category": "phonetic_balance",
+ "text": "Mab, she said.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0348.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Mab, she said.",
+ "raw": {
+ "normalized_reference": "mab, she said.",
+ "normalized_hypothesis": "mab, she said.",
+ "errors": 0,
+ "words": 3,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "mab she said",
+ "normalized_hypothesis": "mab she said",
+ "errors": 0,
+ "words": 3,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0349",
+ "category": "phonetic_balance",
+ "text": "The regret in his voice was provocative of a second burst of laughter.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0349.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The regret in his voice was provocative of a second burst of laughter.",
+ "raw": {
+ "normalized_reference": "the regret in his voice was provocative of a second burst of laughter.",
+ "normalized_hypothesis": "the regret in his voice was provocative of a second burst of laughter.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the regret in his voice was provocative of a 2nd burst of laughter",
+ "normalized_hypothesis": "the regret in his voice was provocative of a 2nd burst of laughter",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0350",
+ "category": "phonetic_balance",
+ "text": "And each year something happened, and I did not go.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0350.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "And each year something happened, and I did not go.",
+ "raw": {
+ "normalized_reference": "and each year something happened, and i did not go.",
+ "normalized_hypothesis": "and each year something happened, and i did not go.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "and each year something happened and i did not go",
+ "normalized_hypothesis": "and each year something happened and i did not go",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0351",
+ "category": "phonetic_balance",
+ "text": "Gad, your letter came just in time.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0351.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Gad, your letter came just in time.",
+ "raw": {
+ "normalized_reference": "gad, your letter came just in time.",
+ "normalized_hypothesis": "gad, your letter came just in time.",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "gad your letter came just in time",
+ "normalized_hypothesis": "gad your letter came just in time",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0352",
+ "category": "phonetic_balance",
+ "text": "But if Pierre did not return, until tomorrow.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0352.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "But if Pierre did not return until tomorrow...",
+ "raw": {
+ "normalized_reference": "but if pierre did not return, until tomorrow.",
+ "normalized_hypothesis": "but if pierre did not return until tomorrow...",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "but if pierre did not return until tomorrow",
+ "normalized_hypothesis": "but if pierre did not return until tomorrow .",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0353",
+ "category": "phonetic_balance",
+ "text": "Billinger may arrive in time.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0353.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Billinger may arrive in time.",
+ "raw": {
+ "normalized_reference": "billinger may arrive in time.",
+ "normalized_hypothesis": "billinger may arrive in time.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "billinger may arrive in time",
+ "normalized_hypothesis": "billinger may arrive in time",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0354",
+ "category": "phonetic_balance",
+ "text": "In such a tumbling of values was no time to sell.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0354.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "In such a tumbling of values was no time to sell.",
+ "raw": {
+ "normalized_reference": "in such a tumbling of values was no time to sell.",
+ "normalized_hypothesis": "in such a tumbling of values was no time to sell.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "in such a tumbling of values was no time to sell",
+ "normalized_hypothesis": "in such a tumbling of values was no time to sell",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0355",
+ "category": "phonetic_balance",
+ "text": "There's not an iota of truth in it.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0355.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "There's not an iota of truth in it.",
+ "raw": {
+ "normalized_reference": "there's not an iota of truth in it.",
+ "normalized_hypothesis": "there's not an iota of truth in it.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "there is not an iota of truth in it",
+ "normalized_hypothesis": "there is not an iota of truth in it",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0356",
+ "category": "phonetic_balance",
+ "text": "It is dog eat dog, and you ate them up.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0356.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It is dog eat dog and you ate them up.",
+ "raw": {
+ "normalized_reference": "it is dog eat dog, and you ate them up.",
+ "normalized_hypothesis": "it is dog eat dog and you ate them up.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it is dog eat dog and you ate them up",
+ "normalized_hypothesis": "it is dog eat dog and you ate them up",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0357",
+ "category": "phonetic_balance",
+ "text": "You can take a vacation on pay.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0357.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "You can take a vacation on pay.",
+ "raw": {
+ "normalized_reference": "you can take a vacation on pay.",
+ "normalized_hypothesis": "you can take a vacation on pay.",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "you can take a vacation on pay",
+ "normalized_hypothesis": "you can take a vacation on pay",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0358",
+ "category": "phonetic_balance",
+ "text": "I have seen myself that one man contemplated by Pascal's philosophic eye.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0358.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I have seen myself that one man contemplated by Pascal's philosophic eye.",
+ "raw": {
+ "normalized_reference": "i have seen myself that one man contemplated by pascal's philosophic eye.",
+ "normalized_hypothesis": "i have seen myself that one man contemplated by pascal's philosophic eye.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i have seen myself that one man contemplated by pascal is philosophic eye",
+ "normalized_hypothesis": "i have seen myself that one man contemplated by pascal is philosophic eye",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0359",
+ "category": "phonetic_balance",
+ "text": "Ah, I had forgotten, he exclaimed.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0359.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "All I had forgotten, he exclaimed.",
+ "raw": {
+ "normalized_reference": "ah, i had forgotten, he exclaimed.",
+ "normalized_hypothesis": "all i had forgotten, he exclaimed.",
+ "errors": 1,
+ "words": 6,
+ "wer": 0.16666666666666666
+ },
+ "semantic": {
+ "normalized_reference": "ah i had forgotten he exclaimed",
+ "normalized_hypothesis": "all i had forgotten he exclaimed",
+ "errors": 1,
+ "words": 6,
+ "wer": 0.16666666666666666
+ }
+ },
+ {
+ "id": "release500_0360",
+ "category": "phonetic_balance",
+ "text": "Providence had delivered him through the maelstrom.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0360.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Providence had delivered him through the maelstrom.",
+ "raw": {
+ "normalized_reference": "providence had delivered him through the maelstrom.",
+ "normalized_hypothesis": "providence had delivered him through the maelstrom.",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "providence had delivered him through the maelstrom",
+ "normalized_hypothesis": "providence had delivered him through the maelstrom",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0361",
+ "category": "phonetic_balance",
+ "text": "She was his now, forever.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0361.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "She was his now forever.",
+ "raw": {
+ "normalized_reference": "she was his now, forever.",
+ "normalized_hypothesis": "she was his now forever.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "she was his now forever",
+ "normalized_hypothesis": "she was his now forever",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0362",
+ "category": "phonetic_balance",
+ "text": "It was unobtrusive, yet it was there.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0362.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It was unobtrusive, yet it was there.",
+ "raw": {
+ "normalized_reference": "it was unobtrusive, yet it was there.",
+ "normalized_hypothesis": "it was unobtrusive, yet it was there.",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it was unobtrusive yet it was there",
+ "normalized_hypothesis": "it was unobtrusive yet it was there",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0363",
+ "category": "phonetic_balance",
+ "text": "Men like Joe Goose dated existence from drunk to drunk.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0363.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Men like Joe Goose dated existence from drunk to drunk.",
+ "raw": {
+ "normalized_reference": "men like joe goose dated existence from drunk to drunk.",
+ "normalized_hypothesis": "men like joe goose dated existence from drunk to drunk.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "men like joe goose dated existence from drunk to drunk",
+ "normalized_hypothesis": "men like joe goose dated existence from drunk to drunk",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0364",
+ "category": "phonetic_balance",
+ "text": "I was completely lost in my work.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0364.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I was completely lost in my work.",
+ "raw": {
+ "normalized_reference": "i was completely lost in my work.",
+ "normalized_hypothesis": "i was completely lost in my work.",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i was completely lost in my work",
+ "normalized_hypothesis": "i was completely lost in my work",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0365",
+ "category": "phonetic_balance",
+ "text": "The very thing, Ernest agreed.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0365.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The very thing Ernest agreed.",
+ "raw": {
+ "normalized_reference": "the very thing, ernest agreed.",
+ "normalized_hypothesis": "the very thing ernest agreed.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the very thing ernest agreed",
+ "normalized_hypothesis": "the very thing ernest agreed",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0366",
+ "category": "phonetic_balance",
+ "text": "He was the leader, and Tudor was his lieutenant.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0366.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He was the leader and Tudor was his lieutenant.",
+ "raw": {
+ "normalized_reference": "he was the leader, and tudor was his lieutenant.",
+ "normalized_hypothesis": "he was the leader and tudor was his lieutenant.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he was the leader and tudor was his lieutenant",
+ "normalized_hypothesis": "he was the leader and tudor was his lieutenant",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0367",
+ "category": "phonetic_balance",
+ "text": "The Fire-Men wore animal skins around their waists and across their shoulders.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0367.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The firemen wore animal skins around their waists and across their shoulders.",
+ "raw": {
+ "normalized_reference": "the fire-men wore animal skins around their waists and across their shoulders.",
+ "normalized_hypothesis": "the firemen wore animal skins around their waists and across their shoulders.",
+ "errors": 2,
+ "words": 13,
+ "wer": 0.15384615384615385
+ },
+ "semantic": {
+ "normalized_reference": "the fire men wore animal skins around their waists and across their shoulders",
+ "normalized_hypothesis": "the firemen wore animal skins around their waists and across their shoulders",
+ "errors": 2,
+ "words": 13,
+ "wer": 0.15384615384615385
+ }
+ },
+ {
+ "id": "release500_0368",
+ "category": "phonetic_balance",
+ "text": "We could throw stones with our feet.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0368.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "We could throw stones with our feet.",
+ "raw": {
+ "normalized_reference": "we could throw stones with our feet.",
+ "normalized_hypothesis": "we could throw stones with our feet.",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "we could throw stones with our feet",
+ "normalized_hypothesis": "we could throw stones with our feet",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0369",
+ "category": "phonetic_balance",
+ "text": "He turned sharply, and faced Gregson across the table.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0369.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He turned sharply and faced Gregson across the table.",
+ "raw": {
+ "normalized_reference": "he turned sharply, and faced gregson across the table.",
+ "normalized_hypothesis": "he turned sharply and faced gregson across the table.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he turned sharply and faced gregson across the table",
+ "normalized_hypothesis": "he turned sharply and faced gregson across the table",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0370",
+ "category": "phonetic_balance",
+ "text": "They are to attack your camp tomorrow night.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0370.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "They are to attack your camp tomorrow night.",
+ "raw": {
+ "normalized_reference": "they are to attack your camp tomorrow night.",
+ "normalized_hypothesis": "they are to attack your camp tomorrow night.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "they are to attack your camp tomorrow night",
+ "normalized_hypothesis": "they are to attack your camp tomorrow night",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0371",
+ "category": "phonetic_balance",
+ "text": "And watch out for wet feet, was his parting advice.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0371.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "And watch out for wet feet was his parting advice.",
+ "raw": {
+ "normalized_reference": "and watch out for wet feet, was his parting advice.",
+ "normalized_hypothesis": "and watch out for wet feet was his parting advice.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "and watch out for wet feet was his parting advice",
+ "normalized_hypothesis": "and watch out for wet feet was his parting advice",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0372",
+ "category": "phonetic_balance",
+ "text": "Two weeks passed, and in that time Thorpe left camp three times.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0372.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Two weeks passed and in that time Thorpe left camp three times.",
+ "raw": {
+ "normalized_reference": "two weeks passed, and in that time thorpe left camp three times.",
+ "normalized_hypothesis": "two weeks passed and in that time thorpe left camp three times.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "2 weeks passed and in that time thorpe left camp 3 times",
+ "normalized_hypothesis": "2 weeks passed and in that time thorpe left camp 3 times",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0373",
+ "category": "phonetic_balance",
+ "text": "But ever his gaze returned to that Crouched Venus on the piano.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0373.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "But ever his gaze returned to that crouched Venus on the piano.",
+ "raw": {
+ "normalized_reference": "but ever his gaze returned to that crouched venus on the piano.",
+ "normalized_hypothesis": "but ever his gaze returned to that crouched venus on the piano.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "but ever his gaze returned to that crouched venus on the piano",
+ "normalized_hypothesis": "but ever his gaze returned to that crouched venus on the piano",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0374",
+ "category": "phonetic_balance",
+ "text": "But she had become an automaton.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0374.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "But she had become an automaton.",
+ "raw": {
+ "normalized_reference": "but she had become an automaton.",
+ "normalized_hypothesis": "but she had become an automaton.",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "but she had become an automaton",
+ "normalized_hypothesis": "but she had become an automaton",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0375",
+ "category": "phonetic_balance",
+ "text": "I do not blame you for anything; remember that.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0375.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I do not blame you for anything. Remember that.",
+ "raw": {
+ "normalized_reference": "i do not blame you for anything; remember that.",
+ "normalized_hypothesis": "i do not blame you for anything. remember that.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i do not blame you for anything remember that",
+ "normalized_hypothesis": "i do not blame you for anything remember that",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0376",
+ "category": "phonetic_balance",
+ "text": "The next thing to watch out for is bed sores.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0376.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The next thing to watch out for is bed sores.",
+ "raw": {
+ "normalized_reference": "the next thing to watch out for is bed sores.",
+ "normalized_hypothesis": "the next thing to watch out for is bed sores.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the next thing to watch out for is bed sores",
+ "normalized_hypothesis": "the next thing to watch out for is bed sores",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0377",
+ "category": "phonetic_balance",
+ "text": "He will follow us soon.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0377.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He will follow us soon.",
+ "raw": {
+ "normalized_reference": "he will follow us soon.",
+ "normalized_hypothesis": "he will follow us soon.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he will follow us soon",
+ "normalized_hypothesis": "he will follow us soon",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0378",
+ "category": "phonetic_balance",
+ "text": "His beady black eyes saw bargains where other men saw bankruptcy.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0378.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "His beady black eyes saw bargains where other men saw bankruptcy.",
+ "raw": {
+ "normalized_reference": "his beady black eyes saw bargains where other men saw bankruptcy.",
+ "normalized_hypothesis": "his beady black eyes saw bargains where other men saw bankruptcy.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "his beady black eyes saw bargains where other men saw bankruptcy",
+ "normalized_hypothesis": "his beady black eyes saw bargains where other men saw bankruptcy",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0379",
+ "category": "phonetic_balance",
+ "text": "They robbed me a few years later.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0379.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "They robbed me a few years later.",
+ "raw": {
+ "normalized_reference": "they robbed me a few years later.",
+ "normalized_hypothesis": "they robbed me a few years later.",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "they robbed me a few years later",
+ "normalized_hypothesis": "they robbed me a few years later",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0380",
+ "category": "phonetic_balance",
+ "text": "Gregson held a lighted match until it burnt his fingertips.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0380.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Gregson held a lighted match until it burnt his fingertips.",
+ "raw": {
+ "normalized_reference": "gregson held a lighted match until it burnt his fingertips.",
+ "normalized_hypothesis": "gregson held a lighted match until it burnt his fingertips.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "gregson held a lighted match until it burnt his fingertips",
+ "normalized_hypothesis": "gregson held a lighted match until it burnt his fingertips",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0381",
+ "category": "phonetic_balance",
+ "text": "A burst of laughter was his reward.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0381.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "A burst of laughter was his reward.",
+ "raw": {
+ "normalized_reference": "a burst of laughter was his reward.",
+ "normalized_hypothesis": "a burst of laughter was his reward.",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "a burst of laughter was his reward",
+ "normalized_hypothesis": "a burst of laughter was his reward",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0382",
+ "category": "phonetic_balance",
+ "text": "He cried, and swung the club wildly.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0382.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He cried and swung the club wildly.",
+ "raw": {
+ "normalized_reference": "he cried, and swung the club wildly.",
+ "normalized_hypothesis": "he cried and swung the club wildly.",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he cried and swung the club wildly",
+ "normalized_hypothesis": "he cried and swung the club wildly",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0383",
+ "category": "phonetic_balance",
+ "text": "Sandel would never become a world champion.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0383.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Sandal would never become a world champion.",
+ "raw": {
+ "normalized_reference": "sandel would never become a world champion.",
+ "normalized_hypothesis": "sandal would never become a world champion.",
+ "errors": 1,
+ "words": 7,
+ "wer": 0.14285714285714285
+ },
+ "semantic": {
+ "normalized_reference": "sandel would never become a world champion",
+ "normalized_hypothesis": "sandal would never become a world champion",
+ "errors": 1,
+ "words": 7,
+ "wer": 0.14285714285714285
+ }
+ },
+ {
+ "id": "release500_0384",
+ "category": "phonetic_balance",
+ "text": "And after the bath a shave would not be bad.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0384.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "And after the bath a shave would not be bad.",
+ "raw": {
+ "normalized_reference": "and after the bath a shave would not be bad.",
+ "normalized_hypothesis": "and after the bath a shave would not be bad.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "and after the bath a shave would not be bad",
+ "normalized_hypothesis": "and after the bath a shave would not be bad",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0385",
+ "category": "phonetic_balance",
+ "text": "They saw each other for the first time in Boston.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0385.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "They saw each other for the first time in Boston.",
+ "raw": {
+ "normalized_reference": "they saw each other for the first time in boston.",
+ "normalized_hypothesis": "they saw each other for the first time in boston.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "they saw each other for the 1st time in boston",
+ "normalized_hypothesis": "they saw each other for the 1st time in boston",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0386",
+ "category": "phonetic_balance",
+ "text": "I had been born with no organic, chemical predisposition toward alcohol.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0386.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I had been born with no organic, chemical predisposition toward alcohol.",
+ "raw": {
+ "normalized_reference": "i had been born with no organic, chemical predisposition toward alcohol.",
+ "normalized_hypothesis": "i had been born with no organic, chemical predisposition toward alcohol.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i had been born with no organic chemical predisposition toward alcohol",
+ "normalized_hypothesis": "i had been born with no organic chemical predisposition toward alcohol",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0387",
+ "category": "phonetic_balance",
+ "text": "They had no fixed values, to be altered by adjectives and adverbs.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0387.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "They had no fixed values to be altered by adjectives and adverbs.",
+ "raw": {
+ "normalized_reference": "they had no fixed values, to be altered by adjectives and adverbs.",
+ "normalized_hypothesis": "they had no fixed values to be altered by adjectives and adverbs.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "they had no fixed values to be altered by adjectives and adverbs",
+ "normalized_hypothesis": "they had no fixed values to be altered by adjectives and adverbs",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0388",
+ "category": "phonetic_balance",
+ "text": "Nor was Elam Harnish an exception.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0388.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Nor was a lamb-harnish an exception.",
+ "raw": {
+ "normalized_reference": "nor was elam harnish an exception.",
+ "normalized_hypothesis": "nor was a lamb-harnish an exception.",
+ "errors": 2,
+ "words": 6,
+ "wer": 0.3333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "nor was elam harnish an exception",
+ "normalized_hypothesis": "nor was a lamb harnish an exception",
+ "errors": 2,
+ "words": 6,
+ "wer": 0.3333333333333333
+ }
+ },
+ {
+ "id": "release500_0389",
+ "category": "phonetic_balance",
+ "text": "He was a wise hyena.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0389.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He was a wise hyena.",
+ "raw": {
+ "normalized_reference": "he was a wise hyena.",
+ "normalized_hypothesis": "he was a wise hyena.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he was a wise hyena",
+ "normalized_hypothesis": "he was a wise hyena",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0390",
+ "category": "phonetic_balance",
+ "text": "And now behold the perversity of things.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0390.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "And now behold the perversity of things.",
+ "raw": {
+ "normalized_reference": "and now behold the perversity of things.",
+ "normalized_hypothesis": "and now behold the perversity of things.",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "and now behold the perversity of things",
+ "normalized_hypothesis": "and now behold the perversity of things",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0391",
+ "category": "phonetic_balance",
+ "text": "The Fire People, like ourselves, lived in caves.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0391.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The fire people, like ourselves, lived in caves.",
+ "raw": {
+ "normalized_reference": "the fire people, like ourselves, lived in caves.",
+ "normalized_hypothesis": "the fire people, like ourselves, lived in caves.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the fire people like ourselves lived in caves",
+ "normalized_hypothesis": "the fire people like ourselves lived in caves",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0392",
+ "category": "phonetic_balance",
+ "text": "To her the bridge was tambo, which is the native for taboo.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0392.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "To her the bridge was Tambo, which is the native for taboo.",
+ "raw": {
+ "normalized_reference": "to her the bridge was tambo, which is the native for taboo.",
+ "normalized_hypothesis": "to her the bridge was tambo, which is the native for taboo.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "to her the bridge was tambo which is the native for taboo",
+ "normalized_hypothesis": "to her the bridge was tambo which is the native for taboo",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0393",
+ "category": "phonetic_balance",
+ "text": "The voyage was our idea of a good time.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0393.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The voyage was our idea of a good time.",
+ "raw": {
+ "normalized_reference": "the voyage was our idea of a good time.",
+ "normalized_hypothesis": "the voyage was our idea of a good time.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the voyage was our idea of a good time",
+ "normalized_hypothesis": "the voyage was our idea of a good time",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0394",
+ "category": "phonetic_balance",
+ "text": "Of course much grumbling went on, and little outbursts were continually occurring.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0394.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Of course much grumbling went on and little outbursts were continually occurring.",
+ "raw": {
+ "normalized_reference": "of course much grumbling went on, and little outbursts were continually occurring.",
+ "normalized_hypothesis": "of course much grumbling went on and little outbursts were continually occurring.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "of course much grumbling went on and little outbursts were continually occurring",
+ "normalized_hypothesis": "of course much grumbling went on and little outbursts were continually occurring",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0395",
+ "category": "phonetic_balance",
+ "text": "Get down and dig in.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0395.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Get down and dig in.",
+ "raw": {
+ "normalized_reference": "get down and dig in.",
+ "normalized_hypothesis": "get down and dig in.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "get down and dig in",
+ "normalized_hypothesis": "get down and dig in",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0396",
+ "category": "phonetic_balance",
+ "text": "It was put together in a casual, helter-skelter sort of way.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0396.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It was put together in a casual, helter-skelter sort of way.",
+ "raw": {
+ "normalized_reference": "it was put together in a casual, helter-skelter sort of way.",
+ "normalized_hypothesis": "it was put together in a casual, helter-skelter sort of way.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it was put together in a casual helter skelter sort of way",
+ "normalized_hypothesis": "it was put together in a casual helter skelter sort of way",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0397",
+ "category": "phonetic_balance",
+ "text": "He deluged me, overwhelmed me with argument.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0397.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He deluged me, overwhelmed me with argument.",
+ "raw": {
+ "normalized_reference": "he deluged me, overwhelmed me with argument.",
+ "normalized_hypothesis": "he deluged me, overwhelmed me with argument.",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he deluged me overwhelmed me with argument",
+ "normalized_hypothesis": "he deluged me overwhelmed me with argument",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0398",
+ "category": "phonetic_balance",
+ "text": "The girl faced him, her eyes shining with sudden fear.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0398.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The girl faced him, her eyes shining with sudden fear.",
+ "raw": {
+ "normalized_reference": "the girl faced him, her eyes shining with sudden fear.",
+ "normalized_hypothesis": "the girl faced him, her eyes shining with sudden fear.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the girl faced him her eyes shining with sudden fear",
+ "normalized_hypothesis": "the girl faced him her eyes shining with sudden fear",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0399",
+ "category": "phonetic_balance",
+ "text": "It will break our hearts and our backs to hoist anchor by hand.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0399.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It will break our hearts and our backs to hoist anchor by hand.",
+ "raw": {
+ "normalized_reference": "it will break our hearts and our backs to hoist anchor by hand.",
+ "normalized_hypothesis": "it will break our hearts and our backs to hoist anchor by hand.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it will break our hearts and our backs to hoist anchor by hand",
+ "normalized_hypothesis": "it will break our hearts and our backs to hoist anchor by hand",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0400",
+ "category": "phonetic_balance",
+ "text": "He made no reply as he waited for Whittemore to continue.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0400.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He made no reply as he waited for Whittemore to continue.",
+ "raw": {
+ "normalized_reference": "he made no reply as he waited for whittemore to continue.",
+ "normalized_hypothesis": "he made no reply as he waited for whittemore to continue.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he made no reply as he waited for whittemore to continue",
+ "normalized_hypothesis": "he made no reply as he waited for whittemore to continue",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0401",
+ "category": "phonetic_balance",
+ "text": "Who the devil gave it to you to be judge and jury.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0401.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Who the devil gave it to you to be judge and jury?",
+ "raw": {
+ "normalized_reference": "who the devil gave it to you to be judge and jury.",
+ "normalized_hypothesis": "who the devil gave it to you to be judge and jury?",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "who the devil gave it to you to be judge and jury",
+ "normalized_hypothesis": "who the devil gave it to you to be judge and jury",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0402",
+ "category": "phonetic_balance",
+ "text": "Daylight found himself charmed and made curious by the little man.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0402.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Daylight found himself charmed and made curious by the little man.",
+ "raw": {
+ "normalized_reference": "daylight found himself charmed and made curious by the little man.",
+ "normalized_hypothesis": "daylight found himself charmed and made curious by the little man.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "daylight found himself charmed and made curious by the little man",
+ "normalized_hypothesis": "daylight found himself charmed and made curious by the little man",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0403",
+ "category": "phonetic_balance",
+ "text": "It's a Yankee, Joan cried.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0403.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It's a Yankee, Joan cried.",
+ "raw": {
+ "normalized_reference": "it's a yankee, joan cried.",
+ "normalized_hypothesis": "it's a yankee, joan cried.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it is a yankee joan cried",
+ "normalized_hypothesis": "it is a yankee joan cried",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0404",
+ "category": "phonetic_balance",
+ "text": "But what they want with your toothbrush is more than I can imagine.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0404.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "But what they want with your toothbrush is more than I can imagine.",
+ "raw": {
+ "normalized_reference": "but what they want with your toothbrush is more than i can imagine.",
+ "normalized_hypothesis": "but what they want with your toothbrush is more than i can imagine.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "but what they want with your toothbrush is more than i can imagine",
+ "normalized_hypothesis": "but what they want with your toothbrush is more than i can imagine",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0405",
+ "category": "phonetic_balance",
+ "text": "They were babbling and chattering all together.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0405.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "They were babbling and chattering all together.",
+ "raw": {
+ "normalized_reference": "they were babbling and chattering all together.",
+ "normalized_hypothesis": "they were babbling and chattering all together.",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "they were babbling and chattering all together",
+ "normalized_hypothesis": "they were babbling and chattering all together",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0406",
+ "category": "phonetic_balance",
+ "text": "Men of Selden's stamp don't stop at women and children.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0406.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Men of Selden's stamp don't stop at women and children.",
+ "raw": {
+ "normalized_reference": "men of selden's stamp don't stop at women and children.",
+ "normalized_hypothesis": "men of selden's stamp don't stop at women and children.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "men of selden is stamp do not stop at women and children",
+ "normalized_hypothesis": "men of selden is stamp do not stop at women and children",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0407",
+ "category": "phonetic_balance",
+ "text": "Cameron looked at his hands with their long, sinewy fingers.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0407.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Cameron looked at his hands with their long, sinewy fingers.",
+ "raw": {
+ "normalized_reference": "cameron looked at his hands with their long, sinewy fingers.",
+ "normalized_hypothesis": "cameron looked at his hands with their long, sinewy fingers.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "cameron looked at his hands with their long sinewy fingers",
+ "normalized_hypothesis": "cameron looked at his hands with their long sinewy fingers",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0408",
+ "category": "phonetic_balance",
+ "text": "Saxon waited, for she knew a fresh idea had struck Billy.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0408.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Saxon waited, for she knew a fresh idea had struck Billy.",
+ "raw": {
+ "normalized_reference": "saxon waited, for she knew a fresh idea had struck billy.",
+ "normalized_hypothesis": "saxon waited, for she knew a fresh idea had struck billy.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "saxon waited for she knew a fresh idea had struck billy",
+ "normalized_hypothesis": "saxon waited for she knew a fresh idea had struck billy",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0409",
+ "category": "phonetic_balance",
+ "text": "He saw only the effect in a general, sketchy way.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0409.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He saw only the effect in a general, sketchy way.",
+ "raw": {
+ "normalized_reference": "he saw only the effect in a general, sketchy way.",
+ "normalized_hypothesis": "he saw only the effect in a general, sketchy way.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he saw only the effect in a general sketchy way",
+ "normalized_hypothesis": "he saw only the effect in a general sketchy way",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0410",
+ "category": "phonetic_balance",
+ "text": "He began to follow the footprints of the dog.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0410.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He began to follow the footprints of the dog.",
+ "raw": {
+ "normalized_reference": "he began to follow the footprints of the dog.",
+ "normalized_hypothesis": "he began to follow the footprints of the dog.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he began to follow the footprints of the dog",
+ "normalized_hypothesis": "he began to follow the footprints of the dog",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0411",
+ "category": "phonetic_balance",
+ "text": "Do you value your hide.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0411.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Do you value your hide?",
+ "raw": {
+ "normalized_reference": "do you value your hide.",
+ "normalized_hypothesis": "do you value your hide?",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "do you value your hide",
+ "normalized_hypothesis": "do you value your hide",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0412",
+ "category": "phonetic_balance",
+ "text": "A rising tide of fat had submerged them.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0412.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "A rising tide of fat had submerged them.",
+ "raw": {
+ "normalized_reference": "a rising tide of fat had submerged them.",
+ "normalized_hypothesis": "a rising tide of fat had submerged them.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "a rising tide of fat had submerged them",
+ "normalized_hypothesis": "a rising tide of fat had submerged them",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0413",
+ "category": "phonetic_balance",
+ "text": "There was nothing on the rock.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0413.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "There was nothing on the rock.",
+ "raw": {
+ "normalized_reference": "there was nothing on the rock.",
+ "normalized_hypothesis": "there was nothing on the rock.",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "there was nothing on the rock",
+ "normalized_hypothesis": "there was nothing on the rock",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0414",
+ "category": "phonetic_balance",
+ "text": "Whatever he guessed he locked away in the taboo room of Naomi.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0414.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Whatever he guessed, he locked away in the taboo room of Naomi.",
+ "raw": {
+ "normalized_reference": "whatever he guessed he locked away in the taboo room of naomi.",
+ "normalized_hypothesis": "whatever he guessed, he locked away in the taboo room of naomi.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "whatever he guessed he locked away in the taboo room of naomi",
+ "normalized_hypothesis": "whatever he guessed he locked away in the taboo room of naomi",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0415",
+ "category": "phonetic_balance",
+ "text": "Will we ever forget it.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0415.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Will we ever forget it?",
+ "raw": {
+ "normalized_reference": "will we ever forget it.",
+ "normalized_hypothesis": "will we ever forget it?",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "will we ever forget it",
+ "normalized_hypothesis": "will we ever forget it",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0416",
+ "category": "phonetic_balance",
+ "text": "Jeanne and Pierre both gazed toward the great rock.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0416.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Jean and Pierre both gazed toward the great rock.",
+ "raw": {
+ "normalized_reference": "jeanne and pierre both gazed toward the great rock.",
+ "normalized_hypothesis": "jean and pierre both gazed toward the great rock.",
+ "errors": 1,
+ "words": 9,
+ "wer": 0.1111111111111111
+ },
+ "semantic": {
+ "normalized_reference": "jeanne and pierre both gazed toward the great rock",
+ "normalized_hypothesis": "jean and pierre both gazed toward the great rock",
+ "errors": 1,
+ "words": 9,
+ "wer": 0.1111111111111111
+ }
+ },
+ {
+ "id": "release500_0417",
+ "category": "phonetic_balance",
+ "text": "Does the old boy often go off at half-cock that way.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0417.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Does the old boy often go off at half-cock that way?",
+ "raw": {
+ "normalized_reference": "does the old boy often go off at half-cock that way.",
+ "normalized_hypothesis": "does the old boy often go off at half-cock that way?",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "does the old boy often go off at half cock that way",
+ "normalized_hypothesis": "does the old boy often go off at half cock that way",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0418",
+ "category": "phonetic_balance",
+ "text": "The temperature dropped to fifty below zero and remained there the whole trip.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0418.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The temperature dropped to 50 below zero and remained there the whole trip.",
+ "raw": {
+ "normalized_reference": "the temperature dropped to fifty below zero and remained there the whole trip.",
+ "normalized_hypothesis": "the temperature dropped to 50 below zero and remained there the whole trip.",
+ "errors": 1,
+ "words": 13,
+ "wer": 0.07692307692307693
+ },
+ "semantic": {
+ "normalized_reference": "the temperature dropped to 50 below 0 and remained there the whole trip",
+ "normalized_hypothesis": "the temperature dropped to 50 below 0 and remained there the whole trip",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0419",
+ "category": "phonetic_balance",
+ "text": "It's worth eight dollars.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0419.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It's worth eight dollars.",
+ "raw": {
+ "normalized_reference": "it's worth eight dollars.",
+ "normalized_hypothesis": "it's worth eight dollars.",
+ "errors": 0,
+ "words": 4,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it is worth $8",
+ "normalized_hypothesis": "it is worth $8",
+ "errors": 0,
+ "words": 4,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0420",
+ "category": "phonetic_balance",
+ "text": "But I did not enjoy it long.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0420.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "But I did not enjoy it long.",
+ "raw": {
+ "normalized_reference": "but i did not enjoy it long.",
+ "normalized_hypothesis": "but i did not enjoy it long.",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "but i did not enjoy it long",
+ "normalized_hypothesis": "but i did not enjoy it long",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0421",
+ "category": "phonetic_balance",
+ "text": "Shorty turned to their employers.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0421.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Shorty turned to their employers.",
+ "raw": {
+ "normalized_reference": "shorty turned to their employers.",
+ "normalized_hypothesis": "shorty turned to their employers.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "shorty turned to their employers",
+ "normalized_hypothesis": "shorty turned to their employers",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0422",
+ "category": "phonetic_balance",
+ "text": "It was the third or fourth time that Philip had heard MacDougall swear.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0422.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It was the third or fourth time that Philip had heard MacDougall swear.",
+ "raw": {
+ "normalized_reference": "it was the third or fourth time that philip had heard macdougall swear.",
+ "normalized_hypothesis": "it was the third or fourth time that philip had heard macdougall swear.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it was the 3rd or 4th time that philip had heard macdougall swear",
+ "normalized_hypothesis": "it was the 3rd or 4th time that philip had heard macdougall swear",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0423",
+ "category": "phonetic_balance",
+ "text": "Ernest saw in the affair the most sinister import.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0423.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "ernest saw in the affair the most sinister import",
+ "raw": {
+ "normalized_reference": "ernest saw in the affair the most sinister import.",
+ "normalized_hypothesis": "ernest saw in the affair the most sinister import",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "ernest saw in the affair the most sinister import",
+ "normalized_hypothesis": "ernest saw in the affair the most sinister import",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0424",
+ "category": "phonetic_balance",
+ "text": "Also, at regular intervals, he would mutter.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0424.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Also at regular intervals he would mutter,",
+ "raw": {
+ "normalized_reference": "also, at regular intervals, he would mutter.",
+ "normalized_hypothesis": "also at regular intervals he would mutter,",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "also at regular intervals he would mutter",
+ "normalized_hypothesis": "also at regular intervals he would mutter",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0425",
+ "category": "phonetic_balance",
+ "text": "He may anticipate the day of his death.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0425.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He may anticipate the day of his death.",
+ "raw": {
+ "normalized_reference": "he may anticipate the day of his death.",
+ "normalized_hypothesis": "he may anticipate the day of his death.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he may anticipate the day of his death",
+ "normalized_hypothesis": "he may anticipate the day of his death",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0426",
+ "category": "phonetic_balance",
+ "text": "I defy any man to get a Solomon Island sore in California.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0426.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I defy any man to get a Solomon Island soar in California.",
+ "raw": {
+ "normalized_reference": "i defy any man to get a solomon island sore in california.",
+ "normalized_hypothesis": "i defy any man to get a solomon island soar in california.",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "i defy any man to get a solomon island sore in california",
+ "normalized_hypothesis": "i defy any man to get a solomon island soar in california",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ }
+ },
+ {
+ "id": "release500_0427",
+ "category": "phonetic_balance",
+ "text": "Now our figuring was all right.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0427.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Now our figuring was all right.",
+ "raw": {
+ "normalized_reference": "now our figuring was all right.",
+ "normalized_hypothesis": "now our figuring was all right.",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "now our figuring was all right",
+ "normalized_hypothesis": "now our figuring was all right",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0428",
+ "category": "phonetic_balance",
+ "text": "MacDougall tapped his forehead suspiciously with a stubby forefinger.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0428.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "MacDougall tapped his forehead suspiciously with a stubby forefinger.",
+ "raw": {
+ "normalized_reference": "macdougall tapped his forehead suspiciously with a stubby forefinger.",
+ "normalized_hypothesis": "macdougall tapped his forehead suspiciously with a stubby forefinger.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "macdougall tapped his forehead suspiciously with a stubby forefinger",
+ "normalized_hypothesis": "macdougall tapped his forehead suspiciously with a stubby forefinger",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0429",
+ "category": "phonetic_balance",
+ "text": "O'Brien took off his coat and bared his right arm.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0429.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "O'Brien took off his coat and bared his right arm.",
+ "raw": {
+ "normalized_reference": "o'brien took off his coat and bared his right arm.",
+ "normalized_hypothesis": "o'brien took off his coat and bared his right arm.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "0 brien took off his coat and bared his right arm",
+ "normalized_hypothesis": "0 brien took off his coat and bared his right arm",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0430",
+ "category": "phonetic_balance",
+ "text": "But it won't continue, she said with easy confidence.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0430.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "But it won't continue, she said with easy confidence.",
+ "raw": {
+ "normalized_reference": "but it won't continue, she said with easy confidence.",
+ "normalized_hypothesis": "but it won't continue, she said with easy confidence.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "but it will not continue she said with easy confidence",
+ "normalized_hypothesis": "but it will not continue she said with easy confidence",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0431",
+ "category": "phonetic_balance",
+ "text": "I was still weak from my prolonged immersion.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0431.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I was still weak from my prolonged immersion.",
+ "raw": {
+ "normalized_reference": "i was still weak from my prolonged immersion.",
+ "normalized_hypothesis": "i was still weak from my prolonged immersion.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i was still weak from my prolonged immersion",
+ "normalized_hypothesis": "i was still weak from my prolonged immersion",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0432",
+ "category": "phonetic_balance",
+ "text": "Oolong was two hundred and fifty miles from the nearest land.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0432.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Oolong was 250 miles from the nearest land.",
+ "raw": {
+ "normalized_reference": "oolong was two hundred and fifty miles from the nearest land.",
+ "normalized_hypothesis": "oolong was 250 miles from the nearest land.",
+ "errors": 4,
+ "words": 11,
+ "wer": 0.36363636363636365
+ },
+ "semantic": {
+ "normalized_reference": "oolong was 250 miles from the nearest land",
+ "normalized_hypothesis": "oolong was 250 miles from the nearest land",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0433",
+ "category": "phonetic_balance",
+ "text": "I just do appreciate it without being able to express my feelings.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0433.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I just do appreciate it without being able to express my feelings.",
+ "raw": {
+ "normalized_reference": "i just do appreciate it without being able to express my feelings.",
+ "normalized_hypothesis": "i just do appreciate it without being able to express my feelings.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i just do appreciate it without being able to express my feelings",
+ "normalized_hypothesis": "i just do appreciate it without being able to express my feelings",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0434",
+ "category": "phonetic_balance",
+ "text": "Now just what do you want to know.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0434.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Now just what do you want to know?",
+ "raw": {
+ "normalized_reference": "now just what do you want to know.",
+ "normalized_hypothesis": "now just what do you want to know?",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "now just what do you want to know",
+ "normalized_hypothesis": "now just what do you want to know",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0435",
+ "category": "phonetic_balance",
+ "text": "But she swung obediently on her heel into the wind.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0435.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "But she swung obediently on her heel into the wind.",
+ "raw": {
+ "normalized_reference": "but she swung obediently on her heel into the wind.",
+ "normalized_hypothesis": "but she swung obediently on her heel into the wind.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "but she swung obediently on her heel into the wind",
+ "normalized_hypothesis": "but she swung obediently on her heel into the wind",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0436",
+ "category": "phonetic_balance",
+ "text": "Ask him, I laughed, then turned to Pasquini.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0436.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Ask him I laughed, then turned to Pasquini.",
+ "raw": {
+ "normalized_reference": "ask him, i laughed, then turned to pasquini.",
+ "normalized_hypothesis": "ask him i laughed, then turned to pasquini.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "ask him i laughed then turned to pasquini",
+ "normalized_hypothesis": "ask him i laughed then turned to pasquini",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0437",
+ "category": "phonetic_balance",
+ "text": "He did not rush in.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0437.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He did not rush in.",
+ "raw": {
+ "normalized_reference": "he did not rush in.",
+ "normalized_hypothesis": "he did not rush in.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he did not rush in",
+ "normalized_hypothesis": "he did not rush in",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0438",
+ "category": "phonetic_balance",
+ "text": "They obeyed him, and went here and there at his commands.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0438.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "They obeyed him and went here and there at his commands.",
+ "raw": {
+ "normalized_reference": "they obeyed him, and went here and there at his commands.",
+ "normalized_hypothesis": "they obeyed him and went here and there at his commands.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "they obeyed him and went here and there at his commands",
+ "normalized_hypothesis": "they obeyed him and went here and there at his commands",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0439",
+ "category": "phonetic_balance",
+ "text": "Tom Spink has a harpoon.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0439.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Tom Spink has a harpoon.",
+ "raw": {
+ "normalized_reference": "tom spink has a harpoon.",
+ "normalized_hypothesis": "tom spink has a harpoon.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "tom spink has a harpoon",
+ "normalized_hypothesis": "tom spink has a harpoon",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0440",
+ "category": "phonetic_balance",
+ "text": "It was my reports from the north which chiefly induced people to buy.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0440.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It was my reports from the north which chiefly induced people to buy.",
+ "raw": {
+ "normalized_reference": "it was my reports from the north which chiefly induced people to buy.",
+ "normalized_hypothesis": "it was my reports from the north which chiefly induced people to buy.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it was my reports from the north which chiefly induced people to buy",
+ "normalized_hypothesis": "it was my reports from the north which chiefly induced people to buy",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0441",
+ "category": "phonetic_balance",
+ "text": "Cherokee identified himself with his instinct.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0441.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Cherokee identified himself with his instinct.",
+ "raw": {
+ "normalized_reference": "cherokee identified himself with his instinct.",
+ "normalized_hypothesis": "cherokee identified himself with his instinct.",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "cherokee identified himself with his instinct",
+ "normalized_hypothesis": "cherokee identified himself with his instinct",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0442",
+ "category": "phonetic_balance",
+ "text": "I graduated last of my class.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0442.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I graduated last of my class.",
+ "raw": {
+ "normalized_reference": "i graduated last of my class.",
+ "normalized_hypothesis": "i graduated last of my class.",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i graduated last of my class",
+ "normalized_hypothesis": "i graduated last of my class",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0443",
+ "category": "phonetic_balance",
+ "text": "It was a gigantic inadequacy.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0443.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It was a gigantic inadequacy.",
+ "raw": {
+ "normalized_reference": "it was a gigantic inadequacy.",
+ "normalized_hypothesis": "it was a gigantic inadequacy.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it was a gigantic inadequacy",
+ "normalized_hypothesis": "it was a gigantic inadequacy",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0444",
+ "category": "phonetic_balance",
+ "text": "Did I possess too much vitality.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0444.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Did I possess too much vitality?",
+ "raw": {
+ "normalized_reference": "did i possess too much vitality.",
+ "normalized_hypothesis": "did i possess too much vitality?",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "did i possess too much vitality",
+ "normalized_hypothesis": "did i possess too much vitality",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0445",
+ "category": "phonetic_balance",
+ "text": "Therefore, hurrah for the game.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0445.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Therefore, hurrah for the game.",
+ "raw": {
+ "normalized_reference": "therefore, hurrah for the game.",
+ "normalized_hypothesis": "therefore, hurrah for the game.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "therefore hurrah for the game",
+ "normalized_hypothesis": "therefore hurrah for the game",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0446",
+ "category": "phonetic_balance",
+ "text": "He had fulfilled his duty and paid properly.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0446.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He had fulfilled his duty and paid properly.",
+ "raw": {
+ "normalized_reference": "he had fulfilled his duty and paid properly.",
+ "normalized_hypothesis": "he had fulfilled his duty and paid properly.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he had fulfilled his duty and paid properly",
+ "normalized_hypothesis": "he had fulfilled his duty and paid properly",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0447",
+ "category": "phonetic_balance",
+ "text": "How can you manage all alone, Mr Young.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0447.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "How can you manage all alone, Mr. Young?",
+ "raw": {
+ "normalized_reference": "how can you manage all alone, mr young.",
+ "normalized_hypothesis": "how can you manage all alone, mr. young?",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "how can you manage all alone mister young",
+ "normalized_hypothesis": "how can you manage all alone mister young",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0448",
+ "category": "phonetic_balance",
+ "text": "Red-Eye never committed a more outrageous deed.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0448.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Red-Eye never committed a more outrageous deed.",
+ "raw": {
+ "normalized_reference": "red-eye never committed a more outrageous deed.",
+ "normalized_hypothesis": "red-eye never committed a more outrageous deed.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "red eye never committed a more outrageous deed",
+ "normalized_hypothesis": "red eye never committed a more outrageous deed",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0449",
+ "category": "phonetic_balance",
+ "text": "He would keep his faith with Gregson for the promised day or two.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0449.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He would keep his faith with Gregson for the promised day or two.",
+ "raw": {
+ "normalized_reference": "he would keep his faith with gregson for the promised day or two.",
+ "normalized_hypothesis": "he would keep his faith with gregson for the promised day or two.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he would keep his faith with gregson for the promised day or 2",
+ "normalized_hypothesis": "he would keep his faith with gregson for the promised day or 2",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0450",
+ "category": "phonetic_balance",
+ "text": "There was one chance, and only one, of saving Jeanne.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0450.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "There was one chance and only one of saving Gene.",
+ "raw": {
+ "normalized_reference": "there was one chance, and only one, of saving jeanne.",
+ "normalized_hypothesis": "there was one chance and only one of saving gene.",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ },
+ "semantic": {
+ "normalized_reference": "there was one chance and only one of saving jeanne",
+ "normalized_hypothesis": "there was one chance and only one of saving gene",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ }
+ },
+ {
+ "id": "release500_0451",
+ "category": "phonetic_balance",
+ "text": "I was beneath the water, suffocating and drowning.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0451.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I was beneath the water, suffocating and drowning.",
+ "raw": {
+ "normalized_reference": "i was beneath the water, suffocating and drowning.",
+ "normalized_hypothesis": "i was beneath the water, suffocating and drowning.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i was beneath the water suffocating and drowning",
+ "normalized_hypothesis": "i was beneath the water suffocating and drowning",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0452",
+ "category": "phonetic_balance",
+ "text": "Give them their choice between a fine or an official whipping.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0452.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Give them their choice between a fine or an official whipping.",
+ "raw": {
+ "normalized_reference": "give them their choice between a fine or an official whipping.",
+ "normalized_hypothesis": "give them their choice between a fine or an official whipping.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "give them their choice between a fine or an official whipping",
+ "normalized_hypothesis": "give them their choice between a fine or an official whipping",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0453",
+ "category": "phonetic_balance",
+ "text": "They must have been swept away by the chaotic currents.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0453.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "They must have been swept away by the chaotic currents.",
+ "raw": {
+ "normalized_reference": "they must have been swept away by the chaotic currents.",
+ "normalized_hypothesis": "they must have been swept away by the chaotic currents.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "they must have been swept away by the chaotic currents",
+ "normalized_hypothesis": "they must have been swept away by the chaotic currents",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0454",
+ "category": "phonetic_balance",
+ "text": "You yellow giant thing of the frost.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0454.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "You yellow giant thing of the frost.",
+ "raw": {
+ "normalized_reference": "you yellow giant thing of the frost.",
+ "normalized_hypothesis": "you yellow giant thing of the frost.",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "you yellow giant thing of the frost",
+ "normalized_hypothesis": "you yellow giant thing of the frost",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0455",
+ "category": "phonetic_balance",
+ "text": "He had peeled off his shirt and was wildly waving it.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0455.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He had peeled off his shirt and was wildly waving it.",
+ "raw": {
+ "normalized_reference": "he had peeled off his shirt and was wildly waving it.",
+ "normalized_hypothesis": "he had peeled off his shirt and was wildly waving it.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he had peeled off his shirt and was wildly waving it",
+ "normalized_hypothesis": "he had peeled off his shirt and was wildly waving it",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0456",
+ "category": "phonetic_balance",
+ "text": "To these he gave castor oil.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0456.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "To these he gave castor oil.",
+ "raw": {
+ "normalized_reference": "to these he gave castor oil.",
+ "normalized_hypothesis": "to these he gave castor oil.",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "to these he gave castor oil",
+ "normalized_hypothesis": "to these he gave castor oil",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0457",
+ "category": "phonetic_balance",
+ "text": "Tudor surveyed him with withering disgust.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0457.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Tudor surveyed him with withering disgust.",
+ "raw": {
+ "normalized_reference": "tudor surveyed him with withering disgust.",
+ "normalized_hypothesis": "tudor surveyed him with withering disgust.",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "tudor surveyed him with withering disgust",
+ "normalized_hypothesis": "tudor surveyed him with withering disgust",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0458",
+ "category": "phonetic_balance",
+ "text": "After all, it was simply a mistake in judgment.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0458.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "After all, it was simply a mistake in judgment.",
+ "raw": {
+ "normalized_reference": "after all, it was simply a mistake in judgment.",
+ "normalized_hypothesis": "after all, it was simply a mistake in judgment.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "after all it was simply a mistake in judgment",
+ "normalized_hypothesis": "after all it was simply a mistake in judgment",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0459",
+ "category": "phonetic_balance",
+ "text": "That is the strange part of it.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0459.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "That is the strange part of it.",
+ "raw": {
+ "normalized_reference": "that is the strange part of it.",
+ "normalized_hypothesis": "that is the strange part of it.",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "that is the strange part of it",
+ "normalized_hypothesis": "that is the strange part of it",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0460",
+ "category": "phonetic_balance",
+ "text": "Not a wheel moved in his empire.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0460.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Not a wheel moved in his empire.",
+ "raw": {
+ "normalized_reference": "not a wheel moved in his empire.",
+ "normalized_hypothesis": "not a wheel moved in his empire.",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "not a wheel moved in his empire",
+ "normalized_hypothesis": "not a wheel moved in his empire",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0461",
+ "category": "phonetic_balance",
+ "text": "Its diameter was not more than two hundred yards.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0461.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Its diameter was not more than 200 yards.",
+ "raw": {
+ "normalized_reference": "its diameter was not more than two hundred yards.",
+ "normalized_hypothesis": "its diameter was not more than 200 yards.",
+ "errors": 2,
+ "words": 9,
+ "wer": 0.2222222222222222
+ },
+ "semantic": {
+ "normalized_reference": "its diameter was not more than 200 yards",
+ "normalized_hypothesis": "its diameter was not more than 200 yards",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0462",
+ "category": "phonetic_balance",
+ "text": "Nimrod replied, with a slight manifestation of sensitiveness.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0462.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Nimrod replied with a slight manifestation of sensitiveness.",
+ "raw": {
+ "normalized_reference": "nimrod replied, with a slight manifestation of sensitiveness.",
+ "normalized_hypothesis": "nimrod replied with a slight manifestation of sensitiveness.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "nimrod replied with a slight manifestation of sensitiveness",
+ "normalized_hypothesis": "nimrod replied with a slight manifestation of sensitiveness",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0463",
+ "category": "phonetic_balance",
+ "text": "I tried to read George Moore last night, and was dreadfully bored.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0463.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I tried to read George Moore last night and was dreadfully bored.",
+ "raw": {
+ "normalized_reference": "i tried to read george moore last night, and was dreadfully bored.",
+ "normalized_hypothesis": "i tried to read george moore last night and was dreadfully bored.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i tried to read george moore last night and was dreadfully bored",
+ "normalized_hypothesis": "i tried to read george moore last night and was dreadfully bored",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0464",
+ "category": "phonetic_balance",
+ "text": "Yes, it was a man who asked, a stranger.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0464.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Yes, it was a man who asked a stranger.",
+ "raw": {
+ "normalized_reference": "yes, it was a man who asked, a stranger.",
+ "normalized_hypothesis": "yes, it was a man who asked a stranger.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "yes it was a man who asked a stranger",
+ "normalized_hypothesis": "yes it was a man who asked a stranger",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0465",
+ "category": "phonetic_balance",
+ "text": "We were now good friends.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0465.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "We were now good friends.",
+ "raw": {
+ "normalized_reference": "we were now good friends.",
+ "normalized_hypothesis": "we were now good friends.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "we were now good friends",
+ "normalized_hypothesis": "we were now good friends",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0466",
+ "category": "phonetic_balance",
+ "text": "Yet, in accordance with Ernest's test of truth, it worked.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0466.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Yet in accordance with Ernest's test of truth, it worked.",
+ "raw": {
+ "normalized_reference": "yet, in accordance with ernest's test of truth, it worked.",
+ "normalized_hypothesis": "yet in accordance with ernest's test of truth, it worked.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "yet in accordance with ernest is test of truth it worked",
+ "normalized_hypothesis": "yet in accordance with ernest is test of truth it worked",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0467",
+ "category": "phonetic_balance",
+ "text": "It's the strap hangers that'll keep us from going under.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0467.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It's the strap hangers that'll keep us from going under.",
+ "raw": {
+ "normalized_reference": "it's the strap hangers that'll keep us from going under.",
+ "normalized_hypothesis": "it's the strap hangers that'll keep us from going under.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it is the strap hangers that will keep us from going under",
+ "normalized_hypothesis": "it is the strap hangers that will keep us from going under",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0468",
+ "category": "phonetic_balance",
+ "text": "I never allow what can't be changed to annoy me.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0468.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I never allow what can't be changed to annoy me.",
+ "raw": {
+ "normalized_reference": "i never allow what can't be changed to annoy me.",
+ "normalized_hypothesis": "i never allow what can't be changed to annoy me.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i never allow what can not be changed to annoy me",
+ "normalized_hypothesis": "i never allow what can not be changed to annoy me",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0469",
+ "category": "phonetic_balance",
+ "text": "After all superfluous flesh is gone what is left is stringy and resistant.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0469.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "After all superfluous flesh is gone, what is left is stringy and resistant.",
+ "raw": {
+ "normalized_reference": "after all superfluous flesh is gone what is left is stringy and resistant.",
+ "normalized_hypothesis": "after all superfluous flesh is gone, what is left is stringy and resistant.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "after all superfluous flesh is gone what is left is stringy and resistant",
+ "normalized_hypothesis": "after all superfluous flesh is gone what is left is stringy and resistant",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0470",
+ "category": "phonetic_balance",
+ "text": "Then he shouted, Shut up.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0470.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Then he shouted, shut up.",
+ "raw": {
+ "normalized_reference": "then he shouted, shut up.",
+ "normalized_hypothesis": "then he shouted, shut up.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "then he shouted shut up",
+ "normalized_hypothesis": "then he shouted shut up",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0471",
+ "category": "phonetic_balance",
+ "text": "He had worshiped her, as Dante might have worshiped Beatrice.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0471.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He had worshipped her as Dante might have worshipped Beatrice.",
+ "raw": {
+ "normalized_reference": "he had worshiped her, as dante might have worshiped beatrice.",
+ "normalized_hypothesis": "he had worshipped her as dante might have worshipped beatrice.",
+ "errors": 2,
+ "words": 10,
+ "wer": 0.2
+ },
+ "semantic": {
+ "normalized_reference": "he had worshiped her as dante might have worshiped beatrice",
+ "normalized_hypothesis": "he had worshipped her as dante might have worshipped beatrice",
+ "errors": 2,
+ "words": 10,
+ "wer": 0.2
+ }
+ },
+ {
+ "id": "release500_0472",
+ "category": "phonetic_balance",
+ "text": "He looked like one who had passed through an uncomfortable hour or two.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0472.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He looked like one who had passed through an uncomfortable hour or two.",
+ "raw": {
+ "normalized_reference": "he looked like one who had passed through an uncomfortable hour or two.",
+ "normalized_hypothesis": "he looked like one who had passed through an uncomfortable hour or two.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he looked like one who had passed through an uncomfortable hour or 2",
+ "normalized_hypothesis": "he looked like one who had passed through an uncomfortable hour or 2",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0473",
+ "category": "phonetic_balance",
+ "text": "The history of our westward-faring race is written in it.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0473.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The history of our westward-faring race is written, in it,",
+ "raw": {
+ "normalized_reference": "the history of our westward-faring race is written in it.",
+ "normalized_hypothesis": "the history of our westward-faring race is written, in it,",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the history of our westward faring race is written in it",
+ "normalized_hypothesis": "the history of our westward faring race is written in it",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0474",
+ "category": "phonetic_balance",
+ "text": "He may desire to escape pain, or to enjoy pleasure.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0474.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He may desire to escape pain or to enjoy pleasure.",
+ "raw": {
+ "normalized_reference": "he may desire to escape pain, or to enjoy pleasure.",
+ "normalized_hypothesis": "he may desire to escape pain or to enjoy pleasure.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he may desire to escape pain or to enjoy pleasure",
+ "normalized_hypothesis": "he may desire to escape pain or to enjoy pleasure",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0475",
+ "category": "phonetic_balance",
+ "text": "And the air was growing chilly.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0475.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "And the air was growing chilly.",
+ "raw": {
+ "normalized_reference": "and the air was growing chilly.",
+ "normalized_hypothesis": "and the air was growing chilly.",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "and the air was growing chilly",
+ "normalized_hypothesis": "and the air was growing chilly",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0476",
+ "category": "phonetic_balance",
+ "text": "Dig in; you're sure good, was Daylight's answer.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0476.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Diggin' you're sure good was daylight's answer.",
+ "raw": {
+ "normalized_reference": "dig in; you're sure good, was daylight's answer.",
+ "normalized_hypothesis": "diggin' you're sure good was daylight's answer.",
+ "errors": 2,
+ "words": 8,
+ "wer": 0.25
+ },
+ "semantic": {
+ "normalized_reference": "dig in you are sure good was daylight is answer",
+ "normalized_hypothesis": "diggin you are sure good was daylight is answer",
+ "errors": 2,
+ "words": 10,
+ "wer": 0.2
+ }
+ },
+ {
+ "id": "release500_0477",
+ "category": "phonetic_balance",
+ "text": "My idea was that he would have more influence over the natives.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0477.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "My idea was that he would have more influence over the natives.",
+ "raw": {
+ "normalized_reference": "my idea was that he would have more influence over the natives.",
+ "normalized_hypothesis": "my idea was that he would have more influence over the natives.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "my idea was that he would have more influence over the natives",
+ "normalized_hypothesis": "my idea was that he would have more influence over the natives",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0478",
+ "category": "phonetic_balance",
+ "text": "Now run along, and tell them to hurry.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0478.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Now run along and tell them to hurry.",
+ "raw": {
+ "normalized_reference": "now run along, and tell them to hurry.",
+ "normalized_hypothesis": "now run along and tell them to hurry.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "now run along and tell them to hurry",
+ "normalized_hypothesis": "now run along and tell them to hurry",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0479",
+ "category": "phonetic_balance",
+ "text": "I have no idea, replied Philip.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0479.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I have no idea, replied Philip.",
+ "raw": {
+ "normalized_reference": "i have no idea, replied philip.",
+ "normalized_hypothesis": "i have no idea, replied philip.",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i have no idea replied philip",
+ "normalized_hypothesis": "i have no idea replied philip",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0480",
+ "category": "phonetic_balance",
+ "text": "For the rest, he was a mere automaton.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0480.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "For the rest, he was a mere automaton.",
+ "raw": {
+ "normalized_reference": "for the rest, he was a mere automaton.",
+ "normalized_hypothesis": "for the rest, he was a mere automaton.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "for the rest he was a mere automaton",
+ "normalized_hypothesis": "for the rest he was a mere automaton",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0481",
+ "category": "phonetic_balance",
+ "text": "Why, the average review is more nauseating than cod liver oil.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0481.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Why the average review is more nauseating than cod liver oil.",
+ "raw": {
+ "normalized_reference": "why, the average review is more nauseating than cod liver oil.",
+ "normalized_hypothesis": "why the average review is more nauseating than cod liver oil.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "why the average review is more nauseating than cod liver oil",
+ "normalized_hypothesis": "why the average review is more nauseating than cod liver oil",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0482",
+ "category": "phonetic_balance",
+ "text": "I have been robbed, sir, I amended.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0482.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I have been robbed, sir, I am mended.",
+ "raw": {
+ "normalized_reference": "i have been robbed, sir, i amended.",
+ "normalized_hypothesis": "i have been robbed, sir, i am mended.",
+ "errors": 2,
+ "words": 7,
+ "wer": 0.2857142857142857
+ },
+ "semantic": {
+ "normalized_reference": "i have been robbed sir i amended",
+ "normalized_hypothesis": "i have been robbed sir i am mended",
+ "errors": 2,
+ "words": 7,
+ "wer": 0.2857142857142857
+ }
+ },
+ {
+ "id": "release500_0483",
+ "category": "phonetic_balance",
+ "text": "He moved away as quietly as he had come.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0483.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He moved away as quietly as he had come.",
+ "raw": {
+ "normalized_reference": "he moved away as quietly as he had come.",
+ "normalized_hypothesis": "he moved away as quietly as he had come.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he moved away as quietly as he had come",
+ "normalized_hypothesis": "he moved away as quietly as he had come",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0484",
+ "category": "phonetic_balance",
+ "text": "A trickle of fresh blood ran over his face.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0484.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "A trickle of fresh blood ran over his face.",
+ "raw": {
+ "normalized_reference": "a trickle of fresh blood ran over his face.",
+ "normalized_hypothesis": "a trickle of fresh blood ran over his face.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "a trickle of fresh blood ran over his face",
+ "normalized_hypothesis": "a trickle of fresh blood ran over his face",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0485",
+ "category": "phonetic_balance",
+ "text": "Suppose you saw me at work through the window.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0485.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Suppose you saw me at work through the window.",
+ "raw": {
+ "normalized_reference": "suppose you saw me at work through the window.",
+ "normalized_hypothesis": "suppose you saw me at work through the window.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "suppose you saw me at work through the window",
+ "normalized_hypothesis": "suppose you saw me at work through the window",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0486",
+ "category": "phonetic_balance",
+ "text": "But who was Eileen's double.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0486.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "But who was Eileen's double?",
+ "raw": {
+ "normalized_reference": "but who was eileen's double.",
+ "normalized_hypothesis": "but who was eileen's double?",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "but who was eileen is double",
+ "normalized_hypothesis": "but who was eileen is double",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0487",
+ "category": "phonetic_balance",
+ "text": "Darkness hid him from Jeanne.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0487.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Darkness hid him from Gene.",
+ "raw": {
+ "normalized_reference": "darkness hid him from jeanne.",
+ "normalized_hypothesis": "darkness hid him from gene.",
+ "errors": 1,
+ "words": 5,
+ "wer": 0.2
+ },
+ "semantic": {
+ "normalized_reference": "darkness hid him from jeanne",
+ "normalized_hypothesis": "darkness hid him from gene",
+ "errors": 1,
+ "words": 5,
+ "wer": 0.2
+ }
+ },
+ {
+ "id": "release500_0488",
+ "category": "phonetic_balance",
+ "text": "By this time Charley was as enraged as the Greek.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0488.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "By this time Charlie was as enraged as the Greek.",
+ "raw": {
+ "normalized_reference": "by this time charley was as enraged as the greek.",
+ "normalized_hypothesis": "by this time charlie was as enraged as the greek.",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ },
+ "semantic": {
+ "normalized_reference": "by this time charley was as enraged as the greek",
+ "normalized_hypothesis": "by this time charlie was as enraged as the greek",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ }
+ },
+ {
+ "id": "release500_0489",
+ "category": "phonetic_balance",
+ "text": "I am writing these lines in Honolulu, Hawaii.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0489.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I am writing these lines in Honolulu, Hawaii.",
+ "raw": {
+ "normalized_reference": "i am writing these lines in honolulu, hawaii.",
+ "normalized_hypothesis": "i am writing these lines in honolulu, hawaii.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i am writing these lines in honolulu hawaii",
+ "normalized_hypothesis": "i am writing these lines in honolulu hawaii",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0490",
+ "category": "phonetic_balance",
+ "text": "Nobody knows how the natives got them.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0490.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Nobody knows how the natives got them.",
+ "raw": {
+ "normalized_reference": "nobody knows how the natives got them.",
+ "normalized_hypothesis": "nobody knows how the natives got them.",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "nobody knows how the natives got them",
+ "normalized_hypothesis": "nobody knows how the natives got them",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0491",
+ "category": "phonetic_balance",
+ "text": "And how would we ever find ourselves.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0491.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "And how would we ever find ourselves?",
+ "raw": {
+ "normalized_reference": "and how would we ever find ourselves.",
+ "normalized_hypothesis": "and how would we ever find ourselves?",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "and how would we ever find ourselves",
+ "normalized_hypothesis": "and how would we ever find ourselves",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0492",
+ "category": "phonetic_balance",
+ "text": "They are coming ashore, whoever they are.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0492.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "They are coming ashore, whoever they are.",
+ "raw": {
+ "normalized_reference": "they are coming ashore, whoever they are.",
+ "normalized_hypothesis": "they are coming ashore, whoever they are.",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "they are coming ashore whoever they are",
+ "normalized_hypothesis": "they are coming ashore whoever they are",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0493",
+ "category": "phonetic_balance",
+ "text": "This is my fifth voyage.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0493.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "This is my fifth voyage.",
+ "raw": {
+ "normalized_reference": "this is my fifth voyage.",
+ "normalized_hypothesis": "this is my fifth voyage.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "this is my 5th voyage",
+ "normalized_hypothesis": "this is my 5th voyage",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0494",
+ "category": "phonetic_balance",
+ "text": "And yet if she came he had no words to say.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0494.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "And yet if she came he had no words to say.",
+ "raw": {
+ "normalized_reference": "and yet if she came he had no words to say.",
+ "normalized_hypothesis": "and yet if she came he had no words to say.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "and yet if she came he had no words to say",
+ "normalized_hypothesis": "and yet if she came he had no words to say",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0495",
+ "category": "phonetic_balance",
+ "text": "You must sleep, he urged.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0495.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "You must sleep, he urged.",
+ "raw": {
+ "normalized_reference": "you must sleep, he urged.",
+ "normalized_hypothesis": "you must sleep, he urged.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "you must sleep he urged",
+ "normalized_hypothesis": "you must sleep he urged",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0496",
+ "category": "phonetic_balance",
+ "text": "Could the incident have anything to do with Jeanne and Pierre.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0496.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Could the incident have anything to do with Jean and Pierre?",
+ "raw": {
+ "normalized_reference": "could the incident have anything to do with jeanne and pierre.",
+ "normalized_hypothesis": "could the incident have anything to do with jean and pierre?",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ },
+ "semantic": {
+ "normalized_reference": "could the incident have anything to do with jeanne and pierre",
+ "normalized_hypothesis": "could the incident have anything to do with jean and pierre",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ }
+ },
+ {
+ "id": "release500_0497",
+ "category": "phonetic_balance",
+ "text": "Each day she became a more vital part of him.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0497.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Each day she became a more vital part of him.",
+ "raw": {
+ "normalized_reference": "each day she became a more vital part of him.",
+ "normalized_hypothesis": "each day she became a more vital part of him.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "each day she became a more vital part of him",
+ "normalized_hypothesis": "each day she became a more vital part of him",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0498",
+ "category": "phonetic_balance",
+ "text": "But why continue the tirade, for tirade it was.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0498.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "But why continue the tirade for tirade it was?",
+ "raw": {
+ "normalized_reference": "but why continue the tirade, for tirade it was.",
+ "normalized_hypothesis": "but why continue the tirade for tirade it was?",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "but why continue the tirade for tirade it was",
+ "normalized_hypothesis": "but why continue the tirade for tirade it was",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0499",
+ "category": "phonetic_balance",
+ "text": "That is what distinguishes all of us from the lower animals.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0499.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "That is what distinguishes all of us from the lower animals.",
+ "raw": {
+ "normalized_reference": "that is what distinguishes all of us from the lower animals.",
+ "normalized_hypothesis": "that is what distinguishes all of us from the lower animals.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "that is what distinguishes all of us from the lower animals",
+ "normalized_hypothesis": "that is what distinguishes all of us from the lower animals",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0500",
+ "category": "phonetic_balance",
+ "text": "There was something pathetic in the girl's attitude now.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0500.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "There was something pathetic in the girl's attitude now.",
+ "raw": {
+ "normalized_reference": "there was something pathetic in the girl's attitude now.",
+ "normalized_hypothesis": "there was something pathetic in the girl's attitude now.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "there was something pathetic in the girl is attitude now",
+ "normalized_hypothesis": "there was something pathetic in the girl is attitude now",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0001",
+ "category": "conversational",
+ "text": "It wasn't until later that I realized what had actually happened.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0001.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It wasn't until later that I realized what had actually happened.",
+ "raw": {
+ "normalized_reference": "it wasn't until later that i realized what had actually happened.",
+ "normalized_hypothesis": "it wasn't until later that i realized what had actually happened.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it was not until later that i realized what had actually happened",
+ "normalized_hypothesis": "it was not until later that i realized what had actually happened",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0002",
+ "category": "descriptive",
+ "text": "The light at that time of afternoon is perfect for about twenty minutes.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0002.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The light at that time of afternoon is perfect for about 20 minutes.",
+ "raw": {
+ "normalized_reference": "the light at that time of afternoon is perfect for about twenty minutes.",
+ "normalized_hypothesis": "the light at that time of afternoon is perfect for about 20 minutes.",
+ "errors": 1,
+ "words": 13,
+ "wer": 0.07692307692307693
+ },
+ "semantic": {
+ "normalized_reference": "the light at that time of afternoon is perfect for about 20 minutes",
+ "normalized_hypothesis": "the light at that time of afternoon is perfect for about 20 minutes",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0003",
+ "category": "emotional",
+ "text": "That was the scariest thing I've ever done, and also maybe the best.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0003.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "That was the scariest thing I've ever done, and also maybe the best.",
+ "raw": {
+ "normalized_reference": "that was the scariest thing i've ever done, and also maybe the best.",
+ "normalized_hypothesis": "that was the scariest thing i've ever done, and also maybe the best.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "that was the scariest thing i have ever done and also maybe the best",
+ "normalized_hypothesis": "that was the scariest thing i have ever done and also maybe the best",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0004",
+ "category": "homographs",
+ "text": "After I record the record, please present the present to the bass player.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0004.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "After I record the record, please present the present to the bass player.",
+ "raw": {
+ "normalized_reference": "after i record the record, please present the present to the bass player.",
+ "normalized_hypothesis": "after i record the record, please present the present to the bass player.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "after i record the record please present the present to the bass player",
+ "normalized_hypothesis": "after i record the record please present the present to the bass player",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0005",
+ "category": "long",
+ "text": "If I'm being completely honest, I had no idea what I was doing for most of it, and I think that might be the most normal thing about the whole experience.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0005.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "If I'm being completely honest, I had no idea what I was doing for most of it, and I think that might be the most normal thing about the whole experience.",
+ "raw": {
+ "normalized_reference": "if i'm being completely honest, i had no idea what i was doing for most of it, and i think that might be the most normal thing about the whole experience.",
+ "normalized_hypothesis": "if i'm being completely honest, i had no idea what i was doing for most of it, and i think that might be the most normal thing about the whole experience.",
+ "errors": 0,
+ "words": 31,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "if i am being completely honest i had no idea what i was doing for most of it and i think that might be the most normal thing about the whole experience",
+ "normalized_hypothesis": "if i am being completely honest i had no idea what i was doing for most of it and i think that might be the most normal thing about the whole experience",
+ "errors": 0,
+ "words": 32,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0006",
+ "category": "long_range",
+ "text": "The committee postponed its decision after discovering that three pages of the report had been printed in the wrong order.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0006.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The committee postponed its decision after discovering that three pages of the report had been printed in the wrong order.",
+ "raw": {
+ "normalized_reference": "the committee postponed its decision after discovering that three pages of the report had been printed in the wrong order.",
+ "normalized_hypothesis": "the committee postponed its decision after discovering that three pages of the report had been printed in the wrong order.",
+ "errors": 0,
+ "words": 20,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the committee postponed its decision after discovering that 3 pages of the report had been printed in the wrong order",
+ "normalized_hypothesis": "the committee postponed its decision after discovering that 3 pages of the report had been printed in the wrong order",
+ "errors": 0,
+ "words": 20,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0007",
+ "category": "names_places",
+ "text": "Gwendolyn photographed the eucalyptus trees outside Ljubljana.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0007.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Gwendolyn photographed the eucalyptus trees outside Ljubljana.",
+ "raw": {
+ "normalized_reference": "gwendolyn photographed the eucalyptus trees outside ljubljana.",
+ "normalized_hypothesis": "gwendolyn photographed the eucalyptus trees outside ljubljana.",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "gwendolyn photographed the eucalyptus trees outside ljubljana",
+ "normalized_hypothesis": "gwendolyn photographed the eucalyptus trees outside ljubljana",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0008",
+ "category": "narrative",
+ "text": "He had this theory, and the more time passes, the more I think he was right.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0008.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He had this theory, and the more time passes, the more I think he was right.",
+ "raw": {
+ "normalized_reference": "he had this theory, and the more time passes, the more i think he was right.",
+ "normalized_hypothesis": "he had this theory, and the more time passes, the more i think he was right.",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he had this theory and the more time passes the more i think he was right",
+ "normalized_hypothesis": "he had this theory and the more time passes the more i think he was right",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0009",
+ "category": "numbers",
+ "text": "The package weighs twelve point six kilograms and arrived on July twenty-first.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0009.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The package weighs 12.6 kilograms and arrived on July 21.",
+ "raw": {
+ "normalized_reference": "the package weighs twelve point six kilograms and arrived on july twenty-first.",
+ "normalized_hypothesis": "the package weighs 12.6 kilograms and arrived on july 21.",
+ "errors": 5,
+ "words": 13,
+ "wer": 0.38461538461538464
+ },
+ "semantic": {
+ "normalized_reference": "the package weighs 12.6 kilograms and arrived on july 21st",
+ "normalized_hypothesis": "the package weighs 12.6 kilograms and arrived on july 21",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ }
+ },
+ {
+ "id": "release500_0010",
+ "category": "punctuation",
+ "text": "First, close the window; second, turn off the lamp; finally, lock the door.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0010.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "First, close the window. Second, turn off the lamp. Finally, lock the door.",
+ "raw": {
+ "normalized_reference": "first, close the window; second, turn off the lamp; finally, lock the door.",
+ "normalized_hypothesis": "first, close the window. second, turn off the lamp. finally, lock the door.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "1st close the window 2nd turn off the lamp finally lock the door",
+ "normalized_hypothesis": "1st close the window 2nd turn off the lamp finally lock the door",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0011",
+ "category": "question",
+ "text": "When did you first realize something was wrong?",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0011.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "When did you first realize something was wrong?",
+ "raw": {
+ "normalized_reference": "when did you first realize something was wrong?",
+ "normalized_hypothesis": "when did you first realize something was wrong?",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "when did you 1st realize something was wrong",
+ "normalized_hypothesis": "when did you 1st realize something was wrong",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0012",
+ "category": "rare_words",
+ "text": "The archaeologist catalogued a Byzantine reliquary and several anachronistic fragments.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0012.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The archaeologists cataloged a Byzantine reliquary and several anachronistic fragments.",
+ "raw": {
+ "normalized_reference": "the archaeologist catalogued a byzantine reliquary and several anachronistic fragments.",
+ "normalized_hypothesis": "the archaeologists cataloged a byzantine reliquary and several anachronistic fragments.",
+ "errors": 2,
+ "words": 10,
+ "wer": 0.2
+ },
+ "semantic": {
+ "normalized_reference": "the archaeologist catalogued a byzantine reliquary and several anachronistic fragments",
+ "normalized_hypothesis": "the archaeologists cataloged a byzantine reliquary and several anachronistic fragments",
+ "errors": 2,
+ "words": 10,
+ "wer": 0.2
+ }
+ },
+ {
+ "id": "release500_0013",
+ "category": "short",
+ "text": "I have no idea.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0013.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I have no idea.",
+ "raw": {
+ "normalized_reference": "i have no idea.",
+ "normalized_hypothesis": "i have no idea.",
+ "errors": 0,
+ "words": 4,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i have no idea",
+ "normalized_hypothesis": "i have no idea",
+ "errors": 0,
+ "words": 4,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0014",
+ "category": "technical",
+ "text": "The system runs on three core components that all have to stay in sync.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0014.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The system runs on three core components that all have to stay in sync.",
+ "raw": {
+ "normalized_reference": "the system runs on three core components that all have to stay in sync.",
+ "normalized_hypothesis": "the system runs on three core components that all have to stay in sync.",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the system runs on 3 core components that all have to stay in sync",
+ "normalized_hypothesis": "the system runs on 3 core components that all have to stay in sync",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0015",
+ "category": "conversational",
+ "text": "We improvised the entire second half and somehow nobody noticed.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0015.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "We improvised the entire second half and somehow nobody noticed.",
+ "raw": {
+ "normalized_reference": "we improvised the entire second half and somehow nobody noticed.",
+ "normalized_hypothesis": "we improvised the entire second half and somehow nobody noticed.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "we improvised the entire 2nd half and somehow nobody noticed",
+ "normalized_hypothesis": "we improvised the entire 2nd half and somehow nobody noticed",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0016",
+ "category": "descriptive",
+ "text": "There's a smell in old buildings that's impossible to describe but immediately familiar.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0016.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "There's a smell in old buildings that's impossible to describe but immediately familiar.",
+ "raw": {
+ "normalized_reference": "there's a smell in old buildings that's impossible to describe but immediately familiar.",
+ "normalized_hypothesis": "there's a smell in old buildings that's impossible to describe but immediately familiar.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "there is a smell in old buildings that is impossible to describe but immediately familiar",
+ "normalized_hypothesis": "there is a smell in old buildings that is impossible to describe but immediately familiar",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0017",
+ "category": "emotional",
+ "text": "I specifically asked for one thing. One thing. And it still didn't happen.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0017.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I specifically asked for one thing. One thing. And it still didn't happen.",
+ "raw": {
+ "normalized_reference": "i specifically asked for one thing. one thing. and it still didn't happen.",
+ "normalized_hypothesis": "i specifically asked for one thing. one thing. and it still didn't happen.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i specifically asked for one thing one thing and it still did not happen",
+ "normalized_hypothesis": "i specifically asked for one thing one thing and it still did not happen",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0018",
+ "category": "homographs",
+ "text": "They refuse to place the refuse beside the produce aisle.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0018.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "They refused to place the refuse beside the produce aisle.",
+ "raw": {
+ "normalized_reference": "they refuse to place the refuse beside the produce aisle.",
+ "normalized_hypothesis": "they refused to place the refuse beside the produce aisle.",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ },
+ "semantic": {
+ "normalized_reference": "they refuse to place the refuse beside the produce aisle",
+ "normalized_hypothesis": "they refused to place the refuse beside the produce aisle",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ }
+ },
+ {
+ "id": "release500_0019",
+ "category": "long",
+ "text": "There were days last year when I genuinely didn't know how I was going to get through it, and somehow I did, and I still don't fully understand how.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0019.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "There were days last year when I genuinely didn't know how I was going to get through it, and somehow I did, and I still don't fully understand how.",
+ "raw": {
+ "normalized_reference": "there were days last year when i genuinely didn't know how i was going to get through it, and somehow i did, and i still don't fully understand how.",
+ "normalized_hypothesis": "there were days last year when i genuinely didn't know how i was going to get through it, and somehow i did, and i still don't fully understand how.",
+ "errors": 0,
+ "words": 29,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "there were days last year when i genuinely did not know how i was going to get through it and somehow i did and i still do not fully understand how",
+ "normalized_hypothesis": "there were days last year when i genuinely did not know how i was going to get through it and somehow i did and i still do not fully understand how",
+ "errors": 0,
+ "words": 31,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0020",
+ "category": "long_range",
+ "text": "Before announcing the winner, the judge thanked the volunteers who had worked through the storm to repair the damaged stage.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0020.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Before announcing the winner, the judge thanked the volunteers who had worked through the storm to repair the damaged stage.",
+ "raw": {
+ "normalized_reference": "before announcing the winner, the judge thanked the volunteers who had worked through the storm to repair the damaged stage.",
+ "normalized_hypothesis": "before announcing the winner, the judge thanked the volunteers who had worked through the storm to repair the damaged stage.",
+ "errors": 0,
+ "words": 20,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "before announcing the winner the judge thanked the volunteers who had worked through the storm to repair the damaged stage",
+ "normalized_hypothesis": "before announcing the winner the judge thanked the volunteers who had worked through the storm to repair the damaged stage",
+ "errors": 0,
+ "words": 20,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0021",
+ "category": "names_places",
+ "text": "Professor Nguyen mailed the fluorescent poster to Reykjavik on Thursday.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0021.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Professor Nguyen mailed the fluorescent poster to Reykjavik on Thursday.",
+ "raw": {
+ "normalized_reference": "professor nguyen mailed the fluorescent poster to reykjavik on thursday.",
+ "normalized_hypothesis": "professor nguyen mailed the fluorescent poster to reykjavik on thursday.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "professor nguyen mailed the fluorescent poster to reykjavik on thursday",
+ "normalized_hypothesis": "professor nguyen mailed the fluorescent poster to reykjavik on thursday",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0022",
+ "category": "narrative",
+ "text": "The plan was simple. The execution was anything but.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0022.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The plan was simple. The execution was anything but.",
+ "raw": {
+ "normalized_reference": "the plan was simple. the execution was anything but.",
+ "normalized_hypothesis": "the plan was simple. the execution was anything but.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the plan was simple the execution was anything but",
+ "normalized_hypothesis": "the plan was simple the execution was anything but",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0023",
+ "category": "numbers",
+ "text": "Her score improved from eighty-two percent to ninety-seven percent in two weeks.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0023.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Her score improved from 82% to 97% in two weeks.",
+ "raw": {
+ "normalized_reference": "her score improved from eighty-two percent to ninety-seven percent in two weeks.",
+ "normalized_hypothesis": "her score improved from 82% to 97% in two weeks.",
+ "errors": 6,
+ "words": 14,
+ "wer": 0.42857142857142855
+ },
+ "semantic": {
+ "normalized_reference": "her score improved from 82% to 97% in 2 weeks",
+ "normalized_hypothesis": "her score improved from 82% to 97% in 2 weeks",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0024",
+ "category": "punctuation",
+ "text": "Why would anyone pack spoons, batteries, and a violin for one short trip?",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0024.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Why would anyone pack spoons, batteries, and a violin for one short trip?",
+ "raw": {
+ "normalized_reference": "why would anyone pack spoons, batteries, and a violin for one short trip?",
+ "normalized_hypothesis": "why would anyone pack spoons, batteries, and a violin for one short trip?",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "why would anyone pack spoons batteries and a violin for one short trip",
+ "normalized_hypothesis": "why would anyone pack spoons batteries and a violin for one short trip",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0025",
+ "category": "question",
+ "text": "How long have you been sitting on this?",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0025.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "How long have you been sitting on this?",
+ "raw": {
+ "normalized_reference": "how long have you been sitting on this?",
+ "normalized_hypothesis": "how long have you been sitting on this?",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "how long have you been sitting on this",
+ "normalized_hypothesis": "how long have you been sitting on this",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0026",
+ "category": "rare_words",
+ "text": "A mischievous chameleon rested beside the chrysanthemum and the bougainvillea.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0026.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "A mischievous chameleon rested beside the chrysanthemum and the bougainvillea.",
+ "raw": {
+ "normalized_reference": "a mischievous chameleon rested beside the chrysanthemum and the bougainvillea.",
+ "normalized_hypothesis": "a mischievous chameleon rested beside the chrysanthemum and the bougainvillea.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "a mischievous chameleon rested beside the chrysanthemum and the bougainvillea",
+ "normalized_hypothesis": "a mischievous chameleon rested beside the chrysanthemum and the bougainvillea",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0027",
+ "category": "short",
+ "text": "We don't talk about that.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0027.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "We don't talk about that.",
+ "raw": {
+ "normalized_reference": "we don't talk about that.",
+ "normalized_hypothesis": "we don't talk about that.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "we do not talk about that",
+ "normalized_hypothesis": "we do not talk about that",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0028",
+ "category": "technical",
+ "text": "Hash the UTF eight payload with S H A two fifty-six before opening the checkpoint.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0028.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Hash the UTF-8 payload with SH-256 before opening the checkpoint.",
+ "raw": {
+ "normalized_reference": "hash the utf eight payload with s h a two fifty-six before opening the checkpoint.",
+ "normalized_hypothesis": "hash the utf-8 payload with sh-256 before opening the checkpoint.",
+ "errors": 7,
+ "words": 16,
+ "wer": 0.4375
+ },
+ "semantic": {
+ "normalized_reference": "hash the utf 8 payload with s h a 256 before opening the checkpoint",
+ "normalized_hypothesis": "hash the utf 8 payload with sh 256 before opening the checkpoint",
+ "errors": 3,
+ "words": 14,
+ "wer": 0.21428571428571427
+ }
+ },
+ {
+ "id": "release500_0029",
+ "category": "conversational",
+ "text": "I thought I had it all figured out, and then this happened.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0029.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I thought I had it all figured out, and then this happened.",
+ "raw": {
+ "normalized_reference": "i thought i had it all figured out, and then this happened.",
+ "normalized_hypothesis": "i thought i had it all figured out, and then this happened.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i thought i had it all figured out and then this happened",
+ "normalized_hypothesis": "i thought i had it all figured out and then this happened",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0030",
+ "category": "descriptive",
+ "text": "The kind of quiet you only get at three in the morning in a city.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0030.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The kind of quiet you only get at 3 in the morning in a city.",
+ "raw": {
+ "normalized_reference": "the kind of quiet you only get at three in the morning in a city.",
+ "normalized_hypothesis": "the kind of quiet you only get at 3 in the morning in a city.",
+ "errors": 1,
+ "words": 15,
+ "wer": 0.06666666666666667
+ },
+ "semantic": {
+ "normalized_reference": "the kind of quiet you only get at 3 in the morning in a city",
+ "normalized_hypothesis": "the kind of quiet you only get at 3 in the morning in a city",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0031",
+ "category": "emotional",
+ "text": "The joy of it was almost overwhelming, like too much good all at once.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0031.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The joy of it was almost overwhelming, like too much good all at once.",
+ "raw": {
+ "normalized_reference": "the joy of it was almost overwhelming, like too much good all at once.",
+ "normalized_hypothesis": "the joy of it was almost overwhelming, like too much good all at once.",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the joy of it was almost overwhelming like too much good all at once",
+ "normalized_hypothesis": "the joy of it was almost overwhelming like too much good all at once",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0032",
+ "category": "homographs",
+ "text": "Please close the door while I stand close to the microphone.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0032.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Please close the door while I stand close to the microphone.",
+ "raw": {
+ "normalized_reference": "please close the door while i stand close to the microphone.",
+ "normalized_hypothesis": "please close the door while i stand close to the microphone.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "please close the door while i stand close to the microphone",
+ "normalized_hypothesis": "please close the door while i stand close to the microphone",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0033",
+ "category": "long",
+ "text": "I think the most underrated skill in the world is knowing when to stop explaining yourself and just let people draw their own conclusions.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0033.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I think the most underrated skill in the world is knowing when to stop explaining yourself and just let people draw their own conclusions.",
+ "raw": {
+ "normalized_reference": "i think the most underrated skill in the world is knowing when to stop explaining yourself and just let people draw their own conclusions.",
+ "normalized_hypothesis": "i think the most underrated skill in the world is knowing when to stop explaining yourself and just let people draw their own conclusions.",
+ "errors": 0,
+ "words": 24,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i think the most underrated skill in the world is knowing when to stop explaining yourself and just let people draw their own conclusions",
+ "normalized_hypothesis": "i think the most underrated skill in the world is knowing when to stop explaining yourself and just let people draw their own conclusions",
+ "errors": 0,
+ "words": 24,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0034",
+ "category": "long_range",
+ "text": "When the final bus disappeared around the corner, Nadia remembered that her umbrella, notebook, and lunch were still inside the quiet office.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0034.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "When the final bus disappeared around the corner, Nadia remembered that her umbrella, notebook, and lunch were still inside the quiet office.",
+ "raw": {
+ "normalized_reference": "when the final bus disappeared around the corner, nadia remembered that her umbrella, notebook, and lunch were still inside the quiet office.",
+ "normalized_hypothesis": "when the final bus disappeared around the corner, nadia remembered that her umbrella, notebook, and lunch were still inside the quiet office.",
+ "errors": 0,
+ "words": 22,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "when the final bus disappeared around the corner nadia remembered that her umbrella notebook and lunch were still inside the quiet office",
+ "normalized_hypothesis": "when the final bus disappeared around the corner nadia remembered that her umbrella notebook and lunch were still inside the quiet office",
+ "errors": 0,
+ "words": 22,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0035",
+ "category": "names_places",
+ "text": "Aurelia and Joaquin crossed the Schuylkill River before noon.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0035.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Aurelia and Joquin crossed the Shiokul River before noon.",
+ "raw": {
+ "normalized_reference": "aurelia and joaquin crossed the schuylkill river before noon.",
+ "normalized_hypothesis": "aurelia and joquin crossed the shiokul river before noon.",
+ "errors": 2,
+ "words": 9,
+ "wer": 0.2222222222222222
+ },
+ "semantic": {
+ "normalized_reference": "aurelia and joaquin crossed the schuylkill river before noon",
+ "normalized_hypothesis": "aurelia and joquin crossed the shiokul river before noon",
+ "errors": 2,
+ "words": 9,
+ "wer": 0.2222222222222222
+ }
+ },
+ {
+ "id": "release500_0036",
+ "category": "narrative",
+ "text": "The village had a name, but nobody used it. Everyone just called it home.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0036.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The village had a name, but nobody used it. Everyone just called it home.",
+ "raw": {
+ "normalized_reference": "the village had a name, but nobody used it. everyone just called it home.",
+ "normalized_hypothesis": "the village had a name, but nobody used it. everyone just called it home.",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the village had a name but nobody used it everyone just called it home",
+ "normalized_hypothesis": "the village had a name but nobody used it everyone just called it home",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0037",
+ "category": "numbers",
+ "text": "Increase the gain by three decibels, wait five seconds, then reduce it by one.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0037.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Increase the gain by 3 decibels, wait 5 seconds, then reduce it by 1.",
+ "raw": {
+ "normalized_reference": "increase the gain by three decibels, wait five seconds, then reduce it by one.",
+ "normalized_hypothesis": "increase the gain by 3 decibels, wait 5 seconds, then reduce it by 1.",
+ "errors": 3,
+ "words": 14,
+ "wer": 0.21428571428571427
+ },
+ "semantic": {
+ "normalized_reference": "increase the gain by 3 decibels wait 5 seconds then reduce it by one",
+ "normalized_hypothesis": "increase the gain by 3 decibels wait 5 seconds then reduce it by one",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0038",
+ "category": "punctuation",
+ "text": "The result was unexpected: clear, stable, and surprisingly natural.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0038.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The result was unexpected, clear, stable, and surprisingly natural.",
+ "raw": {
+ "normalized_reference": "the result was unexpected: clear, stable, and surprisingly natural.",
+ "normalized_hypothesis": "the result was unexpected, clear, stable, and surprisingly natural.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the result was unexpected clear stable and surprisingly natural",
+ "normalized_hypothesis": "the result was unexpected clear stable and surprisingly natural",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0039",
+ "category": "question",
+ "text": "Who told you that was the right way to do it?",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0039.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Who told you that was the right way to do it?",
+ "raw": {
+ "normalized_reference": "who told you that was the right way to do it?",
+ "normalized_hypothesis": "who told you that was the right way to do it?",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "who told you that was the right way to do it",
+ "normalized_hypothesis": "who told you that was the right way to do it",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0040",
+ "category": "rare_words",
+ "text": "The meticulous cartographer examined an iridescent archipelago through binoculars.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0040.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "the meticulous cartographer examined an iridescent archipelago through binoculars.",
+ "raw": {
+ "normalized_reference": "the meticulous cartographer examined an iridescent archipelago through binoculars.",
+ "normalized_hypothesis": "the meticulous cartographer examined an iridescent archipelago through binoculars.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the meticulous cartographer examined an iridescent archipelago through binoculars",
+ "normalized_hypothesis": "the meticulous cartographer examined an iridescent archipelago through binoculars",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0041",
+ "category": "short",
+ "text": "I mean. Yeah. Kind of.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0041.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I mean, no, I know.",
+ "raw": {
+ "normalized_reference": "i mean. yeah. kind of.",
+ "normalized_hypothesis": "i mean, no, i know.",
+ "errors": 3,
+ "words": 5,
+ "wer": 0.6
+ },
+ "semantic": {
+ "normalized_reference": "i mean yeah kind of",
+ "normalized_hypothesis": "i mean no i know",
+ "errors": 3,
+ "words": 5,
+ "wer": 0.6
+ }
+ },
+ {
+ "id": "release500_0042",
+ "category": "technical",
+ "text": "The logs are your best friend when something behaves unexpectedly.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0042.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The logs are your best friend when something behaves unexpectedly.",
+ "raw": {
+ "normalized_reference": "the logs are your best friend when something behaves unexpectedly.",
+ "normalized_hypothesis": "the logs are your best friend when something behaves unexpectedly.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the logs are your best friend when something behaves unexpectedly",
+ "normalized_hypothesis": "the logs are your best friend when something behaves unexpectedly",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0043",
+ "category": "conversational",
+ "text": "He kept talking and I kept nodding and neither of us was listening.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0043.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He kept talking and I kept nodding and neither of us was listening.",
+ "raw": {
+ "normalized_reference": "he kept talking and i kept nodding and neither of us was listening.",
+ "normalized_hypothesis": "he kept talking and i kept nodding and neither of us was listening.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he kept talking and i kept nodding and neither of us was listening",
+ "normalized_hypothesis": "he kept talking and i kept nodding and neither of us was listening",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0044",
+ "category": "descriptive",
+ "text": "It's one of those places where everyone seems to know each other already.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0044.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It's one of those places where everyone seems to know each other already.",
+ "raw": {
+ "normalized_reference": "it's one of those places where everyone seems to know each other already.",
+ "normalized_hypothesis": "it's one of those places where everyone seems to know each other already.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it is one of those places where everyone seems to know each other already",
+ "normalized_hypothesis": "it is one of those places where everyone seems to know each other already",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0045",
+ "category": "emotional",
+ "text": "He said he was proud of me and nobody had ever said that to me before.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0045.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He said he was proud of me and nobody had ever said that to me before.",
+ "raw": {
+ "normalized_reference": "he said he was proud of me and nobody had ever said that to me before.",
+ "normalized_hypothesis": "he said he was proud of me and nobody had ever said that to me before.",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he said he was proud of me and nobody had ever said that to me before",
+ "normalized_hypothesis": "he said he was proud of me and nobody had ever said that to me before",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0046",
+ "category": "homographs",
+ "text": "The minute detail took a minute to explain clearly.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0046.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The minute detail took a minute to explain clearly.",
+ "raw": {
+ "normalized_reference": "the minute detail took a minute to explain clearly.",
+ "normalized_hypothesis": "the minute detail took a minute to explain clearly.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the minute detail took a minute to explain clearly",
+ "normalized_hypothesis": "the minute detail took a minute to explain clearly",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0047",
+ "category": "long",
+ "text": "The strange thing about really wanting something is that getting it never quite feels the way you imagined it would, and not getting it never quite feels as bad.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0047.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The strange thing about really wanting something is that getting it never quite feels the way you imagined it would, and not getting it never quite feels as bad.",
+ "raw": {
+ "normalized_reference": "the strange thing about really wanting something is that getting it never quite feels the way you imagined it would, and not getting it never quite feels as bad.",
+ "normalized_hypothesis": "the strange thing about really wanting something is that getting it never quite feels the way you imagined it would, and not getting it never quite feels as bad.",
+ "errors": 0,
+ "words": 29,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the strange thing about really wanting something is that getting it never quite feels the way you imagined it would and not getting it never quite feels as bad",
+ "normalized_hypothesis": "the strange thing about really wanting something is that getting it never quite feels the way you imagined it would and not getting it never quite feels as bad",
+ "errors": 0,
+ "words": 29,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0048",
+ "category": "long_range",
+ "text": "Even though the recipe looked simple at first, preparing the sauce required patience, careful timing, and a much larger pan.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0048.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Even though the recipe looked simple at first, preparing this sauce required patience, careful timing, and a much larger pan.",
+ "raw": {
+ "normalized_reference": "even though the recipe looked simple at first, preparing the sauce required patience, careful timing, and a much larger pan.",
+ "normalized_hypothesis": "even though the recipe looked simple at first, preparing this sauce required patience, careful timing, and a much larger pan.",
+ "errors": 1,
+ "words": 20,
+ "wer": 0.05
+ },
+ "semantic": {
+ "normalized_reference": "even though the recipe looked simple at 1st preparing the sauce required patience careful timing and a much larger pan",
+ "normalized_hypothesis": "even though the recipe looked simple at 1st preparing this sauce required patience careful timing and a much larger pan",
+ "errors": 1,
+ "words": 20,
+ "wer": 0.05
+ }
+ },
+ {
+ "id": "release500_0049",
+ "category": "names_places",
+ "text": "Xiomara met Siobhan near the old library in Worcestershire.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0049.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Xiomara met Siobhan near the old library in Worcestershire.",
+ "raw": {
+ "normalized_reference": "xiomara met siobhan near the old library in worcestershire.",
+ "normalized_hypothesis": "xiomara met siobhan near the old library in worcestershire.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "xiomara met siobhan near the old library in worcestershire",
+ "normalized_hypothesis": "xiomara met siobhan near the old library in worcestershire",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0050",
+ "category": "narrative",
+ "text": "She walked into the room and everyone went quiet, not because she was loud, but because she wasn't.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0050.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "She walked into the room and everyone went quiet, not because she was loud, but because she wasn't.",
+ "raw": {
+ "normalized_reference": "she walked into the room and everyone went quiet, not because she was loud, but because she wasn't.",
+ "normalized_hypothesis": "she walked into the room and everyone went quiet, not because she was loud, but because she wasn't.",
+ "errors": 0,
+ "words": 18,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "she walked into the room and everyone went quiet not because she was loud but because she was not",
+ "normalized_hypothesis": "she walked into the room and everyone went quiet not because she was loud but because she was not",
+ "errors": 0,
+ "words": 19,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0051",
+ "category": "numbers",
+ "text": "At exactly seven forty-five, Morgan placed eighteen dollars and fifty cents beside the notebook.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0051.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "At exactly 7.45, Morgan placed $18.50 beside the notebook.",
+ "raw": {
+ "normalized_reference": "at exactly seven forty-five, morgan placed eighteen dollars and fifty cents beside the notebook.",
+ "normalized_hypothesis": "at exactly 7.45, morgan placed $18.50 beside the notebook.",
+ "errors": 8,
+ "words": 15,
+ "wer": 0.5333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "at exactly 745 morgan placed $18.50 beside the notebook",
+ "normalized_hypothesis": "at exactly 7.45 morgan placed $18.50 beside the notebook",
+ "errors": 2,
+ "words": 10,
+ "wer": 0.2
+ }
+ },
+ {
+ "id": "release500_0052",
+ "category": "punctuation",
+ "text": "Does punctuation work? Yes: commas, semicolons, and periods should create sensible boundaries.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0052.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Is punctuation work? Yes, commas, semicolons, and periods should create sensible boundaries.",
+ "raw": {
+ "normalized_reference": "does punctuation work? yes: commas, semicolons, and periods should create sensible boundaries.",
+ "normalized_hypothesis": "is punctuation work? yes, commas, semicolons, and periods should create sensible boundaries.",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "does punctuation work yes commas semicolons and periods should create sensible boundaries",
+ "normalized_hypothesis": "is punctuation work yes commas semicolons and periods should create sensible boundaries",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ }
+ },
+ {
+ "id": "release500_0053",
+ "category": "question",
+ "text": "Have you ever just stopped and thought about how strange all of this is?",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0053.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Have you ever just stopped and thought about how strange all of this is?",
+ "raw": {
+ "normalized_reference": "have you ever just stopped and thought about how strange all of this is?",
+ "normalized_hypothesis": "have you ever just stopped and thought about how strange all of this is?",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "have you ever just stopped and thought about how strange all of this is",
+ "normalized_hypothesis": "have you ever just stopped and thought about how strange all of this is",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0054",
+ "category": "rare_words",
+ "text": "Her idiosyncratic explanation sounded plausible but remained fundamentally ambiguous.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0054.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Her idiosyncratic explanation sounded plausible but remained fundamentally ambiguous.",
+ "raw": {
+ "normalized_reference": "her idiosyncratic explanation sounded plausible but remained fundamentally ambiguous.",
+ "normalized_hypothesis": "her idiosyncratic explanation sounded plausible but remained fundamentally ambiguous.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "her idiosyncratic explanation sounded plausible but remained fundamentally ambiguous",
+ "normalized_hypothesis": "her idiosyncratic explanation sounded plausible but remained fundamentally ambiguous",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0055",
+ "category": "short",
+ "text": "That tracks.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0055.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "That tracks.",
+ "raw": {
+ "normalized_reference": "that tracks.",
+ "normalized_hypothesis": "that tracks.",
+ "errors": 0,
+ "words": 2,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "that tracks",
+ "normalized_hypothesis": "that tracks",
+ "errors": 0,
+ "words": 2,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0056",
+ "category": "technical",
+ "text": "The key thing to remember is that order matters here more than you'd think.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0056.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The key thing to remember is that order matters here more than you'd think.",
+ "raw": {
+ "normalized_reference": "the key thing to remember is that order matters here more than you'd think.",
+ "normalized_hypothesis": "the key thing to remember is that order matters here more than you'd think.",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the key thing to remember is that order matters here more than you would think",
+ "normalized_hypothesis": "the key thing to remember is that order matters here more than you would think",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0057",
+ "category": "conversational",
+ "text": "I would not say that again in this room if I were you.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0057.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I would not say that again in this room if I were you.",
+ "raw": {
+ "normalized_reference": "i would not say that again in this room if i were you.",
+ "normalized_hypothesis": "i would not say that again in this room if i were you.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i would not say that again in this room if i were you",
+ "normalized_hypothesis": "i would not say that again in this room if i were you",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0058",
+ "category": "descriptive",
+ "text": "The room was filled with the quiet that means something is about to change.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0058.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The room was filled with the quiet that means something is about to change.",
+ "raw": {
+ "normalized_reference": "the room was filled with the quiet that means something is about to change.",
+ "normalized_hypothesis": "the room was filled with the quiet that means something is about to change.",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the room was filled with the quiet that means something is about to change",
+ "normalized_hypothesis": "the room was filled with the quiet that means something is about to change",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0059",
+ "category": "emotional",
+ "text": "I didn't expect to feel that way about it, but I did.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0059.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I didn't expect to feel that way about it, but I did.",
+ "raw": {
+ "normalized_reference": "i didn't expect to feel that way about it, but i did.",
+ "normalized_hypothesis": "i didn't expect to feel that way about it, but i did.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i did not expect to feel that way about it but i did",
+ "normalized_hypothesis": "i did not expect to feel that way about it but i did",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0060",
+ "category": "homographs",
+ "text": "I object to the strange object that was left near the entrance.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0060.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I object to the strange object that was left near the entrance.",
+ "raw": {
+ "normalized_reference": "i object to the strange object that was left near the entrance.",
+ "normalized_hypothesis": "i object to the strange object that was left near the entrance.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i object to the strange object that was left near the entrance",
+ "normalized_hypothesis": "i object to the strange object that was left near the entrance",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0061",
+ "category": "long",
+ "text": "I've learned more from the things that went wrong than from the things that worked, which is a frustrating way to have to learn things but apparently the only way that stuck.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0061.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I've learned more from the things that went wrong than from the things that worked, which is a frustrating way to have to learn things but apparently the only way that's stuck.",
+ "raw": {
+ "normalized_reference": "i've learned more from the things that went wrong than from the things that worked, which is a frustrating way to have to learn things but apparently the only way that stuck.",
+ "normalized_hypothesis": "i've learned more from the things that went wrong than from the things that worked, which is a frustrating way to have to learn things but apparently the only way that's stuck.",
+ "errors": 1,
+ "words": 32,
+ "wer": 0.03125
+ },
+ "semantic": {
+ "normalized_reference": "i have learned more from the things that went wrong than from the things that worked which is a frustrating way to have to learn things but apparently the only way that stuck",
+ "normalized_hypothesis": "i have learned more from the things that went wrong than from the things that worked which is a frustrating way to have to learn things but apparently the only way that is stuck",
+ "errors": 1,
+ "words": 33,
+ "wer": 0.030303030303030304
+ }
+ },
+ {
+ "id": "release500_0062",
+ "category": "long_range",
+ "text": "Because every nearby hotel was full, the exhausted hikers followed the narrow road until they found a farmhouse with one empty room.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0062.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Because every nearby hotel was full, the exhausted hikers followed the narrow road until they found a farmhouse with one empty room.",
+ "raw": {
+ "normalized_reference": "because every nearby hotel was full, the exhausted hikers followed the narrow road until they found a farmhouse with one empty room.",
+ "normalized_hypothesis": "because every nearby hotel was full, the exhausted hikers followed the narrow road until they found a farmhouse with one empty room.",
+ "errors": 0,
+ "words": 22,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "because every nearby hotel was full the exhausted hikers followed the narrow road until they found a farmhouse with one empty room",
+ "normalized_hypothesis": "because every nearby hotel was full the exhausted hikers followed the narrow road until they found a farmhouse with one empty room",
+ "errors": 0,
+ "words": 22,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0063",
+ "category": "names_places",
+ "text": "The itinerary sends Beatrice from Albuquerque to Des Moines and then to Boise.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0063.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The itinerary sends Beatrice from Albuquerque to Des Moines and then to Boise.",
+ "raw": {
+ "normalized_reference": "the itinerary sends beatrice from albuquerque to des moines and then to boise.",
+ "normalized_hypothesis": "the itinerary sends beatrice from albuquerque to des moines and then to boise.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the itinerary sends beatrice from albuquerque to des moines and then to boise",
+ "normalized_hypothesis": "the itinerary sends beatrice from albuquerque to des moines and then to boise",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0064",
+ "category": "narrative",
+ "text": "We had been driving for six hours when we finally saw the sign for the town.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0064.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "We had been driving for six hours when we finally saw the sign for the town.",
+ "raw": {
+ "normalized_reference": "we had been driving for six hours when we finally saw the sign for the town.",
+ "normalized_hypothesis": "we had been driving for six hours when we finally saw the sign for the town.",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "we had been driving for 6 hours when we finally saw the sign for the town",
+ "normalized_hypothesis": "we had been driving for 6 hours when we finally saw the sign for the town",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0065",
+ "category": "numbers",
+ "text": "Call extension zero four nine after eight thirty in the morning.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0065.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Call extension 049 after 8.30 in the morning.",
+ "raw": {
+ "normalized_reference": "call extension zero four nine after eight thirty in the morning.",
+ "normalized_hypothesis": "call extension 049 after 8.30 in the morning.",
+ "errors": 5,
+ "words": 11,
+ "wer": 0.45454545454545453
+ },
+ "semantic": {
+ "normalized_reference": "call extension 049 after 830 in the morning",
+ "normalized_hypothesis": "call extension 49 after 8.30 in the morning",
+ "errors": 3,
+ "words": 8,
+ "wer": 0.375
+ }
+ },
+ {
+ "id": "release500_0066",
+ "category": "punctuation",
+ "text": "Stop - listen carefully - then answer the question without guessing.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0066.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Stop listen carefully then answer the question without guessing.",
+ "raw": {
+ "normalized_reference": "stop - listen carefully - then answer the question without guessing.",
+ "normalized_hypothesis": "stop listen carefully then answer the question without guessing.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "stop listen carefully then answer the question without guessing",
+ "normalized_hypothesis": "stop listen carefully then answer the question without guessing",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0067",
+ "category": "question",
+ "text": "Does it matter at this point, honestly?",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0067.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Does it matter at this point, honestly?",
+ "raw": {
+ "normalized_reference": "does it matter at this point, honestly?",
+ "normalized_hypothesis": "does it matter at this point, honestly?",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "does it matter at this point honestly",
+ "normalized_hypothesis": "does it matter at this point honestly",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0068",
+ "category": "rare_words",
+ "text": "An entrepreneurial meteorologist described the unprecedented atmospheric phenomenon.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0068.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "an entrepreneurial meteorologist described the unprecedented atmospheric phenomenon.",
+ "raw": {
+ "normalized_reference": "an entrepreneurial meteorologist described the unprecedented atmospheric phenomenon.",
+ "normalized_hypothesis": "an entrepreneurial meteorologist described the unprecedented atmospheric phenomenon.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "an entrepreneurial meteorologist described the unprecedented atmospheric phenomenon",
+ "normalized_hypothesis": "an entrepreneurial meteorologist described the unprecedented atmospheric phenomenon",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0069",
+ "category": "short",
+ "text": "That's not what I expected.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0069.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "That's not what I expected.",
+ "raw": {
+ "normalized_reference": "that's not what i expected.",
+ "normalized_hypothesis": "that's not what i expected.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "that is not what i expected",
+ "normalized_hypothesis": "that is not what i expected",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0070",
+ "category": "technical",
+ "text": "There's a known issue with that version that causes failures under load.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0070.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "There's a known issue with that version that causes failures under load.",
+ "raw": {
+ "normalized_reference": "there's a known issue with that version that causes failures under load.",
+ "normalized_hypothesis": "there's a known issue with that version that causes failures under load.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "there is a known issue with that version that causes failures under load",
+ "normalized_hypothesis": "there is a known issue with that version that causes failures under load",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0071",
+ "category": "conversational",
+ "text": "Don't look now but I think that's him over there.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0071.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Don't look now, but I think that's him over there.",
+ "raw": {
+ "normalized_reference": "don't look now but i think that's him over there.",
+ "normalized_hypothesis": "don't look now, but i think that's him over there.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "do not look now but i think that is him over there",
+ "normalized_hypothesis": "do not look now but i think that is him over there",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0072",
+ "category": "descriptive",
+ "text": "The city at night looks completely different from how it feels during the day.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0072.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The city at night looks completely different from how it feels during the day.",
+ "raw": {
+ "normalized_reference": "the city at night looks completely different from how it feels during the day.",
+ "normalized_hypothesis": "the city at night looks completely different from how it feels during the day.",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the city at night looks completely different from how it feels during the day",
+ "normalized_hypothesis": "the city at night looks completely different from how it feels during the day",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0073",
+ "category": "emotional",
+ "text": "It's the kind of happy that makes you nervous because you know it can't last.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0073.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It's the kind of happy that makes you nervous because you know it can't last.",
+ "raw": {
+ "normalized_reference": "it's the kind of happy that makes you nervous because you know it can't last.",
+ "normalized_hypothesis": "it's the kind of happy that makes you nervous because you know it can't last.",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it is the kind of happy that makes you nervous because you know it can not last",
+ "normalized_hypothesis": "it is the kind of happy that makes you nervous because you know it can not last",
+ "errors": 0,
+ "words": 17,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0074",
+ "category": "homographs",
+ "text": "The dove dove beneath the branch while the wind began to wind the loose cable.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0074.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The dove dove beneath the branch while the wind began to wind the loose cable.",
+ "raw": {
+ "normalized_reference": "the dove dove beneath the branch while the wind began to wind the loose cable.",
+ "normalized_hypothesis": "the dove dove beneath the branch while the wind began to wind the loose cable.",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the dove dove beneath the branch while the wind began to wind the loose cable",
+ "normalized_hypothesis": "the dove dove beneath the branch while the wind began to wind the loose cable",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0075",
+ "category": "long",
+ "text": "I've thought a lot about what I would do differently if I could go back, and the honest answer is probably not as much as I'd like to think.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0075.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I've thought a lot about what I would do differently if I could go back, and the honest answer is probably not as much as I'd like to think.",
+ "raw": {
+ "normalized_reference": "i've thought a lot about what i would do differently if i could go back, and the honest answer is probably not as much as i'd like to think.",
+ "normalized_hypothesis": "i've thought a lot about what i would do differently if i could go back, and the honest answer is probably not as much as i'd like to think.",
+ "errors": 0,
+ "words": 29,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i have thought a lot about what i would do differently if i could go back and the honest answer is probably not as much as i would like to think",
+ "normalized_hypothesis": "i have thought a lot about what i would do differently if i could go back and the honest answer is probably not as much as i would like to think",
+ "errors": 0,
+ "words": 31,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0076",
+ "category": "long_range",
+ "text": "Although the weather changed without warning, the musicians finished packing before anyone opened the enormous glass door.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0076.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Although the weather changed without warning, the musicians finished packing before anyone opened the enormous glass door.",
+ "raw": {
+ "normalized_reference": "although the weather changed without warning, the musicians finished packing before anyone opened the enormous glass door.",
+ "normalized_hypothesis": "although the weather changed without warning, the musicians finished packing before anyone opened the enormous glass door.",
+ "errors": 0,
+ "words": 17,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "although the weather changed without warning the musicians finished packing before anyone opened the enormous glass door",
+ "normalized_hypothesis": "although the weather changed without warning the musicians finished packing before anyone opened the enormous glass door",
+ "errors": 0,
+ "words": 17,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0077",
+ "category": "names_places",
+ "text": "Saskatchewan looked unusually bright beneath the pale February sunrise.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0077.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Saskatchewan looked unusually bright beneath the pale February sunrise.",
+ "raw": {
+ "normalized_reference": "saskatchewan looked unusually bright beneath the pale february sunrise.",
+ "normalized_hypothesis": "saskatchewan looked unusually bright beneath the pale february sunrise.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "saskatchewan looked unusually bright beneath the pale february sunrise",
+ "normalized_hypothesis": "saskatchewan looked unusually bright beneath the pale february sunrise",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0078",
+ "category": "narrative",
+ "text": "She handled it with a grace I genuinely did not expect.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0078.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "She handled it with a grace I genuinely did not expect.",
+ "raw": {
+ "normalized_reference": "she handled it with a grace i genuinely did not expect.",
+ "normalized_hypothesis": "she handled it with a grace i genuinely did not expect.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "she handled it with a grace i genuinely did not expect",
+ "normalized_hypothesis": "she handled it with a grace i genuinely did not expect",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0079",
+ "category": "numbers",
+ "text": "Apartment four hundred seven is across from room twenty-nine on the third floor.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0079.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Apartment 407 is across from room 29 on the third floor.",
+ "raw": {
+ "normalized_reference": "apartment four hundred seven is across from room twenty-nine on the third floor.",
+ "normalized_hypothesis": "apartment 407 is across from room 29 on the third floor.",
+ "errors": 5,
+ "words": 14,
+ "wer": 0.35714285714285715
+ },
+ "semantic": {
+ "normalized_reference": "apartment 407 is across from room 29 on the 3rd floor",
+ "normalized_hypothesis": "apartment 407 is across from room 29 on the 3rd floor",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0080",
+ "category": "punctuation",
+ "text": "\"Not yet,\" Rowan whispered. \"Give me another minute... please.\"",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0080.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Not yet, Rowan whispered. Give me another minute, please.",
+ "raw": {
+ "normalized_reference": "\"not yet,\" rowan whispered. \"give me another minute... please.\"",
+ "normalized_hypothesis": "not yet, rowan whispered. give me another minute, please.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "not yet rowan whispered give me another minute please",
+ "normalized_hypothesis": "not yet rowan whispered give me another minute please",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0081",
+ "category": "question",
+ "text": "What are we actually trying to accomplish here?",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0081.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "What are we actually trying to accomplish here?",
+ "raw": {
+ "normalized_reference": "what are we actually trying to accomplish here?",
+ "normalized_hypothesis": "what are we actually trying to accomplish here?",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "what are we actually trying to accomplish here",
+ "normalized_hypothesis": "what are we actually trying to accomplish here",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0082",
+ "category": "rare_words",
+ "text": "The physician recommended a noninvasive procedure for the recurring neurological symptoms.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0082.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The physician recommended a non-invasive procedure for the recurring neurological symptoms.",
+ "raw": {
+ "normalized_reference": "the physician recommended a noninvasive procedure for the recurring neurological symptoms.",
+ "normalized_hypothesis": "the physician recommended a non-invasive procedure for the recurring neurological symptoms.",
+ "errors": 2,
+ "words": 11,
+ "wer": 0.18181818181818182
+ },
+ "semantic": {
+ "normalized_reference": "the physician recommended a noninvasive procedure for the recurring neurological symptoms",
+ "normalized_hypothesis": "the physician recommended a non invasive procedure for the recurring neurological symptoms",
+ "errors": 2,
+ "words": 11,
+ "wer": 0.18181818181818182
+ }
+ },
+ {
+ "id": "release500_0083",
+ "category": "short",
+ "text": "Are you serious right now?",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0083.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Are you serious right now?",
+ "raw": {
+ "normalized_reference": "are you serious right now?",
+ "normalized_hypothesis": "are you serious right now?",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "are you serious right now",
+ "normalized_hypothesis": "are you serious right now",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0084",
+ "category": "technical",
+ "text": "The performance improvement is real, but it comes with tradeoffs.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0084.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The performance improvement is real, but it comes with trade-offs.",
+ "raw": {
+ "normalized_reference": "the performance improvement is real, but it comes with tradeoffs.",
+ "normalized_hypothesis": "the performance improvement is real, but it comes with trade-offs.",
+ "errors": 2,
+ "words": 10,
+ "wer": 0.2
+ },
+ "semantic": {
+ "normalized_reference": "the performance improvement is real but it comes with tradeoffs",
+ "normalized_hypothesis": "the performance improvement is real but it comes with trade offs",
+ "errors": 2,
+ "words": 10,
+ "wer": 0.2
+ }
+ },
+ {
+ "id": "release500_0085",
+ "category": "conversational",
+ "text": "I don't know. It's just been one of those weeks, you know?",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0085.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I don't know. It's just been one of those weeks, you know.",
+ "raw": {
+ "normalized_reference": "i don't know. it's just been one of those weeks, you know?",
+ "normalized_hypothesis": "i don't know. it's just been one of those weeks, you know.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i do not know it is just been one of those weeks you know",
+ "normalized_hypothesis": "i do not know it is just been one of those weeks you know",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0086",
+ "category": "descriptive",
+ "text": "The moment you stop expecting perfection, things get a lot more interesting.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0086.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The moment you stop expecting perfection, things get a lot more interesting.",
+ "raw": {
+ "normalized_reference": "the moment you stop expecting perfection, things get a lot more interesting.",
+ "normalized_hypothesis": "the moment you stop expecting perfection, things get a lot more interesting.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the moment you stop expecting perfection things get a lot more interesting",
+ "normalized_hypothesis": "the moment you stop expecting perfection things get a lot more interesting",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0087",
+ "category": "emotional",
+ "text": "It's the kind of thing that stays with you even years later.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0087.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It's the kind of thing that stays with you even years later.",
+ "raw": {
+ "normalized_reference": "it's the kind of thing that stays with you even years later.",
+ "normalized_hypothesis": "it's the kind of thing that stays with you even years later.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it is the kind of thing that stays with you even years later",
+ "normalized_hypothesis": "it is the kind of thing that stays with you even years later",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0088",
+ "category": "long",
+ "text": "She told me once that the most important skill she'd developed was learning to disagree with someone without making them feel they were wrong.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0088.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "She told me once that the most important skill she developed was learning to disagree with someone without making them feel they were wrong.",
+ "raw": {
+ "normalized_reference": "she told me once that the most important skill she'd developed was learning to disagree with someone without making them feel they were wrong.",
+ "normalized_hypothesis": "she told me once that the most important skill she developed was learning to disagree with someone without making them feel they were wrong.",
+ "errors": 1,
+ "words": 24,
+ "wer": 0.041666666666666664
+ },
+ "semantic": {
+ "normalized_reference": "she told me once that the most important skill she would developed was learning to disagree with someone without making them feel they were wrong",
+ "normalized_hypothesis": "she told me once that the most important skill she developed was learning to disagree with someone without making them feel they were wrong",
+ "errors": 1,
+ "words": 25,
+ "wer": 0.04
+ }
+ },
+ {
+ "id": "release500_0089",
+ "category": "narrative",
+ "text": "He stood at the edge of the cliff and wondered, not for the first time, how he'd gotten here.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0089.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He stood at the edge of the cliff and wondered, not for the first time, how he'd gotten here.",
+ "raw": {
+ "normalized_reference": "he stood at the edge of the cliff and wondered, not for the first time, how he'd gotten here.",
+ "normalized_hypothesis": "he stood at the edge of the cliff and wondered, not for the first time, how he'd gotten here.",
+ "errors": 0,
+ "words": 19,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he stood at the edge of the cliff and wondered not for the 1st time how he would gotten here",
+ "normalized_hypothesis": "he stood at the edge of the cliff and wondered not for the 1st time how he would gotten here",
+ "errors": 0,
+ "words": 20,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0090",
+ "category": "question",
+ "text": "Why does asking for help feel so hard when you actually need it?",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0090.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Why does asking for help feel so hard when you actually need it?",
+ "raw": {
+ "normalized_reference": "why does asking for help feel so hard when you actually need it?",
+ "normalized_hypothesis": "why does asking for help feel so hard when you actually need it?",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "why does asking for help feel so hard when you actually need it",
+ "normalized_hypothesis": "why does asking for help feel so hard when you actually need it",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0091",
+ "category": "short",
+ "text": "That's fair.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0091.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "That's fair.",
+ "raw": {
+ "normalized_reference": "that's fair.",
+ "normalized_hypothesis": "that's fair.",
+ "errors": 0,
+ "words": 2,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "that is fair",
+ "normalized_hypothesis": "that is fair",
+ "errors": 0,
+ "words": 3,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0092",
+ "category": "technical",
+ "text": "If you run into that error, the fix is usually just restarting the service.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0092.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "If you run into that error, the fix is usually just restarting the service.",
+ "raw": {
+ "normalized_reference": "if you run into that error, the fix is usually just restarting the service.",
+ "normalized_hypothesis": "if you run into that error, the fix is usually just restarting the service.",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "if you run into that error the fix is usually just restarting the service",
+ "normalized_hypothesis": "if you run into that error the fix is usually just restarting the service",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0093",
+ "category": "conversational",
+ "text": "It was a disaster in the best possible way.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0093.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It was a disaster in the best possible way.",
+ "raw": {
+ "normalized_reference": "it was a disaster in the best possible way.",
+ "normalized_hypothesis": "it was a disaster in the best possible way.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it was a disaster in the best possible way",
+ "normalized_hypothesis": "it was a disaster in the best possible way",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0094",
+ "category": "descriptive",
+ "text": "It had the look of a place that used to mean something to someone.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0094.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It had the look of a place that used to mean something to someone.",
+ "raw": {
+ "normalized_reference": "it had the look of a place that used to mean something to someone.",
+ "normalized_hypothesis": "it had the look of a place that used to mean something to someone.",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it had the look of a place that used to mean something to someone",
+ "normalized_hypothesis": "it had the look of a place that used to mean something to someone",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0095",
+ "category": "emotional",
+ "text": "There's this feeling you get when something finally works after weeks of failure.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0095.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "There's this feeling you get when something finally works after weeks of failure.",
+ "raw": {
+ "normalized_reference": "there's this feeling you get when something finally works after weeks of failure.",
+ "normalized_hypothesis": "there's this feeling you get when something finally works after weeks of failure.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "there is this feeling you get when something finally works after weeks of failure",
+ "normalized_hypothesis": "there is this feeling you get when something finally works after weeks of failure",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0096",
+ "category": "long",
+ "text": "He had a theory that most arguments aren't actually about what they seem to be about, and the longer I live the more I think he was onto something.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0096.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He had a theory that most arguments aren't actually about what they seem to be about, and the longer I lived the more I think he was onto something.",
+ "raw": {
+ "normalized_reference": "he had a theory that most arguments aren't actually about what they seem to be about, and the longer i live the more i think he was onto something.",
+ "normalized_hypothesis": "he had a theory that most arguments aren't actually about what they seem to be about, and the longer i lived the more i think he was onto something.",
+ "errors": 1,
+ "words": 29,
+ "wer": 0.034482758620689655
+ },
+ "semantic": {
+ "normalized_reference": "he had a theory that most arguments are not actually about what they seem to be about and the longer i live the more i think he was onto something",
+ "normalized_hypothesis": "he had a theory that most arguments are not actually about what they seem to be about and the longer i lived the more i think he was onto something",
+ "errors": 1,
+ "words": 30,
+ "wer": 0.03333333333333333
+ }
+ },
+ {
+ "id": "release500_0097",
+ "category": "narrative",
+ "text": "The thing nobody tells you about starting over is how ordinary most of it feels.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0097.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The thing nobody tells you about starting over is how ordinary most of it feels.",
+ "raw": {
+ "normalized_reference": "the thing nobody tells you about starting over is how ordinary most of it feels.",
+ "normalized_hypothesis": "the thing nobody tells you about starting over is how ordinary most of it feels.",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the thing nobody tells you about starting over is how ordinary most of it feels",
+ "normalized_hypothesis": "the thing nobody tells you about starting over is how ordinary most of it feels",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0098",
+ "category": "question",
+ "text": "What exactly are we waiting for at this point?",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0098.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "What exactly are we waiting for at this point?",
+ "raw": {
+ "normalized_reference": "what exactly are we waiting for at this point?",
+ "normalized_hypothesis": "what exactly are we waiting for at this point?",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "what exactly are we waiting for at this point",
+ "normalized_hypothesis": "what exactly are we waiting for at this point",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0099",
+ "category": "short",
+ "text": "I stand by what I said.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0099.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I stand by what I say.",
+ "raw": {
+ "normalized_reference": "i stand by what i said.",
+ "normalized_hypothesis": "i stand by what i say.",
+ "errors": 1,
+ "words": 6,
+ "wer": 0.16666666666666666
+ },
+ "semantic": {
+ "normalized_reference": "i stand by what i said",
+ "normalized_hypothesis": "i stand by what i say",
+ "errors": 1,
+ "words": 6,
+ "wer": 0.16666666666666666
+ }
+ },
+ {
+ "id": "release500_0100",
+ "category": "technical",
+ "text": "Double-check that value before you push anything to production.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0100.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Double-check that value before you push anything to production.",
+ "raw": {
+ "normalized_reference": "double-check that value before you push anything to production.",
+ "normalized_hypothesis": "double-check that value before you push anything to production.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "double check that value before you push anything to production",
+ "normalized_hypothesis": "double check that value before you push anything to production",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0101",
+ "category": "conversational",
+ "text": "I understand your point; I just do not agree with the conclusion.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0101.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I understand your point, I just do not agree with the conclusion.",
+ "raw": {
+ "normalized_reference": "i understand your point; i just do not agree with the conclusion.",
+ "normalized_hypothesis": "i understand your point, i just do not agree with the conclusion.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i understand your point i just do not agree with the conclusion",
+ "normalized_hypothesis": "i understand your point i just do not agree with the conclusion",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0102",
+ "category": "descriptive",
+ "text": "The sky that day was a strange shade of orange I'd never seen before.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0102.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The sky that day was a strange shade of orange I'd never seen before.",
+ "raw": {
+ "normalized_reference": "the sky that day was a strange shade of orange i'd never seen before.",
+ "normalized_hypothesis": "the sky that day was a strange shade of orange i'd never seen before.",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the sky that day was a strange shade of orange i would never seen before",
+ "normalized_hypothesis": "the sky that day was a strange shade of orange i would never seen before",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0103",
+ "category": "emotional",
+ "text": "It hurt more than I expected, and I thought I was prepared.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0103.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It hurt more than I expected, and I thought I was prepared.",
+ "raw": {
+ "normalized_reference": "it hurt more than i expected, and i thought i was prepared.",
+ "normalized_hypothesis": "it hurt more than i expected, and i thought i was prepared.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it hurt more than i expected and i thought i was prepared",
+ "normalized_hypothesis": "it hurt more than i expected and i thought i was prepared",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0104",
+ "category": "long",
+ "text": "The hardest part wasn't making the decision, it was sitting with the uncertainty of not knowing whether it was the right one for months afterward.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0104.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The hardest part wasn't making the decision, it was sitting with the uncertainty of not knowing whether it was the right one for months afterward.",
+ "raw": {
+ "normalized_reference": "the hardest part wasn't making the decision, it was sitting with the uncertainty of not knowing whether it was the right one for months afterward.",
+ "normalized_hypothesis": "the hardest part wasn't making the decision, it was sitting with the uncertainty of not knowing whether it was the right one for months afterward.",
+ "errors": 0,
+ "words": 25,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the hardest part was not making the decision it was sitting with the uncertainty of not knowing whether it was the right one for months afterward",
+ "normalized_hypothesis": "the hardest part was not making the decision it was sitting with the uncertainty of not knowing whether it was the right one for months afterward",
+ "errors": 0,
+ "words": 26,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0105",
+ "category": "narrative",
+ "text": "There was this one day last summer when everything just clicked into place.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0105.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "There was this one day last summer when everything just clicked into place.",
+ "raw": {
+ "normalized_reference": "there was this one day last summer when everything just clicked into place.",
+ "normalized_hypothesis": "there was this one day last summer when everything just clicked into place.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "there was this one day last summer when everything just clicked into place",
+ "normalized_hypothesis": "there was this one day last summer when everything just clicked into place",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0106",
+ "category": "question",
+ "text": "Have you ever thought about why we keep doing things the same way when they stop working?",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0106.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Have you ever thought about why we keep doing things the same way when they stop working?",
+ "raw": {
+ "normalized_reference": "have you ever thought about why we keep doing things the same way when they stop working?",
+ "normalized_hypothesis": "have you ever thought about why we keep doing things the same way when they stop working?",
+ "errors": 0,
+ "words": 17,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "have you ever thought about why we keep doing things the same way when they stop working",
+ "normalized_hypothesis": "have you ever thought about why we keep doing things the same way when they stop working",
+ "errors": 0,
+ "words": 17,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0107",
+ "category": "short",
+ "text": "I can't believe you just said that.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0107.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I can't believe you just said that.",
+ "raw": {
+ "normalized_reference": "i can't believe you just said that.",
+ "normalized_hypothesis": "i can't believe you just said that.",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i can not believe you just said that",
+ "normalized_hypothesis": "i can not believe you just said that",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0108",
+ "category": "technical",
+ "text": "You'll want to isolate that variable before drawing any conclusions.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0108.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "you'll want to isolate that variable before drawing any conclusions.",
+ "raw": {
+ "normalized_reference": "you'll want to isolate that variable before drawing any conclusions.",
+ "normalized_hypothesis": "you'll want to isolate that variable before drawing any conclusions.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "you will want to isolate that variable before drawing any conclusions",
+ "normalized_hypothesis": "you will want to isolate that variable before drawing any conclusions",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0109",
+ "category": "conversational",
+ "text": "The meeting that could have been an email lasted two hours.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0109.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The meeting that could have been an email lasted two hours.",
+ "raw": {
+ "normalized_reference": "the meeting that could have been an email lasted two hours.",
+ "normalized_hypothesis": "the meeting that could have been an email lasted two hours.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the meeting that could have been an email lasted 2 hours",
+ "normalized_hypothesis": "the meeting that could have been an email lasted 2 hours",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0110",
+ "category": "descriptive",
+ "text": "There's a specific kind of tired that sleep doesn't fix.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0110.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "There's a specific kind of tired that sleep doesn't fix.",
+ "raw": {
+ "normalized_reference": "there's a specific kind of tired that sleep doesn't fix.",
+ "normalized_hypothesis": "there's a specific kind of tired that sleep doesn't fix.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "there is a specific kind of tired that sleep does not fix",
+ "normalized_hypothesis": "there is a specific kind of tired that sleep does not fix",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0111",
+ "category": "emotional",
+ "text": "Something about that moment broke something open in me that I hadn't realized was closed.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0111.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Something about that moment broke something open in me that I hadn't realized was closed.",
+ "raw": {
+ "normalized_reference": "something about that moment broke something open in me that i hadn't realized was closed.",
+ "normalized_hypothesis": "something about that moment broke something open in me that i hadn't realized was closed.",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "something about that moment broke something open in me that i had not realized was closed",
+ "normalized_hypothesis": "something about that moment broke something open in me that i had not realized was closed",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0112",
+ "category": "long",
+ "text": "The thing about making something good is that you don't usually know it's good while you're making it. You just know you're not done yet.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0112.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The thing about making something good is that you don't usually know it's good while you're making it. You just know you're not done yet.",
+ "raw": {
+ "normalized_reference": "the thing about making something good is that you don't usually know it's good while you're making it. you just know you're not done yet.",
+ "normalized_hypothesis": "the thing about making something good is that you don't usually know it's good while you're making it. you just know you're not done yet.",
+ "errors": 0,
+ "words": 25,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the thing about making something good is that you do not usually know it is good while you are making it you just know you are not done yet",
+ "normalized_hypothesis": "the thing about making something good is that you do not usually know it is good while you are making it you just know you are not done yet",
+ "errors": 0,
+ "words": 29,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0113",
+ "category": "narrative",
+ "text": "She opened the door slowly, as if she expected to find something she wasn't ready for.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0113.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "She opened the door slowly, as if she expected to find something she wasn't ready for.",
+ "raw": {
+ "normalized_reference": "she opened the door slowly, as if she expected to find something she wasn't ready for.",
+ "normalized_hypothesis": "she opened the door slowly, as if she expected to find something she wasn't ready for.",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "she opened the door slowly as if she expected to find something she was not ready for",
+ "normalized_hypothesis": "she opened the door slowly as if she expected to find something she was not ready for",
+ "errors": 0,
+ "words": 17,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0114",
+ "category": "question",
+ "text": "What would you do differently if you could go back?",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0114.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "What would you do differently if you could go back?",
+ "raw": {
+ "normalized_reference": "what would you do differently if you could go back?",
+ "normalized_hypothesis": "what would you do differently if you could go back?",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "what would you do differently if you could go back",
+ "normalized_hypothesis": "what would you do differently if you could go back",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0115",
+ "category": "short",
+ "text": "Okay. Okay okay okay.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0115.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Okay, okay, okay, okay.",
+ "raw": {
+ "normalized_reference": "okay. okay okay okay.",
+ "normalized_hypothesis": "okay, okay, okay, okay.",
+ "errors": 0,
+ "words": 4,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "okay okay okay okay",
+ "normalized_hypothesis": "okay okay okay okay",
+ "errors": 0,
+ "words": 4,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0116",
+ "category": "technical",
+ "text": "The latency goes up significantly at scale, which is expected.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0116.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The latency goes up significantly at scale, which is expected.",
+ "raw": {
+ "normalized_reference": "the latency goes up significantly at scale, which is expected.",
+ "normalized_hypothesis": "the latency goes up significantly at scale, which is expected.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the latency goes up significantly at scale which is expected",
+ "normalized_hypothesis": "the latency goes up significantly at scale which is expected",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0117",
+ "category": "conversational",
+ "text": "We ended up staying until two in the morning talking about nothing.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0117.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "We ended up staying until 2 in the morning talking about nothing.",
+ "raw": {
+ "normalized_reference": "we ended up staying until two in the morning talking about nothing.",
+ "normalized_hypothesis": "we ended up staying until 2 in the morning talking about nothing.",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "we ended up staying until 2 in the morning talking about nothing",
+ "normalized_hypothesis": "we ended up staying until 2 in the morning talking about nothing",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0118",
+ "category": "descriptive",
+ "text": "The rain started slowly, and then all at once, the way it usually does.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0118.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The rain started slowly and then all at once, the way it usually does.",
+ "raw": {
+ "normalized_reference": "the rain started slowly, and then all at once, the way it usually does.",
+ "normalized_hypothesis": "the rain started slowly and then all at once, the way it usually does.",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the rain started slowly and then all at once the way it usually does",
+ "normalized_hypothesis": "the rain started slowly and then all at once the way it usually does",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0119",
+ "category": "emotional",
+ "text": "I honestly did not think I was going to make it through that year.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0119.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I honestly did not think I was going to make it through that year.",
+ "raw": {
+ "normalized_reference": "i honestly did not think i was going to make it through that year.",
+ "normalized_hypothesis": "i honestly did not think i was going to make it through that year.",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i honestly did not think i was going to make it through that year",
+ "normalized_hypothesis": "i honestly did not think i was going to make it through that year",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0120",
+ "category": "long",
+ "text": "There's an assumption that people who seem confident all the time actually feel confident, and in my experience that's almost never true.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0120.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "There's an assumption that people who seem confident all the time actually feel confident, and in my experience that's almost never true.",
+ "raw": {
+ "normalized_reference": "there's an assumption that people who seem confident all the time actually feel confident, and in my experience that's almost never true.",
+ "normalized_hypothesis": "there's an assumption that people who seem confident all the time actually feel confident, and in my experience that's almost never true.",
+ "errors": 0,
+ "words": 22,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "there is an assumption that people who seem confident all the time actually feel confident and in my experience that is almost never true",
+ "normalized_hypothesis": "there is an assumption that people who seem confident all the time actually feel confident and in my experience that is almost never true",
+ "errors": 0,
+ "words": 24,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0121",
+ "category": "narrative",
+ "text": "The moment I opened the door I knew something was different, even if I couldn't say what.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0121.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The moment I opened the door I knew something was different, even if I couldn't say what.",
+ "raw": {
+ "normalized_reference": "the moment i opened the door i knew something was different, even if i couldn't say what.",
+ "normalized_hypothesis": "the moment i opened the door i knew something was different, even if i couldn't say what.",
+ "errors": 0,
+ "words": 17,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the moment i opened the door i knew something was different even if i could not say what",
+ "normalized_hypothesis": "the moment i opened the door i knew something was different even if i could not say what",
+ "errors": 0,
+ "words": 18,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0122",
+ "category": "question",
+ "text": "How do you explain something that doesn't make sense even to you?",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0122.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "How do you explain something that doesn't make sense even to you?",
+ "raw": {
+ "normalized_reference": "how do you explain something that doesn't make sense even to you?",
+ "normalized_hypothesis": "how do you explain something that doesn't make sense even to you?",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "how do you explain something that does not make sense even to you",
+ "normalized_hypothesis": "how do you explain something that does not make sense even to you",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0123",
+ "category": "short",
+ "text": "Hold on a second.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0123.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Hold on a second.",
+ "raw": {
+ "normalized_reference": "hold on a second.",
+ "normalized_hypothesis": "hold on a second.",
+ "errors": 0,
+ "words": 4,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "hold on a 2nd",
+ "normalized_hypothesis": "hold on a 2nd",
+ "errors": 0,
+ "words": 4,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0124",
+ "category": "technical",
+ "text": "A deterministic split prevents duplicate normalized text from leaking into evaluation.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0124.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "A deterministic split prevents duplicate normalized text from leaking into evaluation.",
+ "raw": {
+ "normalized_reference": "a deterministic split prevents duplicate normalized text from leaking into evaluation.",
+ "normalized_hypothesis": "a deterministic split prevents duplicate normalized text from leaking into evaluation.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "a deterministic split prevents duplicate normalized text from leaking into evaluation",
+ "normalized_hypothesis": "a deterministic split prevents duplicate normalized text from leaking into evaluation",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0125",
+ "category": "conversational",
+ "text": "Honestly, that was much better than I expected, but the ending still felt rushed.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0125.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Honestly, that was much better than I expected, but the ending still felt rushed.",
+ "raw": {
+ "normalized_reference": "honestly, that was much better than i expected, but the ending still felt rushed.",
+ "normalized_hypothesis": "honestly, that was much better than i expected, but the ending still felt rushed.",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "honestly that was much better than i expected but the ending still felt rushed",
+ "normalized_hypothesis": "honestly that was much better than i expected but the ending still felt rushed",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0126",
+ "category": "descriptive",
+ "text": "She had a habit of tilting her head slightly when thinking hard about something.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0126.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "She had a habit of tilting her head slightly when thinking hard about something.",
+ "raw": {
+ "normalized_reference": "she had a habit of tilting her head slightly when thinking hard about something.",
+ "normalized_hypothesis": "she had a habit of tilting her head slightly when thinking hard about something.",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "she had a habit of tilting her head slightly when thinking hard about something",
+ "normalized_hypothesis": "she had a habit of tilting her head slightly when thinking hard about something",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0127",
+ "category": "emotional",
+ "text": "I cannot stress enough how much I did not want to be right about this.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0127.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I cannot stress enough how much I did not want to be right about this.",
+ "raw": {
+ "normalized_reference": "i cannot stress enough how much i did not want to be right about this.",
+ "normalized_hypothesis": "i cannot stress enough how much i did not want to be right about this.",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i cannot stress enough how much i did not want to be right about this",
+ "normalized_hypothesis": "i cannot stress enough how much i did not want to be right about this",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0128",
+ "category": "long",
+ "text": "I used to think the goal was to get to a point where nothing was hard anymore, and now I think that point doesn't exist and I'm not sure I'd want it to.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0128.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I used to think the goal was to get to a point where nothing was hard anymore, and now I think that point doesn't exist and I'm not sure I'd want it to.",
+ "raw": {
+ "normalized_reference": "i used to think the goal was to get to a point where nothing was hard anymore, and now i think that point doesn't exist and i'm not sure i'd want it to.",
+ "normalized_hypothesis": "i used to think the goal was to get to a point where nothing was hard anymore, and now i think that point doesn't exist and i'm not sure i'd want it to.",
+ "errors": 0,
+ "words": 33,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i used to think the goal was to get to a point where nothing was hard anymore and now i think that point does not exist and i am not sure i would want it to",
+ "normalized_hypothesis": "i used to think the goal was to get to a point where nothing was hard anymore and now i think that point does not exist and i am not sure i would want it to",
+ "errors": 0,
+ "words": 36,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0129",
+ "category": "narrative",
+ "text": "So this is the part of the story where everything goes completely sideways.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0129.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "So this is the part of the story where everything goes completely sideways.",
+ "raw": {
+ "normalized_reference": "so this is the part of the story where everything goes completely sideways.",
+ "normalized_hypothesis": "so this is the part of the story where everything goes completely sideways.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "so this is the part of the story where everything goes completely sideways",
+ "normalized_hypothesis": "so this is the part of the story where everything goes completely sideways",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0130",
+ "category": "question",
+ "text": "Is there a version of this where it actually works out okay?",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0130.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Is there a version of this where it actually works out okay?",
+ "raw": {
+ "normalized_reference": "is there a version of this where it actually works out okay?",
+ "normalized_hypothesis": "is there a version of this where it actually works out okay?",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "is there a version of this where it actually works out okay",
+ "normalized_hypothesis": "is there a version of this where it actually works out okay",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0131",
+ "category": "short",
+ "text": "Oh no. Oh no no no.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0131.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Oh no, oh no, no, no.",
+ "raw": {
+ "normalized_reference": "oh no. oh no no no.",
+ "normalized_hypothesis": "oh no, oh no, no, no.",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "0 no 0 no no no",
+ "normalized_hypothesis": "0 no 0 no no no",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0132",
+ "category": "technical",
+ "text": "Always test on a copy before modifying the original.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0132.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Always test on a copy before modifying the original.",
+ "raw": {
+ "normalized_reference": "always test on a copy before modifying the original.",
+ "normalized_hypothesis": "always test on a copy before modifying the original.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "always test on a copy before modifying the original",
+ "normalized_hypothesis": "always test on a copy before modifying the original",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0133",
+ "category": "conversational",
+ "text": "I wasn't even supposed to be there. That's the wild part.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0133.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I wasn't even supposed to be there. That's the wild part.",
+ "raw": {
+ "normalized_reference": "i wasn't even supposed to be there. that's the wild part.",
+ "normalized_hypothesis": "i wasn't even supposed to be there. that's the wild part.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i was not even supposed to be there that is the wild part",
+ "normalized_hypothesis": "i was not even supposed to be there that is the wild part",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0134",
+ "category": "descriptive",
+ "text": "It's a small town. Not the kind you leave. The kind you leave and come back to.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0134.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It's a small town. Not the kind you leave. The kind you leave and come back to.",
+ "raw": {
+ "normalized_reference": "it's a small town. not the kind you leave. the kind you leave and come back to.",
+ "normalized_hypothesis": "it's a small town. not the kind you leave. the kind you leave and come back to.",
+ "errors": 0,
+ "words": 17,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it is a small town not the kind you leave the kind you leave and come back to",
+ "normalized_hypothesis": "it is a small town not the kind you leave the kind you leave and come back to",
+ "errors": 0,
+ "words": 18,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0135",
+ "category": "emotional",
+ "text": "She cried. I cried. We all just sort of stood there crying.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0135.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "She cried. I cry. We all just sort of stood there crying.",
+ "raw": {
+ "normalized_reference": "she cried. i cried. we all just sort of stood there crying.",
+ "normalized_hypothesis": "she cried. i cry. we all just sort of stood there crying.",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "she cried i cried we all just sort of stood there crying",
+ "normalized_hypothesis": "she cried i cry we all just sort of stood there crying",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ }
+ },
+ {
+ "id": "release500_0136",
+ "category": "long",
+ "text": "There's a version of being busy that's actually just a very elaborate way of avoiding the one thing you know you should be doing.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0136.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "There's a version of being busy that's actually just a very elaborate way of avoiding the one thing you know you should be doing.",
+ "raw": {
+ "normalized_reference": "there's a version of being busy that's actually just a very elaborate way of avoiding the one thing you know you should be doing.",
+ "normalized_hypothesis": "there's a version of being busy that's actually just a very elaborate way of avoiding the one thing you know you should be doing.",
+ "errors": 0,
+ "words": 24,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "there is a version of being busy that is actually just a very elaborate way of avoiding the one thing you know you should be doing",
+ "normalized_hypothesis": "there is a version of being busy that is actually just a very elaborate way of avoiding the one thing you know you should be doing",
+ "errors": 0,
+ "words": 26,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0137",
+ "category": "narrative",
+ "text": "The letter arrived on a Tuesday, which felt appropriate somehow.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0137.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The letter arrived on a Tuesday, which felt appropriate somehow.",
+ "raw": {
+ "normalized_reference": "the letter arrived on a tuesday, which felt appropriate somehow.",
+ "normalized_hypothesis": "the letter arrived on a tuesday, which felt appropriate somehow.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the letter arrived on a tuesday which felt appropriate somehow",
+ "normalized_hypothesis": "the letter arrived on a tuesday which felt appropriate somehow",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0138",
+ "category": "question",
+ "text": "What exactly did she say when you told her?",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0138.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "What exactly did she say when you told her?",
+ "raw": {
+ "normalized_reference": "what exactly did she say when you told her?",
+ "normalized_hypothesis": "what exactly did she say when you told her?",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "what exactly did she say when you told her",
+ "normalized_hypothesis": "what exactly did she say when you told her",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0139",
+ "category": "short",
+ "text": "No way. Absolutely not.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0139.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "No way. Absolutely not.",
+ "raw": {
+ "normalized_reference": "no way. absolutely not.",
+ "normalized_hypothesis": "no way. absolutely not.",
+ "errors": 0,
+ "words": 4,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "no way absolutely not",
+ "normalized_hypothesis": "no way absolutely not",
+ "errors": 0,
+ "words": 4,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0140",
+ "category": "technical",
+ "text": "The first thing you want to do is make sure everything is backed up.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0140.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The first thing you want to do is make sure everything is backed up.",
+ "raw": {
+ "normalized_reference": "the first thing you want to do is make sure everything is backed up.",
+ "normalized_hypothesis": "the first thing you want to do is make sure everything is backed up.",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the 1st thing you want to do is make sure everything is backed up",
+ "normalized_hypothesis": "the 1st thing you want to do is make sure everything is backed up",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0141",
+ "category": "conversational",
+ "text": "I have the most insane idea and you need to just hear me out.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0141.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I have the most insane idea and you need to just hear me out.",
+ "raw": {
+ "normalized_reference": "i have the most insane idea and you need to just hear me out.",
+ "normalized_hypothesis": "i have the most insane idea and you need to just hear me out.",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i have the most insane idea and you need to just hear me out",
+ "normalized_hypothesis": "i have the most insane idea and you need to just hear me out",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0142",
+ "category": "descriptive",
+ "text": "There's a coffee shop on the corner that's been there longer than anyone can remember.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0142.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "There's a coffee shop on the corner that's been there longer than anyone can remember.",
+ "raw": {
+ "normalized_reference": "there's a coffee shop on the corner that's been there longer than anyone can remember.",
+ "normalized_hypothesis": "there's a coffee shop on the corner that's been there longer than anyone can remember.",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "there is a coffee shop on the corner that has been there longer than anyone can remember",
+ "normalized_hypothesis": "there is a coffee shop on the corner that has been there longer than anyone can remember",
+ "errors": 0,
+ "words": 17,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0143",
+ "category": "emotional",
+ "text": "I kept it together until I got to the car, and then I completely fell apart.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0143.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I kept it together until I got to the car and then I completely fell apart.",
+ "raw": {
+ "normalized_reference": "i kept it together until i got to the car, and then i completely fell apart.",
+ "normalized_hypothesis": "i kept it together until i got to the car and then i completely fell apart.",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i kept it together until i got to the car and then i completely fell apart",
+ "normalized_hypothesis": "i kept it together until i got to the car and then i completely fell apart",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0144",
+ "category": "long",
+ "text": "The older I get, the more I think that most of what we call wisdom is just pattern recognition built from having been wrong enough times to start noticing.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0144.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The older I get, the more I think that most of what we call wisdom is just pattern recognition built from having been wrong enough times to start noticing.",
+ "raw": {
+ "normalized_reference": "the older i get, the more i think that most of what we call wisdom is just pattern recognition built from having been wrong enough times to start noticing.",
+ "normalized_hypothesis": "the older i get, the more i think that most of what we call wisdom is just pattern recognition built from having been wrong enough times to start noticing.",
+ "errors": 0,
+ "words": 29,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the older i get the more i think that most of what we call wisdom is just pattern recognition built from having been wrong enough times to start noticing",
+ "normalized_hypothesis": "the older i get the more i think that most of what we call wisdom is just pattern recognition built from having been wrong enough times to start noticing",
+ "errors": 0,
+ "words": 29,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0145",
+ "category": "narrative",
+ "text": "I remember the exact moment I realized I had made a huge mistake.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0145.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I remember the exact moment I realized I had made a huge mistake.",
+ "raw": {
+ "normalized_reference": "i remember the exact moment i realized i had made a huge mistake.",
+ "normalized_hypothesis": "i remember the exact moment i realized i had made a huge mistake.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i remember the exact moment i realized i had made a huge mistake",
+ "normalized_hypothesis": "i remember the exact moment i realized i had made a huge mistake",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0146",
+ "category": "question",
+ "text": "Do you think it would have gone differently if we'd left earlier?",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0146.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Do you think it would have gone differently if we'd left earlier?",
+ "raw": {
+ "normalized_reference": "do you think it would have gone differently if we'd left earlier?",
+ "normalized_hypothesis": "do you think it would have gone differently if we'd left earlier?",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "do you think it would have gone differently if we would left earlier",
+ "normalized_hypothesis": "do you think it would have gone differently if we would left earlier",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0147",
+ "category": "short",
+ "text": "You have to be kidding me.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0147.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "You have to be kidding me.",
+ "raw": {
+ "normalized_reference": "you have to be kidding me.",
+ "normalized_hypothesis": "you have to be kidding me.",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "you have to be kidding me",
+ "normalized_hypothesis": "you have to be kidding me",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0148",
+ "category": "technical",
+ "text": "If the checksum fails, restart the worker process without deleting the cached tensors.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0148.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "If the checksum fails, restart the worker process without deleting the cache tensors.",
+ "raw": {
+ "normalized_reference": "if the checksum fails, restart the worker process without deleting the cached tensors.",
+ "normalized_hypothesis": "if the checksum fails, restart the worker process without deleting the cache tensors.",
+ "errors": 1,
+ "words": 13,
+ "wer": 0.07692307692307693
+ },
+ "semantic": {
+ "normalized_reference": "if the checksum fails restart the worker process without deleting the cached tensors",
+ "normalized_hypothesis": "if the checksum fails restart the worker process without deleting the cache tensors",
+ "errors": 1,
+ "words": 13,
+ "wer": 0.07692307692307693
+ }
+ },
+ {
+ "id": "release500_0149",
+ "category": "conversational",
+ "text": "I need you to understand how close that was to not working.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0149.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I need you to understand how close that was to not working.",
+ "raw": {
+ "normalized_reference": "i need you to understand how close that was to not working.",
+ "normalized_hypothesis": "i need you to understand how close that was to not working.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i need you to understand how close that was to not working",
+ "normalized_hypothesis": "i need you to understand how close that was to not working",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0150",
+ "category": "descriptive",
+ "text": "It was the sort of morning where the whole world looks rinsed clean.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0150.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It was the sort of morning where the whole world looks rinsed clean.",
+ "raw": {
+ "normalized_reference": "it was the sort of morning where the whole world looks rinsed clean.",
+ "normalized_hypothesis": "it was the sort of morning where the whole world looks rinsed clean.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it was the sort of morning where the whole world looks rinsed clean",
+ "normalized_hypothesis": "it was the sort of morning where the whole world looks rinsed clean",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0151",
+ "category": "emotional",
+ "text": "She started laughing and I started laughing and we couldn't stop for a full minute.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0151.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "She started laughing and I started laughing and we couldn't stop for a full minute.",
+ "raw": {
+ "normalized_reference": "she started laughing and i started laughing and we couldn't stop for a full minute.",
+ "normalized_hypothesis": "she started laughing and i started laughing and we couldn't stop for a full minute.",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "she started laughing and i started laughing and we could not stop for a full minute",
+ "normalized_hypothesis": "she started laughing and i started laughing and we could not stop for a full minute",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0152",
+ "category": "long",
+ "text": "We spent three years building something we were genuinely proud of, and then the market shifted and none of it mattered the way we thought it would.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0152.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "We spent three years building something we were genuinely proud of, and then the market shifted and none of it mattered the way we thought it would.",
+ "raw": {
+ "normalized_reference": "we spent three years building something we were genuinely proud of, and then the market shifted and none of it mattered the way we thought it would.",
+ "normalized_hypothesis": "we spent three years building something we were genuinely proud of, and then the market shifted and none of it mattered the way we thought it would.",
+ "errors": 0,
+ "words": 27,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "we spent 3 years building something we were genuinely proud of and then the market shifted and none of it mattered the way we thought it would",
+ "normalized_hypothesis": "we spent 3 years building something we were genuinely proud of and then the market shifted and none of it mattered the way we thought it would",
+ "errors": 0,
+ "words": 27,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0153",
+ "category": "narrative",
+ "text": "And somehow, despite all of that, it actually turned out okay.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0153.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "And somehow, despite all of that, it actually turned out okay.",
+ "raw": {
+ "normalized_reference": "and somehow, despite all of that, it actually turned out okay.",
+ "normalized_hypothesis": "and somehow, despite all of that, it actually turned out okay.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "and somehow despite all of that it actually turned out okay",
+ "normalized_hypothesis": "and somehow despite all of that it actually turned out okay",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0154",
+ "category": "question",
+ "text": "Is that something you actually want, or something you think you should want?",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0154.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Is that something you actually want or something you think you should want?",
+ "raw": {
+ "normalized_reference": "is that something you actually want, or something you think you should want?",
+ "normalized_hypothesis": "is that something you actually want or something you think you should want?",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "is that something you actually want or something you think you should want",
+ "normalized_hypothesis": "is that something you actually want or something you think you should want",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0155",
+ "category": "short",
+ "text": "This is fine.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0155.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "This is fine.",
+ "raw": {
+ "normalized_reference": "this is fine.",
+ "normalized_hypothesis": "this is fine.",
+ "errors": 0,
+ "words": 3,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "this is fine",
+ "normalized_hypothesis": "this is fine",
+ "errors": 0,
+ "words": 3,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0156",
+ "category": "technical",
+ "text": "The tricky part isn't the setup, it's making sure all dependencies line up.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0156.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The tricky part isn't the setup, it's making sure all dependencies line up.",
+ "raw": {
+ "normalized_reference": "the tricky part isn't the setup, it's making sure all dependencies line up.",
+ "normalized_hypothesis": "the tricky part isn't the setup, it's making sure all dependencies line up.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the tricky part is not the setup it is making sure all dependencies line up",
+ "normalized_hypothesis": "the tricky part is not the setup it is making sure all dependencies line up",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0157",
+ "category": "conversational",
+ "text": "He called it a minor setback. I called it a disaster.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0157.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He called it a minor setback. I called it a disaster.",
+ "raw": {
+ "normalized_reference": "he called it a minor setback. i called it a disaster.",
+ "normalized_hypothesis": "he called it a minor setback. i called it a disaster.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he called it a minor setback i called it a disaster",
+ "normalized_hypothesis": "he called it a minor setback i called it a disaster",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0158",
+ "category": "descriptive",
+ "text": "The whole place smelled like pine and old wood and cold air.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0158.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The whole place smelled like pine and old wood in cold air.",
+ "raw": {
+ "normalized_reference": "the whole place smelled like pine and old wood and cold air.",
+ "normalized_hypothesis": "the whole place smelled like pine and old wood in cold air.",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "the whole place smelled like pine and old wood and cold air",
+ "normalized_hypothesis": "the whole place smelled like pine and old wood in cold air",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ }
+ },
+ {
+ "id": "release500_0159",
+ "category": "emotional",
+ "text": "I've never been so relieved in my entire life.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0159.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I've never been so relieved in my entire life.",
+ "raw": {
+ "normalized_reference": "i've never been so relieved in my entire life.",
+ "normalized_hypothesis": "i've never been so relieved in my entire life.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i have never been so relieved in my entire life",
+ "normalized_hypothesis": "i have never been so relieved in my entire life",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0160",
+ "category": "narrative",
+ "text": "I wrote it all down that night so I wouldn't forget, and I still haven't looked at it.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0160.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I wrote it all down that night so I wouldn't forget, and I still haven't looked at it.",
+ "raw": {
+ "normalized_reference": "i wrote it all down that night so i wouldn't forget, and i still haven't looked at it.",
+ "normalized_hypothesis": "i wrote it all down that night so i wouldn't forget, and i still haven't looked at it.",
+ "errors": 0,
+ "words": 18,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i wrote it all down that night so i would not forget and i still have not looked at it",
+ "normalized_hypothesis": "i wrote it all down that night so i would not forget and i still have not looked at it",
+ "errors": 0,
+ "words": 20,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0161",
+ "category": "short",
+ "text": "That's kind of wild actually.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0161.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "That's kind of wild, actually.",
+ "raw": {
+ "normalized_reference": "that's kind of wild actually.",
+ "normalized_hypothesis": "that's kind of wild, actually.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "that is kind of wild actually",
+ "normalized_hypothesis": "that is kind of wild actually",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0162",
+ "category": "technical",
+ "text": "Version pinning will save you a lot of pain later down the line.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0162.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Version pinning will save you a lot of pain later down the line.",
+ "raw": {
+ "normalized_reference": "version pinning will save you a lot of pain later down the line.",
+ "normalized_hypothesis": "version pinning will save you a lot of pain later down the line.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "version pinning will save you a lot of pain later down the line",
+ "normalized_hypothesis": "version pinning will save you a lot of pain later down the line",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0163",
+ "category": "conversational",
+ "text": "At some point you just have to commit and see what happens.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0163.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "At some point you just have to commit and see what happens.",
+ "raw": {
+ "normalized_reference": "at some point you just have to commit and see what happens.",
+ "normalized_hypothesis": "at some point you just have to commit and see what happens.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "at some point you just have to commit and see what happens",
+ "normalized_hypothesis": "at some point you just have to commit and see what happens",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0164",
+ "category": "emotional",
+ "text": "Nobody told me it was going to be this hard, and honestly, I'm grateful.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0164.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Nobody told me it was going to be this hard, and honestly, I'm grateful.",
+ "raw": {
+ "normalized_reference": "nobody told me it was going to be this hard, and honestly, i'm grateful.",
+ "normalized_hypothesis": "nobody told me it was going to be this hard, and honestly, i'm grateful.",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "nobody told me it was going to be this hard and honestly i am grateful",
+ "normalized_hypothesis": "nobody told me it was going to be this hard and honestly i am grateful",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0165",
+ "category": "narrative",
+ "text": "Chapter seven. In which everything that could go wrong did, and a few things that couldn't.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0165.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Chapter 7, in which everything that could go wrong did, and a few things that couldn't.",
+ "raw": {
+ "normalized_reference": "chapter seven. in which everything that could go wrong did, and a few things that couldn't.",
+ "normalized_hypothesis": "chapter 7, in which everything that could go wrong did, and a few things that couldn't.",
+ "errors": 1,
+ "words": 16,
+ "wer": 0.0625
+ },
+ "semantic": {
+ "normalized_reference": "chapter 7 in which everything that could go wrong did and a few things that could not",
+ "normalized_hypothesis": "chapter 7 in which everything that could go wrong did and a few things that could not",
+ "errors": 0,
+ "words": 17,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0166",
+ "category": "short",
+ "text": "Wait, what?",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0166.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Wait, what?",
+ "raw": {
+ "normalized_reference": "wait, what?",
+ "normalized_hypothesis": "wait, what?",
+ "errors": 0,
+ "words": 2,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "wait what",
+ "normalized_hypothesis": "wait what",
+ "errors": 0,
+ "words": 2,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0167",
+ "category": "technical",
+ "text": "The graphics processor remained idle because the data loader exhausted every central processing core.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0167.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The graphics processor remained idle because the data loader exhausted every central processing core.",
+ "raw": {
+ "normalized_reference": "the graphics processor remained idle because the data loader exhausted every central processing core.",
+ "normalized_hypothesis": "the graphics processor remained idle because the data loader exhausted every central processing core.",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the graphics processor remained idle because the data loader exhausted every central processing core",
+ "normalized_hypothesis": "the graphics processor remained idle because the data loader exhausted every central processing core",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0168",
+ "category": "conversational",
+ "text": "Could you speak a little more slowly? I am trying to write every detail down.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0168.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Could you speak a little more swally? I'm trying to write every detail down.",
+ "raw": {
+ "normalized_reference": "could you speak a little more slowly? i am trying to write every detail down.",
+ "normalized_hypothesis": "could you speak a little more swally? i'm trying to write every detail down.",
+ "errors": 3,
+ "words": 15,
+ "wer": 0.2
+ },
+ "semantic": {
+ "normalized_reference": "could you speak a little more slowly i am trying to write every detail down",
+ "normalized_hypothesis": "could you speak a little more swally i am trying to write every detail down",
+ "errors": 1,
+ "words": 15,
+ "wer": 0.06666666666666667
+ }
+ },
+ {
+ "id": "release500_0169",
+ "category": "emotional",
+ "text": "I was so excited I could barely form a complete sentence.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0169.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I was so excited I could barely form a complete sentence.",
+ "raw": {
+ "normalized_reference": "i was so excited i could barely form a complete sentence.",
+ "normalized_hypothesis": "i was so excited i could barely form a complete sentence.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i was so excited i could barely form a complete sentence",
+ "normalized_hypothesis": "i was so excited i could barely form a complete sentence",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0170",
+ "category": "narrative",
+ "text": "He looked at me like I had said something in a language he didn't speak.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0170.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He looked at me like I had said something in a language he didn't speak.",
+ "raw": {
+ "normalized_reference": "he looked at me like i had said something in a language he didn't speak.",
+ "normalized_hypothesis": "he looked at me like i had said something in a language he didn't speak.",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he looked at me like i had said something in a language he did not speak",
+ "normalized_hypothesis": "he looked at me like i had said something in a language he did not speak",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0171",
+ "category": "short",
+ "text": "That's actually hilarious.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0171.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "That's actually hilarious.",
+ "raw": {
+ "normalized_reference": "that's actually hilarious.",
+ "normalized_hypothesis": "that's actually hilarious.",
+ "errors": 0,
+ "words": 3,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "that is actually hilarious",
+ "normalized_hypothesis": "that is actually hilarious",
+ "errors": 0,
+ "words": 4,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0172",
+ "category": "technical",
+ "text": "The encoder maps phonemes, stress, and phrase boundaries into a shared hidden representation.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0172.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The encoder maps phonemes, stress, and phrase boundaries into a shared hidden representation.",
+ "raw": {
+ "normalized_reference": "the encoder maps phonemes, stress, and phrase boundaries into a shared hidden representation.",
+ "normalized_hypothesis": "the encoder maps phonemes, stress, and phrase boundaries into a shared hidden representation.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the encoder maps phonemes stress and phrase boundaries into a shared hidden representation",
+ "normalized_hypothesis": "the encoder maps phonemes stress and phrase boundaries into a shared hidden representation",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0173",
+ "category": "conversational",
+ "text": "She showed up three hours late and acted like nothing had happened.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0173.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "She showed up three hours late and acted like nothing had happened.",
+ "raw": {
+ "normalized_reference": "she showed up three hours late and acted like nothing had happened.",
+ "normalized_hypothesis": "she showed up three hours late and acted like nothing had happened.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "she showed up 3 hours late and acted like nothing had happened",
+ "normalized_hypothesis": "she showed up 3 hours late and acted like nothing had happened",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0174",
+ "category": "emotional",
+ "text": "Oh my god. This is the best thing I've ever tasted.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0174.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Oh my god, this is the best thing I've ever tasted.",
+ "raw": {
+ "normalized_reference": "oh my god. this is the best thing i've ever tasted.",
+ "normalized_hypothesis": "oh my god, this is the best thing i've ever tasted.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "0 my god this is the best thing i have ever tasted",
+ "normalized_hypothesis": "0 my god this is the best thing i have ever tasted",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0175",
+ "category": "narrative",
+ "text": "The note was short, which was somehow more frightening than a long one would have been.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0175.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The note was short, which was somehow more frightening than a long one would have been.",
+ "raw": {
+ "normalized_reference": "the note was short, which was somehow more frightening than a long one would have been.",
+ "normalized_hypothesis": "the note was short, which was somehow more frightening than a long one would have been.",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the note was short which was somehow more frightening than a long one would have been",
+ "normalized_hypothesis": "the note was short which was somehow more frightening than a long one would have been",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0176",
+ "category": "short",
+ "text": "Noted.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0176.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Noted.",
+ "raw": {
+ "normalized_reference": "noted.",
+ "normalized_hypothesis": "noted.",
+ "errors": 0,
+ "words": 1,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "noted",
+ "normalized_hypothesis": "noted",
+ "errors": 0,
+ "words": 1,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0177",
+ "category": "technical",
+ "text": "It's a simple process once you understand the underlying structure.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0177.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It's a simple process once you understand the underlying structure.",
+ "raw": {
+ "normalized_reference": "it's a simple process once you understand the underlying structure.",
+ "normalized_hypothesis": "it's a simple process once you understand the underlying structure.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it is a simple process once you understand the underlying structure",
+ "normalized_hypothesis": "it is a simple process once you understand the underlying structure",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0178",
+ "category": "conversational",
+ "text": "You have to hear this story because it's almost too good to be true.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0178.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "You have to hear this story because it's almost too good to be true.",
+ "raw": {
+ "normalized_reference": "you have to hear this story because it's almost too good to be true.",
+ "normalized_hypothesis": "you have to hear this story because it's almost too good to be true.",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "you have to hear this story because it is almost too good to be true",
+ "normalized_hypothesis": "you have to hear this story because it is almost too good to be true",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0179",
+ "category": "emotional",
+ "text": "I have never felt more seen in my entire life than in that moment.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0179.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I have never felt more seen in my entire life than in that moment.",
+ "raw": {
+ "normalized_reference": "i have never felt more seen in my entire life than in that moment.",
+ "normalized_hypothesis": "i have never felt more seen in my entire life than in that moment.",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i have never felt more seen in my entire life than in that moment",
+ "normalized_hypothesis": "i have never felt more seen in my entire life than in that moment",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0180",
+ "category": "narrative",
+ "text": "She knew. She had always known. She just waited for the right moment.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0180.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "She knew. She had always known. She just waited for the right moment.",
+ "raw": {
+ "normalized_reference": "she knew. she had always known. she just waited for the right moment.",
+ "normalized_hypothesis": "she knew. she had always known. she just waited for the right moment.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "she knew she had always known she just waited for the right moment",
+ "normalized_hypothesis": "she knew she had always known she just waited for the right moment",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0181",
+ "category": "short",
+ "text": "Yeah, no. Hard pass.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0181.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Y'all know. Hard pass.",
+ "raw": {
+ "normalized_reference": "yeah, no. hard pass.",
+ "normalized_hypothesis": "y'all know. hard pass.",
+ "errors": 2,
+ "words": 4,
+ "wer": 0.5
+ },
+ "semantic": {
+ "normalized_reference": "yeah no hard pass",
+ "normalized_hypothesis": "you all know hard pass",
+ "errors": 3,
+ "words": 4,
+ "wer": 0.75
+ }
+ },
+ {
+ "id": "release500_0182",
+ "category": "technical",
+ "text": "Please verify the sample rate, hop length, and mel filter configuration before inference.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0182.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Please verify the sample rate, hop length, and MEL filter configuration before inference.",
+ "raw": {
+ "normalized_reference": "please verify the sample rate, hop length, and mel filter configuration before inference.",
+ "normalized_hypothesis": "please verify the sample rate, hop length, and mel filter configuration before inference.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "please verify the sample rate hop length and mel filter configuration before inference",
+ "normalized_hypothesis": "please verify the sample rate hop length and mel filter configuration before inference",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0183",
+ "category": "conversational",
+ "text": "That's the worst idea I've ever heard and I'm completely in.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0183.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "That's the worst idea I've ever heard and I'm completely in.",
+ "raw": {
+ "normalized_reference": "that's the worst idea i've ever heard and i'm completely in.",
+ "normalized_hypothesis": "that's the worst idea i've ever heard and i'm completely in.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "that is the worst idea i have ever heard and i am completely in",
+ "normalized_hypothesis": "that is the worst idea i have ever heard and i am completely in",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0184",
+ "category": "emotional",
+ "text": "I remember thinking: this is it. This is the thing I've been waiting for.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0184.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I remember thinking, this is it. This is the thing I've been waiting for.",
+ "raw": {
+ "normalized_reference": "i remember thinking: this is it. this is the thing i've been waiting for.",
+ "normalized_hypothesis": "i remember thinking, this is it. this is the thing i've been waiting for.",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i remember thinking this is it this is the thing i have been waiting for",
+ "normalized_hypothesis": "i remember thinking this is it this is the thing i have been waiting for",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0185",
+ "category": "narrative",
+ "text": "We never talked about it again, which I think was the right call.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0185.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "We never talked about it again, which I think was the right call.",
+ "raw": {
+ "normalized_reference": "we never talked about it again, which i think was the right call.",
+ "normalized_hypothesis": "we never talked about it again, which i think was the right call.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "we never talked about it again which i think was the right call",
+ "normalized_hypothesis": "we never talked about it again which i think was the right call",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0186",
+ "category": "short",
+ "text": "In hindsight, obvious.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0186.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "In hindsight, obvious.",
+ "raw": {
+ "normalized_reference": "in hindsight, obvious.",
+ "normalized_hypothesis": "in hindsight, obvious.",
+ "errors": 0,
+ "words": 3,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "in hindsight obvious",
+ "normalized_hypothesis": "in hindsight obvious",
+ "errors": 0,
+ "words": 3,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0187",
+ "category": "technical",
+ "text": "Once it's configured correctly, it basically runs itself.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0187.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Once it's configured correctly, it basically runs itself.",
+ "raw": {
+ "normalized_reference": "once it's configured correctly, it basically runs itself.",
+ "normalized_hypothesis": "once it's configured correctly, it basically runs itself.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "once it is configured correctly it basically runs itself",
+ "normalized_hypothesis": "once it is configured correctly it basically runs itself",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0188",
+ "category": "conversational",
+ "text": "You got engaged? When? How? Tell me everything right now.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0188.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "You got engaged. What? Oh. Tell me everything right now.",
+ "raw": {
+ "normalized_reference": "you got engaged? when? how? tell me everything right now.",
+ "normalized_hypothesis": "you got engaged. what? oh. tell me everything right now.",
+ "errors": 2,
+ "words": 10,
+ "wer": 0.2
+ },
+ "semantic": {
+ "normalized_reference": "you got engaged when how tell me everything right now",
+ "normalized_hypothesis": "you got engaged what 0 tell me everything right now",
+ "errors": 2,
+ "words": 10,
+ "wer": 0.2
+ }
+ },
+ {
+ "id": "release500_0189",
+ "category": "emotional",
+ "text": "It's hard to explain how much that moment meant to me.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0189.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It's hard to explain how much that moment meant to me.",
+ "raw": {
+ "normalized_reference": "it's hard to explain how much that moment meant to me.",
+ "normalized_hypothesis": "it's hard to explain how much that moment meant to me.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it is hard to explain how much that moment meant to me",
+ "normalized_hypothesis": "it is hard to explain how much that moment meant to me",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0190",
+ "category": "narrative",
+ "text": "He told the story three different times that night and somehow it got better each time.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0190.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He told the story three different times that night and somehow it got better each time.",
+ "raw": {
+ "normalized_reference": "he told the story three different times that night and somehow it got better each time.",
+ "normalized_hypothesis": "he told the story three different times that night and somehow it got better each time.",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he told the story 3 different times that night and somehow it got better each time",
+ "normalized_hypothesis": "he told the story 3 different times that night and somehow it got better each time",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0191",
+ "category": "short",
+ "text": "Right. Moving on.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0191.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Right, moving on.",
+ "raw": {
+ "normalized_reference": "right. moving on.",
+ "normalized_hypothesis": "right, moving on.",
+ "errors": 0,
+ "words": 3,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "right moving on",
+ "normalized_hypothesis": "right moving on",
+ "errors": 0,
+ "words": 3,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0192",
+ "category": "conversational",
+ "text": "No, I did not move your keys; check beside the small ceramic bowl.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0192.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "No, I did not move your keys. Check the side the small ceramic bowl.",
+ "raw": {
+ "normalized_reference": "no, i did not move your keys; check beside the small ceramic bowl.",
+ "normalized_hypothesis": "no, i did not move your keys. check the side the small ceramic bowl.",
+ "errors": 2,
+ "words": 13,
+ "wer": 0.15384615384615385
+ },
+ "semantic": {
+ "normalized_reference": "no i did not move your keys check beside the small ceramic bowl",
+ "normalized_hypothesis": "no i did not move your keys check the side the small ceramic bowl",
+ "errors": 2,
+ "words": 13,
+ "wer": 0.15384615384615385
+ }
+ },
+ {
+ "id": "release500_0193",
+ "category": "emotional",
+ "text": "I was devastated. I'm not going to pretend otherwise.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0193.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I was devastated. I'm not going to pretend otherwise.",
+ "raw": {
+ "normalized_reference": "i was devastated. i'm not going to pretend otherwise.",
+ "normalized_hypothesis": "i was devastated. i'm not going to pretend otherwise.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i was devastated i am not going to pretend otherwise",
+ "normalized_hypothesis": "i was devastated i am not going to pretend otherwise",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0194",
+ "category": "narrative",
+ "text": "That was the last summer before everything changed, though we didn't know it yet.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0194.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "That was the last summer before everything changed, though we didn't know it yet.",
+ "raw": {
+ "normalized_reference": "that was the last summer before everything changed, though we didn't know it yet.",
+ "normalized_hypothesis": "that was the last summer before everything changed, though we didn't know it yet.",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "that was the last summer before everything changed though we did not know it yet",
+ "normalized_hypothesis": "that was the last summer before everything changed though we did not know it yet",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0195",
+ "category": "short",
+ "text": "Bold of you to assume.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0195.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Bold of you to assume.",
+ "raw": {
+ "normalized_reference": "bold of you to assume.",
+ "normalized_hypothesis": "bold of you to assume.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "bold of you to assume",
+ "normalized_hypothesis": "bold of you to assume",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0196",
+ "category": "conversational",
+ "text": "It's one of those decisions that makes total sense in the moment.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0196.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It's one of those decisions that makes total sense in the moment.",
+ "raw": {
+ "normalized_reference": "it's one of those decisions that makes total sense in the moment.",
+ "normalized_hypothesis": "it's one of those decisions that makes total sense in the moment.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it is one of those decisions that makes total sense in the moment",
+ "normalized_hypothesis": "it is one of those decisions that makes total sense in the moment",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0197",
+ "category": "emotional",
+ "text": "I was furious. Not the loud kind. The quiet kind that lasts for days.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0197.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I was furious. Not the loud kind. The quiet kind that lasts for days.",
+ "raw": {
+ "normalized_reference": "i was furious. not the loud kind. the quiet kind that lasts for days.",
+ "normalized_hypothesis": "i was furious. not the loud kind. the quiet kind that lasts for days.",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i was furious not the loud kind the quiet kind that lasts for days",
+ "normalized_hypothesis": "i was furious not the loud kind the quiet kind that lasts for days",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0198",
+ "category": "narrative",
+ "text": "That's the part of the story I usually leave out.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0198.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "That's the part of the story I usually leave out.",
+ "raw": {
+ "normalized_reference": "that's the part of the story i usually leave out.",
+ "normalized_hypothesis": "that's the part of the story i usually leave out.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "that is the part of the story i usually leave out",
+ "normalized_hypothesis": "that is the part of the story i usually leave out",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0199",
+ "category": "short",
+ "text": "I've said too much.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0199.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I said too much.",
+ "raw": {
+ "normalized_reference": "i've said too much.",
+ "normalized_hypothesis": "i said too much.",
+ "errors": 1,
+ "words": 4,
+ "wer": 0.25
+ },
+ "semantic": {
+ "normalized_reference": "i have said too much",
+ "normalized_hypothesis": "i said too much",
+ "errors": 1,
+ "words": 5,
+ "wer": 0.2
+ }
+ },
+ {
+ "id": "release500_0200",
+ "category": "conversational",
+ "text": "I've been faking that I understood this the entire time.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0200.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I've been faking that I understood this the entire time.",
+ "raw": {
+ "normalized_reference": "i've been faking that i understood this the entire time.",
+ "normalized_hypothesis": "i've been faking that i understood this the entire time.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i have been faking that i understood this the entire time",
+ "normalized_hypothesis": "i have been faking that i understood this the entire time",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0201",
+ "category": "phonetic_balance",
+ "text": "Peace be unto you and grace before the Lord.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0201.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Peace be unto you and grace before the Lord.",
+ "raw": {
+ "normalized_reference": "peace be unto you and grace before the lord.",
+ "normalized_hypothesis": "peace be unto you and grace before the lord.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "peace be unto you and grace before the lord",
+ "normalized_hypothesis": "peace be unto you and grace before the lord",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0202",
+ "category": "phonetic_balance",
+ "text": "From now on we're pals.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0202.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "From now on we're pals.",
+ "raw": {
+ "normalized_reference": "from now on we're pals.",
+ "normalized_hypothesis": "from now on we're pals.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "from now on we are pals",
+ "normalized_hypothesis": "from now on we are pals",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0203",
+ "category": "phonetic_balance",
+ "text": "Here we allow our solicitors to look after our legal work.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0203.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Here we allow our solicitors to look after our illegal work.",
+ "raw": {
+ "normalized_reference": "here we allow our solicitors to look after our legal work.",
+ "normalized_hypothesis": "here we allow our solicitors to look after our illegal work.",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ },
+ "semantic": {
+ "normalized_reference": "here we allow our solicitors to look after our legal work",
+ "normalized_hypothesis": "here we allow our solicitors to look after our illegal work",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ }
+ },
+ {
+ "id": "release500_0204",
+ "category": "phonetic_balance",
+ "text": "Her words sent a strange chill through Philip.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0204.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Her words sent a strange chill through Philip.",
+ "raw": {
+ "normalized_reference": "her words sent a strange chill through philip.",
+ "normalized_hypothesis": "her words sent a strange chill through philip.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "her words sent a strange chill through philip",
+ "normalized_hypothesis": "her words sent a strange chill through philip",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0205",
+ "category": "phonetic_balance",
+ "text": "It was introduced by Representative Dick of Ohio.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0205.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It was introduced by Representative Dick of Ohio.",
+ "raw": {
+ "normalized_reference": "it was introduced by representative dick of ohio.",
+ "normalized_hypothesis": "it was introduced by representative dick of ohio.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it was introduced by representative dick of ohio",
+ "normalized_hypothesis": "it was introduced by representative dick of ohio",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0206",
+ "category": "phonetic_balance",
+ "text": "It's the aurora borealis.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0206.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It's the Aurora Borealis.",
+ "raw": {
+ "normalized_reference": "it's the aurora borealis.",
+ "normalized_hypothesis": "it's the aurora borealis.",
+ "errors": 0,
+ "words": 4,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it is the aurora borealis",
+ "normalized_hypothesis": "it is the aurora borealis",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0207",
+ "category": "phonetic_balance",
+ "text": "The boy grew and prospered.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0207.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The boy grew and prospered.",
+ "raw": {
+ "normalized_reference": "the boy grew and prospered.",
+ "normalized_hypothesis": "the boy grew and prospered.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the boy grew and prospered",
+ "normalized_hypothesis": "the boy grew and prospered",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0208",
+ "category": "phonetic_balance",
+ "text": "His voice was passionately rebellious.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0208.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "His voice was passionately rebellious.",
+ "raw": {
+ "normalized_reference": "his voice was passionately rebellious.",
+ "normalized_hypothesis": "his voice was passionately rebellious.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "his voice was passionately rebellious",
+ "normalized_hypothesis": "his voice was passionately rebellious",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0209",
+ "category": "phonetic_balance",
+ "text": "That's where they cut off the Scottish Chiefs and killed all hands.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0209.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "That's where they cut off the sky's chiefs and killed all hands.",
+ "raw": {
+ "normalized_reference": "that's where they cut off the scottish chiefs and killed all hands.",
+ "normalized_hypothesis": "that's where they cut off the sky's chiefs and killed all hands.",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "that is where they cut off the scottish chiefs and killed all hands",
+ "normalized_hypothesis": "that is where they cut off the sky is chiefs and killed all hands",
+ "errors": 2,
+ "words": 13,
+ "wer": 0.15384615384615385
+ }
+ },
+ {
+ "id": "release500_0210",
+ "category": "phonetic_balance",
+ "text": "And Tom King patiently endured.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0210.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "and Tom King patiently endured.",
+ "raw": {
+ "normalized_reference": "and tom king patiently endured.",
+ "normalized_hypothesis": "and tom king patiently endured.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "and tom king patiently endured",
+ "normalized_hypothesis": "and tom king patiently endured",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0211",
+ "category": "phonetic_balance",
+ "text": "From that moment his friendship for Belize turns to hatred and jealousy.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0211.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "From that moment his friendship for Belize turns to hatred and jealousy.",
+ "raw": {
+ "normalized_reference": "from that moment his friendship for belize turns to hatred and jealousy.",
+ "normalized_hypothesis": "from that moment his friendship for belize turns to hatred and jealousy.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "from that moment his friendship for belize turns to hatred and jealousy",
+ "normalized_hypothesis": "from that moment his friendship for belize turns to hatred and jealousy",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0212",
+ "category": "phonetic_balance",
+ "text": "Then you don't believe in altruism.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0212.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "then you don't believe in altruism.",
+ "raw": {
+ "normalized_reference": "then you don't believe in altruism.",
+ "normalized_hypothesis": "then you don't believe in altruism.",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "then you do not believe in altruism",
+ "normalized_hypothesis": "then you do not believe in altruism",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0213",
+ "category": "phonetic_balance",
+ "text": "What Pascal glimpsed with the vision of a seer, I have lived.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0213.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "What Pascal glimpsed with the vision of a seer, I have lived.",
+ "raw": {
+ "normalized_reference": "what pascal glimpsed with the vision of a seer, i have lived.",
+ "normalized_hypothesis": "what pascal glimpsed with the vision of a seer, i have lived.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "what pascal glimpsed with the vision of a seer i have lived",
+ "normalized_hypothesis": "what pascal glimpsed with the vision of a seer i have lived",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0214",
+ "category": "phonetic_balance",
+ "text": "You were making them talk shop, Ruth charged him.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0214.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "You were making them talk shop. Ruth charged him.",
+ "raw": {
+ "normalized_reference": "you were making them talk shop, ruth charged him.",
+ "normalized_hypothesis": "you were making them talk shop. ruth charged him.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "you were making them talk shop ruth charged him",
+ "normalized_hypothesis": "you were making them talk shop ruth charged him",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0215",
+ "category": "phonetic_balance",
+ "text": "Philip snatched at the letter which Gregson held out to him.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0215.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Philip snatched at the letter which Gregson held out to him.",
+ "raw": {
+ "normalized_reference": "philip snatched at the letter which gregson held out to him.",
+ "normalized_hypothesis": "philip snatched at the letter which gregson held out to him.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "philip snatched at the letter which gregson held out to him",
+ "normalized_hypothesis": "philip snatched at the letter which gregson held out to him",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0216",
+ "category": "phonetic_balance",
+ "text": "For a time the exciting thrill of his adventure was gone.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0216.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "For a time the exciting thrill of his adventure was gone.",
+ "raw": {
+ "normalized_reference": "for a time the exciting thrill of his adventure was gone.",
+ "normalized_hypothesis": "for a time the exciting thrill of his adventure was gone.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "for a time the exciting thrill of his adventure was gone",
+ "normalized_hypothesis": "for a time the exciting thrill of his adventure was gone",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0217",
+ "category": "phonetic_balance",
+ "text": "Obviously, it was a disease that could be contracted by contact.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0217.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Obviously, it was a disease that could be contracted by contact.",
+ "raw": {
+ "normalized_reference": "obviously, it was a disease that could be contracted by contact.",
+ "normalized_hypothesis": "obviously, it was a disease that could be contracted by contact.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "obviously it was a disease that could be contracted by contact",
+ "normalized_hypothesis": "obviously it was a disease that could be contracted by contact",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0218",
+ "category": "phonetic_balance",
+ "text": "Blind with rage, he darted in.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0218.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Aligned with rage, he darted in.",
+ "raw": {
+ "normalized_reference": "blind with rage, he darted in.",
+ "normalized_hypothesis": "aligned with rage, he darted in.",
+ "errors": 1,
+ "words": 6,
+ "wer": 0.16666666666666666
+ },
+ "semantic": {
+ "normalized_reference": "blind with rage he darted in",
+ "normalized_hypothesis": "aligned with rage he darted in",
+ "errors": 1,
+ "words": 6,
+ "wer": 0.16666666666666666
+ }
+ },
+ {
+ "id": "release500_0219",
+ "category": "phonetic_balance",
+ "text": "You live on an income which your father earned.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0219.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "You live on an income which your father earned.",
+ "raw": {
+ "normalized_reference": "you live on an income which your father earned.",
+ "normalized_hypothesis": "you live on an income which your father earned.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "you live on an income which your father earned",
+ "normalized_hypothesis": "you live on an income which your father earned",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0220",
+ "category": "phonetic_balance",
+ "text": "Something about it seemed to fascinate him, to challenge his presence.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0220.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Something about it seemed to fascinate him, to challenge his presence.",
+ "raw": {
+ "normalized_reference": "something about it seemed to fascinate him, to challenge his presence.",
+ "normalized_hypothesis": "something about it seemed to fascinate him, to challenge his presence.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "something about it seemed to fascinate him to challenge his presence",
+ "normalized_hypothesis": "something about it seemed to fascinate him to challenge his presence",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0221",
+ "category": "phonetic_balance",
+ "text": "Her beautiful hair was done up in shining coils.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0221.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Her beautiful hair was done up in shining coils.",
+ "raw": {
+ "normalized_reference": "her beautiful hair was done up in shining coils.",
+ "normalized_hypothesis": "her beautiful hair was done up in shining coils.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "her beautiful hair was done up in shining coils",
+ "normalized_hypothesis": "her beautiful hair was done up in shining coils",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0222",
+ "category": "phonetic_balance",
+ "text": "It does, was her audacious answer.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0222.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It does, was her audacious answer.",
+ "raw": {
+ "normalized_reference": "it does, was her audacious answer.",
+ "normalized_hypothesis": "it does, was her audacious answer.",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it does was her audacious answer",
+ "normalized_hypothesis": "it does was her audacious answer",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0223",
+ "category": "phonetic_balance",
+ "text": "He caught himself with a jerk.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0223.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He caught himself with a jerk.",
+ "raw": {
+ "normalized_reference": "he caught himself with a jerk.",
+ "normalized_hypothesis": "he caught himself with a jerk.",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he caught himself with a jerk",
+ "normalized_hypothesis": "he caught himself with a jerk",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0224",
+ "category": "phonetic_balance",
+ "text": "He did not believe in the burning of daylight for such a luxury.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0224.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He did not believe in the burning of daylight for such a luxury.",
+ "raw": {
+ "normalized_reference": "he did not believe in the burning of daylight for such a luxury.",
+ "normalized_hypothesis": "he did not believe in the burning of daylight for such a luxury.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he did not believe in the burning of daylight for such a luxury",
+ "normalized_hypothesis": "he did not believe in the burning of daylight for such a luxury",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0225",
+ "category": "phonetic_balance",
+ "text": "He had comparatively no advantages at first.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0225.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He had comparatively no advantages at first.",
+ "raw": {
+ "normalized_reference": "he had comparatively no advantages at first.",
+ "normalized_hypothesis": "he had comparatively no advantages at first.",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he had comparatively no advantages at 1st",
+ "normalized_hypothesis": "he had comparatively no advantages at 1st",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0226",
+ "category": "phonetic_balance",
+ "text": "Such men believe, when they come together.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0226.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Such men believe, when they come together,",
+ "raw": {
+ "normalized_reference": "such men believe, when they come together.",
+ "normalized_hypothesis": "such men believe, when they come together,",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "such men believe when they come together",
+ "normalized_hypothesis": "such men believe when they come together",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0227",
+ "category": "phonetic_balance",
+ "text": "The ship should be in within a week or ten days.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0227.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The ship should be in within a week or 10 days.",
+ "raw": {
+ "normalized_reference": "the ship should be in within a week or ten days.",
+ "normalized_hypothesis": "the ship should be in within a week or 10 days.",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ },
+ "semantic": {
+ "normalized_reference": "the ship should be in within a week or 10 days",
+ "normalized_hypothesis": "the ship should be in within a week or 10 days",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0228",
+ "category": "phonetic_balance",
+ "text": "Such a dog the wise driver kills, or turns loose.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0228.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Such a dog the wise driver kills or turns loose.",
+ "raw": {
+ "normalized_reference": "such a dog the wise driver kills, or turns loose.",
+ "normalized_hypothesis": "such a dog the wise driver kills or turns loose.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "such a dog the wise driver kills or turns loose",
+ "normalized_hypothesis": "such a dog the wise driver kills or turns loose",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0229",
+ "category": "phonetic_balance",
+ "text": "You have heard always how he was the lover of the Princess Naomi.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0229.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "You have heard always how he was the lover of the princess Naomi.",
+ "raw": {
+ "normalized_reference": "you have heard always how he was the lover of the princess naomi.",
+ "normalized_hypothesis": "you have heard always how he was the lover of the princess naomi.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "you have heard always how he was the lover of the princess naomi",
+ "normalized_hypothesis": "you have heard always how he was the lover of the princess naomi",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0230",
+ "category": "phonetic_balance",
+ "text": "I want to know how all this is possible.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0230.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I want to know how all this is possible.",
+ "raw": {
+ "normalized_reference": "i want to know how all this is possible.",
+ "normalized_hypothesis": "i want to know how all this is possible.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i want to know how all this is possible",
+ "normalized_hypothesis": "i want to know how all this is possible",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0231",
+ "category": "phonetic_balance",
+ "text": "It was my idea to a tee.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0231.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It was my idea to a tee.",
+ "raw": {
+ "normalized_reference": "it was my idea to a tee.",
+ "normalized_hypothesis": "it was my idea to a tee.",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it was my idea to a tee",
+ "normalized_hypothesis": "it was my idea to a tee",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0232",
+ "category": "phonetic_balance",
+ "text": "They likewise are disinclined to being eaten.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0232.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "They likewise are disinclined to being eaten.",
+ "raw": {
+ "normalized_reference": "they likewise are disinclined to being eaten.",
+ "normalized_hypothesis": "they likewise are disinclined to being eaten.",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "they likewise are disinclined to being eaten",
+ "normalized_hypothesis": "they likewise are disinclined to being eaten",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0233",
+ "category": "phonetic_balance",
+ "text": "This state of mind comes of an undue prominence of the ego.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0233.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "This state of mind comes of an undue prominence of the ego.",
+ "raw": {
+ "normalized_reference": "this state of mind comes of an undue prominence of the ego.",
+ "normalized_hypothesis": "this state of mind comes of an undue prominence of the ego.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "this state of mind comes of an undue prominence of the ego",
+ "normalized_hypothesis": "this state of mind comes of an undue prominence of the ego",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0234",
+ "category": "phonetic_balance",
+ "text": "Then it is as I said, Womble announced with finality.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0234.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Then it is as I said, Womble announced with finality.",
+ "raw": {
+ "normalized_reference": "then it is as i said, womble announced with finality.",
+ "normalized_hypothesis": "then it is as i said, womble announced with finality.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "then it is as i said womble announced with finality",
+ "normalized_hypothesis": "then it is as i said womble announced with finality",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0235",
+ "category": "phonetic_balance",
+ "text": "I can see that knife now.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0235.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I can see that knife now.",
+ "raw": {
+ "normalized_reference": "i can see that knife now.",
+ "normalized_hypothesis": "i can see that knife now.",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i can see that knife now",
+ "normalized_hypothesis": "i can see that knife now",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0236",
+ "category": "phonetic_balance",
+ "text": "If I meet her again I shall apologize, said Eileen.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0236.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "If I meet her again I shall apologize, said Ailing.",
+ "raw": {
+ "normalized_reference": "if i meet her again i shall apologize, said eileen.",
+ "normalized_hypothesis": "if i meet her again i shall apologize, said ailing.",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ },
+ "semantic": {
+ "normalized_reference": "if i meet her again i shall apologize said eileen",
+ "normalized_hypothesis": "if i meet her again i shall apologize said ailing",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ }
+ },
+ {
+ "id": "release500_0237",
+ "category": "phonetic_balance",
+ "text": "A wildly exciting time was his during the week preceding Thursday the eighteenth.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0237.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "A wildly exciting time was his during the week preceding Thursday the 18th.",
+ "raw": {
+ "normalized_reference": "a wildly exciting time was his during the week preceding thursday the eighteenth.",
+ "normalized_hypothesis": "a wildly exciting time was his during the week preceding thursday the 18th.",
+ "errors": 1,
+ "words": 13,
+ "wer": 0.07692307692307693
+ },
+ "semantic": {
+ "normalized_reference": "a wildly exciting time was his during the week preceding thursday the 18th",
+ "normalized_hypothesis": "a wildly exciting time was his during the week preceding thursday the 18th",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0238",
+ "category": "phonetic_balance",
+ "text": "In what bucolic school of fence he had been taught was beyond imagining.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0238.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "In what bucolic school of fence he had been taught was beyond imagining.",
+ "raw": {
+ "normalized_reference": "in what bucolic school of fence he had been taught was beyond imagining.",
+ "normalized_hypothesis": "in what bucolic school of fence he had been taught was beyond imagining.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "in what bucolic school of fence he had been taught was beyond imagining",
+ "normalized_hypothesis": "in what bucolic school of fence he had been taught was beyond imagining",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0239",
+ "category": "phonetic_balance",
+ "text": "So she said, the irate skipper dashed on.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0239.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "So she said, the irate skipper dashed on.",
+ "raw": {
+ "normalized_reference": "so she said, the irate skipper dashed on.",
+ "normalized_hypothesis": "so she said, the irate skipper dashed on.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "so she said the irate skipper dashed on",
+ "normalized_hypothesis": "so she said the irate skipper dashed on",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0240",
+ "category": "phonetic_balance",
+ "text": "It is not an attempt to smash the market.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0240.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It is not an attempt to smash the market.",
+ "raw": {
+ "normalized_reference": "it is not an attempt to smash the market.",
+ "normalized_hypothesis": "it is not an attempt to smash the market.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it is not an attempt to smash the market",
+ "normalized_hypothesis": "it is not an attempt to smash the market",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0241",
+ "category": "phonetic_balance",
+ "text": "He destroyed everything that had belonged to the woman.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0241.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He destroyed everything that had belonged to the woman.",
+ "raw": {
+ "normalized_reference": "he destroyed everything that had belonged to the woman.",
+ "normalized_hypothesis": "he destroyed everything that had belonged to the woman.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he destroyed everything that had belonged to the woman",
+ "normalized_hypothesis": "he destroyed everything that had belonged to the woman",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0242",
+ "category": "phonetic_balance",
+ "text": "Did Thorpe go to see any one in Churchill.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0242.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Did Thorpe go to see anyone in Churchill?",
+ "raw": {
+ "normalized_reference": "did thorpe go to see any one in churchill.",
+ "normalized_hypothesis": "did thorpe go to see anyone in churchill?",
+ "errors": 2,
+ "words": 9,
+ "wer": 0.2222222222222222
+ },
+ "semantic": {
+ "normalized_reference": "did thorpe go to see any one in churchill",
+ "normalized_hypothesis": "did thorpe go to see anyone in churchill",
+ "errors": 2,
+ "words": 9,
+ "wer": 0.2222222222222222
+ }
+ },
+ {
+ "id": "release500_0243",
+ "category": "phonetic_balance",
+ "text": "Many other similar disconcerting omissions will be noticed in the Manuscript.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0243.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Many other similar disconcerting omissions will be noticed in the manuscript.",
+ "raw": {
+ "normalized_reference": "many other similar disconcerting omissions will be noticed in the manuscript.",
+ "normalized_hypothesis": "many other similar disconcerting omissions will be noticed in the manuscript.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "many other similar disconcerting omissions will be noticed in the manuscript",
+ "normalized_hypothesis": "many other similar disconcerting omissions will be noticed in the manuscript",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0244",
+ "category": "phonetic_balance",
+ "text": "The ringing of the big bell aroused him.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0244.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The ringing of the big bell aroused him.",
+ "raw": {
+ "normalized_reference": "the ringing of the big bell aroused him.",
+ "normalized_hypothesis": "the ringing of the big bell aroused him.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the ringing of the big bell aroused him",
+ "normalized_hypothesis": "the ringing of the big bell aroused him",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0245",
+ "category": "phonetic_balance",
+ "text": "A sinewy hand, dripping with water, was clutching the rail.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0245.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "A sinewy hand, dripping with water, was clutching the rail.",
+ "raw": {
+ "normalized_reference": "a sinewy hand, dripping with water, was clutching the rail.",
+ "normalized_hypothesis": "a sinewy hand, dripping with water, was clutching the rail.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "a sinewy hand dripping with water was clutching the rail",
+ "normalized_hypothesis": "a sinewy hand dripping with water was clutching the rail",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0246",
+ "category": "phonetic_balance",
+ "text": "Enters now the psychology of the situation.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0246.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Enders now the psychology of the situation.",
+ "raw": {
+ "normalized_reference": "enters now the psychology of the situation.",
+ "normalized_hypothesis": "enders now the psychology of the situation.",
+ "errors": 1,
+ "words": 7,
+ "wer": 0.14285714285714285
+ },
+ "semantic": {
+ "normalized_reference": "enters now the psychology of the situation",
+ "normalized_hypothesis": "enders now the psychology of the situation",
+ "errors": 1,
+ "words": 7,
+ "wer": 0.14285714285714285
+ }
+ },
+ {
+ "id": "release500_0247",
+ "category": "phonetic_balance",
+ "text": "Wash your hands of me.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0247.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Wash your hands of me.",
+ "raw": {
+ "normalized_reference": "wash your hands of me.",
+ "normalized_hypothesis": "wash your hands of me.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "wash your hands of me",
+ "normalized_hypothesis": "wash your hands of me",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0248",
+ "category": "phonetic_balance",
+ "text": "He confessed that the sketch had startled him.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0248.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He confessed that the sketch had startled him.",
+ "raw": {
+ "normalized_reference": "he confessed that the sketch had startled him.",
+ "normalized_hypothesis": "he confessed that the sketch had startled him.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he confessed that the sketch had startled him",
+ "normalized_hypothesis": "he confessed that the sketch had startled him",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0249",
+ "category": "phonetic_balance",
+ "text": "Her face was against his breast.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0249.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Her face was against his breast.",
+ "raw": {
+ "normalized_reference": "her face was against his breast.",
+ "normalized_hypothesis": "her face was against his breast.",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "her face was against his breast",
+ "normalized_hypothesis": "her face was against his breast",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0250",
+ "category": "phonetic_balance",
+ "text": "He was a pariah; a wanderer without a friend or a home.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0250.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He was a pariah, a wanderer without a friend or a home.",
+ "raw": {
+ "normalized_reference": "he was a pariah; a wanderer without a friend or a home.",
+ "normalized_hypothesis": "he was a pariah, a wanderer without a friend or a home.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he was a pariah a wanderer without a friend or a home",
+ "normalized_hypothesis": "he was a pariah a wanderer without a friend or a home",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0251",
+ "category": "phonetic_balance",
+ "text": "A cry of joy burst from Philip's lips.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0251.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "A cry of joy burst from Philip's lips.",
+ "raw": {
+ "normalized_reference": "a cry of joy burst from philip's lips.",
+ "normalized_hypothesis": "a cry of joy burst from philip's lips.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "a cry of joy burst from philip is lips",
+ "normalized_hypothesis": "a cry of joy burst from philip is lips",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0252",
+ "category": "phonetic_balance",
+ "text": "And the big Persian knew of his existence before he did of hers.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0252.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "and the big Persian knew of his existence before he did of hers.",
+ "raw": {
+ "normalized_reference": "and the big persian knew of his existence before he did of hers.",
+ "normalized_hypothesis": "and the big persian knew of his existence before he did of hers.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "and the big persian knew of his existence before he did of hers",
+ "normalized_hypothesis": "and the big persian knew of his existence before he did of hers",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0253",
+ "category": "phonetic_balance",
+ "text": "The men stared into each other's face.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0253.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The men stared into each other's face.",
+ "raw": {
+ "normalized_reference": "the men stared into each other's face.",
+ "normalized_hypothesis": "the men stared into each other's face.",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the men stared into each other is face",
+ "normalized_hypothesis": "the men stared into each other is face",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0254",
+ "category": "phonetic_balance",
+ "text": "It is the fire, partly, she said.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0254.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It is the fire, partly, she said.",
+ "raw": {
+ "normalized_reference": "it is the fire, partly, she said.",
+ "normalized_hypothesis": "it is the fire, partly, she said.",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it is the fire partly she said",
+ "normalized_hypothesis": "it is the fire partly she said",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0255",
+ "category": "phonetic_balance",
+ "text": "It seemed nearer to him since he had seen and talked with Gregson.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0255.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It seemed nearer to him since he had seen and talked with Gregson.",
+ "raw": {
+ "normalized_reference": "it seemed nearer to him since he had seen and talked with gregson.",
+ "normalized_hypothesis": "it seemed nearer to him since he had seen and talked with gregson.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it seemed nearer to him since he had seen and talked with gregson",
+ "normalized_hypothesis": "it seemed nearer to him since he had seen and talked with gregson",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0256",
+ "category": "phonetic_balance",
+ "text": "Among my minor afflictions, I may mention a new and mysterious one.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0256.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Among my minor afflictions, I may mention a new and mysterious one.",
+ "raw": {
+ "normalized_reference": "among my minor afflictions, i may mention a new and mysterious one.",
+ "normalized_hypothesis": "among my minor afflictions, i may mention a new and mysterious one.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "among my minor afflictions i may mention a new and mysterious one",
+ "normalized_hypothesis": "among my minor afflictions i may mention a new and mysterious one",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0257",
+ "category": "phonetic_balance",
+ "text": "Between him and all domestic animals there must be no hostilities.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0257.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Between him and all domestic animals there must be no hostilities.",
+ "raw": {
+ "normalized_reference": "between him and all domestic animals there must be no hostilities.",
+ "normalized_hypothesis": "between him and all domestic animals there must be no hostilities.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "between him and all domestic animals there must be no hostilities",
+ "normalized_hypothesis": "between him and all domestic animals there must be no hostilities",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0258",
+ "category": "phonetic_balance",
+ "text": "With these arguments he convinced himself that he should go on alone.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0258.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "With these arguments he convinced himself that he should go on alone.",
+ "raw": {
+ "normalized_reference": "with these arguments he convinced himself that he should go on alone.",
+ "normalized_hypothesis": "with these arguments he convinced himself that he should go on alone.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "with these arguments he convinced himself that he should go on alone",
+ "normalized_hypothesis": "with these arguments he convinced himself that he should go on alone",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0259",
+ "category": "phonetic_balance",
+ "text": "We are consumed in our own flesh-pots.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0259.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "We are consumed in our own flesh bots.",
+ "raw": {
+ "normalized_reference": "we are consumed in our own flesh-pots.",
+ "normalized_hypothesis": "we are consumed in our own flesh bots.",
+ "errors": 1,
+ "words": 8,
+ "wer": 0.125
+ },
+ "semantic": {
+ "normalized_reference": "we are consumed in our own flesh pots",
+ "normalized_hypothesis": "we are consumed in our own flesh bots",
+ "errors": 1,
+ "words": 8,
+ "wer": 0.125
+ }
+ },
+ {
+ "id": "release500_0260",
+ "category": "phonetic_balance",
+ "text": "Mercedes screamed, cried, laughed, and manifested the chaotic abandonment of hysteria.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0260.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Mercedes screened, cried, laughed, and manifested the chaotic abandonment of hysteria",
+ "raw": {
+ "normalized_reference": "mercedes screamed, cried, laughed, and manifested the chaotic abandonment of hysteria.",
+ "normalized_hypothesis": "mercedes screened, cried, laughed, and manifested the chaotic abandonment of hysteria",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ },
+ "semantic": {
+ "normalized_reference": "mercedes screamed cried laughed and manifested the chaotic abandonment of hysteria",
+ "normalized_hypothesis": "mercedes screened cried laughed and manifested the chaotic abandonment of hysteria",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ }
+ },
+ {
+ "id": "release500_0261",
+ "category": "phonetic_balance",
+ "text": "Philip made no effort to follow.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0261.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Philip made no effort to follow.",
+ "raw": {
+ "normalized_reference": "philip made no effort to follow.",
+ "normalized_hypothesis": "philip made no effort to follow.",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "philip made no effort to follow",
+ "normalized_hypothesis": "philip made no effort to follow",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0262",
+ "category": "phonetic_balance",
+ "text": "She said with chattering teeth.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0262.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "she said with chattering teeth.",
+ "raw": {
+ "normalized_reference": "she said with chattering teeth.",
+ "normalized_hypothesis": "she said with chattering teeth.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "she said with chattering teeth",
+ "normalized_hypothesis": "she said with chattering teeth",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0263",
+ "category": "phonetic_balance",
+ "text": "Halfway around the track one donkey got into an argument with its rider.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0263.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Halfway around the track, one donkey got into an argument with its rider.",
+ "raw": {
+ "normalized_reference": "halfway around the track one donkey got into an argument with its rider.",
+ "normalized_hypothesis": "halfway around the track, one donkey got into an argument with its rider.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "halfway around the track one donkey got into an argument with its rider",
+ "normalized_hypothesis": "halfway around the track one donkey got into an argument with its rider",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0264",
+ "category": "phonetic_balance",
+ "text": "Earth and gravel seemed to fill the pan.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0264.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Earth and gravel seemed to fill the pan.",
+ "raw": {
+ "normalized_reference": "earth and gravel seemed to fill the pan.",
+ "normalized_hypothesis": "earth and gravel seemed to fill the pan.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "earth and gravel seemed to fill the pan",
+ "normalized_hypothesis": "earth and gravel seemed to fill the pan",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0265",
+ "category": "phonetic_balance",
+ "text": "Her achievements with cocoanuts were a revelation.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0265.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Her achievements with coconuts were a revelation.",
+ "raw": {
+ "normalized_reference": "her achievements with cocoanuts were a revelation.",
+ "normalized_hypothesis": "her achievements with coconuts were a revelation.",
+ "errors": 1,
+ "words": 7,
+ "wer": 0.14285714285714285
+ },
+ "semantic": {
+ "normalized_reference": "her achievements with cocoanuts were a revelation",
+ "normalized_hypothesis": "her achievements with coconuts were a revelation",
+ "errors": 1,
+ "words": 7,
+ "wer": 0.14285714285714285
+ }
+ },
+ {
+ "id": "release500_0266",
+ "category": "phonetic_balance",
+ "text": "For a full minute he crouched and listened.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0266.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "For a full minute he crouched and listened.",
+ "raw": {
+ "normalized_reference": "for a full minute he crouched and listened.",
+ "normalized_hypothesis": "for a full minute he crouched and listened.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "for a full minute he crouched and listened",
+ "normalized_hypothesis": "for a full minute he crouched and listened",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0267",
+ "category": "phonetic_balance",
+ "text": "It was edged with ice.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0267.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It was edged with ice.",
+ "raw": {
+ "normalized_reference": "it was edged with ice.",
+ "normalized_hypothesis": "it was edged with ice.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it was edged with ice",
+ "normalized_hypothesis": "it was edged with ice",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0268",
+ "category": "phonetic_balance",
+ "text": "The Portuguese boy passed the Hawaiian.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0268.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The Portuguese boy passed the Hawaiian.",
+ "raw": {
+ "normalized_reference": "the portuguese boy passed the hawaiian.",
+ "normalized_hypothesis": "the portuguese boy passed the hawaiian.",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the portuguese boy passed the hawaiian",
+ "normalized_hypothesis": "the portuguese boy passed the hawaiian",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0269",
+ "category": "phonetic_balance",
+ "text": "But he did not broach it, preferring to mature it carefully.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0269.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "but he did not broach it, preferring to mature it carefully.",
+ "raw": {
+ "normalized_reference": "but he did not broach it, preferring to mature it carefully.",
+ "normalized_hypothesis": "but he did not broach it, preferring to mature it carefully.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "but he did not broach it preferring to mature it carefully",
+ "normalized_hypothesis": "but he did not broach it preferring to mature it carefully",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0270",
+ "category": "phonetic_balance",
+ "text": "Come on, Del Mar challenged.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0270.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Come on, Delmar challenged.",
+ "raw": {
+ "normalized_reference": "come on, del mar challenged.",
+ "normalized_hypothesis": "come on, delmar challenged.",
+ "errors": 2,
+ "words": 5,
+ "wer": 0.4
+ },
+ "semantic": {
+ "normalized_reference": "come on del mar challenged",
+ "normalized_hypothesis": "come on delmar challenged",
+ "errors": 2,
+ "words": 5,
+ "wer": 0.4
+ }
+ },
+ {
+ "id": "release500_0271",
+ "category": "phonetic_balance",
+ "text": "For a full half minute Jeanne looked at him without speaking.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0271.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "For a full half minute Jean looked at him without speaking.",
+ "raw": {
+ "normalized_reference": "for a full half minute jeanne looked at him without speaking.",
+ "normalized_hypothesis": "for a full half minute jean looked at him without speaking.",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ },
+ "semantic": {
+ "normalized_reference": "for a full half minute jeanne looked at him without speaking",
+ "normalized_hypothesis": "for a full half minute jean looked at him without speaking",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ }
+ },
+ {
+ "id": "release500_0272",
+ "category": "phonetic_balance",
+ "text": "He bore no grudges and had few enemies.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0272.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He bore no grudges and had few enemies.",
+ "raw": {
+ "normalized_reference": "he bore no grudges and had few enemies.",
+ "normalized_hypothesis": "he bore no grudges and had few enemies.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he bore no grudges and had few enemies",
+ "normalized_hypothesis": "he bore no grudges and had few enemies",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0273",
+ "category": "phonetic_balance",
+ "text": "I'll go over tomorrow afternoon.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0273.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I'll go over tomorrow afternoon.",
+ "raw": {
+ "normalized_reference": "i'll go over tomorrow afternoon.",
+ "normalized_hypothesis": "i'll go over tomorrow afternoon.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i will go over tomorrow afternoon",
+ "normalized_hypothesis": "i will go over tomorrow afternoon",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0274",
+ "category": "phonetic_balance",
+ "text": "Well, there are better men in Hawaii, that's all.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0274.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Well, there are better men in Hawaii, that's all.",
+ "raw": {
+ "normalized_reference": "well, there are better men in hawaii, that's all.",
+ "normalized_hypothesis": "well, there are better men in hawaii, that's all.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "well there are better men in hawaii that is all",
+ "normalized_hypothesis": "well there are better men in hawaii that is all",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0275",
+ "category": "phonetic_balance",
+ "text": "He was an athlete and a giant.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0275.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He was an athlete and a giant.",
+ "raw": {
+ "normalized_reference": "he was an athlete and a giant.",
+ "normalized_hypothesis": "he was an athlete and a giant.",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he was an athlete and a giant",
+ "normalized_hypothesis": "he was an athlete and a giant",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0276",
+ "category": "phonetic_balance",
+ "text": "McVeigh when he returns from a trip to Honolulu.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0276.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Look at Vae when he returns from a trip to Honolulu.",
+ "raw": {
+ "normalized_reference": "mcveigh when he returns from a trip to honolulu.",
+ "normalized_hypothesis": "look at vae when he returns from a trip to honolulu.",
+ "errors": 3,
+ "words": 9,
+ "wer": 0.3333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "mcveigh when he returns from a trip to honolulu",
+ "normalized_hypothesis": "look at vae when he returns from a trip to honolulu",
+ "errors": 3,
+ "words": 9,
+ "wer": 0.3333333333333333
+ }
+ },
+ {
+ "id": "release500_0277",
+ "category": "phonetic_balance",
+ "text": "You're a devil for fighting, and will surely win.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0277.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "You're a devil for fighting, and will surely win.",
+ "raw": {
+ "normalized_reference": "you're a devil for fighting, and will surely win.",
+ "normalized_hypothesis": "you're a devil for fighting, and will surely win.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "you are a devil for fighting and will surely win",
+ "normalized_hypothesis": "you are a devil for fighting and will surely win",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0278",
+ "category": "phonetic_balance",
+ "text": "All an appearance can know is mirage.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0278.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "All an appearance can know is mirage.",
+ "raw": {
+ "normalized_reference": "all an appearance can know is mirage.",
+ "normalized_hypothesis": "all an appearance can know is mirage.",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "all an appearance can know is mirage",
+ "normalized_hypothesis": "all an appearance can know is mirage",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0279",
+ "category": "phonetic_balance",
+ "text": "You fired me out of your house, in short.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0279.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "You fired me out of your house, in short.",
+ "raw": {
+ "normalized_reference": "you fired me out of your house, in short.",
+ "normalized_hypothesis": "you fired me out of your house, in short.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "you fired me out of your house in short",
+ "normalized_hypothesis": "you fired me out of your house in short",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0280",
+ "category": "phonetic_balance",
+ "text": "He made sure that the magazine was loaded, and resumed his paddling.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0280.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He made sure that the magazine was loaded and resumed his paddling.",
+ "raw": {
+ "normalized_reference": "he made sure that the magazine was loaded, and resumed his paddling.",
+ "normalized_hypothesis": "he made sure that the magazine was loaded and resumed his paddling.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he made sure that the magazine was loaded and resumed his paddling",
+ "normalized_hypothesis": "he made sure that the magazine was loaded and resumed his paddling",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0281",
+ "category": "phonetic_balance",
+ "text": "Change chairs, Daylight commanded.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0281.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Change chairs, Daylight commanded.",
+ "raw": {
+ "normalized_reference": "change chairs, daylight commanded.",
+ "normalized_hypothesis": "change chairs, daylight commanded.",
+ "errors": 0,
+ "words": 4,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "change chairs daylight commanded",
+ "normalized_hypothesis": "change chairs daylight commanded",
+ "errors": 0,
+ "words": 4,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0282",
+ "category": "phonetic_balance",
+ "text": "By that answer my professional medical prestige stood or fell.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0282.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Why'd I answer my professional medical prestige stud or fell?",
+ "raw": {
+ "normalized_reference": "by that answer my professional medical prestige stood or fell.",
+ "normalized_hypothesis": "why'd i answer my professional medical prestige stud or fell?",
+ "errors": 3,
+ "words": 10,
+ "wer": 0.3
+ },
+ "semantic": {
+ "normalized_reference": "by that answer my professional medical prestige stood or fell",
+ "normalized_hypothesis": "why would i answer my professional medical prestige stud or fell",
+ "errors": 4,
+ "words": 10,
+ "wer": 0.4
+ }
+ },
+ {
+ "id": "release500_0283",
+ "category": "phonetic_balance",
+ "text": "The moon had already begun its westward decline.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0283.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The moon had already begun its westward decline.",
+ "raw": {
+ "normalized_reference": "the moon had already begun its westward decline.",
+ "normalized_hypothesis": "the moon had already begun its westward decline.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the moon had already begun its westward decline",
+ "normalized_hypothesis": "the moon had already begun its westward decline",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0284",
+ "category": "phonetic_balance",
+ "text": "At sea, Monday, March 16, 1908.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0284.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "At sea, Monday, March 16, 1908.",
+ "raw": {
+ "normalized_reference": "at sea, monday, march 16, 1908.",
+ "normalized_hypothesis": "at sea, monday, march 16, 1908.",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "at sea monday march 16 1908",
+ "normalized_hypothesis": "at sea monday march 16 1908",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0285",
+ "category": "phonetic_balance",
+ "text": "And wherever I ranged, the way lay along alcohol-drenched roads.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0285.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "And wherever I ranged, the way lay a long alcohol-drenched road's",
+ "raw": {
+ "normalized_reference": "and wherever i ranged, the way lay along alcohol-drenched roads.",
+ "normalized_hypothesis": "and wherever i ranged, the way lay a long alcohol-drenched road's",
+ "errors": 3,
+ "words": 11,
+ "wer": 0.2727272727272727
+ },
+ "semantic": {
+ "normalized_reference": "and wherever i ranged the way lay along alcohol drenched roads",
+ "normalized_hypothesis": "and wherever i ranged the way lay a long alcohol drenched road is",
+ "errors": 4,
+ "words": 11,
+ "wer": 0.36363636363636365
+ }
+ },
+ {
+ "id": "release500_0286",
+ "category": "phonetic_balance",
+ "text": "The date was nearly eighteen years old.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0286.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The date was nearly 18 years old.",
+ "raw": {
+ "normalized_reference": "the date was nearly eighteen years old.",
+ "normalized_hypothesis": "the date was nearly 18 years old.",
+ "errors": 1,
+ "words": 7,
+ "wer": 0.14285714285714285
+ },
+ "semantic": {
+ "normalized_reference": "the date was nearly 18 years old",
+ "normalized_hypothesis": "the date was nearly 18 years old",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0287",
+ "category": "phonetic_balance",
+ "text": "Very early in my life, I separated from my mother.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0287.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Very early in my life, I separated from my mother.",
+ "raw": {
+ "normalized_reference": "very early in my life, i separated from my mother.",
+ "normalized_hypothesis": "very early in my life, i separated from my mother.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "very early in my life i separated from my mother",
+ "normalized_hypothesis": "very early in my life i separated from my mother",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0288",
+ "category": "phonetic_balance",
+ "text": "There was pride and strength, the ring of triumph in his voice.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0288.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "There was pride and strength, the ring of triumph in his voice.",
+ "raw": {
+ "normalized_reference": "there was pride and strength, the ring of triumph in his voice.",
+ "normalized_hypothesis": "there was pride and strength, the ring of triumph in his voice.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "there was pride and strength the ring of triumph in his voice",
+ "normalized_hypothesis": "there was pride and strength the ring of triumph in his voice",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0289",
+ "category": "phonetic_balance",
+ "text": "From my earliest recollection my sleep was a period of terror.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0289.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "From my earliest recollection my sleep was a period of terror.",
+ "raw": {
+ "normalized_reference": "from my earliest recollection my sleep was a period of terror.",
+ "normalized_hypothesis": "from my earliest recollection my sleep was a period of terror.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "from my earliest recollection my sleep was a period of terror",
+ "normalized_hypothesis": "from my earliest recollection my sleep was a period of terror",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0290",
+ "category": "phonetic_balance",
+ "text": "We'll have to watch our chances.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0290.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "We'll have to watch our chances.",
+ "raw": {
+ "normalized_reference": "we'll have to watch our chances.",
+ "normalized_hypothesis": "we'll have to watch our chances.",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "we will have to watch our chances",
+ "normalized_hypothesis": "we will have to watch our chances",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0291",
+ "category": "phonetic_balance",
+ "text": "Close beside him gleamed the white fangs of the wolf-dog.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0291.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Close beside him gleamed the white fangs of the wolf-dog.",
+ "raw": {
+ "normalized_reference": "close beside him gleamed the white fangs of the wolf-dog.",
+ "normalized_hypothesis": "close beside him gleamed the white fangs of the wolf-dog.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "close beside him gleamed the white fangs of the wolf dog",
+ "normalized_hypothesis": "close beside him gleamed the white fangs of the wolf dog",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0292",
+ "category": "phonetic_balance",
+ "text": "Would you be satisfied with that one hundredth part of me.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0292.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Would you be satisfied with that 100th part of me?",
+ "raw": {
+ "normalized_reference": "would you be satisfied with that one hundredth part of me.",
+ "normalized_hypothesis": "would you be satisfied with that 100th part of me?",
+ "errors": 2,
+ "words": 11,
+ "wer": 0.18181818181818182
+ },
+ "semantic": {
+ "normalized_reference": "would you be satisfied with that 100th part of me",
+ "normalized_hypothesis": "would you be satisfied with that 100th part of me",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0293",
+ "category": "phonetic_balance",
+ "text": "Two gigantic owls were tearing at the carcass.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0293.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Two gigantic owls were tearing at the carcass.",
+ "raw": {
+ "normalized_reference": "two gigantic owls were tearing at the carcass.",
+ "normalized_hypothesis": "two gigantic owls were tearing at the carcass.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "2 gigantic owls were tearing at the carcass",
+ "normalized_hypothesis": "2 gigantic owls were tearing at the carcass",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0294",
+ "category": "phonetic_balance",
+ "text": "The lush vegetation of that sheltered spot make a natural shield.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0294.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The lush vegetation of that sheltered spot make a natural shield.",
+ "raw": {
+ "normalized_reference": "the lush vegetation of that sheltered spot make a natural shield.",
+ "normalized_hypothesis": "the lush vegetation of that sheltered spot make a natural shield.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the lush vegetation of that sheltered spot make a natural shield",
+ "normalized_hypothesis": "the lush vegetation of that sheltered spot make a natural shield",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0295",
+ "category": "phonetic_balance",
+ "text": "Below him the shadow was broken into a pool of rippling starlight.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0295.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Below him the shadow was broken into a pool of rippling starlight.",
+ "raw": {
+ "normalized_reference": "below him the shadow was broken into a pool of rippling starlight.",
+ "normalized_hypothesis": "below him the shadow was broken into a pool of rippling starlight.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "below him the shadow was broken into a pool of rippling starlight",
+ "normalized_hypothesis": "below him the shadow was broken into a pool of rippling starlight",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0296",
+ "category": "phonetic_balance",
+ "text": "Lots of men take women buggy riding.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0296.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Lots of men take women buggy riding.",
+ "raw": {
+ "normalized_reference": "lots of men take women buggy riding.",
+ "normalized_hypothesis": "lots of men take women buggy riding.",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "lots of men take women buggy riding",
+ "normalized_hypothesis": "lots of men take women buggy riding",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0297",
+ "category": "phonetic_balance",
+ "text": "But it contributed to the smash.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0297.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "but it contributed to the smash",
+ "raw": {
+ "normalized_reference": "but it contributed to the smash.",
+ "normalized_hypothesis": "but it contributed to the smash",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "but it contributed to the smash",
+ "normalized_hypothesis": "but it contributed to the smash",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0298",
+ "category": "phonetic_balance",
+ "text": "No man ate of the seal meat or the oil.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0298.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "No man ate of the seal meat or the oil.",
+ "raw": {
+ "normalized_reference": "no man ate of the seal meat or the oil.",
+ "normalized_hypothesis": "no man ate of the seal meat or the oil.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "no man ate of the seal meat or the oil",
+ "normalized_hypothesis": "no man ate of the seal meat or the oil",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0299",
+ "category": "phonetic_balance",
+ "text": "You're going in for grab sharing.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0299.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "You're going in for grad sharing.",
+ "raw": {
+ "normalized_reference": "you're going in for grab sharing.",
+ "normalized_hypothesis": "you're going in for grad sharing.",
+ "errors": 1,
+ "words": 6,
+ "wer": 0.16666666666666666
+ },
+ "semantic": {
+ "normalized_reference": "you are going in for grab sharing",
+ "normalized_hypothesis": "you are going in for grad sharing",
+ "errors": 1,
+ "words": 7,
+ "wer": 0.14285714285714285
+ }
+ },
+ {
+ "id": "release500_0300",
+ "category": "phonetic_balance",
+ "text": "He it was that lived to found the family of the Patino.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0300.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He it was that lived to found the family of the Patino.",
+ "raw": {
+ "normalized_reference": "he it was that lived to found the family of the patino.",
+ "normalized_hypothesis": "he it was that lived to found the family of the patino.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he it was that lived to found the family of the patino",
+ "normalized_hypothesis": "he it was that lived to found the family of the patino",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0301",
+ "category": "phonetic_balance",
+ "text": "For such countries nothing remained but reorganization.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0301.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "For such countries nothing remained but reorganization.",
+ "raw": {
+ "normalized_reference": "for such countries nothing remained but reorganization.",
+ "normalized_hypothesis": "for such countries nothing remained but reorganization.",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "for such countries nothing remained but reorganization",
+ "normalized_hypothesis": "for such countries nothing remained but reorganization",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0302",
+ "category": "phonetic_balance",
+ "text": "Stand off butcher and baker and all the rest.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0302.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Standoff, Butcher and Baker and all the rest.",
+ "raw": {
+ "normalized_reference": "stand off butcher and baker and all the rest.",
+ "normalized_hypothesis": "standoff, butcher and baker and all the rest.",
+ "errors": 2,
+ "words": 9,
+ "wer": 0.2222222222222222
+ },
+ "semantic": {
+ "normalized_reference": "stand off butcher and baker and all the rest",
+ "normalized_hypothesis": "standoff butcher and baker and all the rest",
+ "errors": 2,
+ "words": 9,
+ "wer": 0.2222222222222222
+ }
+ },
+ {
+ "id": "release500_0303",
+ "category": "phonetic_balance",
+ "text": "All eyes, however, were staring at him in certitude of expectancy.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0303.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "All eyes, however, were staring at him in certitude of expectancy.",
+ "raw": {
+ "normalized_reference": "all eyes, however, were staring at him in certitude of expectancy.",
+ "normalized_hypothesis": "all eyes, however, were staring at him in certitude of expectancy.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "all eyes however were staring at him in certitude of expectancy",
+ "normalized_hypothesis": "all eyes however were staring at him in certitude of expectancy",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0304",
+ "category": "phonetic_balance",
+ "text": "Two days ago Jeanne learned where her father's men were hiding.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0304.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Two days ago Jean learned where her father's men were hiding.",
+ "raw": {
+ "normalized_reference": "two days ago jeanne learned where her father's men were hiding.",
+ "normalized_hypothesis": "two days ago jean learned where her father's men were hiding.",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ },
+ "semantic": {
+ "normalized_reference": "2 days ago jeanne learned where her father is men were hiding",
+ "normalized_hypothesis": "2 days ago jean learned where her father is men were hiding",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ }
+ },
+ {
+ "id": "release500_0305",
+ "category": "phonetic_balance",
+ "text": "His previous wives had never lived long enough to bear him children.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0305.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "His previous wives had never lived long enough to bear him children.",
+ "raw": {
+ "normalized_reference": "his previous wives had never lived long enough to bear him children.",
+ "normalized_hypothesis": "his previous wives had never lived long enough to bear him children.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "his previous wives had never lived long enough to bear him children",
+ "normalized_hypothesis": "his previous wives had never lived long enough to bear him children",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0306",
+ "category": "phonetic_balance",
+ "text": "But this little defect did not worry him.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0306.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "But this little defect did not worry him.",
+ "raw": {
+ "normalized_reference": "but this little defect did not worry him.",
+ "normalized_hypothesis": "but this little defect did not worry him.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "but this little defect did not worry him",
+ "normalized_hypothesis": "but this little defect did not worry him",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0307",
+ "category": "phonetic_balance",
+ "text": "Eli Harding asked, as Shunk started to follow.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0307.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Lily Harding asked as Shunk started to follow.",
+ "raw": {
+ "normalized_reference": "eli harding asked, as shunk started to follow.",
+ "normalized_hypothesis": "lily harding asked as shunk started to follow.",
+ "errors": 1,
+ "words": 8,
+ "wer": 0.125
+ },
+ "semantic": {
+ "normalized_reference": "eli harding asked as shunk started to follow",
+ "normalized_hypothesis": "lily harding asked as shunk started to follow",
+ "errors": 1,
+ "words": 8,
+ "wer": 0.125
+ }
+ },
+ {
+ "id": "release500_0308",
+ "category": "phonetic_balance",
+ "text": "And this was their sole conversation throughout the meal.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0308.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "And this was their sole conversation throughout the meal.",
+ "raw": {
+ "normalized_reference": "and this was their sole conversation throughout the meal.",
+ "normalized_hypothesis": "and this was their sole conversation throughout the meal.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "and this was their sole conversation throughout the meal",
+ "normalized_hypothesis": "and this was their sole conversation throughout the meal",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0309",
+ "category": "phonetic_balance",
+ "text": "Then, again, Tudor had such an irritating way about him.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0309.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Then, again, Tudor had such an irritating way about him.",
+ "raw": {
+ "normalized_reference": "then, again, tudor had such an irritating way about him.",
+ "normalized_hypothesis": "then, again, tudor had such an irritating way about him.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "then again tudor had such an irritating way about him",
+ "normalized_hypothesis": "then again tudor had such an irritating way about him",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0310",
+ "category": "phonetic_balance",
+ "text": "In it there was something that was almost tragedy.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0310.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "In it there was something that was almost tragedy.",
+ "raw": {
+ "normalized_reference": "in it there was something that was almost tragedy.",
+ "normalized_hypothesis": "in it there was something that was almost tragedy.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "in it there was something that was almost tragedy",
+ "normalized_hypothesis": "in it there was something that was almost tragedy",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0311",
+ "category": "phonetic_balance",
+ "text": "Thirty pounds, said the captain with finality.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0311.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "30 pounds, said the captain with finality.",
+ "raw": {
+ "normalized_reference": "thirty pounds, said the captain with finality.",
+ "normalized_hypothesis": "30 pounds, said the captain with finality.",
+ "errors": 1,
+ "words": 7,
+ "wer": 0.14285714285714285
+ },
+ "semantic": {
+ "normalized_reference": "\u00a330 said the captain with finality",
+ "normalized_hypothesis": "\u00a330 said the captain with finality",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0312",
+ "category": "phonetic_balance",
+ "text": "It was over when he made his way through the ring of spectators.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0312.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It was over when he made his way through the ring of spectators.",
+ "raw": {
+ "normalized_reference": "it was over when he made his way through the ring of spectators.",
+ "normalized_hypothesis": "it was over when he made his way through the ring of spectators.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it was over when he made his way through the ring of spectators",
+ "normalized_hypothesis": "it was over when he made his way through the ring of spectators",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0313",
+ "category": "phonetic_balance",
+ "text": "She had died from cold and starvation.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0313.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "She had died from cold and starvation.",
+ "raw": {
+ "normalized_reference": "she had died from cold and starvation.",
+ "normalized_hypothesis": "she had died from cold and starvation.",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "she had died from cold and starvation",
+ "normalized_hypothesis": "she had died from cold and starvation",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0314",
+ "category": "phonetic_balance",
+ "text": "We fished sharks on Niihau together.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0314.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "We fished sharks on Ni Hao together.",
+ "raw": {
+ "normalized_reference": "we fished sharks on niihau together.",
+ "normalized_hypothesis": "we fished sharks on ni hao together.",
+ "errors": 2,
+ "words": 6,
+ "wer": 0.3333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "we fished sharks on niihau together",
+ "normalized_hypothesis": "we fished sharks on ni hao together",
+ "errors": 2,
+ "words": 6,
+ "wer": 0.3333333333333333
+ }
+ },
+ {
+ "id": "release500_0315",
+ "category": "phonetic_balance",
+ "text": "His slim hands gripped the edges of the table.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0315.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "His slim hands gripped the edges of the table.",
+ "raw": {
+ "normalized_reference": "his slim hands gripped the edges of the table.",
+ "normalized_hypothesis": "his slim hands gripped the edges of the table.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "his slim hands gripped the edges of the table",
+ "normalized_hypothesis": "his slim hands gripped the edges of the table",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0316",
+ "category": "phonetic_balance",
+ "text": "We had been chased by them ourselves, more than once.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0316.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "We had been chased by them ourselves more than once.",
+ "raw": {
+ "normalized_reference": "we had been chased by them ourselves, more than once.",
+ "normalized_hypothesis": "we had been chased by them ourselves more than once.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "we had been chased by them ourselves more than once",
+ "normalized_hypothesis": "we had been chased by them ourselves more than once",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0317",
+ "category": "phonetic_balance",
+ "text": "Or have they already devised one.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0317.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Or have they already devised one?",
+ "raw": {
+ "normalized_reference": "or have they already devised one.",
+ "normalized_hypothesis": "or have they already devised one?",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "or have they already devised one",
+ "normalized_hypothesis": "or have they already devised one",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0318",
+ "category": "phonetic_balance",
+ "text": "I cannot follow you, she said.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0318.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I cannot follow you, she said.",
+ "raw": {
+ "normalized_reference": "i cannot follow you, she said.",
+ "normalized_hypothesis": "i cannot follow you, she said.",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i cannot follow you she said",
+ "normalized_hypothesis": "i cannot follow you she said",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0319",
+ "category": "phonetic_balance",
+ "text": "He went down in midstream, searching the shadows of both shores.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0319.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He went down in midstream, searching the shadows of both shores.",
+ "raw": {
+ "normalized_reference": "he went down in midstream, searching the shadows of both shores.",
+ "normalized_hypothesis": "he went down in midstream, searching the shadows of both shores.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he went down in midstream searching the shadows of both shores",
+ "normalized_hypothesis": "he went down in midstream searching the shadows of both shores",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0320",
+ "category": "phonetic_balance",
+ "text": "I was sick once -- typhoid.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0320.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I was sick once typhoid.",
+ "raw": {
+ "normalized_reference": "i was sick once -- typhoid.",
+ "normalized_hypothesis": "i was sick once typhoid.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i was sick once typhoid",
+ "normalized_hypothesis": "i was sick once typhoid",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0321",
+ "category": "phonetic_balance",
+ "text": "Why not like any railroad station or ferry depot.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0321.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Why not like any railroad station or ferry depot?",
+ "raw": {
+ "normalized_reference": "why not like any railroad station or ferry depot.",
+ "normalized_hypothesis": "why not like any railroad station or ferry depot?",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "why not like any railroad station or ferry depot",
+ "normalized_hypothesis": "why not like any railroad station or ferry depot",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0322",
+ "category": "phonetic_balance",
+ "text": "Red-Eye swung back and forth on the branch farther down.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0322.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Red Eye swung back and forth on the branch further down.",
+ "raw": {
+ "normalized_reference": "red-eye swung back and forth on the branch farther down.",
+ "normalized_hypothesis": "red eye swung back and forth on the branch further down.",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ },
+ "semantic": {
+ "normalized_reference": "red eye swung back and forth on the branch farther down",
+ "normalized_hypothesis": "red eye swung back and forth on the branch further down",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ }
+ },
+ {
+ "id": "release500_0323",
+ "category": "phonetic_balance",
+ "text": "Within himself he called it no longer his own.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0323.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Within himself he called it no longer his own.",
+ "raw": {
+ "normalized_reference": "within himself he called it no longer his own.",
+ "normalized_hypothesis": "within himself he called it no longer his own.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "within himself he called it no longer his own",
+ "normalized_hypothesis": "within himself he called it no longer his own",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0324",
+ "category": "phonetic_balance",
+ "text": "It lasted as a deterrent for two days.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0324.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It lasted as a deterrent for two days.",
+ "raw": {
+ "normalized_reference": "it lasted as a deterrent for two days.",
+ "normalized_hypothesis": "it lasted as a deterrent for two days.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it lasted as a deterrent for 2 days",
+ "normalized_hypothesis": "it lasted as a deterrent for 2 days",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0325",
+ "category": "phonetic_balance",
+ "text": "He loved to play Chinese lottery.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0325.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He loved to play Chinese lottery.",
+ "raw": {
+ "normalized_reference": "he loved to play chinese lottery.",
+ "normalized_hypothesis": "he loved to play chinese lottery.",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he loved to play chinese lottery",
+ "normalized_hypothesis": "he loved to play chinese lottery",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0326",
+ "category": "phonetic_balance",
+ "text": "Lord, but I'm glad to see you again, Phil.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0326.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Lord, but I'm glad to see you again, Phil.",
+ "raw": {
+ "normalized_reference": "lord, but i'm glad to see you again, phil.",
+ "normalized_hypothesis": "lord, but i'm glad to see you again, phil.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "lord but i am glad to see you again phil",
+ "normalized_hypothesis": "lord but i am glad to see you again phil",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0327",
+ "category": "phonetic_balance",
+ "text": "There is not an iota of truth in it, certainly not.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0327.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "There is not an iota of truth in it, certainly not.",
+ "raw": {
+ "normalized_reference": "there is not an iota of truth in it, certainly not.",
+ "normalized_hypothesis": "there is not an iota of truth in it, certainly not.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "there is not an iota of truth in it certainly not",
+ "normalized_hypothesis": "there is not an iota of truth in it certainly not",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0328",
+ "category": "phonetic_balance",
+ "text": "She added, with genuine sympathy in her face and voice.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0328.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "She added, with genuine sympathy in her face and voice.",
+ "raw": {
+ "normalized_reference": "she added, with genuine sympathy in her face and voice.",
+ "normalized_hypothesis": "she added, with genuine sympathy in her face and voice.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "she added with genuine sympathy in her face and voice",
+ "normalized_hypothesis": "she added with genuine sympathy in her face and voice",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0329",
+ "category": "phonetic_balance",
+ "text": "We have plenty of capital ourselves, and yet we want more.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0329.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "We have plenty of capital ourselves and yet we want more.",
+ "raw": {
+ "normalized_reference": "we have plenty of capital ourselves, and yet we want more.",
+ "normalized_hypothesis": "we have plenty of capital ourselves and yet we want more.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "we have plenty of capital ourselves and yet we want more",
+ "normalized_hypothesis": "we have plenty of capital ourselves and yet we want more",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0330",
+ "category": "phonetic_balance",
+ "text": "And now put yourself in my place for a moment.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0330.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "And now put yourself in my place for a moment.",
+ "raw": {
+ "normalized_reference": "and now put yourself in my place for a moment.",
+ "normalized_hypothesis": "and now put yourself in my place for a moment.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "and now put yourself in my place for a moment",
+ "normalized_hypothesis": "and now put yourself in my place for a moment",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0331",
+ "category": "phonetic_balance",
+ "text": "Fast, but endure.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0331.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Fast, but indoor.",
+ "raw": {
+ "normalized_reference": "fast, but endure.",
+ "normalized_hypothesis": "fast, but indoor.",
+ "errors": 1,
+ "words": 3,
+ "wer": 0.3333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "fast but endure",
+ "normalized_hypothesis": "fast but indoor",
+ "errors": 1,
+ "words": 3,
+ "wer": 0.3333333333333333
+ }
+ },
+ {
+ "id": "release500_0332",
+ "category": "phonetic_balance",
+ "text": "You were looking squeamish this afternoon, he began.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0332.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "You were looking squeamish this afternoon, he began.",
+ "raw": {
+ "normalized_reference": "you were looking squeamish this afternoon, he began.",
+ "normalized_hypothesis": "you were looking squeamish this afternoon, he began.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "you were looking squeamish this afternoon he began",
+ "normalized_hypothesis": "you were looking squeamish this afternoon he began",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0333",
+ "category": "phonetic_balance",
+ "text": "It is the nearest refuge.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0333.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It is the nearest refuge.",
+ "raw": {
+ "normalized_reference": "it is the nearest refuge.",
+ "normalized_hypothesis": "it is the nearest refuge.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it is the nearest refuge",
+ "normalized_hypothesis": "it is the nearest refuge",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0334",
+ "category": "phonetic_balance",
+ "text": "I want to die in it.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0334.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I want to die in it.",
+ "raw": {
+ "normalized_reference": "i want to die in it.",
+ "normalized_hypothesis": "i want to die in it.",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i want to die in it",
+ "normalized_hypothesis": "i want to die in it",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0335",
+ "category": "phonetic_balance",
+ "text": "He stopped, and Philip nodded at the horrified question in his eyes.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0335.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He stopped, and Philip nodded at the horrified question in his eyes.",
+ "raw": {
+ "normalized_reference": "he stopped, and philip nodded at the horrified question in his eyes.",
+ "normalized_hypothesis": "he stopped, and philip nodded at the horrified question in his eyes.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he stopped and philip nodded at the horrified question in his eyes",
+ "normalized_hypothesis": "he stopped and philip nodded at the horrified question in his eyes",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0336",
+ "category": "phonetic_balance",
+ "text": "She turned, fearing that Jacques might see what was in her face.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0336.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "She turned, fearing that Jack might see what was in her face.",
+ "raw": {
+ "normalized_reference": "she turned, fearing that jacques might see what was in her face.",
+ "normalized_hypothesis": "she turned, fearing that jack might see what was in her face.",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "she turned fearing that jacques might see what was in her face",
+ "normalized_hypothesis": "she turned fearing that jack might see what was in her face",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ }
+ },
+ {
+ "id": "release500_0337",
+ "category": "phonetic_balance",
+ "text": "It was a curious coincidence.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0337.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It was a curious coincidence.",
+ "raw": {
+ "normalized_reference": "it was a curious coincidence.",
+ "normalized_hypothesis": "it was a curious coincidence.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it was a curious coincidence",
+ "normalized_hypothesis": "it was a curious coincidence",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0338",
+ "category": "phonetic_balance",
+ "text": "Daylight made no answer, and the door closed behind him.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0338.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Daylight made no answer and the door closed behind him.",
+ "raw": {
+ "normalized_reference": "daylight made no answer, and the door closed behind him.",
+ "normalized_hypothesis": "daylight made no answer and the door closed behind him.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "daylight made no answer and the door closed behind him",
+ "normalized_hypothesis": "daylight made no answer and the door closed behind him",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0339",
+ "category": "phonetic_balance",
+ "text": "And he did hurt my arm.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0339.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "And he did hurt my arm.",
+ "raw": {
+ "normalized_reference": "and he did hurt my arm.",
+ "normalized_hypothesis": "and he did hurt my arm.",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "and he did hurt my arm",
+ "normalized_hypothesis": "and he did hurt my arm",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0340",
+ "category": "phonetic_balance",
+ "text": "Dunham, can your boy go along with Jesse.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0340.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Dunham, can your boy go along with Jessie?",
+ "raw": {
+ "normalized_reference": "dunham, can your boy go along with jesse.",
+ "normalized_hypothesis": "dunham, can your boy go along with jessie?",
+ "errors": 1,
+ "words": 8,
+ "wer": 0.125
+ },
+ "semantic": {
+ "normalized_reference": "dunham can your boy go along with jesse",
+ "normalized_hypothesis": "dunham can your boy go along with jessie",
+ "errors": 1,
+ "words": 8,
+ "wer": 0.125
+ }
+ },
+ {
+ "id": "release500_0341",
+ "category": "phonetic_balance",
+ "text": "The creative joy, I murmured.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0341.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The creative joy, I murmured.",
+ "raw": {
+ "normalized_reference": "the creative joy, i murmured.",
+ "normalized_hypothesis": "the creative joy, i murmured.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the creative joy i murmured",
+ "normalized_hypothesis": "the creative joy i murmured",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0342",
+ "category": "phonetic_balance",
+ "text": "He pulled, and the log crashed down to break his back.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0342.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He pulled and the log crashed down to break his back.",
+ "raw": {
+ "normalized_reference": "he pulled, and the log crashed down to break his back.",
+ "normalized_hypothesis": "he pulled and the log crashed down to break his back.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he pulled and the log crashed down to break his back",
+ "normalized_hypothesis": "he pulled and the log crashed down to break his back",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0343",
+ "category": "phonetic_balance",
+ "text": "By golly, the boy wins.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0343.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "By golly, the boy wins.",
+ "raw": {
+ "normalized_reference": "by golly, the boy wins.",
+ "normalized_hypothesis": "by golly, the boy wins.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "by golly the boy wins",
+ "normalized_hypothesis": "by golly the boy wins",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0344",
+ "category": "phonetic_balance",
+ "text": "The boy threw back his head with pride.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0344.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The boy threw back his head with pride.",
+ "raw": {
+ "normalized_reference": "the boy threw back his head with pride.",
+ "normalized_hypothesis": "the boy threw back his head with pride.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the boy threw back his head with pride",
+ "normalized_hypothesis": "the boy threw back his head with pride",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0345",
+ "category": "phonetic_balance",
+ "text": "The last one I knew was an overseer.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0345.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The last one I knew was an overseer.",
+ "raw": {
+ "normalized_reference": "the last one i knew was an overseer.",
+ "normalized_hypothesis": "the last one i knew was an overseer.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the last one i knew was an overseer",
+ "normalized_hypothesis": "the last one i knew was an overseer",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0346",
+ "category": "phonetic_balance",
+ "text": "Two years ago I gave up civilization for this.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0346.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Two years ago, I gave up civilization for this.",
+ "raw": {
+ "normalized_reference": "two years ago i gave up civilization for this.",
+ "normalized_hypothesis": "two years ago, i gave up civilization for this.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "2 years ago i gave up civilization for this",
+ "normalized_hypothesis": "2 years ago i gave up civilization for this",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0347",
+ "category": "phonetic_balance",
+ "text": "The river bared its bosom, and snorting steamboats challenged the wilderness.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0347.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The river bared its bosom, and snorting steamboats challenged the wilderness.",
+ "raw": {
+ "normalized_reference": "the river bared its bosom, and snorting steamboats challenged the wilderness.",
+ "normalized_hypothesis": "the river bared its bosom, and snorting steamboats challenged the wilderness.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the river bared its bosom and snorting steamboats challenged the wilderness",
+ "normalized_hypothesis": "the river bared its bosom and snorting steamboats challenged the wilderness",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0348",
+ "category": "phonetic_balance",
+ "text": "Mab, she said.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0348.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Mab, she said.",
+ "raw": {
+ "normalized_reference": "mab, she said.",
+ "normalized_hypothesis": "mab, she said.",
+ "errors": 0,
+ "words": 3,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "mab she said",
+ "normalized_hypothesis": "mab she said",
+ "errors": 0,
+ "words": 3,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0349",
+ "category": "phonetic_balance",
+ "text": "The regret in his voice was provocative of a second burst of laughter.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0349.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The regret in his voice was provocative of a second burst of laughter.",
+ "raw": {
+ "normalized_reference": "the regret in his voice was provocative of a second burst of laughter.",
+ "normalized_hypothesis": "the regret in his voice was provocative of a second burst of laughter.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the regret in his voice was provocative of a 2nd burst of laughter",
+ "normalized_hypothesis": "the regret in his voice was provocative of a 2nd burst of laughter",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0350",
+ "category": "phonetic_balance",
+ "text": "And each year something happened, and I did not go.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0350.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "And each year something happened and I did not go.",
+ "raw": {
+ "normalized_reference": "and each year something happened, and i did not go.",
+ "normalized_hypothesis": "and each year something happened and i did not go.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "and each year something happened and i did not go",
+ "normalized_hypothesis": "and each year something happened and i did not go",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0351",
+ "category": "phonetic_balance",
+ "text": "Gad, your letter came just in time.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0351.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Gad, your letter came just in time.",
+ "raw": {
+ "normalized_reference": "gad, your letter came just in time.",
+ "normalized_hypothesis": "gad, your letter came just in time.",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "gad your letter came just in time",
+ "normalized_hypothesis": "gad your letter came just in time",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0352",
+ "category": "phonetic_balance",
+ "text": "But if Pierre did not return, until tomorrow.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0352.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "What if Pierre did not return until tomorrow?",
+ "raw": {
+ "normalized_reference": "but if pierre did not return, until tomorrow.",
+ "normalized_hypothesis": "what if pierre did not return until tomorrow?",
+ "errors": 1,
+ "words": 8,
+ "wer": 0.125
+ },
+ "semantic": {
+ "normalized_reference": "but if pierre did not return until tomorrow",
+ "normalized_hypothesis": "what if pierre did not return until tomorrow",
+ "errors": 1,
+ "words": 8,
+ "wer": 0.125
+ }
+ },
+ {
+ "id": "release500_0353",
+ "category": "phonetic_balance",
+ "text": "Billinger may arrive in time.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0353.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Billinger may arrive in time.",
+ "raw": {
+ "normalized_reference": "billinger may arrive in time.",
+ "normalized_hypothesis": "billinger may arrive in time.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "billinger may arrive in time",
+ "normalized_hypothesis": "billinger may arrive in time",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0354",
+ "category": "phonetic_balance",
+ "text": "In such a tumbling of values was no time to sell.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0354.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "In such a tumbling of values was no time to sell.",
+ "raw": {
+ "normalized_reference": "in such a tumbling of values was no time to sell.",
+ "normalized_hypothesis": "in such a tumbling of values was no time to sell.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "in such a tumbling of values was no time to sell",
+ "normalized_hypothesis": "in such a tumbling of values was no time to sell",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0355",
+ "category": "phonetic_balance",
+ "text": "There's not an iota of truth in it.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0355.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "There's not an iota of truth in it.",
+ "raw": {
+ "normalized_reference": "there's not an iota of truth in it.",
+ "normalized_hypothesis": "there's not an iota of truth in it.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "there is not an iota of truth in it",
+ "normalized_hypothesis": "there is not an iota of truth in it",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0356",
+ "category": "phonetic_balance",
+ "text": "It is dog eat dog, and you ate them up.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0356.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It is dog eat dog, and you ate them up.",
+ "raw": {
+ "normalized_reference": "it is dog eat dog, and you ate them up.",
+ "normalized_hypothesis": "it is dog eat dog, and you ate them up.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it is dog eat dog and you ate them up",
+ "normalized_hypothesis": "it is dog eat dog and you ate them up",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0357",
+ "category": "phonetic_balance",
+ "text": "You can take a vacation on pay.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0357.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "You can take a vacation on pay.",
+ "raw": {
+ "normalized_reference": "you can take a vacation on pay.",
+ "normalized_hypothesis": "you can take a vacation on pay.",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "you can take a vacation on pay",
+ "normalized_hypothesis": "you can take a vacation on pay",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0358",
+ "category": "phonetic_balance",
+ "text": "I have seen myself that one man contemplated by Pascal's philosophic eye.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0358.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I have seen myself that one man contemplated by Pascal's philosophic eye.",
+ "raw": {
+ "normalized_reference": "i have seen myself that one man contemplated by pascal's philosophic eye.",
+ "normalized_hypothesis": "i have seen myself that one man contemplated by pascal's philosophic eye.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i have seen myself that one man contemplated by pascal is philosophic eye",
+ "normalized_hypothesis": "i have seen myself that one man contemplated by pascal is philosophic eye",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0359",
+ "category": "phonetic_balance",
+ "text": "Ah, I had forgotten, he exclaimed.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0359.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Ah, I had forgotten, he exclaimed.",
+ "raw": {
+ "normalized_reference": "ah, i had forgotten, he exclaimed.",
+ "normalized_hypothesis": "ah, i had forgotten, he exclaimed.",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "ah i had forgotten he exclaimed",
+ "normalized_hypothesis": "ah i had forgotten he exclaimed",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0360",
+ "category": "phonetic_balance",
+ "text": "Providence had delivered him through the maelstrom.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0360.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Providence had delivered him through the maelstrom.",
+ "raw": {
+ "normalized_reference": "providence had delivered him through the maelstrom.",
+ "normalized_hypothesis": "providence had delivered him through the maelstrom.",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "providence had delivered him through the maelstrom",
+ "normalized_hypothesis": "providence had delivered him through the maelstrom",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0361",
+ "category": "phonetic_balance",
+ "text": "She was his now, forever.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0361.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "She was his now, forever.",
+ "raw": {
+ "normalized_reference": "she was his now, forever.",
+ "normalized_hypothesis": "she was his now, forever.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "she was his now forever",
+ "normalized_hypothesis": "she was his now forever",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0362",
+ "category": "phonetic_balance",
+ "text": "It was unobtrusive, yet it was there.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0362.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It was unobtrusive, yet it was there.",
+ "raw": {
+ "normalized_reference": "it was unobtrusive, yet it was there.",
+ "normalized_hypothesis": "it was unobtrusive, yet it was there.",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it was unobtrusive yet it was there",
+ "normalized_hypothesis": "it was unobtrusive yet it was there",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0363",
+ "category": "phonetic_balance",
+ "text": "Men like Joe Goose dated existence from drunk to drunk.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0363.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Men like Joe Goose dated existence from drunk to drunk.",
+ "raw": {
+ "normalized_reference": "men like joe goose dated existence from drunk to drunk.",
+ "normalized_hypothesis": "men like joe goose dated existence from drunk to drunk.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "men like joe goose dated existence from drunk to drunk",
+ "normalized_hypothesis": "men like joe goose dated existence from drunk to drunk",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0364",
+ "category": "phonetic_balance",
+ "text": "I was completely lost in my work.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0364.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I was completely lost in my work.",
+ "raw": {
+ "normalized_reference": "i was completely lost in my work.",
+ "normalized_hypothesis": "i was completely lost in my work.",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i was completely lost in my work",
+ "normalized_hypothesis": "i was completely lost in my work",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0365",
+ "category": "phonetic_balance",
+ "text": "The very thing, Ernest agreed.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0365.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The very thing, Ernest agreed.",
+ "raw": {
+ "normalized_reference": "the very thing, ernest agreed.",
+ "normalized_hypothesis": "the very thing, ernest agreed.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the very thing ernest agreed",
+ "normalized_hypothesis": "the very thing ernest agreed",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0366",
+ "category": "phonetic_balance",
+ "text": "He was the leader, and Tudor was his lieutenant.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0366.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He was the leader, and Tudor was his lieutenant.",
+ "raw": {
+ "normalized_reference": "he was the leader, and tudor was his lieutenant.",
+ "normalized_hypothesis": "he was the leader, and tudor was his lieutenant.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he was the leader and tudor was his lieutenant",
+ "normalized_hypothesis": "he was the leader and tudor was his lieutenant",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0367",
+ "category": "phonetic_balance",
+ "text": "The Fire-Men wore animal skins around their waists and across their shoulders.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0367.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The firemen wore animal skins around their waists and across their shoulders.",
+ "raw": {
+ "normalized_reference": "the fire-men wore animal skins around their waists and across their shoulders.",
+ "normalized_hypothesis": "the firemen wore animal skins around their waists and across their shoulders.",
+ "errors": 2,
+ "words": 13,
+ "wer": 0.15384615384615385
+ },
+ "semantic": {
+ "normalized_reference": "the fire men wore animal skins around their waists and across their shoulders",
+ "normalized_hypothesis": "the firemen wore animal skins around their waists and across their shoulders",
+ "errors": 2,
+ "words": 13,
+ "wer": 0.15384615384615385
+ }
+ },
+ {
+ "id": "release500_0368",
+ "category": "phonetic_balance",
+ "text": "We could throw stones with our feet.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0368.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "We can throw stones with our feet.",
+ "raw": {
+ "normalized_reference": "we could throw stones with our feet.",
+ "normalized_hypothesis": "we can throw stones with our feet.",
+ "errors": 1,
+ "words": 7,
+ "wer": 0.14285714285714285
+ },
+ "semantic": {
+ "normalized_reference": "we could throw stones with our feet",
+ "normalized_hypothesis": "we can throw stones with our feet",
+ "errors": 1,
+ "words": 7,
+ "wer": 0.14285714285714285
+ }
+ },
+ {
+ "id": "release500_0369",
+ "category": "phonetic_balance",
+ "text": "He turned sharply, and faced Gregson across the table.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0369.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He turned sharply and faced Gregson across the table.",
+ "raw": {
+ "normalized_reference": "he turned sharply, and faced gregson across the table.",
+ "normalized_hypothesis": "he turned sharply and faced gregson across the table.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he turned sharply and faced gregson across the table",
+ "normalized_hypothesis": "he turned sharply and faced gregson across the table",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0370",
+ "category": "phonetic_balance",
+ "text": "They are to attack your camp tomorrow night.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0370.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "They are to attack your camp tomorrow night.",
+ "raw": {
+ "normalized_reference": "they are to attack your camp tomorrow night.",
+ "normalized_hypothesis": "they are to attack your camp tomorrow night.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "they are to attack your camp tomorrow night",
+ "normalized_hypothesis": "they are to attack your camp tomorrow night",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0371",
+ "category": "phonetic_balance",
+ "text": "And watch out for wet feet, was his parting advice.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0371.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "And watch out for wet feet, was his parting advice.",
+ "raw": {
+ "normalized_reference": "and watch out for wet feet, was his parting advice.",
+ "normalized_hypothesis": "and watch out for wet feet, was his parting advice.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "and watch out for wet feet was his parting advice",
+ "normalized_hypothesis": "and watch out for wet feet was his parting advice",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0372",
+ "category": "phonetic_balance",
+ "text": "Two weeks passed, and in that time Thorpe left camp three times.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0372.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Two weeks passed, and in that time Thorpe left camp three times.",
+ "raw": {
+ "normalized_reference": "two weeks passed, and in that time thorpe left camp three times.",
+ "normalized_hypothesis": "two weeks passed, and in that time thorpe left camp three times.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "2 weeks passed and in that time thorpe left camp 3 times",
+ "normalized_hypothesis": "2 weeks passed and in that time thorpe left camp 3 times",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0373",
+ "category": "phonetic_balance",
+ "text": "But ever his gaze returned to that Crouched Venus on the piano.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0373.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Whatever his gaze returned to that crouched Venus on the piano.",
+ "raw": {
+ "normalized_reference": "but ever his gaze returned to that crouched venus on the piano.",
+ "normalized_hypothesis": "whatever his gaze returned to that crouched venus on the piano.",
+ "errors": 2,
+ "words": 12,
+ "wer": 0.16666666666666666
+ },
+ "semantic": {
+ "normalized_reference": "but ever his gaze returned to that crouched venus on the piano",
+ "normalized_hypothesis": "whatever his gaze returned to that crouched venus on the piano",
+ "errors": 2,
+ "words": 12,
+ "wer": 0.16666666666666666
+ }
+ },
+ {
+ "id": "release500_0374",
+ "category": "phonetic_balance",
+ "text": "But she had become an automaton.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0374.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "But she had become an automaton.",
+ "raw": {
+ "normalized_reference": "but she had become an automaton.",
+ "normalized_hypothesis": "but she had become an automaton.",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "but she had become an automaton",
+ "normalized_hypothesis": "but she had become an automaton",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0375",
+ "category": "phonetic_balance",
+ "text": "I do not blame you for anything; remember that.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0375.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I do not blame you for anything. Remember that.",
+ "raw": {
+ "normalized_reference": "i do not blame you for anything; remember that.",
+ "normalized_hypothesis": "i do not blame you for anything. remember that.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i do not blame you for anything remember that",
+ "normalized_hypothesis": "i do not blame you for anything remember that",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0376",
+ "category": "phonetic_balance",
+ "text": "The next thing to watch out for is bed sores.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0376.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The next thing to watch out for is bed sores.",
+ "raw": {
+ "normalized_reference": "the next thing to watch out for is bed sores.",
+ "normalized_hypothesis": "the next thing to watch out for is bed sores.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the next thing to watch out for is bed sores",
+ "normalized_hypothesis": "the next thing to watch out for is bed sores",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0377",
+ "category": "phonetic_balance",
+ "text": "He will follow us soon.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0377.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He will follow us soon.",
+ "raw": {
+ "normalized_reference": "he will follow us soon.",
+ "normalized_hypothesis": "he will follow us soon.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he will follow us soon",
+ "normalized_hypothesis": "he will follow us soon",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0378",
+ "category": "phonetic_balance",
+ "text": "His beady black eyes saw bargains where other men saw bankruptcy.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0378.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "His beady black eyes saw bargains where other men saw bankruptcy.",
+ "raw": {
+ "normalized_reference": "his beady black eyes saw bargains where other men saw bankruptcy.",
+ "normalized_hypothesis": "his beady black eyes saw bargains where other men saw bankruptcy.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "his beady black eyes saw bargains where other men saw bankruptcy",
+ "normalized_hypothesis": "his beady black eyes saw bargains where other men saw bankruptcy",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0379",
+ "category": "phonetic_balance",
+ "text": "They robbed me a few years later.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0379.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "They robbed me a few years later.",
+ "raw": {
+ "normalized_reference": "they robbed me a few years later.",
+ "normalized_hypothesis": "they robbed me a few years later.",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "they robbed me a few years later",
+ "normalized_hypothesis": "they robbed me a few years later",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0380",
+ "category": "phonetic_balance",
+ "text": "Gregson held a lighted match until it burnt his fingertips.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0380.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Gregson held a lighted match until it burnt his fingertips.",
+ "raw": {
+ "normalized_reference": "gregson held a lighted match until it burnt his fingertips.",
+ "normalized_hypothesis": "gregson held a lighted match until it burnt his fingertips.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "gregson held a lighted match until it burnt his fingertips",
+ "normalized_hypothesis": "gregson held a lighted match until it burnt his fingertips",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0381",
+ "category": "phonetic_balance",
+ "text": "A burst of laughter was his reward.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0381.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "A burst of laughter was his reward.",
+ "raw": {
+ "normalized_reference": "a burst of laughter was his reward.",
+ "normalized_hypothesis": "a burst of laughter was his reward.",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "a burst of laughter was his reward",
+ "normalized_hypothesis": "a burst of laughter was his reward",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0382",
+ "category": "phonetic_balance",
+ "text": "He cried, and swung the club wildly.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0382.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He cry and swung the club wildly.",
+ "raw": {
+ "normalized_reference": "he cried, and swung the club wildly.",
+ "normalized_hypothesis": "he cry and swung the club wildly.",
+ "errors": 1,
+ "words": 7,
+ "wer": 0.14285714285714285
+ },
+ "semantic": {
+ "normalized_reference": "he cried and swung the club wildly",
+ "normalized_hypothesis": "he cry and swung the club wildly",
+ "errors": 1,
+ "words": 7,
+ "wer": 0.14285714285714285
+ }
+ },
+ {
+ "id": "release500_0383",
+ "category": "phonetic_balance",
+ "text": "Sandel would never become a world champion.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0383.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Sandal would never become a world champion.",
+ "raw": {
+ "normalized_reference": "sandel would never become a world champion.",
+ "normalized_hypothesis": "sandal would never become a world champion.",
+ "errors": 1,
+ "words": 7,
+ "wer": 0.14285714285714285
+ },
+ "semantic": {
+ "normalized_reference": "sandel would never become a world champion",
+ "normalized_hypothesis": "sandal would never become a world champion",
+ "errors": 1,
+ "words": 7,
+ "wer": 0.14285714285714285
+ }
+ },
+ {
+ "id": "release500_0384",
+ "category": "phonetic_balance",
+ "text": "And after the bath a shave would not be bad.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0384.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "and after the bath the shave would not be bad.",
+ "raw": {
+ "normalized_reference": "and after the bath a shave would not be bad.",
+ "normalized_hypothesis": "and after the bath the shave would not be bad.",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ },
+ "semantic": {
+ "normalized_reference": "and after the bath a shave would not be bad",
+ "normalized_hypothesis": "and after the bath the shave would not be bad",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ }
+ },
+ {
+ "id": "release500_0385",
+ "category": "phonetic_balance",
+ "text": "They saw each other for the first time in Boston.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0385.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "They saw each other for the first time in Boston.",
+ "raw": {
+ "normalized_reference": "they saw each other for the first time in boston.",
+ "normalized_hypothesis": "they saw each other for the first time in boston.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "they saw each other for the 1st time in boston",
+ "normalized_hypothesis": "they saw each other for the 1st time in boston",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0386",
+ "category": "phonetic_balance",
+ "text": "I had been born with no organic, chemical predisposition toward alcohol.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0386.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I had been born with no organic chemical predisposition toward alcohol.",
+ "raw": {
+ "normalized_reference": "i had been born with no organic, chemical predisposition toward alcohol.",
+ "normalized_hypothesis": "i had been born with no organic chemical predisposition toward alcohol.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i had been born with no organic chemical predisposition toward alcohol",
+ "normalized_hypothesis": "i had been born with no organic chemical predisposition toward alcohol",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0387",
+ "category": "phonetic_balance",
+ "text": "They had no fixed values, to be altered by adjectives and adverbs.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0387.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "They had no fixed values to be altered by adjectives and adverbs.",
+ "raw": {
+ "normalized_reference": "they had no fixed values, to be altered by adjectives and adverbs.",
+ "normalized_hypothesis": "they had no fixed values to be altered by adjectives and adverbs.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "they had no fixed values to be altered by adjectives and adverbs",
+ "normalized_hypothesis": "they had no fixed values to be altered by adjectives and adverbs",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0388",
+ "category": "phonetic_balance",
+ "text": "Nor was Elam Harnish an exception.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0388.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "nor was a lamb harnish an exception.",
+ "raw": {
+ "normalized_reference": "nor was elam harnish an exception.",
+ "normalized_hypothesis": "nor was a lamb harnish an exception.",
+ "errors": 2,
+ "words": 6,
+ "wer": 0.3333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "nor was elam harnish an exception",
+ "normalized_hypothesis": "nor was a lamb harnish an exception",
+ "errors": 2,
+ "words": 6,
+ "wer": 0.3333333333333333
+ }
+ },
+ {
+ "id": "release500_0389",
+ "category": "phonetic_balance",
+ "text": "He was a wise hyena.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0389.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He was a white hyena.",
+ "raw": {
+ "normalized_reference": "he was a wise hyena.",
+ "normalized_hypothesis": "he was a white hyena.",
+ "errors": 1,
+ "words": 5,
+ "wer": 0.2
+ },
+ "semantic": {
+ "normalized_reference": "he was a wise hyena",
+ "normalized_hypothesis": "he was a white hyena",
+ "errors": 1,
+ "words": 5,
+ "wer": 0.2
+ }
+ },
+ {
+ "id": "release500_0390",
+ "category": "phonetic_balance",
+ "text": "And now behold the perversity of things.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0390.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "And now behold the perversity of things.",
+ "raw": {
+ "normalized_reference": "and now behold the perversity of things.",
+ "normalized_hypothesis": "and now behold the perversity of things.",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "and now behold the perversity of things",
+ "normalized_hypothesis": "and now behold the perversity of things",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0391",
+ "category": "phonetic_balance",
+ "text": "The Fire People, like ourselves, lived in caves.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0391.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The fire people, like ourselves, lived in caves.",
+ "raw": {
+ "normalized_reference": "the fire people, like ourselves, lived in caves.",
+ "normalized_hypothesis": "the fire people, like ourselves, lived in caves.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the fire people like ourselves lived in caves",
+ "normalized_hypothesis": "the fire people like ourselves lived in caves",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0392",
+ "category": "phonetic_balance",
+ "text": "To her the bridge was tambo, which is the native for taboo.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0392.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "To her the bridge was tambo, which is the native for taboo.",
+ "raw": {
+ "normalized_reference": "to her the bridge was tambo, which is the native for taboo.",
+ "normalized_hypothesis": "to her the bridge was tambo, which is the native for taboo.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "to her the bridge was tambo which is the native for taboo",
+ "normalized_hypothesis": "to her the bridge was tambo which is the native for taboo",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0393",
+ "category": "phonetic_balance",
+ "text": "The voyage was our idea of a good time.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0393.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The voyage was our idea of a good time.",
+ "raw": {
+ "normalized_reference": "the voyage was our idea of a good time.",
+ "normalized_hypothesis": "the voyage was our idea of a good time.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the voyage was our idea of a good time",
+ "normalized_hypothesis": "the voyage was our idea of a good time",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0394",
+ "category": "phonetic_balance",
+ "text": "Of course much grumbling went on, and little outbursts were continually occurring.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0394.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Of course much grumbling went on, and little outbursts were continually occurring.",
+ "raw": {
+ "normalized_reference": "of course much grumbling went on, and little outbursts were continually occurring.",
+ "normalized_hypothesis": "of course much grumbling went on, and little outbursts were continually occurring.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "of course much grumbling went on and little outbursts were continually occurring",
+ "normalized_hypothesis": "of course much grumbling went on and little outbursts were continually occurring",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0395",
+ "category": "phonetic_balance",
+ "text": "Get down and dig in.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0395.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Get down and dig in.",
+ "raw": {
+ "normalized_reference": "get down and dig in.",
+ "normalized_hypothesis": "get down and dig in.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "get down and dig in",
+ "normalized_hypothesis": "get down and dig in",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0396",
+ "category": "phonetic_balance",
+ "text": "It was put together in a casual, helter-skelter sort of way.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0396.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It was put together in a casual, helter-skelter sort of way.",
+ "raw": {
+ "normalized_reference": "it was put together in a casual, helter-skelter sort of way.",
+ "normalized_hypothesis": "it was put together in a casual, helter-skelter sort of way.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it was put together in a casual helter skelter sort of way",
+ "normalized_hypothesis": "it was put together in a casual helter skelter sort of way",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0397",
+ "category": "phonetic_balance",
+ "text": "He deluged me, overwhelmed me with argument.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0397.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He deluged me, overwhelmed me with arguments.",
+ "raw": {
+ "normalized_reference": "he deluged me, overwhelmed me with argument.",
+ "normalized_hypothesis": "he deluged me, overwhelmed me with arguments.",
+ "errors": 1,
+ "words": 7,
+ "wer": 0.14285714285714285
+ },
+ "semantic": {
+ "normalized_reference": "he deluged me overwhelmed me with argument",
+ "normalized_hypothesis": "he deluged me overwhelmed me with arguments",
+ "errors": 1,
+ "words": 7,
+ "wer": 0.14285714285714285
+ }
+ },
+ {
+ "id": "release500_0398",
+ "category": "phonetic_balance",
+ "text": "The girl faced him, her eyes shining with sudden fear.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0398.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The girl faced him, her eyes shining with sudden fear.",
+ "raw": {
+ "normalized_reference": "the girl faced him, her eyes shining with sudden fear.",
+ "normalized_hypothesis": "the girl faced him, her eyes shining with sudden fear.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the girl faced him her eyes shining with sudden fear",
+ "normalized_hypothesis": "the girl faced him her eyes shining with sudden fear",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0399",
+ "category": "phonetic_balance",
+ "text": "It will break our hearts and our backs to hoist anchor by hand.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0399.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It will break our hearts and our backs to hoist anger by hand.",
+ "raw": {
+ "normalized_reference": "it will break our hearts and our backs to hoist anchor by hand.",
+ "normalized_hypothesis": "it will break our hearts and our backs to hoist anger by hand.",
+ "errors": 1,
+ "words": 13,
+ "wer": 0.07692307692307693
+ },
+ "semantic": {
+ "normalized_reference": "it will break our hearts and our backs to hoist anchor by hand",
+ "normalized_hypothesis": "it will break our hearts and our backs to hoist anger by hand",
+ "errors": 1,
+ "words": 13,
+ "wer": 0.07692307692307693
+ }
+ },
+ {
+ "id": "release500_0400",
+ "category": "phonetic_balance",
+ "text": "He made no reply as he waited for Whittemore to continue.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0400.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He made no reply as he waited for Whittemore to continue.",
+ "raw": {
+ "normalized_reference": "he made no reply as he waited for whittemore to continue.",
+ "normalized_hypothesis": "he made no reply as he waited for whittemore to continue.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he made no reply as he waited for whittemore to continue",
+ "normalized_hypothesis": "he made no reply as he waited for whittemore to continue",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0401",
+ "category": "phonetic_balance",
+ "text": "Who the devil gave it to you to be judge and jury.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0401.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Who the devil gave it to you to be judge and jury?",
+ "raw": {
+ "normalized_reference": "who the devil gave it to you to be judge and jury.",
+ "normalized_hypothesis": "who the devil gave it to you to be judge and jury?",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "who the devil gave it to you to be judge and jury",
+ "normalized_hypothesis": "who the devil gave it to you to be judge and jury",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0402",
+ "category": "phonetic_balance",
+ "text": "Daylight found himself charmed and made curious by the little man.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0402.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Daylight found himself charmed and made curious by the little man.",
+ "raw": {
+ "normalized_reference": "daylight found himself charmed and made curious by the little man.",
+ "normalized_hypothesis": "daylight found himself charmed and made curious by the little man.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "daylight found himself charmed and made curious by the little man",
+ "normalized_hypothesis": "daylight found himself charmed and made curious by the little man",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0403",
+ "category": "phonetic_balance",
+ "text": "It's a Yankee, Joan cried.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0403.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It's a Yankee, Joan cried.",
+ "raw": {
+ "normalized_reference": "it's a yankee, joan cried.",
+ "normalized_hypothesis": "it's a yankee, joan cried.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it is a yankee joan cried",
+ "normalized_hypothesis": "it is a yankee joan cried",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0404",
+ "category": "phonetic_balance",
+ "text": "But what they want with your toothbrush is more than I can imagine.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0404.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "But what they want with your toothbrush is more than I can imagine.",
+ "raw": {
+ "normalized_reference": "but what they want with your toothbrush is more than i can imagine.",
+ "normalized_hypothesis": "but what they want with your toothbrush is more than i can imagine.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "but what they want with your toothbrush is more than i can imagine",
+ "normalized_hypothesis": "but what they want with your toothbrush is more than i can imagine",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0405",
+ "category": "phonetic_balance",
+ "text": "They were babbling and chattering all together.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0405.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "They were babbling and chattering all together.",
+ "raw": {
+ "normalized_reference": "they were babbling and chattering all together.",
+ "normalized_hypothesis": "they were babbling and chattering all together.",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "they were babbling and chattering all together",
+ "normalized_hypothesis": "they were babbling and chattering all together",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0406",
+ "category": "phonetic_balance",
+ "text": "Men of Selden's stamp don't stop at women and children.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0406.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Men of Seldon's stamp don't stop at women and children.",
+ "raw": {
+ "normalized_reference": "men of selden's stamp don't stop at women and children.",
+ "normalized_hypothesis": "men of seldon's stamp don't stop at women and children.",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ },
+ "semantic": {
+ "normalized_reference": "men of selden is stamp do not stop at women and children",
+ "normalized_hypothesis": "men of seldon is stamp do not stop at women and children",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ }
+ },
+ {
+ "id": "release500_0407",
+ "category": "phonetic_balance",
+ "text": "Cameron looked at his hands with their long, sinewy fingers.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0407.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Cameron looked at his hands with their long, sinewy fingers.",
+ "raw": {
+ "normalized_reference": "cameron looked at his hands with their long, sinewy fingers.",
+ "normalized_hypothesis": "cameron looked at his hands with their long, sinewy fingers.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "cameron looked at his hands with their long sinewy fingers",
+ "normalized_hypothesis": "cameron looked at his hands with their long sinewy fingers",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0408",
+ "category": "phonetic_balance",
+ "text": "Saxon waited, for she knew a fresh idea had struck Billy.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0408.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Saxon waited, for she knew a fresh idea had struck Billy.",
+ "raw": {
+ "normalized_reference": "saxon waited, for she knew a fresh idea had struck billy.",
+ "normalized_hypothesis": "saxon waited, for she knew a fresh idea had struck billy.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "saxon waited for she knew a fresh idea had struck billy",
+ "normalized_hypothesis": "saxon waited for she knew a fresh idea had struck billy",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0409",
+ "category": "phonetic_balance",
+ "text": "He saw only the effect in a general, sketchy way.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0409.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He saw only the effect in a general, sketchy way.",
+ "raw": {
+ "normalized_reference": "he saw only the effect in a general, sketchy way.",
+ "normalized_hypothesis": "he saw only the effect in a general, sketchy way.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he saw only the effect in a general sketchy way",
+ "normalized_hypothesis": "he saw only the effect in a general sketchy way",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0410",
+ "category": "phonetic_balance",
+ "text": "He began to follow the footprints of the dog.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0410.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "he began to follow the footprints of the dog.",
+ "raw": {
+ "normalized_reference": "he began to follow the footprints of the dog.",
+ "normalized_hypothesis": "he began to follow the footprints of the dog.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he began to follow the footprints of the dog",
+ "normalized_hypothesis": "he began to follow the footprints of the dog",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0411",
+ "category": "phonetic_balance",
+ "text": "Do you value your hide.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0411.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Do you value your hide?",
+ "raw": {
+ "normalized_reference": "do you value your hide.",
+ "normalized_hypothesis": "do you value your hide?",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "do you value your hide",
+ "normalized_hypothesis": "do you value your hide",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0412",
+ "category": "phonetic_balance",
+ "text": "A rising tide of fat had submerged them.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0412.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "A rising tide of fat had submerged them.",
+ "raw": {
+ "normalized_reference": "a rising tide of fat had submerged them.",
+ "normalized_hypothesis": "a rising tide of fat had submerged them.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "a rising tide of fat had submerged them",
+ "normalized_hypothesis": "a rising tide of fat had submerged them",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0413",
+ "category": "phonetic_balance",
+ "text": "There was nothing on the rock.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0413.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "There was nothing on the rock.",
+ "raw": {
+ "normalized_reference": "there was nothing on the rock.",
+ "normalized_hypothesis": "there was nothing on the rock.",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "there was nothing on the rock",
+ "normalized_hypothesis": "there was nothing on the rock",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0414",
+ "category": "phonetic_balance",
+ "text": "Whatever he guessed he locked away in the taboo room of Naomi.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0414.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Whatever he guessed he locked away in the taboo room of Naomi.",
+ "raw": {
+ "normalized_reference": "whatever he guessed he locked away in the taboo room of naomi.",
+ "normalized_hypothesis": "whatever he guessed he locked away in the taboo room of naomi.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "whatever he guessed he locked away in the taboo room of naomi",
+ "normalized_hypothesis": "whatever he guessed he locked away in the taboo room of naomi",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0415",
+ "category": "phonetic_balance",
+ "text": "Will we ever forget it.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0415.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Will we ever forget it?",
+ "raw": {
+ "normalized_reference": "will we ever forget it.",
+ "normalized_hypothesis": "will we ever forget it?",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "will we ever forget it",
+ "normalized_hypothesis": "will we ever forget it",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0416",
+ "category": "phonetic_balance",
+ "text": "Jeanne and Pierre both gazed toward the great rock.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0416.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Jean and Pierre both gaze toward the great rock.",
+ "raw": {
+ "normalized_reference": "jeanne and pierre both gazed toward the great rock.",
+ "normalized_hypothesis": "jean and pierre both gaze toward the great rock.",
+ "errors": 2,
+ "words": 9,
+ "wer": 0.2222222222222222
+ },
+ "semantic": {
+ "normalized_reference": "jeanne and pierre both gazed toward the great rock",
+ "normalized_hypothesis": "jean and pierre both gaze toward the great rock",
+ "errors": 2,
+ "words": 9,
+ "wer": 0.2222222222222222
+ }
+ },
+ {
+ "id": "release500_0417",
+ "category": "phonetic_balance",
+ "text": "Does the old boy often go off at half-cock that way.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0417.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Does the old boy often go off at half cock that way?",
+ "raw": {
+ "normalized_reference": "does the old boy often go off at half-cock that way.",
+ "normalized_hypothesis": "does the old boy often go off at half cock that way?",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "does the old boy often go off at half cock that way",
+ "normalized_hypothesis": "does the old boy often go off at half cock that way",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0418",
+ "category": "phonetic_balance",
+ "text": "The temperature dropped to fifty below zero and remained there the whole trip.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0418.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The temperature dropped to 50 below zero and remained there the whole trip.",
+ "raw": {
+ "normalized_reference": "the temperature dropped to fifty below zero and remained there the whole trip.",
+ "normalized_hypothesis": "the temperature dropped to 50 below zero and remained there the whole trip.",
+ "errors": 1,
+ "words": 13,
+ "wer": 0.07692307692307693
+ },
+ "semantic": {
+ "normalized_reference": "the temperature dropped to 50 below 0 and remained there the whole trip",
+ "normalized_hypothesis": "the temperature dropped to 50 below 0 and remained there the whole trip",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0419",
+ "category": "phonetic_balance",
+ "text": "It's worth eight dollars.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0419.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It's worth $8.",
+ "raw": {
+ "normalized_reference": "it's worth eight dollars.",
+ "normalized_hypothesis": "it's worth $8.",
+ "errors": 2,
+ "words": 4,
+ "wer": 0.5
+ },
+ "semantic": {
+ "normalized_reference": "it is worth $8",
+ "normalized_hypothesis": "it is worth $8",
+ "errors": 0,
+ "words": 4,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0420",
+ "category": "phonetic_balance",
+ "text": "But I did not enjoy it long.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0420.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I did not enjoy it long.",
+ "raw": {
+ "normalized_reference": "but i did not enjoy it long.",
+ "normalized_hypothesis": "i did not enjoy it long.",
+ "errors": 1,
+ "words": 7,
+ "wer": 0.14285714285714285
+ },
+ "semantic": {
+ "normalized_reference": "but i did not enjoy it long",
+ "normalized_hypothesis": "i did not enjoy it long",
+ "errors": 1,
+ "words": 7,
+ "wer": 0.14285714285714285
+ }
+ },
+ {
+ "id": "release500_0421",
+ "category": "phonetic_balance",
+ "text": "Shorty turned to their employers.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0421.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Shorty turned to their employers.",
+ "raw": {
+ "normalized_reference": "shorty turned to their employers.",
+ "normalized_hypothesis": "shorty turned to their employers.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "shorty turned to their employers",
+ "normalized_hypothesis": "shorty turned to their employers",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0422",
+ "category": "phonetic_balance",
+ "text": "It was the third or fourth time that Philip had heard MacDougall swear.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0422.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It was the third or fourth time that Philip had heard Mac Dougal swear.",
+ "raw": {
+ "normalized_reference": "it was the third or fourth time that philip had heard macdougall swear.",
+ "normalized_hypothesis": "it was the third or fourth time that philip had heard mac dougal swear.",
+ "errors": 2,
+ "words": 13,
+ "wer": 0.15384615384615385
+ },
+ "semantic": {
+ "normalized_reference": "it was the 3rd or 4th time that philip had heard macdougall swear",
+ "normalized_hypothesis": "it was the 3rd or 4th time that philip had heard mac dougal swear",
+ "errors": 2,
+ "words": 13,
+ "wer": 0.15384615384615385
+ }
+ },
+ {
+ "id": "release500_0423",
+ "category": "phonetic_balance",
+ "text": "Ernest saw in the affair the most sinister import.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0423.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Ernest saw in the affair the most sinister import.",
+ "raw": {
+ "normalized_reference": "ernest saw in the affair the most sinister import.",
+ "normalized_hypothesis": "ernest saw in the affair the most sinister import.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "ernest saw in the affair the most sinister import",
+ "normalized_hypothesis": "ernest saw in the affair the most sinister import",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0424",
+ "category": "phonetic_balance",
+ "text": "Also, at regular intervals, he would mutter.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0424.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Also, at regular intervals, he would mutter,",
+ "raw": {
+ "normalized_reference": "also, at regular intervals, he would mutter.",
+ "normalized_hypothesis": "also, at regular intervals, he would mutter,",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "also at regular intervals he would mutter",
+ "normalized_hypothesis": "also at regular intervals he would mutter",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0425",
+ "category": "phonetic_balance",
+ "text": "He may anticipate the day of his death.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0425.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He may anticipate the day of his death.",
+ "raw": {
+ "normalized_reference": "he may anticipate the day of his death.",
+ "normalized_hypothesis": "he may anticipate the day of his death.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he may anticipate the day of his death",
+ "normalized_hypothesis": "he may anticipate the day of his death",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0426",
+ "category": "phonetic_balance",
+ "text": "I defy any man to get a Solomon Island sore in California.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0426.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I defy any man to get a Solomon Island soar in California.",
+ "raw": {
+ "normalized_reference": "i defy any man to get a solomon island sore in california.",
+ "normalized_hypothesis": "i defy any man to get a solomon island soar in california.",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "i defy any man to get a solomon island sore in california",
+ "normalized_hypothesis": "i defy any man to get a solomon island soar in california",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ }
+ },
+ {
+ "id": "release500_0427",
+ "category": "phonetic_balance",
+ "text": "Now our figuring was all right.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0427.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Now our figuring was alright.",
+ "raw": {
+ "normalized_reference": "now our figuring was all right.",
+ "normalized_hypothesis": "now our figuring was alright.",
+ "errors": 2,
+ "words": 6,
+ "wer": 0.3333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "now our figuring was all right",
+ "normalized_hypothesis": "now our figuring was alright",
+ "errors": 2,
+ "words": 6,
+ "wer": 0.3333333333333333
+ }
+ },
+ {
+ "id": "release500_0428",
+ "category": "phonetic_balance",
+ "text": "MacDougall tapped his forehead suspiciously with a stubby forefinger.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0428.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "MacDougall tapped his forehead suspiciously with a stubby forefinger.",
+ "raw": {
+ "normalized_reference": "macdougall tapped his forehead suspiciously with a stubby forefinger.",
+ "normalized_hypothesis": "macdougall tapped his forehead suspiciously with a stubby forefinger.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "macdougall tapped his forehead suspiciously with a stubby forefinger",
+ "normalized_hypothesis": "macdougall tapped his forehead suspiciously with a stubby forefinger",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0429",
+ "category": "phonetic_balance",
+ "text": "O'Brien took off his coat and bared his right arm.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0429.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "O'Brien took off his coat and bared his right arm.",
+ "raw": {
+ "normalized_reference": "o'brien took off his coat and bared his right arm.",
+ "normalized_hypothesis": "o'brien took off his coat and bared his right arm.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "0 brien took off his coat and bared his right arm",
+ "normalized_hypothesis": "0 brien took off his coat and bared his right arm",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0430",
+ "category": "phonetic_balance",
+ "text": "But it won't continue, she said with easy confidence.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0430.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "But it won't continue, she said with easy confidence.",
+ "raw": {
+ "normalized_reference": "but it won't continue, she said with easy confidence.",
+ "normalized_hypothesis": "but it won't continue, she said with easy confidence.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "but it will not continue she said with easy confidence",
+ "normalized_hypothesis": "but it will not continue she said with easy confidence",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0431",
+ "category": "phonetic_balance",
+ "text": "I was still weak from my prolonged immersion.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0431.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I was still weak from my prolonged immersion.",
+ "raw": {
+ "normalized_reference": "i was still weak from my prolonged immersion.",
+ "normalized_hypothesis": "i was still weak from my prolonged immersion.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i was still weak from my prolonged immersion",
+ "normalized_hypothesis": "i was still weak from my prolonged immersion",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0432",
+ "category": "phonetic_balance",
+ "text": "Oolong was two hundred and fifty miles from the nearest land.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0432.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "almost 250 miles from the nearest land.",
+ "raw": {
+ "normalized_reference": "oolong was two hundred and fifty miles from the nearest land.",
+ "normalized_hypothesis": "almost 250 miles from the nearest land.",
+ "errors": 6,
+ "words": 11,
+ "wer": 0.5454545454545454
+ },
+ "semantic": {
+ "normalized_reference": "oolong was 250 miles from the nearest land",
+ "normalized_hypothesis": "almost 250 miles from the nearest land",
+ "errors": 2,
+ "words": 8,
+ "wer": 0.25
+ }
+ },
+ {
+ "id": "release500_0433",
+ "category": "phonetic_balance",
+ "text": "I just do appreciate it without being able to express my feelings.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0433.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I just do appreciate it without being able to express my feelings.",
+ "raw": {
+ "normalized_reference": "i just do appreciate it without being able to express my feelings.",
+ "normalized_hypothesis": "i just do appreciate it without being able to express my feelings.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i just do appreciate it without being able to express my feelings",
+ "normalized_hypothesis": "i just do appreciate it without being able to express my feelings",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0434",
+ "category": "phonetic_balance",
+ "text": "Now just what do you want to know.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0434.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Now, just what do you want to know?",
+ "raw": {
+ "normalized_reference": "now just what do you want to know.",
+ "normalized_hypothesis": "now, just what do you want to know?",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "now just what do you want to know",
+ "normalized_hypothesis": "now just what do you want to know",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0435",
+ "category": "phonetic_balance",
+ "text": "But she swung obediently on her heel into the wind.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0435.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "but she's flung obediently on her heel into the wind.",
+ "raw": {
+ "normalized_reference": "but she swung obediently on her heel into the wind.",
+ "normalized_hypothesis": "but she's flung obediently on her heel into the wind.",
+ "errors": 2,
+ "words": 10,
+ "wer": 0.2
+ },
+ "semantic": {
+ "normalized_reference": "but she swung obediently on her heel into the wind",
+ "normalized_hypothesis": "but she is flung obediently on her heel into the wind",
+ "errors": 2,
+ "words": 10,
+ "wer": 0.2
+ }
+ },
+ {
+ "id": "release500_0436",
+ "category": "phonetic_balance",
+ "text": "Ask him, I laughed, then turned to Pasquini.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0436.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Ask him, I laughed, then turned to Pasquini.",
+ "raw": {
+ "normalized_reference": "ask him, i laughed, then turned to pasquini.",
+ "normalized_hypothesis": "ask him, i laughed, then turned to pasquini.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "ask him i laughed then turned to pasquini",
+ "normalized_hypothesis": "ask him i laughed then turned to pasquini",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0437",
+ "category": "phonetic_balance",
+ "text": "He did not rush in.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0437.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He did not rush in.",
+ "raw": {
+ "normalized_reference": "he did not rush in.",
+ "normalized_hypothesis": "he did not rush in.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he did not rush in",
+ "normalized_hypothesis": "he did not rush in",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0438",
+ "category": "phonetic_balance",
+ "text": "They obeyed him, and went here and there at his commands.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0438.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "They obeyed him and went here and there at his commands.",
+ "raw": {
+ "normalized_reference": "they obeyed him, and went here and there at his commands.",
+ "normalized_hypothesis": "they obeyed him and went here and there at his commands.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "they obeyed him and went here and there at his commands",
+ "normalized_hypothesis": "they obeyed him and went here and there at his commands",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0439",
+ "category": "phonetic_balance",
+ "text": "Tom Spink has a harpoon.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0439.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Tom Spank has a harpoon.",
+ "raw": {
+ "normalized_reference": "tom spink has a harpoon.",
+ "normalized_hypothesis": "tom spank has a harpoon.",
+ "errors": 1,
+ "words": 5,
+ "wer": 0.2
+ },
+ "semantic": {
+ "normalized_reference": "tom spink has a harpoon",
+ "normalized_hypothesis": "tom spank has a harpoon",
+ "errors": 1,
+ "words": 5,
+ "wer": 0.2
+ }
+ },
+ {
+ "id": "release500_0440",
+ "category": "phonetic_balance",
+ "text": "It was my reports from the north which chiefly induced people to buy.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0440.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It was my reports from the north which chiefly induced people to buy.",
+ "raw": {
+ "normalized_reference": "it was my reports from the north which chiefly induced people to buy.",
+ "normalized_hypothesis": "it was my reports from the north which chiefly induced people to buy.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it was my reports from the north which chiefly induced people to buy",
+ "normalized_hypothesis": "it was my reports from the north which chiefly induced people to buy",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0441",
+ "category": "phonetic_balance",
+ "text": "Cherokee identified himself with his instinct.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0441.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Cherokee identified himself with his instincts.",
+ "raw": {
+ "normalized_reference": "cherokee identified himself with his instinct.",
+ "normalized_hypothesis": "cherokee identified himself with his instincts.",
+ "errors": 1,
+ "words": 6,
+ "wer": 0.16666666666666666
+ },
+ "semantic": {
+ "normalized_reference": "cherokee identified himself with his instinct",
+ "normalized_hypothesis": "cherokee identified himself with his instincts",
+ "errors": 1,
+ "words": 6,
+ "wer": 0.16666666666666666
+ }
+ },
+ {
+ "id": "release500_0442",
+ "category": "phonetic_balance",
+ "text": "I graduated last of my class.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0442.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I graduated last of my class.",
+ "raw": {
+ "normalized_reference": "i graduated last of my class.",
+ "normalized_hypothesis": "i graduated last of my class.",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i graduated last of my class",
+ "normalized_hypothesis": "i graduated last of my class",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0443",
+ "category": "phonetic_balance",
+ "text": "It was a gigantic inadequacy.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0443.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It was a gigantic inadequacy.",
+ "raw": {
+ "normalized_reference": "it was a gigantic inadequacy.",
+ "normalized_hypothesis": "it was a gigantic inadequacy.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it was a gigantic inadequacy",
+ "normalized_hypothesis": "it was a gigantic inadequacy",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0444",
+ "category": "phonetic_balance",
+ "text": "Did I possess too much vitality.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0444.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Did I possess too much vitality?",
+ "raw": {
+ "normalized_reference": "did i possess too much vitality.",
+ "normalized_hypothesis": "did i possess too much vitality?",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "did i possess too much vitality",
+ "normalized_hypothesis": "did i possess too much vitality",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0445",
+ "category": "phonetic_balance",
+ "text": "Therefore, hurrah for the game.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0445.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Therefore, hurrah for the game.",
+ "raw": {
+ "normalized_reference": "therefore, hurrah for the game.",
+ "normalized_hypothesis": "therefore, hurrah for the game.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "therefore hurrah for the game",
+ "normalized_hypothesis": "therefore hurrah for the game",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0446",
+ "category": "phonetic_balance",
+ "text": "He had fulfilled his duty and paid properly.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0446.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He had fulfilled his duty and paid properly.",
+ "raw": {
+ "normalized_reference": "he had fulfilled his duty and paid properly.",
+ "normalized_hypothesis": "he had fulfilled his duty and paid properly.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he had fulfilled his duty and paid properly",
+ "normalized_hypothesis": "he had fulfilled his duty and paid properly",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0447",
+ "category": "phonetic_balance",
+ "text": "How can you manage all alone, Mr Young.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0447.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "How can you manage all alone, Mr. Young?",
+ "raw": {
+ "normalized_reference": "how can you manage all alone, mr young.",
+ "normalized_hypothesis": "how can you manage all alone, mr. young?",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "how can you manage all alone mister young",
+ "normalized_hypothesis": "how can you manage all alone mister young",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0448",
+ "category": "phonetic_balance",
+ "text": "Red-Eye never committed a more outrageous deed.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0448.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Red Eye never committed a more outrageous deed.",
+ "raw": {
+ "normalized_reference": "red-eye never committed a more outrageous deed.",
+ "normalized_hypothesis": "red eye never committed a more outrageous deed.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "red eye never committed a more outrageous deed",
+ "normalized_hypothesis": "red eye never committed a more outrageous deed",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0449",
+ "category": "phonetic_balance",
+ "text": "He would keep his faith with Gregson for the promised day or two.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0449.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He would keep his faith with Gregson for the promised day or two.",
+ "raw": {
+ "normalized_reference": "he would keep his faith with gregson for the promised day or two.",
+ "normalized_hypothesis": "he would keep his faith with gregson for the promised day or two.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he would keep his faith with gregson for the promised day or 2",
+ "normalized_hypothesis": "he would keep his faith with gregson for the promised day or 2",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0450",
+ "category": "phonetic_balance",
+ "text": "There was one chance, and only one, of saving Jeanne.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0450.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "There was one chance and only one of saving Jean.",
+ "raw": {
+ "normalized_reference": "there was one chance, and only one, of saving jeanne.",
+ "normalized_hypothesis": "there was one chance and only one of saving jean.",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ },
+ "semantic": {
+ "normalized_reference": "there was one chance and only one of saving jeanne",
+ "normalized_hypothesis": "there was one chance and only one of saving jean",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ }
+ },
+ {
+ "id": "release500_0451",
+ "category": "phonetic_balance",
+ "text": "I was beneath the water, suffocating and drowning.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0451.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I was beneath the water, suffocating and drowning.",
+ "raw": {
+ "normalized_reference": "i was beneath the water, suffocating and drowning.",
+ "normalized_hypothesis": "i was beneath the water, suffocating and drowning.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i was beneath the water suffocating and drowning",
+ "normalized_hypothesis": "i was beneath the water suffocating and drowning",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0452",
+ "category": "phonetic_balance",
+ "text": "Give them their choice between a fine or an official whipping.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0452.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Give them their choice between a fine or an official whipping.",
+ "raw": {
+ "normalized_reference": "give them their choice between a fine or an official whipping.",
+ "normalized_hypothesis": "give them their choice between a fine or an official whipping.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "give them their choice between a fine or an official whipping",
+ "normalized_hypothesis": "give them their choice between a fine or an official whipping",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0453",
+ "category": "phonetic_balance",
+ "text": "They must have been swept away by the chaotic currents.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0453.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "They must have been swept away by the chaotic currents.",
+ "raw": {
+ "normalized_reference": "they must have been swept away by the chaotic currents.",
+ "normalized_hypothesis": "they must have been swept away by the chaotic currents.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "they must have been swept away by the chaotic currents",
+ "normalized_hypothesis": "they must have been swept away by the chaotic currents",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0454",
+ "category": "phonetic_balance",
+ "text": "You yellow giant thing of the frost.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0454.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "You yellow giant thing of the frost.",
+ "raw": {
+ "normalized_reference": "you yellow giant thing of the frost.",
+ "normalized_hypothesis": "you yellow giant thing of the frost.",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "you yellow giant thing of the frost",
+ "normalized_hypothesis": "you yellow giant thing of the frost",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0455",
+ "category": "phonetic_balance",
+ "text": "He had peeled off his shirt and was wildly waving it.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0455.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He had peeled off his shirt and was wildly waving it.",
+ "raw": {
+ "normalized_reference": "he had peeled off his shirt and was wildly waving it.",
+ "normalized_hypothesis": "he had peeled off his shirt and was wildly waving it.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he had peeled off his shirt and was wildly waving it",
+ "normalized_hypothesis": "he had peeled off his shirt and was wildly waving it",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0456",
+ "category": "phonetic_balance",
+ "text": "To these he gave castor oil.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0456.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "To these he gave castor oil.",
+ "raw": {
+ "normalized_reference": "to these he gave castor oil.",
+ "normalized_hypothesis": "to these he gave castor oil.",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "to these he gave castor oil",
+ "normalized_hypothesis": "to these he gave castor oil",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0457",
+ "category": "phonetic_balance",
+ "text": "Tudor surveyed him with withering disgust.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0457.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Tudor surveyed him with withering disgust.",
+ "raw": {
+ "normalized_reference": "tudor surveyed him with withering disgust.",
+ "normalized_hypothesis": "tudor surveyed him with withering disgust.",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "tudor surveyed him with withering disgust",
+ "normalized_hypothesis": "tudor surveyed him with withering disgust",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0458",
+ "category": "phonetic_balance",
+ "text": "After all, it was simply a mistake in judgment.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0458.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "After all, it was simply a mistake in judgment.",
+ "raw": {
+ "normalized_reference": "after all, it was simply a mistake in judgment.",
+ "normalized_hypothesis": "after all, it was simply a mistake in judgment.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "after all it was simply a mistake in judgment",
+ "normalized_hypothesis": "after all it was simply a mistake in judgment",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0459",
+ "category": "phonetic_balance",
+ "text": "That is the strange part of it.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0459.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "That is the strange part of it.",
+ "raw": {
+ "normalized_reference": "that is the strange part of it.",
+ "normalized_hypothesis": "that is the strange part of it.",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "that is the strange part of it",
+ "normalized_hypothesis": "that is the strange part of it",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0460",
+ "category": "phonetic_balance",
+ "text": "Not a wheel moved in his empire.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0460.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Not a wheel moved in his empire.",
+ "raw": {
+ "normalized_reference": "not a wheel moved in his empire.",
+ "normalized_hypothesis": "not a wheel moved in his empire.",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "not a wheel moved in his empire",
+ "normalized_hypothesis": "not a wheel moved in his empire",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0461",
+ "category": "phonetic_balance",
+ "text": "Its diameter was not more than two hundred yards.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0461.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Its diameter was not more than 200 yards.",
+ "raw": {
+ "normalized_reference": "its diameter was not more than two hundred yards.",
+ "normalized_hypothesis": "its diameter was not more than 200 yards.",
+ "errors": 2,
+ "words": 9,
+ "wer": 0.2222222222222222
+ },
+ "semantic": {
+ "normalized_reference": "its diameter was not more than 200 yards",
+ "normalized_hypothesis": "its diameter was not more than 200 yards",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0462",
+ "category": "phonetic_balance",
+ "text": "Nimrod replied, with a slight manifestation of sensitiveness.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0462.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Nimrod replied with a slight manifestation of sensitiveness.",
+ "raw": {
+ "normalized_reference": "nimrod replied, with a slight manifestation of sensitiveness.",
+ "normalized_hypothesis": "nimrod replied with a slight manifestation of sensitiveness.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "nimrod replied with a slight manifestation of sensitiveness",
+ "normalized_hypothesis": "nimrod replied with a slight manifestation of sensitiveness",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0463",
+ "category": "phonetic_balance",
+ "text": "I tried to read George Moore last night, and was dreadfully bored.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0463.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I tried to read George Moore last night and was dreadfully bored.",
+ "raw": {
+ "normalized_reference": "i tried to read george moore last night, and was dreadfully bored.",
+ "normalized_hypothesis": "i tried to read george moore last night and was dreadfully bored.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i tried to read george moore last night and was dreadfully bored",
+ "normalized_hypothesis": "i tried to read george moore last night and was dreadfully bored",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0464",
+ "category": "phonetic_balance",
+ "text": "Yes, it was a man who asked, a stranger.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0464.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Yes, it was a man who asked a stranger.",
+ "raw": {
+ "normalized_reference": "yes, it was a man who asked, a stranger.",
+ "normalized_hypothesis": "yes, it was a man who asked a stranger.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "yes it was a man who asked a stranger",
+ "normalized_hypothesis": "yes it was a man who asked a stranger",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0465",
+ "category": "phonetic_balance",
+ "text": "We were now good friends.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0465.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "We were now good friends.",
+ "raw": {
+ "normalized_reference": "we were now good friends.",
+ "normalized_hypothesis": "we were now good friends.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "we were now good friends",
+ "normalized_hypothesis": "we were now good friends",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0466",
+ "category": "phonetic_balance",
+ "text": "Yet, in accordance with Ernest's test of truth, it worked.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0466.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Yes, in accordance with Ernest's test of truth, it worked.",
+ "raw": {
+ "normalized_reference": "yet, in accordance with ernest's test of truth, it worked.",
+ "normalized_hypothesis": "yes, in accordance with ernest's test of truth, it worked.",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ },
+ "semantic": {
+ "normalized_reference": "yet in accordance with ernest is test of truth it worked",
+ "normalized_hypothesis": "yes in accordance with ernest is test of truth it worked",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ }
+ },
+ {
+ "id": "release500_0467",
+ "category": "phonetic_balance",
+ "text": "It's the strap hangers that'll keep us from going under.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0467.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It's the strap hangers that'll keep us from going under.",
+ "raw": {
+ "normalized_reference": "it's the strap hangers that'll keep us from going under.",
+ "normalized_hypothesis": "it's the strap hangers that'll keep us from going under.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it is the strap hangers that will keep us from going under",
+ "normalized_hypothesis": "it is the strap hangers that will keep us from going under",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0468",
+ "category": "phonetic_balance",
+ "text": "I never allow what can't be changed to annoy me.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0468.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I never allow what can't be changed to annoy me.",
+ "raw": {
+ "normalized_reference": "i never allow what can't be changed to annoy me.",
+ "normalized_hypothesis": "i never allow what can't be changed to annoy me.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i never allow what can not be changed to annoy me",
+ "normalized_hypothesis": "i never allow what can not be changed to annoy me",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0469",
+ "category": "phonetic_balance",
+ "text": "After all superfluous flesh is gone what is left is stringy and resistant.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0469.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "After all superfluous flesh is gone what is left is stringy and resistant.",
+ "raw": {
+ "normalized_reference": "after all superfluous flesh is gone what is left is stringy and resistant.",
+ "normalized_hypothesis": "after all superfluous flesh is gone what is left is stringy and resistant.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "after all superfluous flesh is gone what is left is stringy and resistant",
+ "normalized_hypothesis": "after all superfluous flesh is gone what is left is stringy and resistant",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0470",
+ "category": "phonetic_balance",
+ "text": "Then he shouted, Shut up.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0470.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Then he shouted, shut up.",
+ "raw": {
+ "normalized_reference": "then he shouted, shut up.",
+ "normalized_hypothesis": "then he shouted, shut up.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "then he shouted shut up",
+ "normalized_hypothesis": "then he shouted shut up",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0471",
+ "category": "phonetic_balance",
+ "text": "He had worshiped her, as Dante might have worshiped Beatrice.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0471.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He had worshipped her, as Dante might have worshipped Beatrice.",
+ "raw": {
+ "normalized_reference": "he had worshiped her, as dante might have worshiped beatrice.",
+ "normalized_hypothesis": "he had worshipped her, as dante might have worshipped beatrice.",
+ "errors": 2,
+ "words": 10,
+ "wer": 0.2
+ },
+ "semantic": {
+ "normalized_reference": "he had worshiped her as dante might have worshiped beatrice",
+ "normalized_hypothesis": "he had worshipped her as dante might have worshipped beatrice",
+ "errors": 2,
+ "words": 10,
+ "wer": 0.2
+ }
+ },
+ {
+ "id": "release500_0472",
+ "category": "phonetic_balance",
+ "text": "He looked like one who had passed through an uncomfortable hour or two.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0472.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He looked like one who had passed through an uncomfortable hour or two.",
+ "raw": {
+ "normalized_reference": "he looked like one who had passed through an uncomfortable hour or two.",
+ "normalized_hypothesis": "he looked like one who had passed through an uncomfortable hour or two.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he looked like one who had passed through an uncomfortable hour or 2",
+ "normalized_hypothesis": "he looked like one who had passed through an uncomfortable hour or 2",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0473",
+ "category": "phonetic_balance",
+ "text": "The history of our westward-faring race is written in it.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0473.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The history of our westward-faring race is written in it.",
+ "raw": {
+ "normalized_reference": "the history of our westward-faring race is written in it.",
+ "normalized_hypothesis": "the history of our westward-faring race is written in it.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the history of our westward faring race is written in it",
+ "normalized_hypothesis": "the history of our westward faring race is written in it",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0474",
+ "category": "phonetic_balance",
+ "text": "He may desire to escape pain, or to enjoy pleasure.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0474.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He may desire to escape pain or to enjoy pleasure.",
+ "raw": {
+ "normalized_reference": "he may desire to escape pain, or to enjoy pleasure.",
+ "normalized_hypothesis": "he may desire to escape pain or to enjoy pleasure.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he may desire to escape pain or to enjoy pleasure",
+ "normalized_hypothesis": "he may desire to escape pain or to enjoy pleasure",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0475",
+ "category": "phonetic_balance",
+ "text": "And the air was growing chilly.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0475.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "and the air was growing chilly.",
+ "raw": {
+ "normalized_reference": "and the air was growing chilly.",
+ "normalized_hypothesis": "and the air was growing chilly.",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "and the air was growing chilly",
+ "normalized_hypothesis": "and the air was growing chilly",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0476",
+ "category": "phonetic_balance",
+ "text": "Dig in; you're sure good, was Daylight's answer.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0476.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Dig in, you're sure good with Staylight's answer.",
+ "raw": {
+ "normalized_reference": "dig in; you're sure good, was daylight's answer.",
+ "normalized_hypothesis": "dig in, you're sure good with staylight's answer.",
+ "errors": 2,
+ "words": 8,
+ "wer": 0.25
+ },
+ "semantic": {
+ "normalized_reference": "dig in you are sure good was daylight is answer",
+ "normalized_hypothesis": "dig in you are sure good with staylight is answer",
+ "errors": 2,
+ "words": 10,
+ "wer": 0.2
+ }
+ },
+ {
+ "id": "release500_0477",
+ "category": "phonetic_balance",
+ "text": "My idea was that he would have more influence over the natives.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0477.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "My idea was that he would have more influence over the natives.",
+ "raw": {
+ "normalized_reference": "my idea was that he would have more influence over the natives.",
+ "normalized_hypothesis": "my idea was that he would have more influence over the natives.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "my idea was that he would have more influence over the natives",
+ "normalized_hypothesis": "my idea was that he would have more influence over the natives",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0478",
+ "category": "phonetic_balance",
+ "text": "Now run along, and tell them to hurry.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0478.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Now run along and tell them to hurry.",
+ "raw": {
+ "normalized_reference": "now run along, and tell them to hurry.",
+ "normalized_hypothesis": "now run along and tell them to hurry.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "now run along and tell them to hurry",
+ "normalized_hypothesis": "now run along and tell them to hurry",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0479",
+ "category": "phonetic_balance",
+ "text": "I have no idea, replied Philip.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0479.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I have no idea, replied Philip.",
+ "raw": {
+ "normalized_reference": "i have no idea, replied philip.",
+ "normalized_hypothesis": "i have no idea, replied philip.",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i have no idea replied philip",
+ "normalized_hypothesis": "i have no idea replied philip",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0480",
+ "category": "phonetic_balance",
+ "text": "For the rest, he was a mere automaton.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0480.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "For the rest, he was a mere automaton.",
+ "raw": {
+ "normalized_reference": "for the rest, he was a mere automaton.",
+ "normalized_hypothesis": "for the rest, he was a mere automaton.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "for the rest he was a mere automaton",
+ "normalized_hypothesis": "for the rest he was a mere automaton",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0481",
+ "category": "phonetic_balance",
+ "text": "Why, the average review is more nauseating than cod liver oil.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0481.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Why the average review is more nauseating than cod liver oil.",
+ "raw": {
+ "normalized_reference": "why, the average review is more nauseating than cod liver oil.",
+ "normalized_hypothesis": "why the average review is more nauseating than cod liver oil.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "why the average review is more nauseating than cod liver oil",
+ "normalized_hypothesis": "why the average review is more nauseating than cod liver oil",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0482",
+ "category": "phonetic_balance",
+ "text": "I have been robbed, sir, I amended.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0482.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I have been robbed, sir, I am ended.",
+ "raw": {
+ "normalized_reference": "i have been robbed, sir, i amended.",
+ "normalized_hypothesis": "i have been robbed, sir, i am ended.",
+ "errors": 2,
+ "words": 7,
+ "wer": 0.2857142857142857
+ },
+ "semantic": {
+ "normalized_reference": "i have been robbed sir i amended",
+ "normalized_hypothesis": "i have been robbed sir i am ended",
+ "errors": 2,
+ "words": 7,
+ "wer": 0.2857142857142857
+ }
+ },
+ {
+ "id": "release500_0483",
+ "category": "phonetic_balance",
+ "text": "He moved away as quietly as he had come.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0483.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He moved away as quietly as he had come.",
+ "raw": {
+ "normalized_reference": "he moved away as quietly as he had come.",
+ "normalized_hypothesis": "he moved away as quietly as he had come.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he moved away as quietly as he had come",
+ "normalized_hypothesis": "he moved away as quietly as he had come",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0484",
+ "category": "phonetic_balance",
+ "text": "A trickle of fresh blood ran over his face.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0484.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "A trickle of fresh blood ran over his face.",
+ "raw": {
+ "normalized_reference": "a trickle of fresh blood ran over his face.",
+ "normalized_hypothesis": "a trickle of fresh blood ran over his face.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "a trickle of fresh blood ran over his face",
+ "normalized_hypothesis": "a trickle of fresh blood ran over his face",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0485",
+ "category": "phonetic_balance",
+ "text": "Suppose you saw me at work through the window.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0485.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Suppose you saw me at work through the window.",
+ "raw": {
+ "normalized_reference": "suppose you saw me at work through the window.",
+ "normalized_hypothesis": "suppose you saw me at work through the window.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "suppose you saw me at work through the window",
+ "normalized_hypothesis": "suppose you saw me at work through the window",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0486",
+ "category": "phonetic_balance",
+ "text": "But who was Eileen's double.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0486.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "But who is Eilean Stubble?",
+ "raw": {
+ "normalized_reference": "but who was eileen's double.",
+ "normalized_hypothesis": "but who is eilean stubble?",
+ "errors": 3,
+ "words": 5,
+ "wer": 0.6
+ },
+ "semantic": {
+ "normalized_reference": "but who was eileen is double",
+ "normalized_hypothesis": "but who is eilean stubble",
+ "errors": 4,
+ "words": 6,
+ "wer": 0.6666666666666666
+ }
+ },
+ {
+ "id": "release500_0487",
+ "category": "phonetic_balance",
+ "text": "Darkness hid him from Jeanne.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0487.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Darkus hid him from Jean.",
+ "raw": {
+ "normalized_reference": "darkness hid him from jeanne.",
+ "normalized_hypothesis": "darkus hid him from jean.",
+ "errors": 2,
+ "words": 5,
+ "wer": 0.4
+ },
+ "semantic": {
+ "normalized_reference": "darkness hid him from jeanne",
+ "normalized_hypothesis": "darkus hid him from jean",
+ "errors": 2,
+ "words": 5,
+ "wer": 0.4
+ }
+ },
+ {
+ "id": "release500_0488",
+ "category": "phonetic_balance",
+ "text": "By this time Charley was as enraged as the Greek.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0488.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "At his time Charlie was as enraged as the Greek.",
+ "raw": {
+ "normalized_reference": "by this time charley was as enraged as the greek.",
+ "normalized_hypothesis": "at his time charlie was as enraged as the greek.",
+ "errors": 3,
+ "words": 10,
+ "wer": 0.3
+ },
+ "semantic": {
+ "normalized_reference": "by this time charley was as enraged as the greek",
+ "normalized_hypothesis": "at his time charlie was as enraged as the greek",
+ "errors": 3,
+ "words": 10,
+ "wer": 0.3
+ }
+ },
+ {
+ "id": "release500_0489",
+ "category": "phonetic_balance",
+ "text": "I am writing these lines in Honolulu, Hawaii.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0489.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I am writing these lines in Honolulu, Hawaii.",
+ "raw": {
+ "normalized_reference": "i am writing these lines in honolulu, hawaii.",
+ "normalized_hypothesis": "i am writing these lines in honolulu, hawaii.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i am writing these lines in honolulu hawaii",
+ "normalized_hypothesis": "i am writing these lines in honolulu hawaii",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0490",
+ "category": "phonetic_balance",
+ "text": "Nobody knows how the natives got them.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0490.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Nobody knows how the natives got them.",
+ "raw": {
+ "normalized_reference": "nobody knows how the natives got them.",
+ "normalized_hypothesis": "nobody knows how the natives got them.",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "nobody knows how the natives got them",
+ "normalized_hypothesis": "nobody knows how the natives got them",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0491",
+ "category": "phonetic_balance",
+ "text": "And how would we ever find ourselves.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0491.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "And how would we ever find ourselves?",
+ "raw": {
+ "normalized_reference": "and how would we ever find ourselves.",
+ "normalized_hypothesis": "and how would we ever find ourselves?",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "and how would we ever find ourselves",
+ "normalized_hypothesis": "and how would we ever find ourselves",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0492",
+ "category": "phonetic_balance",
+ "text": "They are coming ashore, whoever they are.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0492.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "They are coming ashore, whoever they are.",
+ "raw": {
+ "normalized_reference": "they are coming ashore, whoever they are.",
+ "normalized_hypothesis": "they are coming ashore, whoever they are.",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "they are coming ashore whoever they are",
+ "normalized_hypothesis": "they are coming ashore whoever they are",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0493",
+ "category": "phonetic_balance",
+ "text": "This is my fifth voyage.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0493.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "This is my fifth voyage.",
+ "raw": {
+ "normalized_reference": "this is my fifth voyage.",
+ "normalized_hypothesis": "this is my fifth voyage.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "this is my 5th voyage",
+ "normalized_hypothesis": "this is my 5th voyage",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0494",
+ "category": "phonetic_balance",
+ "text": "And yet if she came he had no words to say.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0494.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "And yet if she came he had no words to say.",
+ "raw": {
+ "normalized_reference": "and yet if she came he had no words to say.",
+ "normalized_hypothesis": "and yet if she came he had no words to say.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "and yet if she came he had no words to say",
+ "normalized_hypothesis": "and yet if she came he had no words to say",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0495",
+ "category": "phonetic_balance",
+ "text": "You must sleep, he urged.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0495.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "You must sleep, he urged.",
+ "raw": {
+ "normalized_reference": "you must sleep, he urged.",
+ "normalized_hypothesis": "you must sleep, he urged.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "you must sleep he urged",
+ "normalized_hypothesis": "you must sleep he urged",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0496",
+ "category": "phonetic_balance",
+ "text": "Could the incident have anything to do with Jeanne and Pierre.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0496.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Could the incident have anything to do with Jean and Pierre?",
+ "raw": {
+ "normalized_reference": "could the incident have anything to do with jeanne and pierre.",
+ "normalized_hypothesis": "could the incident have anything to do with jean and pierre?",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ },
+ "semantic": {
+ "normalized_reference": "could the incident have anything to do with jeanne and pierre",
+ "normalized_hypothesis": "could the incident have anything to do with jean and pierre",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ }
+ },
+ {
+ "id": "release500_0497",
+ "category": "phonetic_balance",
+ "text": "Each day she became a more vital part of him.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0497.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Each day she became a more vital part of him.",
+ "raw": {
+ "normalized_reference": "each day she became a more vital part of him.",
+ "normalized_hypothesis": "each day she became a more vital part of him.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "each day she became a more vital part of him",
+ "normalized_hypothesis": "each day she became a more vital part of him",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0498",
+ "category": "phonetic_balance",
+ "text": "But why continue the tirade, for tirade it was.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0498.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "But why continue the tirade? For tirade it was.",
+ "raw": {
+ "normalized_reference": "but why continue the tirade, for tirade it was.",
+ "normalized_hypothesis": "but why continue the tirade? for tirade it was.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "but why continue the tirade for tirade it was",
+ "normalized_hypothesis": "but why continue the tirade for tirade it was",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0499",
+ "category": "phonetic_balance",
+ "text": "That is what distinguishes all of us from the lower animals.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0499.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "That is what distinguishes all of us from the lower animals.",
+ "raw": {
+ "normalized_reference": "that is what distinguishes all of us from the lower animals.",
+ "normalized_hypothesis": "that is what distinguishes all of us from the lower animals.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "that is what distinguishes all of us from the lower animals",
+ "normalized_hypothesis": "that is what distinguishes all of us from the lower animals",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0500",
+ "category": "phonetic_balance",
+ "text": "There was something pathetic in the girl's attitude now.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0500.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "There was something pathetic in the girl's attitude now.",
+ "raw": {
+ "normalized_reference": "there was something pathetic in the girl's attitude now.",
+ "normalized_hypothesis": "there was something pathetic in the girl's attitude now.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "there was something pathetic in the girl is attitude now",
+ "normalized_hypothesis": "there was something pathetic in the girl is attitude now",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0001",
+ "category": "conversational",
+ "text": "It wasn't until later that I realized what had actually happened.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0001.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It wasn't until later that I realized what had actually happened.",
+ "raw": {
+ "normalized_reference": "it wasn't until later that i realized what had actually happened.",
+ "normalized_hypothesis": "it wasn't until later that i realized what had actually happened.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it was not until later that i realized what had actually happened",
+ "normalized_hypothesis": "it was not until later that i realized what had actually happened",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0002",
+ "category": "descriptive",
+ "text": "The light at that time of afternoon is perfect for about twenty minutes.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0002.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The light at that time of afternoon is perfect for about 20 minutes.",
+ "raw": {
+ "normalized_reference": "the light at that time of afternoon is perfect for about twenty minutes.",
+ "normalized_hypothesis": "the light at that time of afternoon is perfect for about 20 minutes.",
+ "errors": 1,
+ "words": 13,
+ "wer": 0.07692307692307693
+ },
+ "semantic": {
+ "normalized_reference": "the light at that time of afternoon is perfect for about 20 minutes",
+ "normalized_hypothesis": "the light at that time of afternoon is perfect for about 20 minutes",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0003",
+ "category": "emotional",
+ "text": "That was the scariest thing I've ever done, and also maybe the best.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0003.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "That was the scariest thing I've ever done, and also maybe the best.",
+ "raw": {
+ "normalized_reference": "that was the scariest thing i've ever done, and also maybe the best.",
+ "normalized_hypothesis": "that was the scariest thing i've ever done, and also maybe the best.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "that was the scariest thing i have ever done and also maybe the best",
+ "normalized_hypothesis": "that was the scariest thing i have ever done and also maybe the best",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0004",
+ "category": "homographs",
+ "text": "After I record the record, please present the present to the bass player.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0004.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "After I record the record, please present the present to the bass player.",
+ "raw": {
+ "normalized_reference": "after i record the record, please present the present to the bass player.",
+ "normalized_hypothesis": "after i record the record, please present the present to the bass player.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "after i record the record please present the present to the bass player",
+ "normalized_hypothesis": "after i record the record please present the present to the bass player",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0005",
+ "category": "long",
+ "text": "If I'm being completely honest, I had no idea what I was doing for most of it, and I think that might be the most normal thing about the whole experience.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0005.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "If I'm being completely honest, I had no idea what I was doing for most of it, and I think that might be the most normal thing about the whole experience.",
+ "raw": {
+ "normalized_reference": "if i'm being completely honest, i had no idea what i was doing for most of it, and i think that might be the most normal thing about the whole experience.",
+ "normalized_hypothesis": "if i'm being completely honest, i had no idea what i was doing for most of it, and i think that might be the most normal thing about the whole experience.",
+ "errors": 0,
+ "words": 31,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "if i am being completely honest i had no idea what i was doing for most of it and i think that might be the most normal thing about the whole experience",
+ "normalized_hypothesis": "if i am being completely honest i had no idea what i was doing for most of it and i think that might be the most normal thing about the whole experience",
+ "errors": 0,
+ "words": 32,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0006",
+ "category": "long_range",
+ "text": "The committee postponed its decision after discovering that three pages of the report had been printed in the wrong order.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0006.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The committee postponed its decision after discovering that three pages of the report had been printed in the wrong order.",
+ "raw": {
+ "normalized_reference": "the committee postponed its decision after discovering that three pages of the report had been printed in the wrong order.",
+ "normalized_hypothesis": "the committee postponed its decision after discovering that three pages of the report had been printed in the wrong order.",
+ "errors": 0,
+ "words": 20,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the committee postponed its decision after discovering that 3 pages of the report had been printed in the wrong order",
+ "normalized_hypothesis": "the committee postponed its decision after discovering that 3 pages of the report had been printed in the wrong order",
+ "errors": 0,
+ "words": 20,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0007",
+ "category": "names_places",
+ "text": "Gwendolyn photographed the eucalyptus trees outside Ljubljana.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0007.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Wendelin photographed the eucalyptus trees outside Ljubljana.",
+ "raw": {
+ "normalized_reference": "gwendolyn photographed the eucalyptus trees outside ljubljana.",
+ "normalized_hypothesis": "wendelin photographed the eucalyptus trees outside ljubljana.",
+ "errors": 1,
+ "words": 7,
+ "wer": 0.14285714285714285
+ },
+ "semantic": {
+ "normalized_reference": "gwendolyn photographed the eucalyptus trees outside ljubljana",
+ "normalized_hypothesis": "wendelin photographed the eucalyptus trees outside ljubljana",
+ "errors": 1,
+ "words": 7,
+ "wer": 0.14285714285714285
+ }
+ },
+ {
+ "id": "release500_0008",
+ "category": "narrative",
+ "text": "He had this theory, and the more time passes, the more I think he was right.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0008.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He had this theory. And the more time passes, the more I think he was right.",
+ "raw": {
+ "normalized_reference": "he had this theory, and the more time passes, the more i think he was right.",
+ "normalized_hypothesis": "he had this theory. and the more time passes, the more i think he was right.",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he had this theory and the more time passes the more i think he was right",
+ "normalized_hypothesis": "he had this theory and the more time passes the more i think he was right",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0009",
+ "category": "numbers",
+ "text": "The package weighs twelve point six kilograms and arrived on July twenty-first.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0009.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The package weighs 12.6 kilograms and arrived on July 21.",
+ "raw": {
+ "normalized_reference": "the package weighs twelve point six kilograms and arrived on july twenty-first.",
+ "normalized_hypothesis": "the package weighs 12.6 kilograms and arrived on july 21.",
+ "errors": 5,
+ "words": 13,
+ "wer": 0.38461538461538464
+ },
+ "semantic": {
+ "normalized_reference": "the package weighs 12.6 kilograms and arrived on july 21st",
+ "normalized_hypothesis": "the package weighs 12.6 kilograms and arrived on july 21",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ }
+ },
+ {
+ "id": "release500_0010",
+ "category": "punctuation",
+ "text": "First, close the window; second, turn off the lamp; finally, lock the door.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0010.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "First, close the window. Second, turn off the lamp. Finally, lock the door.",
+ "raw": {
+ "normalized_reference": "first, close the window; second, turn off the lamp; finally, lock the door.",
+ "normalized_hypothesis": "first, close the window. second, turn off the lamp. finally, lock the door.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "1st close the window 2nd turn off the lamp finally lock the door",
+ "normalized_hypothesis": "1st close the window 2nd turn off the lamp finally lock the door",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0011",
+ "category": "question",
+ "text": "When did you first realize something was wrong?",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0011.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "When did you first realize something was wrong?",
+ "raw": {
+ "normalized_reference": "when did you first realize something was wrong?",
+ "normalized_hypothesis": "when did you first realize something was wrong?",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "when did you 1st realize something was wrong",
+ "normalized_hypothesis": "when did you 1st realize something was wrong",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0012",
+ "category": "rare_words",
+ "text": "The archaeologist catalogued a Byzantine reliquary and several anachronistic fragments.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0012.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The archaeologist catalogued a Byzantine reliquary and several anachronistic fragments.",
+ "raw": {
+ "normalized_reference": "the archaeologist catalogued a byzantine reliquary and several anachronistic fragments.",
+ "normalized_hypothesis": "the archaeologist catalogued a byzantine reliquary and several anachronistic fragments.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the archaeologist catalogued a byzantine reliquary and several anachronistic fragments",
+ "normalized_hypothesis": "the archaeologist catalogued a byzantine reliquary and several anachronistic fragments",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0013",
+ "category": "short",
+ "text": "I have no idea.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0013.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I have no idea.",
+ "raw": {
+ "normalized_reference": "i have no idea.",
+ "normalized_hypothesis": "i have no idea.",
+ "errors": 0,
+ "words": 4,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i have no idea",
+ "normalized_hypothesis": "i have no idea",
+ "errors": 0,
+ "words": 4,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0014",
+ "category": "technical",
+ "text": "The system runs on three core components that all have to stay in sync.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0014.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The system runs on three core components that all have to stay in sync.",
+ "raw": {
+ "normalized_reference": "the system runs on three core components that all have to stay in sync.",
+ "normalized_hypothesis": "the system runs on three core components that all have to stay in sync.",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the system runs on 3 core components that all have to stay in sync",
+ "normalized_hypothesis": "the system runs on 3 core components that all have to stay in sync",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0015",
+ "category": "conversational",
+ "text": "We improvised the entire second half and somehow nobody noticed.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0015.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "We improvised the entire second half and somehow nobody noticed.",
+ "raw": {
+ "normalized_reference": "we improvised the entire second half and somehow nobody noticed.",
+ "normalized_hypothesis": "we improvised the entire second half and somehow nobody noticed.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "we improvised the entire 2nd half and somehow nobody noticed",
+ "normalized_hypothesis": "we improvised the entire 2nd half and somehow nobody noticed",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0016",
+ "category": "descriptive",
+ "text": "There's a smell in old buildings that's impossible to describe but immediately familiar.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0016.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "There's a smell in old buildings that's impossible to describe but immediately familiar.",
+ "raw": {
+ "normalized_reference": "there's a smell in old buildings that's impossible to describe but immediately familiar.",
+ "normalized_hypothesis": "there's a smell in old buildings that's impossible to describe but immediately familiar.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "there is a smell in old buildings that is impossible to describe but immediately familiar",
+ "normalized_hypothesis": "there is a smell in old buildings that is impossible to describe but immediately familiar",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0017",
+ "category": "emotional",
+ "text": "I specifically asked for one thing. One thing. And it still didn't happen.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0017.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I specifically asked for one thing, one thing, and it still didn't happen.",
+ "raw": {
+ "normalized_reference": "i specifically asked for one thing. one thing. and it still didn't happen.",
+ "normalized_hypothesis": "i specifically asked for one thing, one thing, and it still didn't happen.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i specifically asked for one thing one thing and it still did not happen",
+ "normalized_hypothesis": "i specifically asked for one thing one thing and it still did not happen",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0018",
+ "category": "homographs",
+ "text": "They refuse to place the refuse beside the produce aisle.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0018.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "They refused to place the refuse beside the produce aisle.",
+ "raw": {
+ "normalized_reference": "they refuse to place the refuse beside the produce aisle.",
+ "normalized_hypothesis": "they refused to place the refuse beside the produce aisle.",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ },
+ "semantic": {
+ "normalized_reference": "they refuse to place the refuse beside the produce aisle",
+ "normalized_hypothesis": "they refused to place the refuse beside the produce aisle",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ }
+ },
+ {
+ "id": "release500_0019",
+ "category": "long",
+ "text": "There were days last year when I genuinely didn't know how I was going to get through it, and somehow I did, and I still don't fully understand how.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0019.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Lower days last year when I genuinely didn't know how I was going to get through it, and somehow I did, and I still don't fully understand how.",
+ "raw": {
+ "normalized_reference": "there were days last year when i genuinely didn't know how i was going to get through it, and somehow i did, and i still don't fully understand how.",
+ "normalized_hypothesis": "lower days last year when i genuinely didn't know how i was going to get through it, and somehow i did, and i still don't fully understand how.",
+ "errors": 2,
+ "words": 29,
+ "wer": 0.06896551724137931
+ },
+ "semantic": {
+ "normalized_reference": "there were days last year when i genuinely did not know how i was going to get through it and somehow i did and i still do not fully understand how",
+ "normalized_hypothesis": "lower days last year when i genuinely did not know how i was going to get through it and somehow i did and i still do not fully understand how",
+ "errors": 2,
+ "words": 31,
+ "wer": 0.06451612903225806
+ }
+ },
+ {
+ "id": "release500_0020",
+ "category": "long_range",
+ "text": "Before announcing the winner, the judge thanked the volunteers who had worked through the storm to repair the damaged stage.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0020.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Before announcing the winner, the judge thanked the volunteers who had worked through the storm to repair the damaged stage.",
+ "raw": {
+ "normalized_reference": "before announcing the winner, the judge thanked the volunteers who had worked through the storm to repair the damaged stage.",
+ "normalized_hypothesis": "before announcing the winner, the judge thanked the volunteers who had worked through the storm to repair the damaged stage.",
+ "errors": 0,
+ "words": 20,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "before announcing the winner the judge thanked the volunteers who had worked through the storm to repair the damaged stage",
+ "normalized_hypothesis": "before announcing the winner the judge thanked the volunteers who had worked through the storm to repair the damaged stage",
+ "errors": 0,
+ "words": 20,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0021",
+ "category": "names_places",
+ "text": "Professor Nguyen mailed the fluorescent poster to Reykjavik on Thursday.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0021.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Professor Nguyen mailed the fluorescent poster to Reykjavik on Thursday.",
+ "raw": {
+ "normalized_reference": "professor nguyen mailed the fluorescent poster to reykjavik on thursday.",
+ "normalized_hypothesis": "professor nguyen mailed the fluorescent poster to reykjavik on thursday.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "professor nguyen mailed the fluorescent poster to reykjavik on thursday",
+ "normalized_hypothesis": "professor nguyen mailed the fluorescent poster to reykjavik on thursday",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0022",
+ "category": "narrative",
+ "text": "The plan was simple. The execution was anything but.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0022.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The plan was simple. The execution was anything but.",
+ "raw": {
+ "normalized_reference": "the plan was simple. the execution was anything but.",
+ "normalized_hypothesis": "the plan was simple. the execution was anything but.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the plan was simple the execution was anything but",
+ "normalized_hypothesis": "the plan was simple the execution was anything but",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0023",
+ "category": "numbers",
+ "text": "Her score improved from eighty-two percent to ninety-seven percent in two weeks.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0023.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Her score improved from 82% to 97% in two weeks.",
+ "raw": {
+ "normalized_reference": "her score improved from eighty-two percent to ninety-seven percent in two weeks.",
+ "normalized_hypothesis": "her score improved from 82% to 97% in two weeks.",
+ "errors": 6,
+ "words": 14,
+ "wer": 0.42857142857142855
+ },
+ "semantic": {
+ "normalized_reference": "her score improved from 82% to 97% in 2 weeks",
+ "normalized_hypothesis": "her score improved from 82% to 97% in 2 weeks",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0024",
+ "category": "punctuation",
+ "text": "Why would anyone pack spoons, batteries, and a violin for one short trip?",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0024.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Why would anyone pack spoons, batteries, and a violin for one short trip?",
+ "raw": {
+ "normalized_reference": "why would anyone pack spoons, batteries, and a violin for one short trip?",
+ "normalized_hypothesis": "why would anyone pack spoons, batteries, and a violin for one short trip?",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "why would anyone pack spoons batteries and a violin for one short trip",
+ "normalized_hypothesis": "why would anyone pack spoons batteries and a violin for one short trip",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0025",
+ "category": "question",
+ "text": "How long have you been sitting on this?",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0025.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "How long have you been sitting on this?",
+ "raw": {
+ "normalized_reference": "how long have you been sitting on this?",
+ "normalized_hypothesis": "how long have you been sitting on this?",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "how long have you been sitting on this",
+ "normalized_hypothesis": "how long have you been sitting on this",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0026",
+ "category": "rare_words",
+ "text": "A mischievous chameleon rested beside the chrysanthemum and the bougainvillea.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0026.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "A mischievous chameleon rested beside the chrysanthemum and the bougainvillea.",
+ "raw": {
+ "normalized_reference": "a mischievous chameleon rested beside the chrysanthemum and the bougainvillea.",
+ "normalized_hypothesis": "a mischievous chameleon rested beside the chrysanthemum and the bougainvillea.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "a mischievous chameleon rested beside the chrysanthemum and the bougainvillea",
+ "normalized_hypothesis": "a mischievous chameleon rested beside the chrysanthemum and the bougainvillea",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0027",
+ "category": "short",
+ "text": "We don't talk about that.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0027.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "We don't talk about that.",
+ "raw": {
+ "normalized_reference": "we don't talk about that.",
+ "normalized_hypothesis": "we don't talk about that.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "we do not talk about that",
+ "normalized_hypothesis": "we do not talk about that",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0028",
+ "category": "technical",
+ "text": "Hash the UTF eight payload with S H A two fifty-six before opening the checkpoint.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0028.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Hash the UTF-8 payload with SH-256 before opening the checkpoint.",
+ "raw": {
+ "normalized_reference": "hash the utf eight payload with s h a two fifty-six before opening the checkpoint.",
+ "normalized_hypothesis": "hash the utf-8 payload with sh-256 before opening the checkpoint.",
+ "errors": 7,
+ "words": 16,
+ "wer": 0.4375
+ },
+ "semantic": {
+ "normalized_reference": "hash the utf 8 payload with s h a 256 before opening the checkpoint",
+ "normalized_hypothesis": "hash the utf 8 payload with sh 256 before opening the checkpoint",
+ "errors": 3,
+ "words": 14,
+ "wer": 0.21428571428571427
+ }
+ },
+ {
+ "id": "release500_0029",
+ "category": "conversational",
+ "text": "I thought I had it all figured out, and then this happened.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0029.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I thought I had it all figured out, and then this happened.",
+ "raw": {
+ "normalized_reference": "i thought i had it all figured out, and then this happened.",
+ "normalized_hypothesis": "i thought i had it all figured out, and then this happened.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i thought i had it all figured out and then this happened",
+ "normalized_hypothesis": "i thought i had it all figured out and then this happened",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0030",
+ "category": "descriptive",
+ "text": "The kind of quiet you only get at three in the morning in a city.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0030.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The kind of quiet you only get at three in the morning in a city.",
+ "raw": {
+ "normalized_reference": "the kind of quiet you only get at three in the morning in a city.",
+ "normalized_hypothesis": "the kind of quiet you only get at three in the morning in a city.",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the kind of quiet you only get at 3 in the morning in a city",
+ "normalized_hypothesis": "the kind of quiet you only get at 3 in the morning in a city",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0031",
+ "category": "emotional",
+ "text": "The joy of it was almost overwhelming, like too much good all at once.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0031.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The joy of it was almost overwhelming, like too much could all at once.",
+ "raw": {
+ "normalized_reference": "the joy of it was almost overwhelming, like too much good all at once.",
+ "normalized_hypothesis": "the joy of it was almost overwhelming, like too much could all at once.",
+ "errors": 1,
+ "words": 14,
+ "wer": 0.07142857142857142
+ },
+ "semantic": {
+ "normalized_reference": "the joy of it was almost overwhelming like too much good all at once",
+ "normalized_hypothesis": "the joy of it was almost overwhelming like too much could all at once",
+ "errors": 1,
+ "words": 14,
+ "wer": 0.07142857142857142
+ }
+ },
+ {
+ "id": "release500_0032",
+ "category": "homographs",
+ "text": "Please close the door while I stand close to the microphone.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0032.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Please close the door while I stand close to the microphone.",
+ "raw": {
+ "normalized_reference": "please close the door while i stand close to the microphone.",
+ "normalized_hypothesis": "please close the door while i stand close to the microphone.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "please close the door while i stand close to the microphone",
+ "normalized_hypothesis": "please close the door while i stand close to the microphone",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0033",
+ "category": "long",
+ "text": "I think the most underrated skill in the world is knowing when to stop explaining yourself and just let people draw their own conclusions.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0033.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I think the most underrated skill in the world is knowing when to stop explaining yourself and just let people draw their own conclusions.",
+ "raw": {
+ "normalized_reference": "i think the most underrated skill in the world is knowing when to stop explaining yourself and just let people draw their own conclusions.",
+ "normalized_hypothesis": "i think the most underrated skill in the world is knowing when to stop explaining yourself and just let people draw their own conclusions.",
+ "errors": 0,
+ "words": 24,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i think the most underrated skill in the world is knowing when to stop explaining yourself and just let people draw their own conclusions",
+ "normalized_hypothesis": "i think the most underrated skill in the world is knowing when to stop explaining yourself and just let people draw their own conclusions",
+ "errors": 0,
+ "words": 24,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0034",
+ "category": "long_range",
+ "text": "When the final bus disappeared around the corner, Nadia remembered that her umbrella, notebook, and lunch were still inside the quiet office.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0034.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "When the final bus disappeared around the corner, Nadia remembered that her umbrella, notebook, and lunch were still inside the quiet office.",
+ "raw": {
+ "normalized_reference": "when the final bus disappeared around the corner, nadia remembered that her umbrella, notebook, and lunch were still inside the quiet office.",
+ "normalized_hypothesis": "when the final bus disappeared around the corner, nadia remembered that her umbrella, notebook, and lunch were still inside the quiet office.",
+ "errors": 0,
+ "words": 22,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "when the final bus disappeared around the corner nadia remembered that her umbrella notebook and lunch were still inside the quiet office",
+ "normalized_hypothesis": "when the final bus disappeared around the corner nadia remembered that her umbrella notebook and lunch were still inside the quiet office",
+ "errors": 0,
+ "words": 22,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0035",
+ "category": "names_places",
+ "text": "Aurelia and Joaquin crossed the Schuylkill River before noon.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0035.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Aurelia and Joaquin crossed the Shiokil River before noon.",
+ "raw": {
+ "normalized_reference": "aurelia and joaquin crossed the schuylkill river before noon.",
+ "normalized_hypothesis": "aurelia and joaquin crossed the shiokil river before noon.",
+ "errors": 1,
+ "words": 9,
+ "wer": 0.1111111111111111
+ },
+ "semantic": {
+ "normalized_reference": "aurelia and joaquin crossed the schuylkill river before noon",
+ "normalized_hypothesis": "aurelia and joaquin crossed the shiokil river before noon",
+ "errors": 1,
+ "words": 9,
+ "wer": 0.1111111111111111
+ }
+ },
+ {
+ "id": "release500_0036",
+ "category": "narrative",
+ "text": "The village had a name, but nobody used it. Everyone just called it home.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0036.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The village had a name, but nobody used it. Everyone just called it home.",
+ "raw": {
+ "normalized_reference": "the village had a name, but nobody used it. everyone just called it home.",
+ "normalized_hypothesis": "the village had a name, but nobody used it. everyone just called it home.",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the village had a name but nobody used it everyone just called it home",
+ "normalized_hypothesis": "the village had a name but nobody used it everyone just called it home",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0037",
+ "category": "numbers",
+ "text": "Increase the gain by three decibels, wait five seconds, then reduce it by one.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0037.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Increase the gain by 3 decibels, wait 5 seconds, then reduce it by 1.",
+ "raw": {
+ "normalized_reference": "increase the gain by three decibels, wait five seconds, then reduce it by one.",
+ "normalized_hypothesis": "increase the gain by 3 decibels, wait 5 seconds, then reduce it by 1.",
+ "errors": 3,
+ "words": 14,
+ "wer": 0.21428571428571427
+ },
+ "semantic": {
+ "normalized_reference": "increase the gain by 3 decibels wait 5 seconds then reduce it by one",
+ "normalized_hypothesis": "increase the gain by 3 decibels wait 5 seconds then reduce it by one",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0038",
+ "category": "punctuation",
+ "text": "The result was unexpected: clear, stable, and surprisingly natural.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0038.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The result was unexpected, clear, stable, and surprisingly natural.",
+ "raw": {
+ "normalized_reference": "the result was unexpected: clear, stable, and surprisingly natural.",
+ "normalized_hypothesis": "the result was unexpected, clear, stable, and surprisingly natural.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the result was unexpected clear stable and surprisingly natural",
+ "normalized_hypothesis": "the result was unexpected clear stable and surprisingly natural",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0039",
+ "category": "question",
+ "text": "Who told you that was the right way to do it?",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0039.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Who told you that was the right way to do it?",
+ "raw": {
+ "normalized_reference": "who told you that was the right way to do it?",
+ "normalized_hypothesis": "who told you that was the right way to do it?",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "who told you that was the right way to do it",
+ "normalized_hypothesis": "who told you that was the right way to do it",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0040",
+ "category": "rare_words",
+ "text": "The meticulous cartographer examined an iridescent archipelago through binoculars.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0040.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The meticulous cartographer examined an iridescent archipelago through binoculars.",
+ "raw": {
+ "normalized_reference": "the meticulous cartographer examined an iridescent archipelago through binoculars.",
+ "normalized_hypothesis": "the meticulous cartographer examined an iridescent archipelago through binoculars.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the meticulous cartographer examined an iridescent archipelago through binoculars",
+ "normalized_hypothesis": "the meticulous cartographer examined an iridescent archipelago through binoculars",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0041",
+ "category": "short",
+ "text": "I mean. Yeah. Kind of.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0041.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I mean, yeah, kind of.",
+ "raw": {
+ "normalized_reference": "i mean. yeah. kind of.",
+ "normalized_hypothesis": "i mean, yeah, kind of.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i mean yeah kind of",
+ "normalized_hypothesis": "i mean yeah kind of",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0042",
+ "category": "technical",
+ "text": "The logs are your best friend when something behaves unexpectedly.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0042.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The logs are your best friend when something behaves unexpectedly.",
+ "raw": {
+ "normalized_reference": "the logs are your best friend when something behaves unexpectedly.",
+ "normalized_hypothesis": "the logs are your best friend when something behaves unexpectedly.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the logs are your best friend when something behaves unexpectedly",
+ "normalized_hypothesis": "the logs are your best friend when something behaves unexpectedly",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0043",
+ "category": "conversational",
+ "text": "He kept talking and I kept nodding and neither of us was listening.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0043.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He kept talking, and I kept nodding, and neither of us was listening.",
+ "raw": {
+ "normalized_reference": "he kept talking and i kept nodding and neither of us was listening.",
+ "normalized_hypothesis": "he kept talking, and i kept nodding, and neither of us was listening.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he kept talking and i kept nodding and neither of us was listening",
+ "normalized_hypothesis": "he kept talking and i kept nodding and neither of us was listening",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0044",
+ "category": "descriptive",
+ "text": "It's one of those places where everyone seems to know each other already.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0044.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It's one of those places where everyone seems to know each other already.",
+ "raw": {
+ "normalized_reference": "it's one of those places where everyone seems to know each other already.",
+ "normalized_hypothesis": "it's one of those places where everyone seems to know each other already.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it is one of those places where everyone seems to know each other already",
+ "normalized_hypothesis": "it is one of those places where everyone seems to know each other already",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0045",
+ "category": "emotional",
+ "text": "He said he was proud of me and nobody had ever said that to me before.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0045.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He said he was proud of me, and nobody had ever said that to me before.",
+ "raw": {
+ "normalized_reference": "he said he was proud of me and nobody had ever said that to me before.",
+ "normalized_hypothesis": "he said he was proud of me, and nobody had ever said that to me before.",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he said he was proud of me and nobody had ever said that to me before",
+ "normalized_hypothesis": "he said he was proud of me and nobody had ever said that to me before",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0046",
+ "category": "homographs",
+ "text": "The minute detail took a minute to explain clearly.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0046.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The minute detail took a minute to explain clearly.",
+ "raw": {
+ "normalized_reference": "the minute detail took a minute to explain clearly.",
+ "normalized_hypothesis": "the minute detail took a minute to explain clearly.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the minute detail took a minute to explain clearly",
+ "normalized_hypothesis": "the minute detail took a minute to explain clearly",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0047",
+ "category": "long",
+ "text": "The strange thing about really wanting something is that getting it never quite feels the way you imagined it would, and not getting it never quite feels as bad.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0047.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The strange thing about really wanting something is that getting it never quite feels the way you imagined it would, and not getting it never quite feels as bad.",
+ "raw": {
+ "normalized_reference": "the strange thing about really wanting something is that getting it never quite feels the way you imagined it would, and not getting it never quite feels as bad.",
+ "normalized_hypothesis": "the strange thing about really wanting something is that getting it never quite feels the way you imagined it would, and not getting it never quite feels as bad.",
+ "errors": 0,
+ "words": 29,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the strange thing about really wanting something is that getting it never quite feels the way you imagined it would and not getting it never quite feels as bad",
+ "normalized_hypothesis": "the strange thing about really wanting something is that getting it never quite feels the way you imagined it would and not getting it never quite feels as bad",
+ "errors": 0,
+ "words": 29,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0048",
+ "category": "long_range",
+ "text": "Even though the recipe looked simple at first, preparing the sauce required patience, careful timing, and a much larger pan.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0048.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Even though the recipe looked simple at first, preparing the sauce required patience, careful timing, and a much larger pan.",
+ "raw": {
+ "normalized_reference": "even though the recipe looked simple at first, preparing the sauce required patience, careful timing, and a much larger pan.",
+ "normalized_hypothesis": "even though the recipe looked simple at first, preparing the sauce required patience, careful timing, and a much larger pan.",
+ "errors": 0,
+ "words": 20,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "even though the recipe looked simple at 1st preparing the sauce required patience careful timing and a much larger pan",
+ "normalized_hypothesis": "even though the recipe looked simple at 1st preparing the sauce required patience careful timing and a much larger pan",
+ "errors": 0,
+ "words": 20,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0049",
+ "category": "names_places",
+ "text": "Xiomara met Siobhan near the old library in Worcestershire.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0049.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Xiomara met Siobhan near the old library in Worcestershire.",
+ "raw": {
+ "normalized_reference": "xiomara met siobhan near the old library in worcestershire.",
+ "normalized_hypothesis": "xiomara met siobhan near the old library in worcestershire.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "xiomara met siobhan near the old library in worcestershire",
+ "normalized_hypothesis": "xiomara met siobhan near the old library in worcestershire",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0050",
+ "category": "narrative",
+ "text": "She walked into the room and everyone went quiet, not because she was loud, but because she wasn't.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0050.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "She walked into the room and everyone went quiet, not because she was loud, but because she wasn't.",
+ "raw": {
+ "normalized_reference": "she walked into the room and everyone went quiet, not because she was loud, but because she wasn't.",
+ "normalized_hypothesis": "she walked into the room and everyone went quiet, not because she was loud, but because she wasn't.",
+ "errors": 0,
+ "words": 18,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "she walked into the room and everyone went quiet not because she was loud but because she was not",
+ "normalized_hypothesis": "she walked into the room and everyone went quiet not because she was loud but because she was not",
+ "errors": 0,
+ "words": 19,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0051",
+ "category": "numbers",
+ "text": "At exactly seven forty-five, Morgan placed eighteen dollars and fifty cents beside the notebook.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0051.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "At exactly $7.45, Morgan placed $18.50 beside the notebook.",
+ "raw": {
+ "normalized_reference": "at exactly seven forty-five, morgan placed eighteen dollars and fifty cents beside the notebook.",
+ "normalized_hypothesis": "at exactly $7.45, morgan placed $18.50 beside the notebook.",
+ "errors": 8,
+ "words": 15,
+ "wer": 0.5333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "at exactly 745 morgan placed $18.50 beside the notebook",
+ "normalized_hypothesis": "at exactly $7.45 morgan placed $18.50 beside the notebook",
+ "errors": 2,
+ "words": 10,
+ "wer": 0.2
+ }
+ },
+ {
+ "id": "release500_0052",
+ "category": "punctuation",
+ "text": "Does punctuation work? Yes: commas, semicolons, and periods should create sensible boundaries.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0052.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Does punctuation work? Yes, commas, semicolons, and periods should create sensible boundaries.",
+ "raw": {
+ "normalized_reference": "does punctuation work? yes: commas, semicolons, and periods should create sensible boundaries.",
+ "normalized_hypothesis": "does punctuation work? yes, commas, semicolons, and periods should create sensible boundaries.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "does punctuation work yes commas semicolons and periods should create sensible boundaries",
+ "normalized_hypothesis": "does punctuation work yes commas semicolons and periods should create sensible boundaries",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0053",
+ "category": "question",
+ "text": "Have you ever just stopped and thought about how strange all of this is?",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0053.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Have you ever just stopped and thought about how strange all of this is?",
+ "raw": {
+ "normalized_reference": "have you ever just stopped and thought about how strange all of this is?",
+ "normalized_hypothesis": "have you ever just stopped and thought about how strange all of this is?",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "have you ever just stopped and thought about how strange all of this is",
+ "normalized_hypothesis": "have you ever just stopped and thought about how strange all of this is",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0054",
+ "category": "rare_words",
+ "text": "Her idiosyncratic explanation sounded plausible but remained fundamentally ambiguous.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0054.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Her idiosyncratic explanation sounded plausible, but remained fundamentally ambiguous.",
+ "raw": {
+ "normalized_reference": "her idiosyncratic explanation sounded plausible but remained fundamentally ambiguous.",
+ "normalized_hypothesis": "her idiosyncratic explanation sounded plausible, but remained fundamentally ambiguous.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "her idiosyncratic explanation sounded plausible but remained fundamentally ambiguous",
+ "normalized_hypothesis": "her idiosyncratic explanation sounded plausible but remained fundamentally ambiguous",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0055",
+ "category": "short",
+ "text": "That tracks.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0055.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "That tracks.",
+ "raw": {
+ "normalized_reference": "that tracks.",
+ "normalized_hypothesis": "that tracks.",
+ "errors": 0,
+ "words": 2,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "that tracks",
+ "normalized_hypothesis": "that tracks",
+ "errors": 0,
+ "words": 2,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0056",
+ "category": "technical",
+ "text": "The key thing to remember is that order matters here more than you'd think.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0056.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The key thing to remember is that order matters here more than you'd think.",
+ "raw": {
+ "normalized_reference": "the key thing to remember is that order matters here more than you'd think.",
+ "normalized_hypothesis": "the key thing to remember is that order matters here more than you'd think.",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the key thing to remember is that order matters here more than you would think",
+ "normalized_hypothesis": "the key thing to remember is that order matters here more than you would think",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0057",
+ "category": "conversational",
+ "text": "I would not say that again in this room if I were you.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0057.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I would not say that again in this room if I were you.",
+ "raw": {
+ "normalized_reference": "i would not say that again in this room if i were you.",
+ "normalized_hypothesis": "i would not say that again in this room if i were you.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i would not say that again in this room if i were you",
+ "normalized_hypothesis": "i would not say that again in this room if i were you",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0058",
+ "category": "descriptive",
+ "text": "The room was filled with the quiet that means something is about to change.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0058.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The room was filled with the quiet that means something is about to change.",
+ "raw": {
+ "normalized_reference": "the room was filled with the quiet that means something is about to change.",
+ "normalized_hypothesis": "the room was filled with the quiet that means something is about to change.",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the room was filled with the quiet that means something is about to change",
+ "normalized_hypothesis": "the room was filled with the quiet that means something is about to change",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0059",
+ "category": "emotional",
+ "text": "I didn't expect to feel that way about it, but I did.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0059.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I didn't expect to feel that way about it, but I did.",
+ "raw": {
+ "normalized_reference": "i didn't expect to feel that way about it, but i did.",
+ "normalized_hypothesis": "i didn't expect to feel that way about it, but i did.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i did not expect to feel that way about it but i did",
+ "normalized_hypothesis": "i did not expect to feel that way about it but i did",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0060",
+ "category": "homographs",
+ "text": "I object to the strange object that was left near the entrance.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0060.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I object to the strange object that was left near the entrance",
+ "raw": {
+ "normalized_reference": "i object to the strange object that was left near the entrance.",
+ "normalized_hypothesis": "i object to the strange object that was left near the entrance",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i object to the strange object that was left near the entrance",
+ "normalized_hypothesis": "i object to the strange object that was left near the entrance",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0061",
+ "category": "long",
+ "text": "I've learned more from the things that went wrong than from the things that worked, which is a frustrating way to have to learn things but apparently the only way that stuck.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0061.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I've learned more from the things that went wrong than from the things that worked, which is a frustrating way to have to learn things, but apparently the only way that stuck.",
+ "raw": {
+ "normalized_reference": "i've learned more from the things that went wrong than from the things that worked, which is a frustrating way to have to learn things but apparently the only way that stuck.",
+ "normalized_hypothesis": "i've learned more from the things that went wrong than from the things that worked, which is a frustrating way to have to learn things, but apparently the only way that stuck.",
+ "errors": 0,
+ "words": 32,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i have learned more from the things that went wrong than from the things that worked which is a frustrating way to have to learn things but apparently the only way that stuck",
+ "normalized_hypothesis": "i have learned more from the things that went wrong than from the things that worked which is a frustrating way to have to learn things but apparently the only way that stuck",
+ "errors": 0,
+ "words": 33,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0062",
+ "category": "long_range",
+ "text": "Because every nearby hotel was full, the exhausted hikers followed the narrow road until they found a farmhouse with one empty room.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0062.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Because every nearby hotel was full, the exhausted hikers followed the narrow road until they found a farmhouse with one empty room.",
+ "raw": {
+ "normalized_reference": "because every nearby hotel was full, the exhausted hikers followed the narrow road until they found a farmhouse with one empty room.",
+ "normalized_hypothesis": "because every nearby hotel was full, the exhausted hikers followed the narrow road until they found a farmhouse with one empty room.",
+ "errors": 0,
+ "words": 22,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "because every nearby hotel was full the exhausted hikers followed the narrow road until they found a farmhouse with one empty room",
+ "normalized_hypothesis": "because every nearby hotel was full the exhausted hikers followed the narrow road until they found a farmhouse with one empty room",
+ "errors": 0,
+ "words": 22,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0063",
+ "category": "names_places",
+ "text": "The itinerary sends Beatrice from Albuquerque to Des Moines and then to Boise.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0063.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The itinerary sends Beatrice from Albuquerque to Des Moines and then to Boise.",
+ "raw": {
+ "normalized_reference": "the itinerary sends beatrice from albuquerque to des moines and then to boise.",
+ "normalized_hypothesis": "the itinerary sends beatrice from albuquerque to des moines and then to boise.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the itinerary sends beatrice from albuquerque to des moines and then to boise",
+ "normalized_hypothesis": "the itinerary sends beatrice from albuquerque to des moines and then to boise",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0064",
+ "category": "narrative",
+ "text": "We had been driving for six hours when we finally saw the sign for the town.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0064.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "We had been driving for six hours when we finally saw the sign for the town.",
+ "raw": {
+ "normalized_reference": "we had been driving for six hours when we finally saw the sign for the town.",
+ "normalized_hypothesis": "we had been driving for six hours when we finally saw the sign for the town.",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "we had been driving for 6 hours when we finally saw the sign for the town",
+ "normalized_hypothesis": "we had been driving for 6 hours when we finally saw the sign for the town",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0065",
+ "category": "numbers",
+ "text": "Call extension zero four nine after eight thirty in the morning.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0065.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Hall extension 049 after 8.30 in the morning.",
+ "raw": {
+ "normalized_reference": "call extension zero four nine after eight thirty in the morning.",
+ "normalized_hypothesis": "hall extension 049 after 8.30 in the morning.",
+ "errors": 6,
+ "words": 11,
+ "wer": 0.5454545454545454
+ },
+ "semantic": {
+ "normalized_reference": "call extension 049 after 830 in the morning",
+ "normalized_hypothesis": "hall extension 49 after 8.30 in the morning",
+ "errors": 4,
+ "words": 8,
+ "wer": 0.5
+ }
+ },
+ {
+ "id": "release500_0066",
+ "category": "punctuation",
+ "text": "Stop - listen carefully - then answer the question without guessing.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0066.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Stop, listen carefully, then answer the question without guessing.",
+ "raw": {
+ "normalized_reference": "stop - listen carefully - then answer the question without guessing.",
+ "normalized_hypothesis": "stop, listen carefully, then answer the question without guessing.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "stop listen carefully then answer the question without guessing",
+ "normalized_hypothesis": "stop listen carefully then answer the question without guessing",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0067",
+ "category": "question",
+ "text": "Does it matter at this point, honestly?",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0067.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Does it matter at this point? Honestly.",
+ "raw": {
+ "normalized_reference": "does it matter at this point, honestly?",
+ "normalized_hypothesis": "does it matter at this point? honestly.",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "does it matter at this point honestly",
+ "normalized_hypothesis": "does it matter at this point honestly",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0068",
+ "category": "rare_words",
+ "text": "An entrepreneurial meteorologist described the unprecedented atmospheric phenomenon.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0068.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "An entrepreneurial meteorologist described the unprecedented atmospheric phenomenon.",
+ "raw": {
+ "normalized_reference": "an entrepreneurial meteorologist described the unprecedented atmospheric phenomenon.",
+ "normalized_hypothesis": "an entrepreneurial meteorologist described the unprecedented atmospheric phenomenon.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "an entrepreneurial meteorologist described the unprecedented atmospheric phenomenon",
+ "normalized_hypothesis": "an entrepreneurial meteorologist described the unprecedented atmospheric phenomenon",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0069",
+ "category": "short",
+ "text": "That's not what I expected.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0069.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "That's not what I expected.",
+ "raw": {
+ "normalized_reference": "that's not what i expected.",
+ "normalized_hypothesis": "that's not what i expected.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "that is not what i expected",
+ "normalized_hypothesis": "that is not what i expected",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0070",
+ "category": "technical",
+ "text": "There's a known issue with that version that causes failures under load.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0070.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "There's a known issue with that version that causes failures under load.",
+ "raw": {
+ "normalized_reference": "there's a known issue with that version that causes failures under load.",
+ "normalized_hypothesis": "there's a known issue with that version that causes failures under load.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "there is a known issue with that version that causes failures under load",
+ "normalized_hypothesis": "there is a known issue with that version that causes failures under load",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0071",
+ "category": "conversational",
+ "text": "Don't look now but I think that's him over there.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0071.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Don't look now, but I think that's him over there.",
+ "raw": {
+ "normalized_reference": "don't look now but i think that's him over there.",
+ "normalized_hypothesis": "don't look now, but i think that's him over there.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "do not look now but i think that is him over there",
+ "normalized_hypothesis": "do not look now but i think that is him over there",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0072",
+ "category": "descriptive",
+ "text": "The city at night looks completely different from how it feels during the day.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0072.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The city at night looks completely different from how it feels during the day.",
+ "raw": {
+ "normalized_reference": "the city at night looks completely different from how it feels during the day.",
+ "normalized_hypothesis": "the city at night looks completely different from how it feels during the day.",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the city at night looks completely different from how it feels during the day",
+ "normalized_hypothesis": "the city at night looks completely different from how it feels during the day",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0073",
+ "category": "emotional",
+ "text": "It's the kind of happy that makes you nervous because you know it can't last.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0073.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It's the kind of happy that makes you nervous because you know it can't last.",
+ "raw": {
+ "normalized_reference": "it's the kind of happy that makes you nervous because you know it can't last.",
+ "normalized_hypothesis": "it's the kind of happy that makes you nervous because you know it can't last.",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it is the kind of happy that makes you nervous because you know it can not last",
+ "normalized_hypothesis": "it is the kind of happy that makes you nervous because you know it can not last",
+ "errors": 0,
+ "words": 17,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0074",
+ "category": "homographs",
+ "text": "The dove dove beneath the branch while the wind began to wind the loose cable.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0074.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The dove dove beneath the branch while the wind began to wind the loose cable.",
+ "raw": {
+ "normalized_reference": "the dove dove beneath the branch while the wind began to wind the loose cable.",
+ "normalized_hypothesis": "the dove dove beneath the branch while the wind began to wind the loose cable.",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the dove dove beneath the branch while the wind began to wind the loose cable",
+ "normalized_hypothesis": "the dove dove beneath the branch while the wind began to wind the loose cable",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0075",
+ "category": "long",
+ "text": "I've thought a lot about what I would do differently if I could go back, and the honest answer is probably not as much as I'd like to think.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0075.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I've thought a lot about what I would do differently if I could go back, and the honest answer is probably not as much as I'd like to think.",
+ "raw": {
+ "normalized_reference": "i've thought a lot about what i would do differently if i could go back, and the honest answer is probably not as much as i'd like to think.",
+ "normalized_hypothesis": "i've thought a lot about what i would do differently if i could go back, and the honest answer is probably not as much as i'd like to think.",
+ "errors": 0,
+ "words": 29,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i have thought a lot about what i would do differently if i could go back and the honest answer is probably not as much as i would like to think",
+ "normalized_hypothesis": "i have thought a lot about what i would do differently if i could go back and the honest answer is probably not as much as i would like to think",
+ "errors": 0,
+ "words": 31,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0076",
+ "category": "long_range",
+ "text": "Although the weather changed without warning, the musicians finished packing before anyone opened the enormous glass door.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0076.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Although the weather changed without warning, the musicians finished packing before anyone opened the enormous glass door.",
+ "raw": {
+ "normalized_reference": "although the weather changed without warning, the musicians finished packing before anyone opened the enormous glass door.",
+ "normalized_hypothesis": "although the weather changed without warning, the musicians finished packing before anyone opened the enormous glass door.",
+ "errors": 0,
+ "words": 17,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "although the weather changed without warning the musicians finished packing before anyone opened the enormous glass door",
+ "normalized_hypothesis": "although the weather changed without warning the musicians finished packing before anyone opened the enormous glass door",
+ "errors": 0,
+ "words": 17,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0077",
+ "category": "names_places",
+ "text": "Saskatchewan looked unusually bright beneath the pale February sunrise.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0077.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "sasvichuin looked unusually bright beneath the pale february sunrise",
+ "raw": {
+ "normalized_reference": "saskatchewan looked unusually bright beneath the pale february sunrise.",
+ "normalized_hypothesis": "sasvichuin looked unusually bright beneath the pale february sunrise",
+ "errors": 1,
+ "words": 9,
+ "wer": 0.1111111111111111
+ },
+ "semantic": {
+ "normalized_reference": "saskatchewan looked unusually bright beneath the pale february sunrise",
+ "normalized_hypothesis": "sasvichuin looked unusually bright beneath the pale february sunrise",
+ "errors": 1,
+ "words": 9,
+ "wer": 0.1111111111111111
+ }
+ },
+ {
+ "id": "release500_0078",
+ "category": "narrative",
+ "text": "She handled it with a grace I genuinely did not expect.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0078.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "She handled it with a grace I genuinely did not expect.",
+ "raw": {
+ "normalized_reference": "she handled it with a grace i genuinely did not expect.",
+ "normalized_hypothesis": "she handled it with a grace i genuinely did not expect.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "she handled it with a grace i genuinely did not expect",
+ "normalized_hypothesis": "she handled it with a grace i genuinely did not expect",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0079",
+ "category": "numbers",
+ "text": "Apartment four hundred seven is across from room twenty-nine on the third floor.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0079.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Apartment 407 is across from room 29 on the third floor.",
+ "raw": {
+ "normalized_reference": "apartment four hundred seven is across from room twenty-nine on the third floor.",
+ "normalized_hypothesis": "apartment 407 is across from room 29 on the third floor.",
+ "errors": 5,
+ "words": 14,
+ "wer": 0.35714285714285715
+ },
+ "semantic": {
+ "normalized_reference": "apartment 407 is across from room 29 on the 3rd floor",
+ "normalized_hypothesis": "apartment 407 is across from room 29 on the 3rd floor",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0080",
+ "category": "punctuation",
+ "text": "\"Not yet,\" Rowan whispered. \"Give me another minute... please.\"",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0080.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Not yet, Rowan whispered. Give me another minute, please.",
+ "raw": {
+ "normalized_reference": "\"not yet,\" rowan whispered. \"give me another minute... please.\"",
+ "normalized_hypothesis": "not yet, rowan whispered. give me another minute, please.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "not yet rowan whispered give me another minute please",
+ "normalized_hypothesis": "not yet rowan whispered give me another minute please",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0081",
+ "category": "question",
+ "text": "What are we actually trying to accomplish here?",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0081.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "What are we actually trying to accomplish here?",
+ "raw": {
+ "normalized_reference": "what are we actually trying to accomplish here?",
+ "normalized_hypothesis": "what are we actually trying to accomplish here?",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "what are we actually trying to accomplish here",
+ "normalized_hypothesis": "what are we actually trying to accomplish here",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0082",
+ "category": "rare_words",
+ "text": "The physician recommended a noninvasive procedure for the recurring neurological symptoms.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0082.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The physician recommended a non-invasive procedure for the recurring neurological symptoms.",
+ "raw": {
+ "normalized_reference": "the physician recommended a noninvasive procedure for the recurring neurological symptoms.",
+ "normalized_hypothesis": "the physician recommended a non-invasive procedure for the recurring neurological symptoms.",
+ "errors": 2,
+ "words": 11,
+ "wer": 0.18181818181818182
+ },
+ "semantic": {
+ "normalized_reference": "the physician recommended a noninvasive procedure for the recurring neurological symptoms",
+ "normalized_hypothesis": "the physician recommended a non invasive procedure for the recurring neurological symptoms",
+ "errors": 2,
+ "words": 11,
+ "wer": 0.18181818181818182
+ }
+ },
+ {
+ "id": "release500_0083",
+ "category": "short",
+ "text": "Are you serious right now?",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0083.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Are you serious right now?",
+ "raw": {
+ "normalized_reference": "are you serious right now?",
+ "normalized_hypothesis": "are you serious right now?",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "are you serious right now",
+ "normalized_hypothesis": "are you serious right now",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0084",
+ "category": "technical",
+ "text": "The performance improvement is real, but it comes with tradeoffs.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0084.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The performance improvement is real, but it comes with Trados.",
+ "raw": {
+ "normalized_reference": "the performance improvement is real, but it comes with tradeoffs.",
+ "normalized_hypothesis": "the performance improvement is real, but it comes with trados.",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ },
+ "semantic": {
+ "normalized_reference": "the performance improvement is real but it comes with tradeoffs",
+ "normalized_hypothesis": "the performance improvement is real but it comes with trados",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ }
+ },
+ {
+ "id": "release500_0085",
+ "category": "conversational",
+ "text": "I don't know. It's just been one of those weeks, you know?",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0085.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I don't know. It's just been one of those weeks, you know.",
+ "raw": {
+ "normalized_reference": "i don't know. it's just been one of those weeks, you know?",
+ "normalized_hypothesis": "i don't know. it's just been one of those weeks, you know.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i do not know it is just been one of those weeks you know",
+ "normalized_hypothesis": "i do not know it is just been one of those weeks you know",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0086",
+ "category": "descriptive",
+ "text": "The moment you stop expecting perfection, things get a lot more interesting.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0086.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The moment you stop expecting perfection, things get a lot more interesting.",
+ "raw": {
+ "normalized_reference": "the moment you stop expecting perfection, things get a lot more interesting.",
+ "normalized_hypothesis": "the moment you stop expecting perfection, things get a lot more interesting.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the moment you stop expecting perfection things get a lot more interesting",
+ "normalized_hypothesis": "the moment you stop expecting perfection things get a lot more interesting",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0087",
+ "category": "emotional",
+ "text": "It's the kind of thing that stays with you even years later.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0087.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It's the kind of thing that stays with you even years later.",
+ "raw": {
+ "normalized_reference": "it's the kind of thing that stays with you even years later.",
+ "normalized_hypothesis": "it's the kind of thing that stays with you even years later.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it is the kind of thing that stays with you even years later",
+ "normalized_hypothesis": "it is the kind of thing that stays with you even years later",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0088",
+ "category": "long",
+ "text": "She told me once that the most important skill she'd developed was learning to disagree with someone without making them feel they were wrong.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0088.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "She told me once that the most important skill she'd developed was learning to disagree with someone without making them feel they were wrong.",
+ "raw": {
+ "normalized_reference": "she told me once that the most important skill she'd developed was learning to disagree with someone without making them feel they were wrong.",
+ "normalized_hypothesis": "she told me once that the most important skill she'd developed was learning to disagree with someone without making them feel they were wrong.",
+ "errors": 0,
+ "words": 24,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "she told me once that the most important skill she would developed was learning to disagree with someone without making them feel they were wrong",
+ "normalized_hypothesis": "she told me once that the most important skill she would developed was learning to disagree with someone without making them feel they were wrong",
+ "errors": 0,
+ "words": 25,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0089",
+ "category": "narrative",
+ "text": "He stood at the edge of the cliff and wondered, not for the first time, how he'd gotten here.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0089.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He stood at the edge of the cliff and wondered, not for the first time, how he'd gotten here.",
+ "raw": {
+ "normalized_reference": "he stood at the edge of the cliff and wondered, not for the first time, how he'd gotten here.",
+ "normalized_hypothesis": "he stood at the edge of the cliff and wondered, not for the first time, how he'd gotten here.",
+ "errors": 0,
+ "words": 19,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he stood at the edge of the cliff and wondered not for the 1st time how he would gotten here",
+ "normalized_hypothesis": "he stood at the edge of the cliff and wondered not for the 1st time how he would gotten here",
+ "errors": 0,
+ "words": 20,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0090",
+ "category": "question",
+ "text": "Why does asking for help feel so hard when you actually need it?",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0090.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Why does asking for help feel so hard when you actually need it?",
+ "raw": {
+ "normalized_reference": "why does asking for help feel so hard when you actually need it?",
+ "normalized_hypothesis": "why does asking for help feel so hard when you actually need it?",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "why does asking for help feel so hard when you actually need it",
+ "normalized_hypothesis": "why does asking for help feel so hard when you actually need it",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0091",
+ "category": "short",
+ "text": "That's fair.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0091.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "That's fair.",
+ "raw": {
+ "normalized_reference": "that's fair.",
+ "normalized_hypothesis": "that's fair.",
+ "errors": 0,
+ "words": 2,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "that is fair",
+ "normalized_hypothesis": "that is fair",
+ "errors": 0,
+ "words": 3,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0092",
+ "category": "technical",
+ "text": "If you run into that error, the fix is usually just restarting the service.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0092.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "If you run into that error, the fix is usually just restarting the service.",
+ "raw": {
+ "normalized_reference": "if you run into that error, the fix is usually just restarting the service.",
+ "normalized_hypothesis": "if you run into that error, the fix is usually just restarting the service.",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "if you run into that error the fix is usually just restarting the service",
+ "normalized_hypothesis": "if you run into that error the fix is usually just restarting the service",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0093",
+ "category": "conversational",
+ "text": "It was a disaster in the best possible way.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0093.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It was a disaster in the best possible way.",
+ "raw": {
+ "normalized_reference": "it was a disaster in the best possible way.",
+ "normalized_hypothesis": "it was a disaster in the best possible way.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it was a disaster in the best possible way",
+ "normalized_hypothesis": "it was a disaster in the best possible way",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0094",
+ "category": "descriptive",
+ "text": "It had the look of a place that used to mean something to someone.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0094.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It had the look of a place that used to mean something to someone.",
+ "raw": {
+ "normalized_reference": "it had the look of a place that used to mean something to someone.",
+ "normalized_hypothesis": "it had the look of a place that used to mean something to someone.",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it had the look of a place that used to mean something to someone",
+ "normalized_hypothesis": "it had the look of a place that used to mean something to someone",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0095",
+ "category": "emotional",
+ "text": "There's this feeling you get when something finally works after weeks of failure.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0095.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "There's this feeling you get when something finally works after weeks of failure.",
+ "raw": {
+ "normalized_reference": "there's this feeling you get when something finally works after weeks of failure.",
+ "normalized_hypothesis": "there's this feeling you get when something finally works after weeks of failure.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "there is this feeling you get when something finally works after weeks of failure",
+ "normalized_hypothesis": "there is this feeling you get when something finally works after weeks of failure",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0096",
+ "category": "long",
+ "text": "He had a theory that most arguments aren't actually about what they seem to be about, and the longer I live the more I think he was onto something.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0096.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He had a theory that most arguments aren't actually about what they seem to be about, and the longer I lived the more I think he was onto something.",
+ "raw": {
+ "normalized_reference": "he had a theory that most arguments aren't actually about what they seem to be about, and the longer i live the more i think he was onto something.",
+ "normalized_hypothesis": "he had a theory that most arguments aren't actually about what they seem to be about, and the longer i lived the more i think he was onto something.",
+ "errors": 1,
+ "words": 29,
+ "wer": 0.034482758620689655
+ },
+ "semantic": {
+ "normalized_reference": "he had a theory that most arguments are not actually about what they seem to be about and the longer i live the more i think he was onto something",
+ "normalized_hypothesis": "he had a theory that most arguments are not actually about what they seem to be about and the longer i lived the more i think he was onto something",
+ "errors": 1,
+ "words": 30,
+ "wer": 0.03333333333333333
+ }
+ },
+ {
+ "id": "release500_0097",
+ "category": "narrative",
+ "text": "The thing nobody tells you about starting over is how ordinary most of it feels.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0097.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The thing nobody tells you about starting over is how ordinary most of it feels.",
+ "raw": {
+ "normalized_reference": "the thing nobody tells you about starting over is how ordinary most of it feels.",
+ "normalized_hypothesis": "the thing nobody tells you about starting over is how ordinary most of it feels.",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the thing nobody tells you about starting over is how ordinary most of it feels",
+ "normalized_hypothesis": "the thing nobody tells you about starting over is how ordinary most of it feels",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0098",
+ "category": "question",
+ "text": "What exactly are we waiting for at this point?",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0098.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "What exactly are we waiting for at this point?",
+ "raw": {
+ "normalized_reference": "what exactly are we waiting for at this point?",
+ "normalized_hypothesis": "what exactly are we waiting for at this point?",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "what exactly are we waiting for at this point",
+ "normalized_hypothesis": "what exactly are we waiting for at this point",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0099",
+ "category": "short",
+ "text": "I stand by what I said.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0099.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I stand by what I said.",
+ "raw": {
+ "normalized_reference": "i stand by what i said.",
+ "normalized_hypothesis": "i stand by what i said.",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i stand by what i said",
+ "normalized_hypothesis": "i stand by what i said",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0100",
+ "category": "technical",
+ "text": "Double-check that value before you push anything to production.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0100.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Double check that value before you push anything to production.",
+ "raw": {
+ "normalized_reference": "double-check that value before you push anything to production.",
+ "normalized_hypothesis": "double check that value before you push anything to production.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "double check that value before you push anything to production",
+ "normalized_hypothesis": "double check that value before you push anything to production",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0101",
+ "category": "conversational",
+ "text": "I understand your point; I just do not agree with the conclusion.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0101.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I understand your point, I just do not agree with the conclusion.",
+ "raw": {
+ "normalized_reference": "i understand your point; i just do not agree with the conclusion.",
+ "normalized_hypothesis": "i understand your point, i just do not agree with the conclusion.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i understand your point i just do not agree with the conclusion",
+ "normalized_hypothesis": "i understand your point i just do not agree with the conclusion",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0102",
+ "category": "descriptive",
+ "text": "The sky that day was a strange shade of orange I'd never seen before.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0102.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The sky that day was a strange shade of orange I'd never seen before.",
+ "raw": {
+ "normalized_reference": "the sky that day was a strange shade of orange i'd never seen before.",
+ "normalized_hypothesis": "the sky that day was a strange shade of orange i'd never seen before.",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the sky that day was a strange shade of orange i would never seen before",
+ "normalized_hypothesis": "the sky that day was a strange shade of orange i would never seen before",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0103",
+ "category": "emotional",
+ "text": "It hurt more than I expected, and I thought I was prepared.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0103.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It hurt more than I expected, and I thought I was prepared.",
+ "raw": {
+ "normalized_reference": "it hurt more than i expected, and i thought i was prepared.",
+ "normalized_hypothesis": "it hurt more than i expected, and i thought i was prepared.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it hurt more than i expected and i thought i was prepared",
+ "normalized_hypothesis": "it hurt more than i expected and i thought i was prepared",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0104",
+ "category": "long",
+ "text": "The hardest part wasn't making the decision, it was sitting with the uncertainty of not knowing whether it was the right one for months afterward.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0104.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The hardest part was making the decision. It was sitting with the uncertainty of not knowing whether it was the right one for months afterward.",
+ "raw": {
+ "normalized_reference": "the hardest part wasn't making the decision, it was sitting with the uncertainty of not knowing whether it was the right one for months afterward.",
+ "normalized_hypothesis": "the hardest part was making the decision. it was sitting with the uncertainty of not knowing whether it was the right one for months afterward.",
+ "errors": 1,
+ "words": 25,
+ "wer": 0.04
+ },
+ "semantic": {
+ "normalized_reference": "the hardest part was not making the decision it was sitting with the uncertainty of not knowing whether it was the right one for months afterward",
+ "normalized_hypothesis": "the hardest part was making the decision it was sitting with the uncertainty of not knowing whether it was the right one for months afterward",
+ "errors": 1,
+ "words": 26,
+ "wer": 0.038461538461538464
+ }
+ },
+ {
+ "id": "release500_0105",
+ "category": "narrative",
+ "text": "There was this one day last summer when everything just clicked into place.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0105.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "There was this one day last summer when everything just clicked into place.",
+ "raw": {
+ "normalized_reference": "there was this one day last summer when everything just clicked into place.",
+ "normalized_hypothesis": "there was this one day last summer when everything just clicked into place.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "there was this one day last summer when everything just clicked into place",
+ "normalized_hypothesis": "there was this one day last summer when everything just clicked into place",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0106",
+ "category": "question",
+ "text": "Have you ever thought about why we keep doing things the same way when they stop working?",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0106.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Have you ever thought about why we keep doing things the same way when they stop working?",
+ "raw": {
+ "normalized_reference": "have you ever thought about why we keep doing things the same way when they stop working?",
+ "normalized_hypothesis": "have you ever thought about why we keep doing things the same way when they stop working?",
+ "errors": 0,
+ "words": 17,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "have you ever thought about why we keep doing things the same way when they stop working",
+ "normalized_hypothesis": "have you ever thought about why we keep doing things the same way when they stop working",
+ "errors": 0,
+ "words": 17,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0107",
+ "category": "short",
+ "text": "I can't believe you just said that.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0107.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I can't believe you just said that.",
+ "raw": {
+ "normalized_reference": "i can't believe you just said that.",
+ "normalized_hypothesis": "i can't believe you just said that.",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i can not believe you just said that",
+ "normalized_hypothesis": "i can not believe you just said that",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0108",
+ "category": "technical",
+ "text": "You'll want to isolate that variable before drawing any conclusions.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0108.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "You'll want to isolate that variable before drawing any conclusions.",
+ "raw": {
+ "normalized_reference": "you'll want to isolate that variable before drawing any conclusions.",
+ "normalized_hypothesis": "you'll want to isolate that variable before drawing any conclusions.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "you will want to isolate that variable before drawing any conclusions",
+ "normalized_hypothesis": "you will want to isolate that variable before drawing any conclusions",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0109",
+ "category": "conversational",
+ "text": "The meeting that could have been an email lasted two hours.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0109.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The meeting that could have been an email lasted two hours.",
+ "raw": {
+ "normalized_reference": "the meeting that could have been an email lasted two hours.",
+ "normalized_hypothesis": "the meeting that could have been an email lasted two hours.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the meeting that could have been an email lasted 2 hours",
+ "normalized_hypothesis": "the meeting that could have been an email lasted 2 hours",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0110",
+ "category": "descriptive",
+ "text": "There's a specific kind of tired that sleep doesn't fix.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0110.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "There's a specific kind of tire that sleep doesn't fix.",
+ "raw": {
+ "normalized_reference": "there's a specific kind of tired that sleep doesn't fix.",
+ "normalized_hypothesis": "there's a specific kind of tire that sleep doesn't fix.",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ },
+ "semantic": {
+ "normalized_reference": "there is a specific kind of tired that sleep does not fix",
+ "normalized_hypothesis": "there is a specific kind of tire that sleep does not fix",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ }
+ },
+ {
+ "id": "release500_0111",
+ "category": "emotional",
+ "text": "Something about that moment broke something open in me that I hadn't realized was closed.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0111.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Something about that moment broke something open in me that I hadn't realized was closed.",
+ "raw": {
+ "normalized_reference": "something about that moment broke something open in me that i hadn't realized was closed.",
+ "normalized_hypothesis": "something about that moment broke something open in me that i hadn't realized was closed.",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "something about that moment broke something open in me that i had not realized was closed",
+ "normalized_hypothesis": "something about that moment broke something open in me that i had not realized was closed",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0112",
+ "category": "long",
+ "text": "The thing about making something good is that you don't usually know it's good while you're making it. You just know you're not done yet.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0112.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The thing about making something good is that you don't usually know it's good while you're making it. You just know you're not done yet.",
+ "raw": {
+ "normalized_reference": "the thing about making something good is that you don't usually know it's good while you're making it. you just know you're not done yet.",
+ "normalized_hypothesis": "the thing about making something good is that you don't usually know it's good while you're making it. you just know you're not done yet.",
+ "errors": 0,
+ "words": 25,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the thing about making something good is that you do not usually know it is good while you are making it you just know you are not done yet",
+ "normalized_hypothesis": "the thing about making something good is that you do not usually know it is good while you are making it you just know you are not done yet",
+ "errors": 0,
+ "words": 29,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0113",
+ "category": "narrative",
+ "text": "She opened the door slowly, as if she expected to find something she wasn't ready for.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0113.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "She opened the door slowly, as if she expected to find something she wasn't ready for.",
+ "raw": {
+ "normalized_reference": "she opened the door slowly, as if she expected to find something she wasn't ready for.",
+ "normalized_hypothesis": "she opened the door slowly, as if she expected to find something she wasn't ready for.",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "she opened the door slowly as if she expected to find something she was not ready for",
+ "normalized_hypothesis": "she opened the door slowly as if she expected to find something she was not ready for",
+ "errors": 0,
+ "words": 17,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0114",
+ "category": "question",
+ "text": "What would you do differently if you could go back?",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0114.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "What would you do differently if you could throwback?",
+ "raw": {
+ "normalized_reference": "what would you do differently if you could go back?",
+ "normalized_hypothesis": "what would you do differently if you could throwback?",
+ "errors": 2,
+ "words": 10,
+ "wer": 0.2
+ },
+ "semantic": {
+ "normalized_reference": "what would you do differently if you could go back",
+ "normalized_hypothesis": "what would you do differently if you could throwback",
+ "errors": 2,
+ "words": 10,
+ "wer": 0.2
+ }
+ },
+ {
+ "id": "release500_0115",
+ "category": "short",
+ "text": "Okay. Okay okay okay.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0115.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Okay, okay, okay, okay.",
+ "raw": {
+ "normalized_reference": "okay. okay okay okay.",
+ "normalized_hypothesis": "okay, okay, okay, okay.",
+ "errors": 0,
+ "words": 4,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "okay okay okay okay",
+ "normalized_hypothesis": "okay okay okay okay",
+ "errors": 0,
+ "words": 4,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0116",
+ "category": "technical",
+ "text": "The latency goes up significantly at scale, which is expected.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0116.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The latency goes up significantly at scale, which is expected.",
+ "raw": {
+ "normalized_reference": "the latency goes up significantly at scale, which is expected.",
+ "normalized_hypothesis": "the latency goes up significantly at scale, which is expected.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the latency goes up significantly at scale which is expected",
+ "normalized_hypothesis": "the latency goes up significantly at scale which is expected",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0117",
+ "category": "conversational",
+ "text": "We ended up staying until two in the morning talking about nothing.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0117.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "We ended up staying until two in the morning talking about nothing.",
+ "raw": {
+ "normalized_reference": "we ended up staying until two in the morning talking about nothing.",
+ "normalized_hypothesis": "we ended up staying until two in the morning talking about nothing.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "we ended up staying until 2 in the morning talking about nothing",
+ "normalized_hypothesis": "we ended up staying until 2 in the morning talking about nothing",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0118",
+ "category": "descriptive",
+ "text": "The rain started slowly, and then all at once, the way it usually does.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0118.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The rain started slowly, and then all at once, the way it usually does.",
+ "raw": {
+ "normalized_reference": "the rain started slowly, and then all at once, the way it usually does.",
+ "normalized_hypothesis": "the rain started slowly, and then all at once, the way it usually does.",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the rain started slowly and then all at once the way it usually does",
+ "normalized_hypothesis": "the rain started slowly and then all at once the way it usually does",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0119",
+ "category": "emotional",
+ "text": "I honestly did not think I was going to make it through that year.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0119.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I honestly did not think I was going to make it through that year.",
+ "raw": {
+ "normalized_reference": "i honestly did not think i was going to make it through that year.",
+ "normalized_hypothesis": "i honestly did not think i was going to make it through that year.",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i honestly did not think i was going to make it through that year",
+ "normalized_hypothesis": "i honestly did not think i was going to make it through that year",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0120",
+ "category": "long",
+ "text": "There's an assumption that people who seem confident all the time actually feel confident, and in my experience that's almost never true.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0120.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "There's an assumption that people who seem confident all the time actually feel confident. And in my experience, that's almost never true.",
+ "raw": {
+ "normalized_reference": "there's an assumption that people who seem confident all the time actually feel confident, and in my experience that's almost never true.",
+ "normalized_hypothesis": "there's an assumption that people who seem confident all the time actually feel confident. and in my experience, that's almost never true.",
+ "errors": 0,
+ "words": 22,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "there is an assumption that people who seem confident all the time actually feel confident and in my experience that is almost never true",
+ "normalized_hypothesis": "there is an assumption that people who seem confident all the time actually feel confident and in my experience that is almost never true",
+ "errors": 0,
+ "words": 24,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0121",
+ "category": "narrative",
+ "text": "The moment I opened the door I knew something was different, even if I couldn't say what.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0121.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The moment I opened the door I knew something was different, even if I couldn't say what.",
+ "raw": {
+ "normalized_reference": "the moment i opened the door i knew something was different, even if i couldn't say what.",
+ "normalized_hypothesis": "the moment i opened the door i knew something was different, even if i couldn't say what.",
+ "errors": 0,
+ "words": 17,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the moment i opened the door i knew something was different even if i could not say what",
+ "normalized_hypothesis": "the moment i opened the door i knew something was different even if i could not say what",
+ "errors": 0,
+ "words": 18,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0122",
+ "category": "question",
+ "text": "How do you explain something that doesn't make sense even to you?",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0122.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "How do you explain something that doesn't make sense even to you?",
+ "raw": {
+ "normalized_reference": "how do you explain something that doesn't make sense even to you?",
+ "normalized_hypothesis": "how do you explain something that doesn't make sense even to you?",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "how do you explain something that does not make sense even to you",
+ "normalized_hypothesis": "how do you explain something that does not make sense even to you",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0123",
+ "category": "short",
+ "text": "Hold on a second.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0123.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Hold on a second.",
+ "raw": {
+ "normalized_reference": "hold on a second.",
+ "normalized_hypothesis": "hold on a second.",
+ "errors": 0,
+ "words": 4,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "hold on a 2nd",
+ "normalized_hypothesis": "hold on a 2nd",
+ "errors": 0,
+ "words": 4,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0124",
+ "category": "technical",
+ "text": "A deterministic split prevents duplicate normalized text from leaking into evaluation.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0124.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "A deterministic split prevents duplicate normalized text from leaking into evaluation.",
+ "raw": {
+ "normalized_reference": "a deterministic split prevents duplicate normalized text from leaking into evaluation.",
+ "normalized_hypothesis": "a deterministic split prevents duplicate normalized text from leaking into evaluation.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "a deterministic split prevents duplicate normalized text from leaking into evaluation",
+ "normalized_hypothesis": "a deterministic split prevents duplicate normalized text from leaking into evaluation",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0125",
+ "category": "conversational",
+ "text": "Honestly, that was much better than I expected, but the ending still felt rushed.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0125.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Honestly, that was much better than I expected, but the ending still felt rushed.",
+ "raw": {
+ "normalized_reference": "honestly, that was much better than i expected, but the ending still felt rushed.",
+ "normalized_hypothesis": "honestly, that was much better than i expected, but the ending still felt rushed.",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "honestly that was much better than i expected but the ending still felt rushed",
+ "normalized_hypothesis": "honestly that was much better than i expected but the ending still felt rushed",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0126",
+ "category": "descriptive",
+ "text": "She had a habit of tilting her head slightly when thinking hard about something.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0126.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "She had a habit of tilting her head slightly when thinking hard about something.",
+ "raw": {
+ "normalized_reference": "she had a habit of tilting her head slightly when thinking hard about something.",
+ "normalized_hypothesis": "she had a habit of tilting her head slightly when thinking hard about something.",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "she had a habit of tilting her head slightly when thinking hard about something",
+ "normalized_hypothesis": "she had a habit of tilting her head slightly when thinking hard about something",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0127",
+ "category": "emotional",
+ "text": "I cannot stress enough how much I did not want to be right about this.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0127.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I cannot stress enough how much I did not want to be right about this.",
+ "raw": {
+ "normalized_reference": "i cannot stress enough how much i did not want to be right about this.",
+ "normalized_hypothesis": "i cannot stress enough how much i did not want to be right about this.",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i cannot stress enough how much i did not want to be right about this",
+ "normalized_hypothesis": "i cannot stress enough how much i did not want to be right about this",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0128",
+ "category": "long",
+ "text": "I used to think the goal was to get to a point where nothing was hard anymore, and now I think that point doesn't exist and I'm not sure I'd want it to.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0128.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I used to think the goal was to get to a point where nothing was hard anymore, and now I think that point doesn't exist and I'm not sure I wanted to.",
+ "raw": {
+ "normalized_reference": "i used to think the goal was to get to a point where nothing was hard anymore, and now i think that point doesn't exist and i'm not sure i'd want it to.",
+ "normalized_hypothesis": "i used to think the goal was to get to a point where nothing was hard anymore, and now i think that point doesn't exist and i'm not sure i wanted to.",
+ "errors": 3,
+ "words": 33,
+ "wer": 0.09090909090909091
+ },
+ "semantic": {
+ "normalized_reference": "i used to think the goal was to get to a point where nothing was hard anymore and now i think that point does not exist and i am not sure i would want it to",
+ "normalized_hypothesis": "i used to think the goal was to get to a point where nothing was hard anymore and now i think that point does not exist and i am not sure i wanted to",
+ "errors": 3,
+ "words": 36,
+ "wer": 0.08333333333333333
+ }
+ },
+ {
+ "id": "release500_0129",
+ "category": "narrative",
+ "text": "So this is the part of the story where everything goes completely sideways.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0129.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "So this is the part of the story where everything goes completely sideways.",
+ "raw": {
+ "normalized_reference": "so this is the part of the story where everything goes completely sideways.",
+ "normalized_hypothesis": "so this is the part of the story where everything goes completely sideways.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "so this is the part of the story where everything goes completely sideways",
+ "normalized_hypothesis": "so this is the part of the story where everything goes completely sideways",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0130",
+ "category": "question",
+ "text": "Is there a version of this where it actually works out okay?",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0130.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Is there a version of this where it actually works out okay?",
+ "raw": {
+ "normalized_reference": "is there a version of this where it actually works out okay?",
+ "normalized_hypothesis": "is there a version of this where it actually works out okay?",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "is there a version of this where it actually works out okay",
+ "normalized_hypothesis": "is there a version of this where it actually works out okay",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0131",
+ "category": "short",
+ "text": "Oh no. Oh no no no.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0131.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Oh no, oh no no no.",
+ "raw": {
+ "normalized_reference": "oh no. oh no no no.",
+ "normalized_hypothesis": "oh no, oh no no no.",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "0 no 0 no no no",
+ "normalized_hypothesis": "0 no 0 no no no",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0132",
+ "category": "technical",
+ "text": "Always test on a copy before modifying the original.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0132.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Always test on a copy before modifying the original.",
+ "raw": {
+ "normalized_reference": "always test on a copy before modifying the original.",
+ "normalized_hypothesis": "always test on a copy before modifying the original.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "always test on a copy before modifying the original",
+ "normalized_hypothesis": "always test on a copy before modifying the original",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0133",
+ "category": "conversational",
+ "text": "I wasn't even supposed to be there. That's the wild part.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0133.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I wasn't even supposed to be there. That's the wild part.",
+ "raw": {
+ "normalized_reference": "i wasn't even supposed to be there. that's the wild part.",
+ "normalized_hypothesis": "i wasn't even supposed to be there. that's the wild part.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i was not even supposed to be there that is the wild part",
+ "normalized_hypothesis": "i was not even supposed to be there that is the wild part",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0134",
+ "category": "descriptive",
+ "text": "It's a small town. Not the kind you leave. The kind you leave and come back to.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0134.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It's a small town. Not the kind you leave. The kind you leave and come back to.",
+ "raw": {
+ "normalized_reference": "it's a small town. not the kind you leave. the kind you leave and come back to.",
+ "normalized_hypothesis": "it's a small town. not the kind you leave. the kind you leave and come back to.",
+ "errors": 0,
+ "words": 17,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it is a small town not the kind you leave the kind you leave and come back to",
+ "normalized_hypothesis": "it is a small town not the kind you leave the kind you leave and come back to",
+ "errors": 0,
+ "words": 18,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0135",
+ "category": "emotional",
+ "text": "She cried. I cried. We all just sort of stood there crying.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0135.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "She cried, I cried, we all just sort of stood there crying.",
+ "raw": {
+ "normalized_reference": "she cried. i cried. we all just sort of stood there crying.",
+ "normalized_hypothesis": "she cried, i cried, we all just sort of stood there crying.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "she cried i cried we all just sort of stood there crying",
+ "normalized_hypothesis": "she cried i cried we all just sort of stood there crying",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0136",
+ "category": "long",
+ "text": "There's a version of being busy that's actually just a very elaborate way of avoiding the one thing you know you should be doing.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0136.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "There's a version of being busy that's actually just a very elaborate way of avoiding the one thing you know you should be doing.",
+ "raw": {
+ "normalized_reference": "there's a version of being busy that's actually just a very elaborate way of avoiding the one thing you know you should be doing.",
+ "normalized_hypothesis": "there's a version of being busy that's actually just a very elaborate way of avoiding the one thing you know you should be doing.",
+ "errors": 0,
+ "words": 24,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "there is a version of being busy that is actually just a very elaborate way of avoiding the one thing you know you should be doing",
+ "normalized_hypothesis": "there is a version of being busy that is actually just a very elaborate way of avoiding the one thing you know you should be doing",
+ "errors": 0,
+ "words": 26,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0137",
+ "category": "narrative",
+ "text": "The letter arrived on a Tuesday, which felt appropriate somehow.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0137.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The letter arrived on a Tuesday, which felt appropriate somehow.",
+ "raw": {
+ "normalized_reference": "the letter arrived on a tuesday, which felt appropriate somehow.",
+ "normalized_hypothesis": "the letter arrived on a tuesday, which felt appropriate somehow.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the letter arrived on a tuesday which felt appropriate somehow",
+ "normalized_hypothesis": "the letter arrived on a tuesday which felt appropriate somehow",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0138",
+ "category": "question",
+ "text": "What exactly did she say when you told her?",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0138.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "What exactly did she say when you told her?",
+ "raw": {
+ "normalized_reference": "what exactly did she say when you told her?",
+ "normalized_hypothesis": "what exactly did she say when you told her?",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "what exactly did she say when you told her",
+ "normalized_hypothesis": "what exactly did she say when you told her",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0139",
+ "category": "short",
+ "text": "No way. Absolutely not.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0139.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "No way, absolutely not.",
+ "raw": {
+ "normalized_reference": "no way. absolutely not.",
+ "normalized_hypothesis": "no way, absolutely not.",
+ "errors": 0,
+ "words": 4,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "no way absolutely not",
+ "normalized_hypothesis": "no way absolutely not",
+ "errors": 0,
+ "words": 4,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0140",
+ "category": "technical",
+ "text": "The first thing you want to do is make sure everything is backed up.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0140.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The first thing you want to do is make sure everything is backed up.",
+ "raw": {
+ "normalized_reference": "the first thing you want to do is make sure everything is backed up.",
+ "normalized_hypothesis": "the first thing you want to do is make sure everything is backed up.",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the 1st thing you want to do is make sure everything is backed up",
+ "normalized_hypothesis": "the 1st thing you want to do is make sure everything is backed up",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0141",
+ "category": "conversational",
+ "text": "I have the most insane idea and you need to just hear me out.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0141.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I have the most insane idea, and you need to just hear me out.",
+ "raw": {
+ "normalized_reference": "i have the most insane idea and you need to just hear me out.",
+ "normalized_hypothesis": "i have the most insane idea, and you need to just hear me out.",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i have the most insane idea and you need to just hear me out",
+ "normalized_hypothesis": "i have the most insane idea and you need to just hear me out",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0142",
+ "category": "descriptive",
+ "text": "There's a coffee shop on the corner that's been there longer than anyone can remember.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0142.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "There's a coffee shop on the corner that's been there longer than anyone can remember.",
+ "raw": {
+ "normalized_reference": "there's a coffee shop on the corner that's been there longer than anyone can remember.",
+ "normalized_hypothesis": "there's a coffee shop on the corner that's been there longer than anyone can remember.",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "there is a coffee shop on the corner that has been there longer than anyone can remember",
+ "normalized_hypothesis": "there is a coffee shop on the corner that has been there longer than anyone can remember",
+ "errors": 0,
+ "words": 17,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0143",
+ "category": "emotional",
+ "text": "I kept it together until I got to the car, and then I completely fell apart.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0143.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I kept it together until I got to the car, and then I completely fell apart.",
+ "raw": {
+ "normalized_reference": "i kept it together until i got to the car, and then i completely fell apart.",
+ "normalized_hypothesis": "i kept it together until i got to the car, and then i completely fell apart.",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i kept it together until i got to the car and then i completely fell apart",
+ "normalized_hypothesis": "i kept it together until i got to the car and then i completely fell apart",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0144",
+ "category": "long",
+ "text": "The older I get, the more I think that most of what we call wisdom is just pattern recognition built from having been wrong enough times to start noticing.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0144.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The older I get, the more I think that most of what we call wisdom is just pattern recognition built from having been wrong enough times to start noticing.",
+ "raw": {
+ "normalized_reference": "the older i get, the more i think that most of what we call wisdom is just pattern recognition built from having been wrong enough times to start noticing.",
+ "normalized_hypothesis": "the older i get, the more i think that most of what we call wisdom is just pattern recognition built from having been wrong enough times to start noticing.",
+ "errors": 0,
+ "words": 29,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the older i get the more i think that most of what we call wisdom is just pattern recognition built from having been wrong enough times to start noticing",
+ "normalized_hypothesis": "the older i get the more i think that most of what we call wisdom is just pattern recognition built from having been wrong enough times to start noticing",
+ "errors": 0,
+ "words": 29,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0145",
+ "category": "narrative",
+ "text": "I remember the exact moment I realized I had made a huge mistake.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0145.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I remember the exact moment I realized I had made a huge mistake.",
+ "raw": {
+ "normalized_reference": "i remember the exact moment i realized i had made a huge mistake.",
+ "normalized_hypothesis": "i remember the exact moment i realized i had made a huge mistake.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i remember the exact moment i realized i had made a huge mistake",
+ "normalized_hypothesis": "i remember the exact moment i realized i had made a huge mistake",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0146",
+ "category": "question",
+ "text": "Do you think it would have gone differently if we'd left earlier?",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0146.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Do you think it would have gone differently if we'd left earlier?",
+ "raw": {
+ "normalized_reference": "do you think it would have gone differently if we'd left earlier?",
+ "normalized_hypothesis": "do you think it would have gone differently if we'd left earlier?",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "do you think it would have gone differently if we would left earlier",
+ "normalized_hypothesis": "do you think it would have gone differently if we would left earlier",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0147",
+ "category": "short",
+ "text": "You have to be kidding me.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0147.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "You have to be kidding me.",
+ "raw": {
+ "normalized_reference": "you have to be kidding me.",
+ "normalized_hypothesis": "you have to be kidding me.",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "you have to be kidding me",
+ "normalized_hypothesis": "you have to be kidding me",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0148",
+ "category": "technical",
+ "text": "If the checksum fails, restart the worker process without deleting the cached tensors.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0148.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "If the checksum fails, restart the worker process without deleting the cache tensors.",
+ "raw": {
+ "normalized_reference": "if the checksum fails, restart the worker process without deleting the cached tensors.",
+ "normalized_hypothesis": "if the checksum fails, restart the worker process without deleting the cache tensors.",
+ "errors": 1,
+ "words": 13,
+ "wer": 0.07692307692307693
+ },
+ "semantic": {
+ "normalized_reference": "if the checksum fails restart the worker process without deleting the cached tensors",
+ "normalized_hypothesis": "if the checksum fails restart the worker process without deleting the cache tensors",
+ "errors": 1,
+ "words": 13,
+ "wer": 0.07692307692307693
+ }
+ },
+ {
+ "id": "release500_0149",
+ "category": "conversational",
+ "text": "I need you to understand how close that was to not working.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0149.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I need you to understand how close that was to not working.",
+ "raw": {
+ "normalized_reference": "i need you to understand how close that was to not working.",
+ "normalized_hypothesis": "i need you to understand how close that was to not working.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i need you to understand how close that was to not working",
+ "normalized_hypothesis": "i need you to understand how close that was to not working",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0150",
+ "category": "descriptive",
+ "text": "It was the sort of morning where the whole world looks rinsed clean.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0150.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It was the sort of morning where the whole world looks rinsed clean,",
+ "raw": {
+ "normalized_reference": "it was the sort of morning where the whole world looks rinsed clean.",
+ "normalized_hypothesis": "it was the sort of morning where the whole world looks rinsed clean,",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it was the sort of morning where the whole world looks rinsed clean",
+ "normalized_hypothesis": "it was the sort of morning where the whole world looks rinsed clean",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0151",
+ "category": "emotional",
+ "text": "She started laughing and I started laughing and we couldn't stop for a full minute.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0151.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "She started laughing and I started laughing and we couldn't stop for a full minute.",
+ "raw": {
+ "normalized_reference": "she started laughing and i started laughing and we couldn't stop for a full minute.",
+ "normalized_hypothesis": "she started laughing and i started laughing and we couldn't stop for a full minute.",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "she started laughing and i started laughing and we could not stop for a full minute",
+ "normalized_hypothesis": "she started laughing and i started laughing and we could not stop for a full minute",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0152",
+ "category": "long",
+ "text": "We spent three years building something we were genuinely proud of, and then the market shifted and none of it mattered the way we thought it would.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0152.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "We spent three years building something we were genuinely proud of, and then the market shifted, and none of it mattered the way we thought it would.",
+ "raw": {
+ "normalized_reference": "we spent three years building something we were genuinely proud of, and then the market shifted and none of it mattered the way we thought it would.",
+ "normalized_hypothesis": "we spent three years building something we were genuinely proud of, and then the market shifted, and none of it mattered the way we thought it would.",
+ "errors": 0,
+ "words": 27,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "we spent 3 years building something we were genuinely proud of and then the market shifted and none of it mattered the way we thought it would",
+ "normalized_hypothesis": "we spent 3 years building something we were genuinely proud of and then the market shifted and none of it mattered the way we thought it would",
+ "errors": 0,
+ "words": 27,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0153",
+ "category": "narrative",
+ "text": "And somehow, despite all of that, it actually turned out okay.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0153.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "And somehow, despite all of that, it actually turned out okay.",
+ "raw": {
+ "normalized_reference": "and somehow, despite all of that, it actually turned out okay.",
+ "normalized_hypothesis": "and somehow, despite all of that, it actually turned out okay.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "and somehow despite all of that it actually turned out okay",
+ "normalized_hypothesis": "and somehow despite all of that it actually turned out okay",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0154",
+ "category": "question",
+ "text": "Is that something you actually want, or something you think you should want?",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0154.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Is that something you actually want? Or something you think you should want?",
+ "raw": {
+ "normalized_reference": "is that something you actually want, or something you think you should want?",
+ "normalized_hypothesis": "is that something you actually want? or something you think you should want?",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "is that something you actually want or something you think you should want",
+ "normalized_hypothesis": "is that something you actually want or something you think you should want",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0155",
+ "category": "short",
+ "text": "This is fine.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0155.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "This is fine.",
+ "raw": {
+ "normalized_reference": "this is fine.",
+ "normalized_hypothesis": "this is fine.",
+ "errors": 0,
+ "words": 3,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "this is fine",
+ "normalized_hypothesis": "this is fine",
+ "errors": 0,
+ "words": 3,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0156",
+ "category": "technical",
+ "text": "The tricky part isn't the setup, it's making sure all dependencies line up.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0156.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The tricky part isn't the setup, it's making sure all dependencies line up.",
+ "raw": {
+ "normalized_reference": "the tricky part isn't the setup, it's making sure all dependencies line up.",
+ "normalized_hypothesis": "the tricky part isn't the setup, it's making sure all dependencies line up.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the tricky part is not the setup it is making sure all dependencies line up",
+ "normalized_hypothesis": "the tricky part is not the setup it is making sure all dependencies line up",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0157",
+ "category": "conversational",
+ "text": "He called it a minor setback. I called it a disaster.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0157.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He called it a minor setback. I called it a disaster.",
+ "raw": {
+ "normalized_reference": "he called it a minor setback. i called it a disaster.",
+ "normalized_hypothesis": "he called it a minor setback. i called it a disaster.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he called it a minor setback i called it a disaster",
+ "normalized_hypothesis": "he called it a minor setback i called it a disaster",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0158",
+ "category": "descriptive",
+ "text": "The whole place smelled like pine and old wood and cold air.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0158.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The whole place smelled like pine and old wood and cold air.",
+ "raw": {
+ "normalized_reference": "the whole place smelled like pine and old wood and cold air.",
+ "normalized_hypothesis": "the whole place smelled like pine and old wood and cold air.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the whole place smelled like pine and old wood and cold air",
+ "normalized_hypothesis": "the whole place smelled like pine and old wood and cold air",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0159",
+ "category": "emotional",
+ "text": "I've never been so relieved in my entire life.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0159.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I've never been so relieved in my entire life.",
+ "raw": {
+ "normalized_reference": "i've never been so relieved in my entire life.",
+ "normalized_hypothesis": "i've never been so relieved in my entire life.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i have never been so relieved in my entire life",
+ "normalized_hypothesis": "i have never been so relieved in my entire life",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0160",
+ "category": "narrative",
+ "text": "I wrote it all down that night so I wouldn't forget, and I still haven't looked at it.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0160.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I wrote it all down that night so I wouldn't forget, and I still haven't looked at it.",
+ "raw": {
+ "normalized_reference": "i wrote it all down that night so i wouldn't forget, and i still haven't looked at it.",
+ "normalized_hypothesis": "i wrote it all down that night so i wouldn't forget, and i still haven't looked at it.",
+ "errors": 0,
+ "words": 18,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i wrote it all down that night so i would not forget and i still have not looked at it",
+ "normalized_hypothesis": "i wrote it all down that night so i would not forget and i still have not looked at it",
+ "errors": 0,
+ "words": 20,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0161",
+ "category": "short",
+ "text": "That's kind of wild actually.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0161.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "That's kind of wild actually.",
+ "raw": {
+ "normalized_reference": "that's kind of wild actually.",
+ "normalized_hypothesis": "that's kind of wild actually.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "that is kind of wild actually",
+ "normalized_hypothesis": "that is kind of wild actually",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0162",
+ "category": "technical",
+ "text": "Version pinning will save you a lot of pain later down the line.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0162.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Version pinning will save you a lot of pain later down the line.",
+ "raw": {
+ "normalized_reference": "version pinning will save you a lot of pain later down the line.",
+ "normalized_hypothesis": "version pinning will save you a lot of pain later down the line.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "version pinning will save you a lot of pain later down the line",
+ "normalized_hypothesis": "version pinning will save you a lot of pain later down the line",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0163",
+ "category": "conversational",
+ "text": "At some point you just have to commit and see what happens.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0163.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "At some point you just have to commit and see what happens.",
+ "raw": {
+ "normalized_reference": "at some point you just have to commit and see what happens.",
+ "normalized_hypothesis": "at some point you just have to commit and see what happens.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "at some point you just have to commit and see what happens",
+ "normalized_hypothesis": "at some point you just have to commit and see what happens",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0164",
+ "category": "emotional",
+ "text": "Nobody told me it was going to be this hard, and honestly, I'm grateful.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0164.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Nobody told me it was going to be this hard, and honestly, I'm grateful.",
+ "raw": {
+ "normalized_reference": "nobody told me it was going to be this hard, and honestly, i'm grateful.",
+ "normalized_hypothesis": "nobody told me it was going to be this hard, and honestly, i'm grateful.",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "nobody told me it was going to be this hard and honestly i am grateful",
+ "normalized_hypothesis": "nobody told me it was going to be this hard and honestly i am grateful",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0165",
+ "category": "narrative",
+ "text": "Chapter seven. In which everything that could go wrong did, and a few things that couldn't.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0165.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Chapter 7. In which everything that could go wrong did, and a few things that couldn't.",
+ "raw": {
+ "normalized_reference": "chapter seven. in which everything that could go wrong did, and a few things that couldn't.",
+ "normalized_hypothesis": "chapter 7. in which everything that could go wrong did, and a few things that couldn't.",
+ "errors": 1,
+ "words": 16,
+ "wer": 0.0625
+ },
+ "semantic": {
+ "normalized_reference": "chapter 7 in which everything that could go wrong did and a few things that could not",
+ "normalized_hypothesis": "chapter 7 in which everything that could go wrong did and a few things that could not",
+ "errors": 0,
+ "words": 17,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0166",
+ "category": "short",
+ "text": "Wait, what?",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0166.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Wait, what?",
+ "raw": {
+ "normalized_reference": "wait, what?",
+ "normalized_hypothesis": "wait, what?",
+ "errors": 0,
+ "words": 2,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "wait what",
+ "normalized_hypothesis": "wait what",
+ "errors": 0,
+ "words": 2,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0167",
+ "category": "technical",
+ "text": "The graphics processor remained idle because the data loader exhausted every central processing core.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0167.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The graphics processor remained idle because the data loader exhausted every central processing core.",
+ "raw": {
+ "normalized_reference": "the graphics processor remained idle because the data loader exhausted every central processing core.",
+ "normalized_hypothesis": "the graphics processor remained idle because the data loader exhausted every central processing core.",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the graphics processor remained idle because the data loader exhausted every central processing core",
+ "normalized_hypothesis": "the graphics processor remained idle because the data loader exhausted every central processing core",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0168",
+ "category": "conversational",
+ "text": "Could you speak a little more slowly? I am trying to write every detail down.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0168.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Could you speak a little more slowly? I am trying to write every detail down.",
+ "raw": {
+ "normalized_reference": "could you speak a little more slowly? i am trying to write every detail down.",
+ "normalized_hypothesis": "could you speak a little more slowly? i am trying to write every detail down.",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "could you speak a little more slowly i am trying to write every detail down",
+ "normalized_hypothesis": "could you speak a little more slowly i am trying to write every detail down",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0169",
+ "category": "emotional",
+ "text": "I was so excited I could barely form a complete sentence.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0169.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I was so excited I could barely form a complete sentence.",
+ "raw": {
+ "normalized_reference": "i was so excited i could barely form a complete sentence.",
+ "normalized_hypothesis": "i was so excited i could barely form a complete sentence.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i was so excited i could barely form a complete sentence",
+ "normalized_hypothesis": "i was so excited i could barely form a complete sentence",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0170",
+ "category": "narrative",
+ "text": "He looked at me like I had said something in a language he didn't speak.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0170.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He looked at me like I had said something in a language he didn't speak.",
+ "raw": {
+ "normalized_reference": "he looked at me like i had said something in a language he didn't speak.",
+ "normalized_hypothesis": "he looked at me like i had said something in a language he didn't speak.",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he looked at me like i had said something in a language he did not speak",
+ "normalized_hypothesis": "he looked at me like i had said something in a language he did not speak",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0171",
+ "category": "short",
+ "text": "That's actually hilarious.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0171.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "that's actually hilarious",
+ "raw": {
+ "normalized_reference": "that's actually hilarious.",
+ "normalized_hypothesis": "that's actually hilarious",
+ "errors": 0,
+ "words": 3,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "that is actually hilarious",
+ "normalized_hypothesis": "that is actually hilarious",
+ "errors": 0,
+ "words": 4,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0172",
+ "category": "technical",
+ "text": "The encoder maps phonemes, stress, and phrase boundaries into a shared hidden representation.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0172.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The encoder maps phonemes, stress, and phrase boundaries into a shared hidden representation.",
+ "raw": {
+ "normalized_reference": "the encoder maps phonemes, stress, and phrase boundaries into a shared hidden representation.",
+ "normalized_hypothesis": "the encoder maps phonemes, stress, and phrase boundaries into a shared hidden representation.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the encoder maps phonemes stress and phrase boundaries into a shared hidden representation",
+ "normalized_hypothesis": "the encoder maps phonemes stress and phrase boundaries into a shared hidden representation",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0173",
+ "category": "conversational",
+ "text": "She showed up three hours late and acted like nothing had happened.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0173.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "She showed up three hours late and acted like nothing had happened.",
+ "raw": {
+ "normalized_reference": "she showed up three hours late and acted like nothing had happened.",
+ "normalized_hypothesis": "she showed up three hours late and acted like nothing had happened.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "she showed up 3 hours late and acted like nothing had happened",
+ "normalized_hypothesis": "she showed up 3 hours late and acted like nothing had happened",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0174",
+ "category": "emotional",
+ "text": "Oh my god. This is the best thing I've ever tasted.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0174.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Oh my god, this is the best thing I've ever tasted.",
+ "raw": {
+ "normalized_reference": "oh my god. this is the best thing i've ever tasted.",
+ "normalized_hypothesis": "oh my god, this is the best thing i've ever tasted.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "0 my god this is the best thing i have ever tasted",
+ "normalized_hypothesis": "0 my god this is the best thing i have ever tasted",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0175",
+ "category": "narrative",
+ "text": "The note was short, which was somehow more frightening than a long one would have been.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0175.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The note was short, which was somehow more frightening than a long one would have been.",
+ "raw": {
+ "normalized_reference": "the note was short, which was somehow more frightening than a long one would have been.",
+ "normalized_hypothesis": "the note was short, which was somehow more frightening than a long one would have been.",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the note was short which was somehow more frightening than a long one would have been",
+ "normalized_hypothesis": "the note was short which was somehow more frightening than a long one would have been",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0176",
+ "category": "short",
+ "text": "Noted.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0176.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Noted.",
+ "raw": {
+ "normalized_reference": "noted.",
+ "normalized_hypothesis": "noted.",
+ "errors": 0,
+ "words": 1,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "noted",
+ "normalized_hypothesis": "noted",
+ "errors": 0,
+ "words": 1,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0177",
+ "category": "technical",
+ "text": "It's a simple process once you understand the underlying structure.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0177.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It's a simple process once you understand the underlying structure.",
+ "raw": {
+ "normalized_reference": "it's a simple process once you understand the underlying structure.",
+ "normalized_hypothesis": "it's a simple process once you understand the underlying structure.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it is a simple process once you understand the underlying structure",
+ "normalized_hypothesis": "it is a simple process once you understand the underlying structure",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0178",
+ "category": "conversational",
+ "text": "You have to hear this story because it's almost too good to be true.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0178.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "You have to hear this story because it's almost too good to be true.",
+ "raw": {
+ "normalized_reference": "you have to hear this story because it's almost too good to be true.",
+ "normalized_hypothesis": "you have to hear this story because it's almost too good to be true.",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "you have to hear this story because it is almost too good to be true",
+ "normalized_hypothesis": "you have to hear this story because it is almost too good to be true",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0179",
+ "category": "emotional",
+ "text": "I have never felt more seen in my entire life than in that moment.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0179.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I have never felt more seen in my entire life than in that moment.",
+ "raw": {
+ "normalized_reference": "i have never felt more seen in my entire life than in that moment.",
+ "normalized_hypothesis": "i have never felt more seen in my entire life than in that moment.",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i have never felt more seen in my entire life than in that moment",
+ "normalized_hypothesis": "i have never felt more seen in my entire life than in that moment",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0180",
+ "category": "narrative",
+ "text": "She knew. She had always known. She just waited for the right moment.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0180.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "She knew. She had always known. She just waited for the right moment.",
+ "raw": {
+ "normalized_reference": "she knew. she had always known. she just waited for the right moment.",
+ "normalized_hypothesis": "she knew. she had always known. she just waited for the right moment.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "she knew she had always known she just waited for the right moment",
+ "normalized_hypothesis": "she knew she had always known she just waited for the right moment",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0181",
+ "category": "short",
+ "text": "Yeah, no. Hard pass.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0181.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Yay, no. Hot bass.",
+ "raw": {
+ "normalized_reference": "yeah, no. hard pass.",
+ "normalized_hypothesis": "yay, no. hot bass.",
+ "errors": 3,
+ "words": 4,
+ "wer": 0.75
+ },
+ "semantic": {
+ "normalized_reference": "yeah no hard pass",
+ "normalized_hypothesis": "yay no hot bass",
+ "errors": 3,
+ "words": 4,
+ "wer": 0.75
+ }
+ },
+ {
+ "id": "release500_0182",
+ "category": "technical",
+ "text": "Please verify the sample rate, hop length, and mel filter configuration before inference.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0182.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Please verify the sample rate, hop length, and MEL filter configuration before inference.",
+ "raw": {
+ "normalized_reference": "please verify the sample rate, hop length, and mel filter configuration before inference.",
+ "normalized_hypothesis": "please verify the sample rate, hop length, and mel filter configuration before inference.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "please verify the sample rate hop length and mel filter configuration before inference",
+ "normalized_hypothesis": "please verify the sample rate hop length and mel filter configuration before inference",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0183",
+ "category": "conversational",
+ "text": "That's the worst idea I've ever heard and I'm completely in.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0183.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "That's the worst idea I've ever heard and I'm completely in.",
+ "raw": {
+ "normalized_reference": "that's the worst idea i've ever heard and i'm completely in.",
+ "normalized_hypothesis": "that's the worst idea i've ever heard and i'm completely in.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "that is the worst idea i have ever heard and i am completely in",
+ "normalized_hypothesis": "that is the worst idea i have ever heard and i am completely in",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0184",
+ "category": "emotional",
+ "text": "I remember thinking: this is it. This is the thing I've been waiting for.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0184.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I remember thinking, this is it. This is the thing I've been waiting for.",
+ "raw": {
+ "normalized_reference": "i remember thinking: this is it. this is the thing i've been waiting for.",
+ "normalized_hypothesis": "i remember thinking, this is it. this is the thing i've been waiting for.",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i remember thinking this is it this is the thing i have been waiting for",
+ "normalized_hypothesis": "i remember thinking this is it this is the thing i have been waiting for",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0185",
+ "category": "narrative",
+ "text": "We never talked about it again, which I think was the right call.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0185.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "We never talked about it again, which I think was the right call.",
+ "raw": {
+ "normalized_reference": "we never talked about it again, which i think was the right call.",
+ "normalized_hypothesis": "we never talked about it again, which i think was the right call.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "we never talked about it again which i think was the right call",
+ "normalized_hypothesis": "we never talked about it again which i think was the right call",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0186",
+ "category": "short",
+ "text": "In hindsight, obvious.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0186.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "In hindsight, obvious.",
+ "raw": {
+ "normalized_reference": "in hindsight, obvious.",
+ "normalized_hypothesis": "in hindsight, obvious.",
+ "errors": 0,
+ "words": 3,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "in hindsight obvious",
+ "normalized_hypothesis": "in hindsight obvious",
+ "errors": 0,
+ "words": 3,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0187",
+ "category": "technical",
+ "text": "Once it's configured correctly, it basically runs itself.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0187.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Once it's configured correctly, it basically runs itself.",
+ "raw": {
+ "normalized_reference": "once it's configured correctly, it basically runs itself.",
+ "normalized_hypothesis": "once it's configured correctly, it basically runs itself.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "once it is configured correctly it basically runs itself",
+ "normalized_hypothesis": "once it is configured correctly it basically runs itself",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0188",
+ "category": "conversational",
+ "text": "You got engaged? When? How? Tell me everything right now.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0188.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "You got engaged. Wang. Au. Tell me everything right now.",
+ "raw": {
+ "normalized_reference": "you got engaged? when? how? tell me everything right now.",
+ "normalized_hypothesis": "you got engaged. wang. au. tell me everything right now.",
+ "errors": 2,
+ "words": 10,
+ "wer": 0.2
+ },
+ "semantic": {
+ "normalized_reference": "you got engaged when how tell me everything right now",
+ "normalized_hypothesis": "you got engaged wang au tell me everything right now",
+ "errors": 2,
+ "words": 10,
+ "wer": 0.2
+ }
+ },
+ {
+ "id": "release500_0189",
+ "category": "emotional",
+ "text": "It's hard to explain how much that moment meant to me.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0189.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It's hard to explain how much that moment meant to me.",
+ "raw": {
+ "normalized_reference": "it's hard to explain how much that moment meant to me.",
+ "normalized_hypothesis": "it's hard to explain how much that moment meant to me.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it is hard to explain how much that moment meant to me",
+ "normalized_hypothesis": "it is hard to explain how much that moment meant to me",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0190",
+ "category": "narrative",
+ "text": "He told the story three different times that night and somehow it got better each time.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0190.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He told the story three different times that night, and somehow it got better each time.",
+ "raw": {
+ "normalized_reference": "he told the story three different times that night and somehow it got better each time.",
+ "normalized_hypothesis": "he told the story three different times that night, and somehow it got better each time.",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he told the story 3 different times that night and somehow it got better each time",
+ "normalized_hypothesis": "he told the story 3 different times that night and somehow it got better each time",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0191",
+ "category": "short",
+ "text": "Right. Moving on.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0191.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Right, moving on.",
+ "raw": {
+ "normalized_reference": "right. moving on.",
+ "normalized_hypothesis": "right, moving on.",
+ "errors": 0,
+ "words": 3,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "right moving on",
+ "normalized_hypothesis": "right moving on",
+ "errors": 0,
+ "words": 3,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0192",
+ "category": "conversational",
+ "text": "No, I did not move your keys; check beside the small ceramic bowl.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0192.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "No, I did not move your keys. Check beside the small ceramic bowl.",
+ "raw": {
+ "normalized_reference": "no, i did not move your keys; check beside the small ceramic bowl.",
+ "normalized_hypothesis": "no, i did not move your keys. check beside the small ceramic bowl.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "no i did not move your keys check beside the small ceramic bowl",
+ "normalized_hypothesis": "no i did not move your keys check beside the small ceramic bowl",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0193",
+ "category": "emotional",
+ "text": "I was devastated. I'm not going to pretend otherwise.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0193.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I was devastated. I'm not going to pretend otherwise.",
+ "raw": {
+ "normalized_reference": "i was devastated. i'm not going to pretend otherwise.",
+ "normalized_hypothesis": "i was devastated. i'm not going to pretend otherwise.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i was devastated i am not going to pretend otherwise",
+ "normalized_hypothesis": "i was devastated i am not going to pretend otherwise",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0194",
+ "category": "narrative",
+ "text": "That was the last summer before everything changed, though we didn't know it yet.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0194.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "That was the last summer before everything changed, though we didn't know it yet.",
+ "raw": {
+ "normalized_reference": "that was the last summer before everything changed, though we didn't know it yet.",
+ "normalized_hypothesis": "that was the last summer before everything changed, though we didn't know it yet.",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "that was the last summer before everything changed though we did not know it yet",
+ "normalized_hypothesis": "that was the last summer before everything changed though we did not know it yet",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0195",
+ "category": "short",
+ "text": "Bold of you to assume.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0195.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Bold of you to assume.",
+ "raw": {
+ "normalized_reference": "bold of you to assume.",
+ "normalized_hypothesis": "bold of you to assume.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "bold of you to assume",
+ "normalized_hypothesis": "bold of you to assume",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0196",
+ "category": "conversational",
+ "text": "It's one of those decisions that makes total sense in the moment.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0196.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It's one of those decisions that makes total sense in the moment.",
+ "raw": {
+ "normalized_reference": "it's one of those decisions that makes total sense in the moment.",
+ "normalized_hypothesis": "it's one of those decisions that makes total sense in the moment.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it is one of those decisions that makes total sense in the moment",
+ "normalized_hypothesis": "it is one of those decisions that makes total sense in the moment",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0197",
+ "category": "emotional",
+ "text": "I was furious. Not the loud kind. The quiet kind that lasts for days.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0197.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I was furious, not the loud kind, the quiet kind that lasts for days.",
+ "raw": {
+ "normalized_reference": "i was furious. not the loud kind. the quiet kind that lasts for days.",
+ "normalized_hypothesis": "i was furious, not the loud kind, the quiet kind that lasts for days.",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i was furious not the loud kind the quiet kind that lasts for days",
+ "normalized_hypothesis": "i was furious not the loud kind the quiet kind that lasts for days",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0198",
+ "category": "narrative",
+ "text": "That's the part of the story I usually leave out.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0198.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "That's the part of the story I usually leave out.",
+ "raw": {
+ "normalized_reference": "that's the part of the story i usually leave out.",
+ "normalized_hypothesis": "that's the part of the story i usually leave out.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "that is the part of the story i usually leave out",
+ "normalized_hypothesis": "that is the part of the story i usually leave out",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0199",
+ "category": "short",
+ "text": "I've said too much.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0199.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I've said too much.",
+ "raw": {
+ "normalized_reference": "i've said too much.",
+ "normalized_hypothesis": "i've said too much.",
+ "errors": 0,
+ "words": 4,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i have said too much",
+ "normalized_hypothesis": "i have said too much",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0200",
+ "category": "conversational",
+ "text": "I've been faking that I understood this the entire time.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0200.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I've been faking that I understood this the entire time.",
+ "raw": {
+ "normalized_reference": "i've been faking that i understood this the entire time.",
+ "normalized_hypothesis": "i've been faking that i understood this the entire time.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i have been faking that i understood this the entire time",
+ "normalized_hypothesis": "i have been faking that i understood this the entire time",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0201",
+ "category": "phonetic_balance",
+ "text": "Peace be unto you and grace before the Lord.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0201.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Peace be unto you, and grace before the Lord.",
+ "raw": {
+ "normalized_reference": "peace be unto you and grace before the lord.",
+ "normalized_hypothesis": "peace be unto you, and grace before the lord.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "peace be unto you and grace before the lord",
+ "normalized_hypothesis": "peace be unto you and grace before the lord",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0202",
+ "category": "phonetic_balance",
+ "text": "From now on we're pals.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0202.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "From now on, we're pals.",
+ "raw": {
+ "normalized_reference": "from now on we're pals.",
+ "normalized_hypothesis": "from now on, we're pals.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "from now on we are pals",
+ "normalized_hypothesis": "from now on we are pals",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0203",
+ "category": "phonetic_balance",
+ "text": "Here we allow our solicitors to look after our legal work.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0203.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Here we allow our solicitors to look after our legal work",
+ "raw": {
+ "normalized_reference": "here we allow our solicitors to look after our legal work.",
+ "normalized_hypothesis": "here we allow our solicitors to look after our legal work",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "here we allow our solicitors to look after our legal work",
+ "normalized_hypothesis": "here we allow our solicitors to look after our legal work",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0204",
+ "category": "phonetic_balance",
+ "text": "Her words sent a strange chill through Philip.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0204.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Her words sent a strange chill through Philip.",
+ "raw": {
+ "normalized_reference": "her words sent a strange chill through philip.",
+ "normalized_hypothesis": "her words sent a strange chill through philip.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "her words sent a strange chill through philip",
+ "normalized_hypothesis": "her words sent a strange chill through philip",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0205",
+ "category": "phonetic_balance",
+ "text": "It was introduced by Representative Dick of Ohio.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0205.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It was introduced by Representative Dick of Ohio",
+ "raw": {
+ "normalized_reference": "it was introduced by representative dick of ohio.",
+ "normalized_hypothesis": "it was introduced by representative dick of ohio",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it was introduced by representative dick of ohio",
+ "normalized_hypothesis": "it was introduced by representative dick of ohio",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0206",
+ "category": "phonetic_balance",
+ "text": "It's the aurora borealis.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0206.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It's the aura borealis.",
+ "raw": {
+ "normalized_reference": "it's the aurora borealis.",
+ "normalized_hypothesis": "it's the aura borealis.",
+ "errors": 1,
+ "words": 4,
+ "wer": 0.25
+ },
+ "semantic": {
+ "normalized_reference": "it is the aurora borealis",
+ "normalized_hypothesis": "it is the aura borealis",
+ "errors": 1,
+ "words": 5,
+ "wer": 0.2
+ }
+ },
+ {
+ "id": "release500_0207",
+ "category": "phonetic_balance",
+ "text": "The boy grew and prospered.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0207.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The boy grew and prospered.",
+ "raw": {
+ "normalized_reference": "the boy grew and prospered.",
+ "normalized_hypothesis": "the boy grew and prospered.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the boy grew and prospered",
+ "normalized_hypothesis": "the boy grew and prospered",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0208",
+ "category": "phonetic_balance",
+ "text": "His voice was passionately rebellious.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0208.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "His voice was passionately rebellious.",
+ "raw": {
+ "normalized_reference": "his voice was passionately rebellious.",
+ "normalized_hypothesis": "his voice was passionately rebellious.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "his voice was passionately rebellious",
+ "normalized_hypothesis": "his voice was passionately rebellious",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0209",
+ "category": "phonetic_balance",
+ "text": "That's where they cut off the Scottish Chiefs and killed all hands.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0209.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "That's where they cut off the Scottish chiefs and killed all hands.",
+ "raw": {
+ "normalized_reference": "that's where they cut off the scottish chiefs and killed all hands.",
+ "normalized_hypothesis": "that's where they cut off the scottish chiefs and killed all hands.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "that is where they cut off the scottish chiefs and killed all hands",
+ "normalized_hypothesis": "that is where they cut off the scottish chiefs and killed all hands",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0210",
+ "category": "phonetic_balance",
+ "text": "And Tom King patiently endured.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0210.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "and Tom King patiently endured.",
+ "raw": {
+ "normalized_reference": "and tom king patiently endured.",
+ "normalized_hypothesis": "and tom king patiently endured.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "and tom king patiently endured",
+ "normalized_hypothesis": "and tom king patiently endured",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0211",
+ "category": "phonetic_balance",
+ "text": "From that moment his friendship for Belize turns to hatred and jealousy.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0211.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "From that moment his friendship for Belize turns to hatred and jealousy.",
+ "raw": {
+ "normalized_reference": "from that moment his friendship for belize turns to hatred and jealousy.",
+ "normalized_hypothesis": "from that moment his friendship for belize turns to hatred and jealousy.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "from that moment his friendship for belize turns to hatred and jealousy",
+ "normalized_hypothesis": "from that moment his friendship for belize turns to hatred and jealousy",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0212",
+ "category": "phonetic_balance",
+ "text": "Then you don't believe in altruism.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0212.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "then you don't believe in altruism.",
+ "raw": {
+ "normalized_reference": "then you don't believe in altruism.",
+ "normalized_hypothesis": "then you don't believe in altruism.",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "then you do not believe in altruism",
+ "normalized_hypothesis": "then you do not believe in altruism",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0213",
+ "category": "phonetic_balance",
+ "text": "What Pascal glimpsed with the vision of a seer, I have lived.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0213.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "what passable glimpsed with the vision of a seer, I have lived.",
+ "raw": {
+ "normalized_reference": "what pascal glimpsed with the vision of a seer, i have lived.",
+ "normalized_hypothesis": "what passable glimpsed with the vision of a seer, i have lived.",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "what pascal glimpsed with the vision of a seer i have lived",
+ "normalized_hypothesis": "what passable glimpsed with the vision of a seer i have lived",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ }
+ },
+ {
+ "id": "release500_0214",
+ "category": "phonetic_balance",
+ "text": "You were making them talk shop, Ruth charged him.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0214.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "You were making them talk shop, Ruth charged him.",
+ "raw": {
+ "normalized_reference": "you were making them talk shop, ruth charged him.",
+ "normalized_hypothesis": "you were making them talk shop, ruth charged him.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "you were making them talk shop ruth charged him",
+ "normalized_hypothesis": "you were making them talk shop ruth charged him",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0215",
+ "category": "phonetic_balance",
+ "text": "Philip snatched at the letter which Gregson held out to him.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0215.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Philip snatched at the letter which Gregson held out to him.",
+ "raw": {
+ "normalized_reference": "philip snatched at the letter which gregson held out to him.",
+ "normalized_hypothesis": "philip snatched at the letter which gregson held out to him.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "philip snatched at the letter which gregson held out to him",
+ "normalized_hypothesis": "philip snatched at the letter which gregson held out to him",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0216",
+ "category": "phonetic_balance",
+ "text": "For a time the exciting thrill of his adventure was gone.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0216.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "For a time, the exciting thrill of his adventure was gone.",
+ "raw": {
+ "normalized_reference": "for a time the exciting thrill of his adventure was gone.",
+ "normalized_hypothesis": "for a time, the exciting thrill of his adventure was gone.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "for a time the exciting thrill of his adventure was gone",
+ "normalized_hypothesis": "for a time the exciting thrill of his adventure was gone",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0217",
+ "category": "phonetic_balance",
+ "text": "Obviously, it was a disease that could be contracted by contact.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0217.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Obviously, it was a disease that could be contracted by contact.",
+ "raw": {
+ "normalized_reference": "obviously, it was a disease that could be contracted by contact.",
+ "normalized_hypothesis": "obviously, it was a disease that could be contracted by contact.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "obviously it was a disease that could be contracted by contact",
+ "normalized_hypothesis": "obviously it was a disease that could be contracted by contact",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0218",
+ "category": "phonetic_balance",
+ "text": "Blind with rage, he darted in.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0218.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Lined with rage, he darted in.",
+ "raw": {
+ "normalized_reference": "blind with rage, he darted in.",
+ "normalized_hypothesis": "lined with rage, he darted in.",
+ "errors": 1,
+ "words": 6,
+ "wer": 0.16666666666666666
+ },
+ "semantic": {
+ "normalized_reference": "blind with rage he darted in",
+ "normalized_hypothesis": "lined with rage he darted in",
+ "errors": 1,
+ "words": 6,
+ "wer": 0.16666666666666666
+ }
+ },
+ {
+ "id": "release500_0219",
+ "category": "phonetic_balance",
+ "text": "You live on an income which your father earned.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0219.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "You live on an income which your father earned.",
+ "raw": {
+ "normalized_reference": "you live on an income which your father earned.",
+ "normalized_hypothesis": "you live on an income which your father earned.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "you live on an income which your father earned",
+ "normalized_hypothesis": "you live on an income which your father earned",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0220",
+ "category": "phonetic_balance",
+ "text": "Something about it seemed to fascinate him, to challenge his presence.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0220.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Something about it seemed to fascinate him, to challenge his presence.",
+ "raw": {
+ "normalized_reference": "something about it seemed to fascinate him, to challenge his presence.",
+ "normalized_hypothesis": "something about it seemed to fascinate him, to challenge his presence.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "something about it seemed to fascinate him to challenge his presence",
+ "normalized_hypothesis": "something about it seemed to fascinate him to challenge his presence",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0221",
+ "category": "phonetic_balance",
+ "text": "Her beautiful hair was done up in shining coils.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0221.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Her beautiful hair was done up in shining coils.",
+ "raw": {
+ "normalized_reference": "her beautiful hair was done up in shining coils.",
+ "normalized_hypothesis": "her beautiful hair was done up in shining coils.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "her beautiful hair was done up in shining coils",
+ "normalized_hypothesis": "her beautiful hair was done up in shining coils",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0222",
+ "category": "phonetic_balance",
+ "text": "It does, was her audacious answer.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0222.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It does was her audacious answer.",
+ "raw": {
+ "normalized_reference": "it does, was her audacious answer.",
+ "normalized_hypothesis": "it does was her audacious answer.",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it does was her audacious answer",
+ "normalized_hypothesis": "it does was her audacious answer",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0223",
+ "category": "phonetic_balance",
+ "text": "He caught himself with a jerk.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0223.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He caught himself with a jerk.",
+ "raw": {
+ "normalized_reference": "he caught himself with a jerk.",
+ "normalized_hypothesis": "he caught himself with a jerk.",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he caught himself with a jerk",
+ "normalized_hypothesis": "he caught himself with a jerk",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0224",
+ "category": "phonetic_balance",
+ "text": "He did not believe in the burning of daylight for such a luxury.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0224.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He did not believe in the burning of daylight for such a luxury.",
+ "raw": {
+ "normalized_reference": "he did not believe in the burning of daylight for such a luxury.",
+ "normalized_hypothesis": "he did not believe in the burning of daylight for such a luxury.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he did not believe in the burning of daylight for such a luxury",
+ "normalized_hypothesis": "he did not believe in the burning of daylight for such a luxury",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0225",
+ "category": "phonetic_balance",
+ "text": "He had comparatively no advantages at first.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0225.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He had comparatively no advantages at first.",
+ "raw": {
+ "normalized_reference": "he had comparatively no advantages at first.",
+ "normalized_hypothesis": "he had comparatively no advantages at first.",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he had comparatively no advantages at 1st",
+ "normalized_hypothesis": "he had comparatively no advantages at 1st",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0226",
+ "category": "phonetic_balance",
+ "text": "Such men believe, when they come together.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0226.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Such men believe when they come together.",
+ "raw": {
+ "normalized_reference": "such men believe, when they come together.",
+ "normalized_hypothesis": "such men believe when they come together.",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "such men believe when they come together",
+ "normalized_hypothesis": "such men believe when they come together",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0227",
+ "category": "phonetic_balance",
+ "text": "The ship should be in within a week or ten days.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0227.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The ship should be in within a week or ten days.",
+ "raw": {
+ "normalized_reference": "the ship should be in within a week or ten days.",
+ "normalized_hypothesis": "the ship should be in within a week or ten days.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the ship should be in within a week or 10 days",
+ "normalized_hypothesis": "the ship should be in within a week or 10 days",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0228",
+ "category": "phonetic_balance",
+ "text": "Such a dog the wise driver kills, or turns loose.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0228.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Such a dog the wise driver kills or turns loose.",
+ "raw": {
+ "normalized_reference": "such a dog the wise driver kills, or turns loose.",
+ "normalized_hypothesis": "such a dog the wise driver kills or turns loose.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "such a dog the wise driver kills or turns loose",
+ "normalized_hypothesis": "such a dog the wise driver kills or turns loose",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0229",
+ "category": "phonetic_balance",
+ "text": "You have heard always how he was the lover of the Princess Naomi.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0229.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "You have heard always how he was the lover of the Princess Naomi.",
+ "raw": {
+ "normalized_reference": "you have heard always how he was the lover of the princess naomi.",
+ "normalized_hypothesis": "you have heard always how he was the lover of the princess naomi.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "you have heard always how he was the lover of the princess naomi",
+ "normalized_hypothesis": "you have heard always how he was the lover of the princess naomi",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0230",
+ "category": "phonetic_balance",
+ "text": "I want to know how all this is possible.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0230.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I want to know how all this is possible.",
+ "raw": {
+ "normalized_reference": "i want to know how all this is possible.",
+ "normalized_hypothesis": "i want to know how all this is possible.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i want to know how all this is possible",
+ "normalized_hypothesis": "i want to know how all this is possible",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0231",
+ "category": "phonetic_balance",
+ "text": "It was my idea to a tee.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0231.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It was my idea to a tee.",
+ "raw": {
+ "normalized_reference": "it was my idea to a tee.",
+ "normalized_hypothesis": "it was my idea to a tee.",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it was my idea to a tee",
+ "normalized_hypothesis": "it was my idea to a tee",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0232",
+ "category": "phonetic_balance",
+ "text": "They likewise are disinclined to being eaten.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0232.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "They likewise are disinclined to being eaten.",
+ "raw": {
+ "normalized_reference": "they likewise are disinclined to being eaten.",
+ "normalized_hypothesis": "they likewise are disinclined to being eaten.",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "they likewise are disinclined to being eaten",
+ "normalized_hypothesis": "they likewise are disinclined to being eaten",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0233",
+ "category": "phonetic_balance",
+ "text": "This state of mind comes of an undue prominence of the ego.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0233.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "This state of mind comes of an undue prominence of the ego.",
+ "raw": {
+ "normalized_reference": "this state of mind comes of an undue prominence of the ego.",
+ "normalized_hypothesis": "this state of mind comes of an undue prominence of the ego.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "this state of mind comes of an undue prominence of the ego",
+ "normalized_hypothesis": "this state of mind comes of an undue prominence of the ego",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0234",
+ "category": "phonetic_balance",
+ "text": "Then it is as I said, Womble announced with finality.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0234.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Then it is as I said, Womble announced with finality.",
+ "raw": {
+ "normalized_reference": "then it is as i said, womble announced with finality.",
+ "normalized_hypothesis": "then it is as i said, womble announced with finality.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "then it is as i said womble announced with finality",
+ "normalized_hypothesis": "then it is as i said womble announced with finality",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0235",
+ "category": "phonetic_balance",
+ "text": "I can see that knife now.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0235.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I can see that knife now.",
+ "raw": {
+ "normalized_reference": "i can see that knife now.",
+ "normalized_hypothesis": "i can see that knife now.",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i can see that knife now",
+ "normalized_hypothesis": "i can see that knife now",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0236",
+ "category": "phonetic_balance",
+ "text": "If I meet her again I shall apologize, said Eileen.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0236.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "If I meet her again I shall apologize, said Eileen.",
+ "raw": {
+ "normalized_reference": "if i meet her again i shall apologize, said eileen.",
+ "normalized_hypothesis": "if i meet her again i shall apologize, said eileen.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "if i meet her again i shall apologize said eileen",
+ "normalized_hypothesis": "if i meet her again i shall apologize said eileen",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0237",
+ "category": "phonetic_balance",
+ "text": "A wildly exciting time was his during the week preceding Thursday the eighteenth.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0237.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "A wildly exciting time was his during the week preceding Thursday the 18th.",
+ "raw": {
+ "normalized_reference": "a wildly exciting time was his during the week preceding thursday the eighteenth.",
+ "normalized_hypothesis": "a wildly exciting time was his during the week preceding thursday the 18th.",
+ "errors": 1,
+ "words": 13,
+ "wer": 0.07692307692307693
+ },
+ "semantic": {
+ "normalized_reference": "a wildly exciting time was his during the week preceding thursday the 18th",
+ "normalized_hypothesis": "a wildly exciting time was his during the week preceding thursday the 18th",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0238",
+ "category": "phonetic_balance",
+ "text": "In what bucolic school of fence he had been taught was beyond imagining.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0238.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "In what bucolic school, a fence he had been taught was beyond imagining.",
+ "raw": {
+ "normalized_reference": "in what bucolic school of fence he had been taught was beyond imagining.",
+ "normalized_hypothesis": "in what bucolic school, a fence he had been taught was beyond imagining.",
+ "errors": 1,
+ "words": 13,
+ "wer": 0.07692307692307693
+ },
+ "semantic": {
+ "normalized_reference": "in what bucolic school of fence he had been taught was beyond imagining",
+ "normalized_hypothesis": "in what bucolic school a fence he had been taught was beyond imagining",
+ "errors": 1,
+ "words": 13,
+ "wer": 0.07692307692307693
+ }
+ },
+ {
+ "id": "release500_0239",
+ "category": "phonetic_balance",
+ "text": "So she said, the irate skipper dashed on.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0239.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "So she said, the irate skipper dashed on.",
+ "raw": {
+ "normalized_reference": "so she said, the irate skipper dashed on.",
+ "normalized_hypothesis": "so she said, the irate skipper dashed on.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "so she said the irate skipper dashed on",
+ "normalized_hypothesis": "so she said the irate skipper dashed on",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0240",
+ "category": "phonetic_balance",
+ "text": "It is not an attempt to smash the market.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0240.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It is not an attempt to smash the market.",
+ "raw": {
+ "normalized_reference": "it is not an attempt to smash the market.",
+ "normalized_hypothesis": "it is not an attempt to smash the market.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it is not an attempt to smash the market",
+ "normalized_hypothesis": "it is not an attempt to smash the market",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0241",
+ "category": "phonetic_balance",
+ "text": "He destroyed everything that had belonged to the woman.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0241.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He destroyed everything that had belonged to the woman.",
+ "raw": {
+ "normalized_reference": "he destroyed everything that had belonged to the woman.",
+ "normalized_hypothesis": "he destroyed everything that had belonged to the woman.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he destroyed everything that had belonged to the woman",
+ "normalized_hypothesis": "he destroyed everything that had belonged to the woman",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0242",
+ "category": "phonetic_balance",
+ "text": "Did Thorpe go to see any one in Churchill.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0242.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Did Thorpe go to see anyone in Churchill?",
+ "raw": {
+ "normalized_reference": "did thorpe go to see any one in churchill.",
+ "normalized_hypothesis": "did thorpe go to see anyone in churchill?",
+ "errors": 2,
+ "words": 9,
+ "wer": 0.2222222222222222
+ },
+ "semantic": {
+ "normalized_reference": "did thorpe go to see any one in churchill",
+ "normalized_hypothesis": "did thorpe go to see anyone in churchill",
+ "errors": 2,
+ "words": 9,
+ "wer": 0.2222222222222222
+ }
+ },
+ {
+ "id": "release500_0243",
+ "category": "phonetic_balance",
+ "text": "Many other similar disconcerting omissions will be noticed in the Manuscript.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0243.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Many other similar disconcerting omissions will be noticed in the manuscript.",
+ "raw": {
+ "normalized_reference": "many other similar disconcerting omissions will be noticed in the manuscript.",
+ "normalized_hypothesis": "many other similar disconcerting omissions will be noticed in the manuscript.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "many other similar disconcerting omissions will be noticed in the manuscript",
+ "normalized_hypothesis": "many other similar disconcerting omissions will be noticed in the manuscript",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0244",
+ "category": "phonetic_balance",
+ "text": "The ringing of the big bell aroused him.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0244.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The ringing of the big bell rouse him.",
+ "raw": {
+ "normalized_reference": "the ringing of the big bell aroused him.",
+ "normalized_hypothesis": "the ringing of the big bell rouse him.",
+ "errors": 1,
+ "words": 8,
+ "wer": 0.125
+ },
+ "semantic": {
+ "normalized_reference": "the ringing of the big bell aroused him",
+ "normalized_hypothesis": "the ringing of the big bell rouse him",
+ "errors": 1,
+ "words": 8,
+ "wer": 0.125
+ }
+ },
+ {
+ "id": "release500_0245",
+ "category": "phonetic_balance",
+ "text": "A sinewy hand, dripping with water, was clutching the rail.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0245.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "A sinewy hand, dripping with water, was clutching the rail.",
+ "raw": {
+ "normalized_reference": "a sinewy hand, dripping with water, was clutching the rail.",
+ "normalized_hypothesis": "a sinewy hand, dripping with water, was clutching the rail.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "a sinewy hand dripping with water was clutching the rail",
+ "normalized_hypothesis": "a sinewy hand dripping with water was clutching the rail",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0246",
+ "category": "phonetic_balance",
+ "text": "Enters now the psychology of the situation.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0246.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "enters now the psychology of the situation.",
+ "raw": {
+ "normalized_reference": "enters now the psychology of the situation.",
+ "normalized_hypothesis": "enters now the psychology of the situation.",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "enters now the psychology of the situation",
+ "normalized_hypothesis": "enters now the psychology of the situation",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0247",
+ "category": "phonetic_balance",
+ "text": "Wash your hands of me.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0247.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Wash your hands of me.",
+ "raw": {
+ "normalized_reference": "wash your hands of me.",
+ "normalized_hypothesis": "wash your hands of me.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "wash your hands of me",
+ "normalized_hypothesis": "wash your hands of me",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0248",
+ "category": "phonetic_balance",
+ "text": "He confessed that the sketch had startled him.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0248.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He confessed that the sketch had startled him.",
+ "raw": {
+ "normalized_reference": "he confessed that the sketch had startled him.",
+ "normalized_hypothesis": "he confessed that the sketch had startled him.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he confessed that the sketch had startled him",
+ "normalized_hypothesis": "he confessed that the sketch had startled him",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0249",
+ "category": "phonetic_balance",
+ "text": "Her face was against his breast.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0249.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Her face was against his breast.",
+ "raw": {
+ "normalized_reference": "her face was against his breast.",
+ "normalized_hypothesis": "her face was against his breast.",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "her face was against his breast",
+ "normalized_hypothesis": "her face was against his breast",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0250",
+ "category": "phonetic_balance",
+ "text": "He was a pariah; a wanderer without a friend or a home.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0250.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He was a pariah, a wanderer without a friend or a home.",
+ "raw": {
+ "normalized_reference": "he was a pariah; a wanderer without a friend or a home.",
+ "normalized_hypothesis": "he was a pariah, a wanderer without a friend or a home.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he was a pariah a wanderer without a friend or a home",
+ "normalized_hypothesis": "he was a pariah a wanderer without a friend or a home",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0251",
+ "category": "phonetic_balance",
+ "text": "A cry of joy burst from Philip's lips.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0251.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "A cry of joy burst from Philip's lips.",
+ "raw": {
+ "normalized_reference": "a cry of joy burst from philip's lips.",
+ "normalized_hypothesis": "a cry of joy burst from philip's lips.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "a cry of joy burst from philip is lips",
+ "normalized_hypothesis": "a cry of joy burst from philip is lips",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0252",
+ "category": "phonetic_balance",
+ "text": "And the big Persian knew of his existence before he did of hers.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0252.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "and the big Persian knew of his existence before he did of hers.",
+ "raw": {
+ "normalized_reference": "and the big persian knew of his existence before he did of hers.",
+ "normalized_hypothesis": "and the big persian knew of his existence before he did of hers.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "and the big persian knew of his existence before he did of hers",
+ "normalized_hypothesis": "and the big persian knew of his existence before he did of hers",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0253",
+ "category": "phonetic_balance",
+ "text": "The men stared into each other's face.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0253.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The men stared into each other's face.",
+ "raw": {
+ "normalized_reference": "the men stared into each other's face.",
+ "normalized_hypothesis": "the men stared into each other's face.",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the men stared into each other is face",
+ "normalized_hypothesis": "the men stared into each other is face",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0254",
+ "category": "phonetic_balance",
+ "text": "It is the fire, partly, she said.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0254.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It is the fire, partly, she said.",
+ "raw": {
+ "normalized_reference": "it is the fire, partly, she said.",
+ "normalized_hypothesis": "it is the fire, partly, she said.",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it is the fire partly she said",
+ "normalized_hypothesis": "it is the fire partly she said",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0255",
+ "category": "phonetic_balance",
+ "text": "It seemed nearer to him since he had seen and talked with Gregson.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0255.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It seemed nearer to him since he had seen and talked with Gregson.",
+ "raw": {
+ "normalized_reference": "it seemed nearer to him since he had seen and talked with gregson.",
+ "normalized_hypothesis": "it seemed nearer to him since he had seen and talked with gregson.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it seemed nearer to him since he had seen and talked with gregson",
+ "normalized_hypothesis": "it seemed nearer to him since he had seen and talked with gregson",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0256",
+ "category": "phonetic_balance",
+ "text": "Among my minor afflictions, I may mention a new and mysterious one.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0256.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Among my minor afflictions, I may mention a new and mysterious one.",
+ "raw": {
+ "normalized_reference": "among my minor afflictions, i may mention a new and mysterious one.",
+ "normalized_hypothesis": "among my minor afflictions, i may mention a new and mysterious one.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "among my minor afflictions i may mention a new and mysterious one",
+ "normalized_hypothesis": "among my minor afflictions i may mention a new and mysterious one",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0257",
+ "category": "phonetic_balance",
+ "text": "Between him and all domestic animals there must be no hostilities.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0257.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Between him and all domestic animals there must be no hostilities.",
+ "raw": {
+ "normalized_reference": "between him and all domestic animals there must be no hostilities.",
+ "normalized_hypothesis": "between him and all domestic animals there must be no hostilities.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "between him and all domestic animals there must be no hostilities",
+ "normalized_hypothesis": "between him and all domestic animals there must be no hostilities",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0258",
+ "category": "phonetic_balance",
+ "text": "With these arguments he convinced himself that he should go on alone.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0258.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "With these arguments he convinced himself that he should go on alone.",
+ "raw": {
+ "normalized_reference": "with these arguments he convinced himself that he should go on alone.",
+ "normalized_hypothesis": "with these arguments he convinced himself that he should go on alone.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "with these arguments he convinced himself that he should go on alone",
+ "normalized_hypothesis": "with these arguments he convinced himself that he should go on alone",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0259",
+ "category": "phonetic_balance",
+ "text": "We are consumed in our own flesh-pots.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0259.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "we are consumed in our own flesh parts.",
+ "raw": {
+ "normalized_reference": "we are consumed in our own flesh-pots.",
+ "normalized_hypothesis": "we are consumed in our own flesh parts.",
+ "errors": 1,
+ "words": 8,
+ "wer": 0.125
+ },
+ "semantic": {
+ "normalized_reference": "we are consumed in our own flesh pots",
+ "normalized_hypothesis": "we are consumed in our own flesh parts",
+ "errors": 1,
+ "words": 8,
+ "wer": 0.125
+ }
+ },
+ {
+ "id": "release500_0260",
+ "category": "phonetic_balance",
+ "text": "Mercedes screamed, cried, laughed, and manifested the chaotic abandonment of hysteria.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0260.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Mercedes screamed, cried, laughed, and manifested the chaotic abandonment of hysteria.",
+ "raw": {
+ "normalized_reference": "mercedes screamed, cried, laughed, and manifested the chaotic abandonment of hysteria.",
+ "normalized_hypothesis": "mercedes screamed, cried, laughed, and manifested the chaotic abandonment of hysteria.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "mercedes screamed cried laughed and manifested the chaotic abandonment of hysteria",
+ "normalized_hypothesis": "mercedes screamed cried laughed and manifested the chaotic abandonment of hysteria",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0261",
+ "category": "phonetic_balance",
+ "text": "Philip made no effort to follow.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0261.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Philip made no effort to follow.",
+ "raw": {
+ "normalized_reference": "philip made no effort to follow.",
+ "normalized_hypothesis": "philip made no effort to follow.",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "philip made no effort to follow",
+ "normalized_hypothesis": "philip made no effort to follow",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0262",
+ "category": "phonetic_balance",
+ "text": "She said with chattering teeth.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0262.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "she said with chattering teeth.",
+ "raw": {
+ "normalized_reference": "she said with chattering teeth.",
+ "normalized_hypothesis": "she said with chattering teeth.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "she said with chattering teeth",
+ "normalized_hypothesis": "she said with chattering teeth",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0263",
+ "category": "phonetic_balance",
+ "text": "Halfway around the track one donkey got into an argument with its rider.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0263.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Halfway around the track, one donkey got into an argument with its rider.",
+ "raw": {
+ "normalized_reference": "halfway around the track one donkey got into an argument with its rider.",
+ "normalized_hypothesis": "halfway around the track, one donkey got into an argument with its rider.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "halfway around the track one donkey got into an argument with its rider",
+ "normalized_hypothesis": "halfway around the track one donkey got into an argument with its rider",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0264",
+ "category": "phonetic_balance",
+ "text": "Earth and gravel seemed to fill the pan.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0264.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Earth and gravel seemed to fill the pan.",
+ "raw": {
+ "normalized_reference": "earth and gravel seemed to fill the pan.",
+ "normalized_hypothesis": "earth and gravel seemed to fill the pan.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "earth and gravel seemed to fill the pan",
+ "normalized_hypothesis": "earth and gravel seemed to fill the pan",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0265",
+ "category": "phonetic_balance",
+ "text": "Her achievements with cocoanuts were a revelation.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0265.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Her achievements with coconuts were a revelation.",
+ "raw": {
+ "normalized_reference": "her achievements with cocoanuts were a revelation.",
+ "normalized_hypothesis": "her achievements with coconuts were a revelation.",
+ "errors": 1,
+ "words": 7,
+ "wer": 0.14285714285714285
+ },
+ "semantic": {
+ "normalized_reference": "her achievements with cocoanuts were a revelation",
+ "normalized_hypothesis": "her achievements with coconuts were a revelation",
+ "errors": 1,
+ "words": 7,
+ "wer": 0.14285714285714285
+ }
+ },
+ {
+ "id": "release500_0266",
+ "category": "phonetic_balance",
+ "text": "For a full minute he crouched and listened.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0266.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "For a full minute he crouched and listened.",
+ "raw": {
+ "normalized_reference": "for a full minute he crouched and listened.",
+ "normalized_hypothesis": "for a full minute he crouched and listened.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "for a full minute he crouched and listened",
+ "normalized_hypothesis": "for a full minute he crouched and listened",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0267",
+ "category": "phonetic_balance",
+ "text": "It was edged with ice.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0267.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It was edged with ice.",
+ "raw": {
+ "normalized_reference": "it was edged with ice.",
+ "normalized_hypothesis": "it was edged with ice.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it was edged with ice",
+ "normalized_hypothesis": "it was edged with ice",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0268",
+ "category": "phonetic_balance",
+ "text": "The Portuguese boy passed the Hawaiian.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0268.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The Portuguese boy passed the Hawaiian.",
+ "raw": {
+ "normalized_reference": "the portuguese boy passed the hawaiian.",
+ "normalized_hypothesis": "the portuguese boy passed the hawaiian.",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the portuguese boy passed the hawaiian",
+ "normalized_hypothesis": "the portuguese boy passed the hawaiian",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0269",
+ "category": "phonetic_balance",
+ "text": "But he did not broach it, preferring to mature it carefully.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0269.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "but he did not broach it, preferring to mature it carefully.",
+ "raw": {
+ "normalized_reference": "but he did not broach it, preferring to mature it carefully.",
+ "normalized_hypothesis": "but he did not broach it, preferring to mature it carefully.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "but he did not broach it preferring to mature it carefully",
+ "normalized_hypothesis": "but he did not broach it preferring to mature it carefully",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0270",
+ "category": "phonetic_balance",
+ "text": "Come on, Del Mar challenged.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0270.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Come on. Delmar challenged.",
+ "raw": {
+ "normalized_reference": "come on, del mar challenged.",
+ "normalized_hypothesis": "come on. delmar challenged.",
+ "errors": 2,
+ "words": 5,
+ "wer": 0.4
+ },
+ "semantic": {
+ "normalized_reference": "come on del mar challenged",
+ "normalized_hypothesis": "come on delmar challenged",
+ "errors": 2,
+ "words": 5,
+ "wer": 0.4
+ }
+ },
+ {
+ "id": "release500_0271",
+ "category": "phonetic_balance",
+ "text": "For a full half minute Jeanne looked at him without speaking.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0271.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "For a full half-minute Jean looked at him without speaking.",
+ "raw": {
+ "normalized_reference": "for a full half minute jeanne looked at him without speaking.",
+ "normalized_hypothesis": "for a full half-minute jean looked at him without speaking.",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ },
+ "semantic": {
+ "normalized_reference": "for a full half minute jeanne looked at him without speaking",
+ "normalized_hypothesis": "for a full half minute jean looked at him without speaking",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ }
+ },
+ {
+ "id": "release500_0272",
+ "category": "phonetic_balance",
+ "text": "He bore no grudges and had few enemies.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0272.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He bore no grudges and had few enemies.",
+ "raw": {
+ "normalized_reference": "he bore no grudges and had few enemies.",
+ "normalized_hypothesis": "he bore no grudges and had few enemies.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he bore no grudges and had few enemies",
+ "normalized_hypothesis": "he bore no grudges and had few enemies",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0273",
+ "category": "phonetic_balance",
+ "text": "I'll go over tomorrow afternoon.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0273.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I'll go over tomorrow afternoon.",
+ "raw": {
+ "normalized_reference": "i'll go over tomorrow afternoon.",
+ "normalized_hypothesis": "i'll go over tomorrow afternoon.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i will go over tomorrow afternoon",
+ "normalized_hypothesis": "i will go over tomorrow afternoon",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0274",
+ "category": "phonetic_balance",
+ "text": "Well, there are better men in Hawaii, that's all.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0274.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Well, there are better men in Hawaii. That's all.",
+ "raw": {
+ "normalized_reference": "well, there are better men in hawaii, that's all.",
+ "normalized_hypothesis": "well, there are better men in hawaii. that's all.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "well there are better men in hawaii that is all",
+ "normalized_hypothesis": "well there are better men in hawaii that is all",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0275",
+ "category": "phonetic_balance",
+ "text": "He was an athlete and a giant.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0275.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He was an athlete and a giant.",
+ "raw": {
+ "normalized_reference": "he was an athlete and a giant.",
+ "normalized_hypothesis": "he was an athlete and a giant.",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he was an athlete and a giant",
+ "normalized_hypothesis": "he was an athlete and a giant",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0276",
+ "category": "phonetic_balance",
+ "text": "McVeigh when he returns from a trip to Honolulu.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0276.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "McVeigh when he returns from a trip to Honolulu.",
+ "raw": {
+ "normalized_reference": "mcveigh when he returns from a trip to honolulu.",
+ "normalized_hypothesis": "mcveigh when he returns from a trip to honolulu.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "mcveigh when he returns from a trip to honolulu",
+ "normalized_hypothesis": "mcveigh when he returns from a trip to honolulu",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0277",
+ "category": "phonetic_balance",
+ "text": "You're a devil for fighting, and will surely win.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0277.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "You're a devil for fighting, and will surely win.",
+ "raw": {
+ "normalized_reference": "you're a devil for fighting, and will surely win.",
+ "normalized_hypothesis": "you're a devil for fighting, and will surely win.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "you are a devil for fighting and will surely win",
+ "normalized_hypothesis": "you are a devil for fighting and will surely win",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0278",
+ "category": "phonetic_balance",
+ "text": "All an appearance can know is mirage.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0278.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "All an appearance can know is mirage.",
+ "raw": {
+ "normalized_reference": "all an appearance can know is mirage.",
+ "normalized_hypothesis": "all an appearance can know is mirage.",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "all an appearance can know is mirage",
+ "normalized_hypothesis": "all an appearance can know is mirage",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0279",
+ "category": "phonetic_balance",
+ "text": "You fired me out of your house, in short.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0279.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "You fired me out of your house, in short.",
+ "raw": {
+ "normalized_reference": "you fired me out of your house, in short.",
+ "normalized_hypothesis": "you fired me out of your house, in short.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "you fired me out of your house in short",
+ "normalized_hypothesis": "you fired me out of your house in short",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0280",
+ "category": "phonetic_balance",
+ "text": "He made sure that the magazine was loaded, and resumed his paddling.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0280.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He made sure that the magazine was loaded and resumed his paddling.",
+ "raw": {
+ "normalized_reference": "he made sure that the magazine was loaded, and resumed his paddling.",
+ "normalized_hypothesis": "he made sure that the magazine was loaded and resumed his paddling.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he made sure that the magazine was loaded and resumed his paddling",
+ "normalized_hypothesis": "he made sure that the magazine was loaded and resumed his paddling",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0281",
+ "category": "phonetic_balance",
+ "text": "Change chairs, Daylight commanded.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0281.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Change chairs, Daylight commanded.",
+ "raw": {
+ "normalized_reference": "change chairs, daylight commanded.",
+ "normalized_hypothesis": "change chairs, daylight commanded.",
+ "errors": 0,
+ "words": 4,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "change chairs daylight commanded",
+ "normalized_hypothesis": "change chairs daylight commanded",
+ "errors": 0,
+ "words": 4,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0282",
+ "category": "phonetic_balance",
+ "text": "By that answer my professional medical prestige stood or fell.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0282.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "By that answer my professional medical prestige stood or fell.",
+ "raw": {
+ "normalized_reference": "by that answer my professional medical prestige stood or fell.",
+ "normalized_hypothesis": "by that answer my professional medical prestige stood or fell.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "by that answer my professional medical prestige stood or fell",
+ "normalized_hypothesis": "by that answer my professional medical prestige stood or fell",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0283",
+ "category": "phonetic_balance",
+ "text": "The moon had already begun its westward decline.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0283.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The moon had already begun its westward decline.",
+ "raw": {
+ "normalized_reference": "the moon had already begun its westward decline.",
+ "normalized_hypothesis": "the moon had already begun its westward decline.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the moon had already begun its westward decline",
+ "normalized_hypothesis": "the moon had already begun its westward decline",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0284",
+ "category": "phonetic_balance",
+ "text": "At sea, Monday, March 16, 1908.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0284.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "At sea, Monday, March 16, 1908.",
+ "raw": {
+ "normalized_reference": "at sea, monday, march 16, 1908.",
+ "normalized_hypothesis": "at sea, monday, march 16, 1908.",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "at sea monday march 16 1908",
+ "normalized_hypothesis": "at sea monday march 16 1908",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0285",
+ "category": "phonetic_balance",
+ "text": "And wherever I ranged, the way lay along alcohol-drenched roads.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0285.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "And wherever I ranged, the way lay a long alcohol-drenched road.",
+ "raw": {
+ "normalized_reference": "and wherever i ranged, the way lay along alcohol-drenched roads.",
+ "normalized_hypothesis": "and wherever i ranged, the way lay a long alcohol-drenched road.",
+ "errors": 3,
+ "words": 11,
+ "wer": 0.2727272727272727
+ },
+ "semantic": {
+ "normalized_reference": "and wherever i ranged the way lay along alcohol drenched roads",
+ "normalized_hypothesis": "and wherever i ranged the way lay a long alcohol drenched road",
+ "errors": 3,
+ "words": 11,
+ "wer": 0.2727272727272727
+ }
+ },
+ {
+ "id": "release500_0286",
+ "category": "phonetic_balance",
+ "text": "The date was nearly eighteen years old.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0286.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The date was nearly 18 years old.",
+ "raw": {
+ "normalized_reference": "the date was nearly eighteen years old.",
+ "normalized_hypothesis": "the date was nearly 18 years old.",
+ "errors": 1,
+ "words": 7,
+ "wer": 0.14285714285714285
+ },
+ "semantic": {
+ "normalized_reference": "the date was nearly 18 years old",
+ "normalized_hypothesis": "the date was nearly 18 years old",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0287",
+ "category": "phonetic_balance",
+ "text": "Very early in my life, I separated from my mother.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0287.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Very early in my life, I separated from my mother.",
+ "raw": {
+ "normalized_reference": "very early in my life, i separated from my mother.",
+ "normalized_hypothesis": "very early in my life, i separated from my mother.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "very early in my life i separated from my mother",
+ "normalized_hypothesis": "very early in my life i separated from my mother",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0288",
+ "category": "phonetic_balance",
+ "text": "There was pride and strength, the ring of triumph in his voice.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0288.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "There was pride and strength, the ring of triumph in his voice.",
+ "raw": {
+ "normalized_reference": "there was pride and strength, the ring of triumph in his voice.",
+ "normalized_hypothesis": "there was pride and strength, the ring of triumph in his voice.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "there was pride and strength the ring of triumph in his voice",
+ "normalized_hypothesis": "there was pride and strength the ring of triumph in his voice",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0289",
+ "category": "phonetic_balance",
+ "text": "From my earliest recollection my sleep was a period of terror.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0289.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "From my earliest recollection my sleep was a period of terror.",
+ "raw": {
+ "normalized_reference": "from my earliest recollection my sleep was a period of terror.",
+ "normalized_hypothesis": "from my earliest recollection my sleep was a period of terror.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "from my earliest recollection my sleep was a period of terror",
+ "normalized_hypothesis": "from my earliest recollection my sleep was a period of terror",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0290",
+ "category": "phonetic_balance",
+ "text": "We'll have to watch our chances.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0290.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "We'll have to watch our chances.",
+ "raw": {
+ "normalized_reference": "we'll have to watch our chances.",
+ "normalized_hypothesis": "we'll have to watch our chances.",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "we will have to watch our chances",
+ "normalized_hypothesis": "we will have to watch our chances",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0291",
+ "category": "phonetic_balance",
+ "text": "Close beside him gleamed the white fangs of the wolf-dog.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0291.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Close beside him gleamed the white fangs of the wolf-dog.",
+ "raw": {
+ "normalized_reference": "close beside him gleamed the white fangs of the wolf-dog.",
+ "normalized_hypothesis": "close beside him gleamed the white fangs of the wolf-dog.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "close beside him gleamed the white fangs of the wolf dog",
+ "normalized_hypothesis": "close beside him gleamed the white fangs of the wolf dog",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0292",
+ "category": "phonetic_balance",
+ "text": "Would you be satisfied with that one hundredth part of me.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0292.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Would you be satisfied with that one hundredth part of me?",
+ "raw": {
+ "normalized_reference": "would you be satisfied with that one hundredth part of me.",
+ "normalized_hypothesis": "would you be satisfied with that one hundredth part of me?",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "would you be satisfied with that 100th part of me",
+ "normalized_hypothesis": "would you be satisfied with that 100th part of me",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0293",
+ "category": "phonetic_balance",
+ "text": "Two gigantic owls were tearing at the carcass.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0293.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Two gigantic owls were tearing at the carcass.",
+ "raw": {
+ "normalized_reference": "two gigantic owls were tearing at the carcass.",
+ "normalized_hypothesis": "two gigantic owls were tearing at the carcass.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "2 gigantic owls were tearing at the carcass",
+ "normalized_hypothesis": "2 gigantic owls were tearing at the carcass",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0294",
+ "category": "phonetic_balance",
+ "text": "The lush vegetation of that sheltered spot make a natural shield.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0294.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The lush vegetation of that sheltered spot make a natural shield.",
+ "raw": {
+ "normalized_reference": "the lush vegetation of that sheltered spot make a natural shield.",
+ "normalized_hypothesis": "the lush vegetation of that sheltered spot make a natural shield.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the lush vegetation of that sheltered spot make a natural shield",
+ "normalized_hypothesis": "the lush vegetation of that sheltered spot make a natural shield",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0295",
+ "category": "phonetic_balance",
+ "text": "Below him the shadow was broken into a pool of rippling starlight.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0295.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Below him the shadow was broken into a pool of rippling starlight.",
+ "raw": {
+ "normalized_reference": "below him the shadow was broken into a pool of rippling starlight.",
+ "normalized_hypothesis": "below him the shadow was broken into a pool of rippling starlight.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "below him the shadow was broken into a pool of rippling starlight",
+ "normalized_hypothesis": "below him the shadow was broken into a pool of rippling starlight",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0296",
+ "category": "phonetic_balance",
+ "text": "Lots of men take women buggy riding.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0296.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Lots of men take women buggy riding.",
+ "raw": {
+ "normalized_reference": "lots of men take women buggy riding.",
+ "normalized_hypothesis": "lots of men take women buggy riding.",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "lots of men take women buggy riding",
+ "normalized_hypothesis": "lots of men take women buggy riding",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0297",
+ "category": "phonetic_balance",
+ "text": "But it contributed to the smash.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0297.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "but it contributed to the smash.",
+ "raw": {
+ "normalized_reference": "but it contributed to the smash.",
+ "normalized_hypothesis": "but it contributed to the smash.",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "but it contributed to the smash",
+ "normalized_hypothesis": "but it contributed to the smash",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0298",
+ "category": "phonetic_balance",
+ "text": "No man ate of the seal meat or the oil.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0298.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "No man ate of the seal meat or the oil.",
+ "raw": {
+ "normalized_reference": "no man ate of the seal meat or the oil.",
+ "normalized_hypothesis": "no man ate of the seal meat or the oil.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "no man ate of the seal meat or the oil",
+ "normalized_hypothesis": "no man ate of the seal meat or the oil",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0299",
+ "category": "phonetic_balance",
+ "text": "You're going in for grab sharing.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0299.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "You're going in for grouch sharing.",
+ "raw": {
+ "normalized_reference": "you're going in for grab sharing.",
+ "normalized_hypothesis": "you're going in for grouch sharing.",
+ "errors": 1,
+ "words": 6,
+ "wer": 0.16666666666666666
+ },
+ "semantic": {
+ "normalized_reference": "you are going in for grab sharing",
+ "normalized_hypothesis": "you are going in for grouch sharing",
+ "errors": 1,
+ "words": 7,
+ "wer": 0.14285714285714285
+ }
+ },
+ {
+ "id": "release500_0300",
+ "category": "phonetic_balance",
+ "text": "He it was that lived to found the family of the Patino.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0300.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He it was that lived to found the family of the Patino.",
+ "raw": {
+ "normalized_reference": "he it was that lived to found the family of the patino.",
+ "normalized_hypothesis": "he it was that lived to found the family of the patino.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he it was that lived to found the family of the patino",
+ "normalized_hypothesis": "he it was that lived to found the family of the patino",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0301",
+ "category": "phonetic_balance",
+ "text": "For such countries nothing remained but reorganization.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0301.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "For such countries nothing remained but reorganization.",
+ "raw": {
+ "normalized_reference": "for such countries nothing remained but reorganization.",
+ "normalized_hypothesis": "for such countries nothing remained but reorganization.",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "for such countries nothing remained but reorganization",
+ "normalized_hypothesis": "for such countries nothing remained but reorganization",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0302",
+ "category": "phonetic_balance",
+ "text": "Stand off butcher and baker and all the rest.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0302.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Standoff Butcher and Baker and all the rest.",
+ "raw": {
+ "normalized_reference": "stand off butcher and baker and all the rest.",
+ "normalized_hypothesis": "standoff butcher and baker and all the rest.",
+ "errors": 2,
+ "words": 9,
+ "wer": 0.2222222222222222
+ },
+ "semantic": {
+ "normalized_reference": "stand off butcher and baker and all the rest",
+ "normalized_hypothesis": "standoff butcher and baker and all the rest",
+ "errors": 2,
+ "words": 9,
+ "wer": 0.2222222222222222
+ }
+ },
+ {
+ "id": "release500_0303",
+ "category": "phonetic_balance",
+ "text": "All eyes, however, were staring at him in certitude of expectancy.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0303.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "All eyes, however, were staring at him in certitude of expectancy.",
+ "raw": {
+ "normalized_reference": "all eyes, however, were staring at him in certitude of expectancy.",
+ "normalized_hypothesis": "all eyes, however, were staring at him in certitude of expectancy.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "all eyes however were staring at him in certitude of expectancy",
+ "normalized_hypothesis": "all eyes however were staring at him in certitude of expectancy",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0304",
+ "category": "phonetic_balance",
+ "text": "Two days ago Jeanne learned where her father's men were hiding.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0304.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Two days ago, Jean learned where her father's men were hiding.",
+ "raw": {
+ "normalized_reference": "two days ago jeanne learned where her father's men were hiding.",
+ "normalized_hypothesis": "two days ago, jean learned where her father's men were hiding.",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ },
+ "semantic": {
+ "normalized_reference": "2 days ago jeanne learned where her father is men were hiding",
+ "normalized_hypothesis": "2 days ago jean learned where her father is men were hiding",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ }
+ },
+ {
+ "id": "release500_0305",
+ "category": "phonetic_balance",
+ "text": "His previous wives had never lived long enough to bear him children.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0305.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "His previous wives had never lived long enough to bear him children.",
+ "raw": {
+ "normalized_reference": "his previous wives had never lived long enough to bear him children.",
+ "normalized_hypothesis": "his previous wives had never lived long enough to bear him children.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "his previous wives had never lived long enough to bear him children",
+ "normalized_hypothesis": "his previous wives had never lived long enough to bear him children",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0306",
+ "category": "phonetic_balance",
+ "text": "But this little defect did not worry him.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0306.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "But this little defect did not worry him.",
+ "raw": {
+ "normalized_reference": "but this little defect did not worry him.",
+ "normalized_hypothesis": "but this little defect did not worry him.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "but this little defect did not worry him",
+ "normalized_hypothesis": "but this little defect did not worry him",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0307",
+ "category": "phonetic_balance",
+ "text": "Eli Harding asked, as Shunk started to follow.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0307.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Billy Harding asked as Shunk started to follow.",
+ "raw": {
+ "normalized_reference": "eli harding asked, as shunk started to follow.",
+ "normalized_hypothesis": "billy harding asked as shunk started to follow.",
+ "errors": 1,
+ "words": 8,
+ "wer": 0.125
+ },
+ "semantic": {
+ "normalized_reference": "eli harding asked as shunk started to follow",
+ "normalized_hypothesis": "billy harding asked as shunk started to follow",
+ "errors": 1,
+ "words": 8,
+ "wer": 0.125
+ }
+ },
+ {
+ "id": "release500_0308",
+ "category": "phonetic_balance",
+ "text": "And this was their sole conversation throughout the meal.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0308.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "and this was their sole conversation throughout the meal.",
+ "raw": {
+ "normalized_reference": "and this was their sole conversation throughout the meal.",
+ "normalized_hypothesis": "and this was their sole conversation throughout the meal.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "and this was their sole conversation throughout the meal",
+ "normalized_hypothesis": "and this was their sole conversation throughout the meal",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0309",
+ "category": "phonetic_balance",
+ "text": "Then, again, Tudor had such an irritating way about him.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0309.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Then, again, Tudor had such an irritating way about him.",
+ "raw": {
+ "normalized_reference": "then, again, tudor had such an irritating way about him.",
+ "normalized_hypothesis": "then, again, tudor had such an irritating way about him.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "then again tudor had such an irritating way about him",
+ "normalized_hypothesis": "then again tudor had such an irritating way about him",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0310",
+ "category": "phonetic_balance",
+ "text": "In it there was something that was almost tragedy.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0310.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "In it there was something that was almost tragedy.",
+ "raw": {
+ "normalized_reference": "in it there was something that was almost tragedy.",
+ "normalized_hypothesis": "in it there was something that was almost tragedy.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "in it there was something that was almost tragedy",
+ "normalized_hypothesis": "in it there was something that was almost tragedy",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0311",
+ "category": "phonetic_balance",
+ "text": "Thirty pounds, said the captain with finality.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0311.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Thirty pounds, said the captain with finality.",
+ "raw": {
+ "normalized_reference": "thirty pounds, said the captain with finality.",
+ "normalized_hypothesis": "thirty pounds, said the captain with finality.",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "\u00a330 said the captain with finality",
+ "normalized_hypothesis": "\u00a330 said the captain with finality",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0312",
+ "category": "phonetic_balance",
+ "text": "It was over when he made his way through the ring of spectators.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0312.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It was over when he made his way through the ring of spectators.",
+ "raw": {
+ "normalized_reference": "it was over when he made his way through the ring of spectators.",
+ "normalized_hypothesis": "it was over when he made his way through the ring of spectators.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it was over when he made his way through the ring of spectators",
+ "normalized_hypothesis": "it was over when he made his way through the ring of spectators",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0313",
+ "category": "phonetic_balance",
+ "text": "She had died from cold and starvation.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0313.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "She had died from cold and starvation.",
+ "raw": {
+ "normalized_reference": "she had died from cold and starvation.",
+ "normalized_hypothesis": "she had died from cold and starvation.",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "she had died from cold and starvation",
+ "normalized_hypothesis": "she had died from cold and starvation",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0314",
+ "category": "phonetic_balance",
+ "text": "We fished sharks on Niihau together.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0314.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "We fished sharks on Ni Hao together.",
+ "raw": {
+ "normalized_reference": "we fished sharks on niihau together.",
+ "normalized_hypothesis": "we fished sharks on ni hao together.",
+ "errors": 2,
+ "words": 6,
+ "wer": 0.3333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "we fished sharks on niihau together",
+ "normalized_hypothesis": "we fished sharks on ni hao together",
+ "errors": 2,
+ "words": 6,
+ "wer": 0.3333333333333333
+ }
+ },
+ {
+ "id": "release500_0315",
+ "category": "phonetic_balance",
+ "text": "His slim hands gripped the edges of the table.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0315.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "His slim hands gripped the edges of the table.",
+ "raw": {
+ "normalized_reference": "his slim hands gripped the edges of the table.",
+ "normalized_hypothesis": "his slim hands gripped the edges of the table.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "his slim hands gripped the edges of the table",
+ "normalized_hypothesis": "his slim hands gripped the edges of the table",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0316",
+ "category": "phonetic_balance",
+ "text": "We had been chased by them ourselves, more than once.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0316.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "We had been chased by them ourselves, more than once.",
+ "raw": {
+ "normalized_reference": "we had been chased by them ourselves, more than once.",
+ "normalized_hypothesis": "we had been chased by them ourselves, more than once.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "we had been chased by them ourselves more than once",
+ "normalized_hypothesis": "we had been chased by them ourselves more than once",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0317",
+ "category": "phonetic_balance",
+ "text": "Or have they already devised one.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0317.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "or have they already devised one?",
+ "raw": {
+ "normalized_reference": "or have they already devised one.",
+ "normalized_hypothesis": "or have they already devised one?",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "or have they already devised one",
+ "normalized_hypothesis": "or have they already devised one",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0318",
+ "category": "phonetic_balance",
+ "text": "I cannot follow you, she said.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0318.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I cannot follow you, she said.",
+ "raw": {
+ "normalized_reference": "i cannot follow you, she said.",
+ "normalized_hypothesis": "i cannot follow you, she said.",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i cannot follow you she said",
+ "normalized_hypothesis": "i cannot follow you she said",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0319",
+ "category": "phonetic_balance",
+ "text": "He went down in midstream, searching the shadows of both shores.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0319.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He went down in midstream, searching the shadows of both shores.",
+ "raw": {
+ "normalized_reference": "he went down in midstream, searching the shadows of both shores.",
+ "normalized_hypothesis": "he went down in midstream, searching the shadows of both shores.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he went down in midstream searching the shadows of both shores",
+ "normalized_hypothesis": "he went down in midstream searching the shadows of both shores",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0320",
+ "category": "phonetic_balance",
+ "text": "I was sick once -- typhoid.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0320.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I was sick once typhoid.",
+ "raw": {
+ "normalized_reference": "i was sick once -- typhoid.",
+ "normalized_hypothesis": "i was sick once typhoid.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i was sick once typhoid",
+ "normalized_hypothesis": "i was sick once typhoid",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0321",
+ "category": "phonetic_balance",
+ "text": "Why not like any railroad station or ferry depot.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0321.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Why not like any railroad station or ferry depot?",
+ "raw": {
+ "normalized_reference": "why not like any railroad station or ferry depot.",
+ "normalized_hypothesis": "why not like any railroad station or ferry depot?",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "why not like any railroad station or ferry depot",
+ "normalized_hypothesis": "why not like any railroad station or ferry depot",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0322",
+ "category": "phonetic_balance",
+ "text": "Red-Eye swung back and forth on the branch farther down.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0322.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Red Eye swung back and forth on the branch farther down.",
+ "raw": {
+ "normalized_reference": "red-eye swung back and forth on the branch farther down.",
+ "normalized_hypothesis": "red eye swung back and forth on the branch farther down.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "red eye swung back and forth on the branch farther down",
+ "normalized_hypothesis": "red eye swung back and forth on the branch farther down",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0323",
+ "category": "phonetic_balance",
+ "text": "Within himself he called it no longer his own.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0323.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Within himself he called it no longer his own.",
+ "raw": {
+ "normalized_reference": "within himself he called it no longer his own.",
+ "normalized_hypothesis": "within himself he called it no longer his own.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "within himself he called it no longer his own",
+ "normalized_hypothesis": "within himself he called it no longer his own",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0324",
+ "category": "phonetic_balance",
+ "text": "It lasted as a deterrent for two days.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0324.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It lasted as a deterrent for two days.",
+ "raw": {
+ "normalized_reference": "it lasted as a deterrent for two days.",
+ "normalized_hypothesis": "it lasted as a deterrent for two days.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it lasted as a deterrent for 2 days",
+ "normalized_hypothesis": "it lasted as a deterrent for 2 days",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0325",
+ "category": "phonetic_balance",
+ "text": "He loved to play Chinese lottery.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0325.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He loved to play Chinese lottery.",
+ "raw": {
+ "normalized_reference": "he loved to play chinese lottery.",
+ "normalized_hypothesis": "he loved to play chinese lottery.",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he loved to play chinese lottery",
+ "normalized_hypothesis": "he loved to play chinese lottery",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0326",
+ "category": "phonetic_balance",
+ "text": "Lord, but I'm glad to see you again, Phil.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0326.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Lord, but I'm glad to see you again, Phil.",
+ "raw": {
+ "normalized_reference": "lord, but i'm glad to see you again, phil.",
+ "normalized_hypothesis": "lord, but i'm glad to see you again, phil.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "lord but i am glad to see you again phil",
+ "normalized_hypothesis": "lord but i am glad to see you again phil",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0327",
+ "category": "phonetic_balance",
+ "text": "There is not an iota of truth in it, certainly not.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0327.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "There is not an iota of truth in it, certainly not.",
+ "raw": {
+ "normalized_reference": "there is not an iota of truth in it, certainly not.",
+ "normalized_hypothesis": "there is not an iota of truth in it, certainly not.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "there is not an iota of truth in it certainly not",
+ "normalized_hypothesis": "there is not an iota of truth in it certainly not",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0328",
+ "category": "phonetic_balance",
+ "text": "She added, with genuine sympathy in her face and voice.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0328.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "she added with genuine sympathy in her face and voice.",
+ "raw": {
+ "normalized_reference": "she added, with genuine sympathy in her face and voice.",
+ "normalized_hypothesis": "she added with genuine sympathy in her face and voice.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "she added with genuine sympathy in her face and voice",
+ "normalized_hypothesis": "she added with genuine sympathy in her face and voice",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0329",
+ "category": "phonetic_balance",
+ "text": "We have plenty of capital ourselves, and yet we want more.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0329.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "We have plenty of capital ourselves, and yet we want more.",
+ "raw": {
+ "normalized_reference": "we have plenty of capital ourselves, and yet we want more.",
+ "normalized_hypothesis": "we have plenty of capital ourselves, and yet we want more.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "we have plenty of capital ourselves and yet we want more",
+ "normalized_hypothesis": "we have plenty of capital ourselves and yet we want more",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0330",
+ "category": "phonetic_balance",
+ "text": "And now put yourself in my place for a moment.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0330.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "And now put yourself in my place for a moment.",
+ "raw": {
+ "normalized_reference": "and now put yourself in my place for a moment.",
+ "normalized_hypothesis": "and now put yourself in my place for a moment.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "and now put yourself in my place for a moment",
+ "normalized_hypothesis": "and now put yourself in my place for a moment",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0331",
+ "category": "phonetic_balance",
+ "text": "Fast, but endure.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0331.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Fast, but endure.",
+ "raw": {
+ "normalized_reference": "fast, but endure.",
+ "normalized_hypothesis": "fast, but endure.",
+ "errors": 0,
+ "words": 3,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "fast but endure",
+ "normalized_hypothesis": "fast but endure",
+ "errors": 0,
+ "words": 3,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0332",
+ "category": "phonetic_balance",
+ "text": "You were looking squeamish this afternoon, he began.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0332.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "You were looking squeamish this afternoon, he began.",
+ "raw": {
+ "normalized_reference": "you were looking squeamish this afternoon, he began.",
+ "normalized_hypothesis": "you were looking squeamish this afternoon, he began.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "you were looking squeamish this afternoon he began",
+ "normalized_hypothesis": "you were looking squeamish this afternoon he began",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0333",
+ "category": "phonetic_balance",
+ "text": "It is the nearest refuge.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0333.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It is the nearest refuge.",
+ "raw": {
+ "normalized_reference": "it is the nearest refuge.",
+ "normalized_hypothesis": "it is the nearest refuge.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it is the nearest refuge",
+ "normalized_hypothesis": "it is the nearest refuge",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0334",
+ "category": "phonetic_balance",
+ "text": "I want to die in it.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0334.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I want to die in it.",
+ "raw": {
+ "normalized_reference": "i want to die in it.",
+ "normalized_hypothesis": "i want to die in it.",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i want to die in it",
+ "normalized_hypothesis": "i want to die in it",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0335",
+ "category": "phonetic_balance",
+ "text": "He stopped, and Philip nodded at the horrified question in his eyes.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0335.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He stopped, and Philip nodded at the horrified question in his eyes.",
+ "raw": {
+ "normalized_reference": "he stopped, and philip nodded at the horrified question in his eyes.",
+ "normalized_hypothesis": "he stopped, and philip nodded at the horrified question in his eyes.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he stopped and philip nodded at the horrified question in his eyes",
+ "normalized_hypothesis": "he stopped and philip nodded at the horrified question in his eyes",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0336",
+ "category": "phonetic_balance",
+ "text": "She turned, fearing that Jacques might see what was in her face.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0336.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "She turned, fearing that Jack might see what was in her face.",
+ "raw": {
+ "normalized_reference": "she turned, fearing that jacques might see what was in her face.",
+ "normalized_hypothesis": "she turned, fearing that jack might see what was in her face.",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "she turned fearing that jacques might see what was in her face",
+ "normalized_hypothesis": "she turned fearing that jack might see what was in her face",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ }
+ },
+ {
+ "id": "release500_0337",
+ "category": "phonetic_balance",
+ "text": "It was a curious coincidence.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0337.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It was a curious coincidence.",
+ "raw": {
+ "normalized_reference": "it was a curious coincidence.",
+ "normalized_hypothesis": "it was a curious coincidence.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it was a curious coincidence",
+ "normalized_hypothesis": "it was a curious coincidence",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0338",
+ "category": "phonetic_balance",
+ "text": "Daylight made no answer, and the door closed behind him.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0338.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Daylight made no answer, and the door closed behind him.",
+ "raw": {
+ "normalized_reference": "daylight made no answer, and the door closed behind him.",
+ "normalized_hypothesis": "daylight made no answer, and the door closed behind him.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "daylight made no answer and the door closed behind him",
+ "normalized_hypothesis": "daylight made no answer and the door closed behind him",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0339",
+ "category": "phonetic_balance",
+ "text": "And he did hurt my arm.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0339.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "and he did hurt my arm.",
+ "raw": {
+ "normalized_reference": "and he did hurt my arm.",
+ "normalized_hypothesis": "and he did hurt my arm.",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "and he did hurt my arm",
+ "normalized_hypothesis": "and he did hurt my arm",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0340",
+ "category": "phonetic_balance",
+ "text": "Dunham, can your boy go along with Jesse.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0340.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Dunham, can your boy go along with Jessie?",
+ "raw": {
+ "normalized_reference": "dunham, can your boy go along with jesse.",
+ "normalized_hypothesis": "dunham, can your boy go along with jessie?",
+ "errors": 1,
+ "words": 8,
+ "wer": 0.125
+ },
+ "semantic": {
+ "normalized_reference": "dunham can your boy go along with jesse",
+ "normalized_hypothesis": "dunham can your boy go along with jessie",
+ "errors": 1,
+ "words": 8,
+ "wer": 0.125
+ }
+ },
+ {
+ "id": "release500_0341",
+ "category": "phonetic_balance",
+ "text": "The creative joy, I murmured.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0341.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The creative joy, I murmured.",
+ "raw": {
+ "normalized_reference": "the creative joy, i murmured.",
+ "normalized_hypothesis": "the creative joy, i murmured.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the creative joy i murmured",
+ "normalized_hypothesis": "the creative joy i murmured",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0342",
+ "category": "phonetic_balance",
+ "text": "He pulled, and the log crashed down to break his back.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0342.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He pulled, and the log crashed down to break his back.",
+ "raw": {
+ "normalized_reference": "he pulled, and the log crashed down to break his back.",
+ "normalized_hypothesis": "he pulled, and the log crashed down to break his back.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he pulled and the log crashed down to break his back",
+ "normalized_hypothesis": "he pulled and the log crashed down to break his back",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0343",
+ "category": "phonetic_balance",
+ "text": "By golly, the boy wins.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0343.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "By golly, the boy wins.",
+ "raw": {
+ "normalized_reference": "by golly, the boy wins.",
+ "normalized_hypothesis": "by golly, the boy wins.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "by golly the boy wins",
+ "normalized_hypothesis": "by golly the boy wins",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0344",
+ "category": "phonetic_balance",
+ "text": "The boy threw back his head with pride.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0344.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The boy threw back his head with pride.",
+ "raw": {
+ "normalized_reference": "the boy threw back his head with pride.",
+ "normalized_hypothesis": "the boy threw back his head with pride.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the boy threw back his head with pride",
+ "normalized_hypothesis": "the boy threw back his head with pride",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0345",
+ "category": "phonetic_balance",
+ "text": "The last one I knew was an overseer.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0345.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The last one I knew was an overseer.",
+ "raw": {
+ "normalized_reference": "the last one i knew was an overseer.",
+ "normalized_hypothesis": "the last one i knew was an overseer.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the last one i knew was an overseer",
+ "normalized_hypothesis": "the last one i knew was an overseer",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0346",
+ "category": "phonetic_balance",
+ "text": "Two years ago I gave up civilization for this.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0346.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Two years ago I gave up civilization for this.",
+ "raw": {
+ "normalized_reference": "two years ago i gave up civilization for this.",
+ "normalized_hypothesis": "two years ago i gave up civilization for this.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "2 years ago i gave up civilization for this",
+ "normalized_hypothesis": "2 years ago i gave up civilization for this",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0347",
+ "category": "phonetic_balance",
+ "text": "The river bared its bosom, and snorting steamboats challenged the wilderness.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0347.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The river bared its bosom, and snorting steamboats challenged the wilderness.",
+ "raw": {
+ "normalized_reference": "the river bared its bosom, and snorting steamboats challenged the wilderness.",
+ "normalized_hypothesis": "the river bared its bosom, and snorting steamboats challenged the wilderness.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the river bared its bosom and snorting steamboats challenged the wilderness",
+ "normalized_hypothesis": "the river bared its bosom and snorting steamboats challenged the wilderness",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0348",
+ "category": "phonetic_balance",
+ "text": "Mab, she said.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0348.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "\"'Bab,' she said.",
+ "raw": {
+ "normalized_reference": "mab, she said.",
+ "normalized_hypothesis": "\"'bab,' she said.",
+ "errors": 2,
+ "words": 3,
+ "wer": 0.6666666666666666
+ },
+ "semantic": {
+ "normalized_reference": "mab she said",
+ "normalized_hypothesis": "bab she said",
+ "errors": 1,
+ "words": 3,
+ "wer": 0.3333333333333333
+ }
+ },
+ {
+ "id": "release500_0349",
+ "category": "phonetic_balance",
+ "text": "The regret in his voice was provocative of a second burst of laughter.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0349.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The regret in his voice was provocative of a second burst of laughter.",
+ "raw": {
+ "normalized_reference": "the regret in his voice was provocative of a second burst of laughter.",
+ "normalized_hypothesis": "the regret in his voice was provocative of a second burst of laughter.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the regret in his voice was provocative of a 2nd burst of laughter",
+ "normalized_hypothesis": "the regret in his voice was provocative of a 2nd burst of laughter",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0350",
+ "category": "phonetic_balance",
+ "text": "And each year something happened, and I did not go.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0350.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "And each year something happened, and I did not go.",
+ "raw": {
+ "normalized_reference": "and each year something happened, and i did not go.",
+ "normalized_hypothesis": "and each year something happened, and i did not go.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "and each year something happened and i did not go",
+ "normalized_hypothesis": "and each year something happened and i did not go",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0351",
+ "category": "phonetic_balance",
+ "text": "Gad, your letter came just in time.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0351.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Gad, your letter came just in time.",
+ "raw": {
+ "normalized_reference": "gad, your letter came just in time.",
+ "normalized_hypothesis": "gad, your letter came just in time.",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "gad your letter came just in time",
+ "normalized_hypothesis": "gad your letter came just in time",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0352",
+ "category": "phonetic_balance",
+ "text": "But if Pierre did not return, until tomorrow.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0352.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "But if Pierre did not return until tomorrow,",
+ "raw": {
+ "normalized_reference": "but if pierre did not return, until tomorrow.",
+ "normalized_hypothesis": "but if pierre did not return until tomorrow,",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "but if pierre did not return until tomorrow",
+ "normalized_hypothesis": "but if pierre did not return until tomorrow",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0353",
+ "category": "phonetic_balance",
+ "text": "Billinger may arrive in time.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0353.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Bilingue may arrive in time.",
+ "raw": {
+ "normalized_reference": "billinger may arrive in time.",
+ "normalized_hypothesis": "bilingue may arrive in time.",
+ "errors": 1,
+ "words": 5,
+ "wer": 0.2
+ },
+ "semantic": {
+ "normalized_reference": "billinger may arrive in time",
+ "normalized_hypothesis": "bilingue may arrive in time",
+ "errors": 1,
+ "words": 5,
+ "wer": 0.2
+ }
+ },
+ {
+ "id": "release500_0354",
+ "category": "phonetic_balance",
+ "text": "In such a tumbling of values was no time to sell.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0354.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "In such a tumbling of values was no time to sell.",
+ "raw": {
+ "normalized_reference": "in such a tumbling of values was no time to sell.",
+ "normalized_hypothesis": "in such a tumbling of values was no time to sell.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "in such a tumbling of values was no time to sell",
+ "normalized_hypothesis": "in such a tumbling of values was no time to sell",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0355",
+ "category": "phonetic_balance",
+ "text": "There's not an iota of truth in it.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0355.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "There's not an iota of truth in it.",
+ "raw": {
+ "normalized_reference": "there's not an iota of truth in it.",
+ "normalized_hypothesis": "there's not an iota of truth in it.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "there is not an iota of truth in it",
+ "normalized_hypothesis": "there is not an iota of truth in it",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0356",
+ "category": "phonetic_balance",
+ "text": "It is dog eat dog, and you ate them up.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0356.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It is dog eat dog, and you ate them up.",
+ "raw": {
+ "normalized_reference": "it is dog eat dog, and you ate them up.",
+ "normalized_hypothesis": "it is dog eat dog, and you ate them up.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it is dog eat dog and you ate them up",
+ "normalized_hypothesis": "it is dog eat dog and you ate them up",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0357",
+ "category": "phonetic_balance",
+ "text": "You can take a vacation on pay.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0357.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "you can take a vacation on pay.",
+ "raw": {
+ "normalized_reference": "you can take a vacation on pay.",
+ "normalized_hypothesis": "you can take a vacation on pay.",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "you can take a vacation on pay",
+ "normalized_hypothesis": "you can take a vacation on pay",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0358",
+ "category": "phonetic_balance",
+ "text": "I have seen myself that one man contemplated by Pascal's philosophic eye.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0358.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I have seen myself that one man contemplated by Pascal's philosophic eye.",
+ "raw": {
+ "normalized_reference": "i have seen myself that one man contemplated by pascal's philosophic eye.",
+ "normalized_hypothesis": "i have seen myself that one man contemplated by pascal's philosophic eye.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i have seen myself that one man contemplated by pascal is philosophic eye",
+ "normalized_hypothesis": "i have seen myself that one man contemplated by pascal is philosophic eye",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0359",
+ "category": "phonetic_balance",
+ "text": "Ah, I had forgotten, he exclaimed.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0359.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Ah, I had forgotten, he exclaimed.",
+ "raw": {
+ "normalized_reference": "ah, i had forgotten, he exclaimed.",
+ "normalized_hypothesis": "ah, i had forgotten, he exclaimed.",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "ah i had forgotten he exclaimed",
+ "normalized_hypothesis": "ah i had forgotten he exclaimed",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0360",
+ "category": "phonetic_balance",
+ "text": "Providence had delivered him through the maelstrom.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0360.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Providence had delivered him through the maelstrom.",
+ "raw": {
+ "normalized_reference": "providence had delivered him through the maelstrom.",
+ "normalized_hypothesis": "providence had delivered him through the maelstrom.",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "providence had delivered him through the maelstrom",
+ "normalized_hypothesis": "providence had delivered him through the maelstrom",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0361",
+ "category": "phonetic_balance",
+ "text": "She was his now, forever.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0361.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "She was his now, forever.",
+ "raw": {
+ "normalized_reference": "she was his now, forever.",
+ "normalized_hypothesis": "she was his now, forever.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "she was his now forever",
+ "normalized_hypothesis": "she was his now forever",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0362",
+ "category": "phonetic_balance",
+ "text": "It was unobtrusive, yet it was there.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0362.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It was unobtrusive, yet it was there.",
+ "raw": {
+ "normalized_reference": "it was unobtrusive, yet it was there.",
+ "normalized_hypothesis": "it was unobtrusive, yet it was there.",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it was unobtrusive yet it was there",
+ "normalized_hypothesis": "it was unobtrusive yet it was there",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0363",
+ "category": "phonetic_balance",
+ "text": "Men like Joe Goose dated existence from drunk to drunk.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0363.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Men like Joe Goose dated existence from drunk to drunk.",
+ "raw": {
+ "normalized_reference": "men like joe goose dated existence from drunk to drunk.",
+ "normalized_hypothesis": "men like joe goose dated existence from drunk to drunk.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "men like joe goose dated existence from drunk to drunk",
+ "normalized_hypothesis": "men like joe goose dated existence from drunk to drunk",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0364",
+ "category": "phonetic_balance",
+ "text": "I was completely lost in my work.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0364.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I was completely lost in my work.",
+ "raw": {
+ "normalized_reference": "i was completely lost in my work.",
+ "normalized_hypothesis": "i was completely lost in my work.",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i was completely lost in my work",
+ "normalized_hypothesis": "i was completely lost in my work",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0365",
+ "category": "phonetic_balance",
+ "text": "The very thing, Ernest agreed.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0365.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The very thing, Ernest agreed.",
+ "raw": {
+ "normalized_reference": "the very thing, ernest agreed.",
+ "normalized_hypothesis": "the very thing, ernest agreed.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the very thing ernest agreed",
+ "normalized_hypothesis": "the very thing ernest agreed",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0366",
+ "category": "phonetic_balance",
+ "text": "He was the leader, and Tudor was his lieutenant.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0366.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He was the leader, and Tudor was his lieutenant.",
+ "raw": {
+ "normalized_reference": "he was the leader, and tudor was his lieutenant.",
+ "normalized_hypothesis": "he was the leader, and tudor was his lieutenant.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he was the leader and tudor was his lieutenant",
+ "normalized_hypothesis": "he was the leader and tudor was his lieutenant",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0367",
+ "category": "phonetic_balance",
+ "text": "The Fire-Men wore animal skins around their waists and across their shoulders.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0367.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The firemen wore animal skins around their waists and across their shoulders.",
+ "raw": {
+ "normalized_reference": "the fire-men wore animal skins around their waists and across their shoulders.",
+ "normalized_hypothesis": "the firemen wore animal skins around their waists and across their shoulders.",
+ "errors": 2,
+ "words": 13,
+ "wer": 0.15384615384615385
+ },
+ "semantic": {
+ "normalized_reference": "the fire men wore animal skins around their waists and across their shoulders",
+ "normalized_hypothesis": "the firemen wore animal skins around their waists and across their shoulders",
+ "errors": 2,
+ "words": 13,
+ "wer": 0.15384615384615385
+ }
+ },
+ {
+ "id": "release500_0368",
+ "category": "phonetic_balance",
+ "text": "We could throw stones with our feet.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0368.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "We could throw stones with our feet",
+ "raw": {
+ "normalized_reference": "we could throw stones with our feet.",
+ "normalized_hypothesis": "we could throw stones with our feet",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "we could throw stones with our feet",
+ "normalized_hypothesis": "we could throw stones with our feet",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0369",
+ "category": "phonetic_balance",
+ "text": "He turned sharply, and faced Gregson across the table.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0369.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He turned sharply and faced Gregson across the table.",
+ "raw": {
+ "normalized_reference": "he turned sharply, and faced gregson across the table.",
+ "normalized_hypothesis": "he turned sharply and faced gregson across the table.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he turned sharply and faced gregson across the table",
+ "normalized_hypothesis": "he turned sharply and faced gregson across the table",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0370",
+ "category": "phonetic_balance",
+ "text": "They are to attack your camp tomorrow night.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0370.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "They are to attack your camp tomorrow night.",
+ "raw": {
+ "normalized_reference": "they are to attack your camp tomorrow night.",
+ "normalized_hypothesis": "they are to attack your camp tomorrow night.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "they are to attack your camp tomorrow night",
+ "normalized_hypothesis": "they are to attack your camp tomorrow night",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0371",
+ "category": "phonetic_balance",
+ "text": "And watch out for wet feet, was his parting advice.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0371.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "and watch out for wet feet, was his parting advice.",
+ "raw": {
+ "normalized_reference": "and watch out for wet feet, was his parting advice.",
+ "normalized_hypothesis": "and watch out for wet feet, was his parting advice.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "and watch out for wet feet was his parting advice",
+ "normalized_hypothesis": "and watch out for wet feet was his parting advice",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0372",
+ "category": "phonetic_balance",
+ "text": "Two weeks passed, and in that time Thorpe left camp three times.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0372.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Two weeks passed, and in that time, Thorpe left camp three times.",
+ "raw": {
+ "normalized_reference": "two weeks passed, and in that time thorpe left camp three times.",
+ "normalized_hypothesis": "two weeks passed, and in that time, thorpe left camp three times.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "2 weeks passed and in that time thorpe left camp 3 times",
+ "normalized_hypothesis": "2 weeks passed and in that time thorpe left camp 3 times",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0373",
+ "category": "phonetic_balance",
+ "text": "But ever his gaze returned to that Crouched Venus on the piano.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0373.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "But ever his gaze returned to that crouched Venus on the piano.",
+ "raw": {
+ "normalized_reference": "but ever his gaze returned to that crouched venus on the piano.",
+ "normalized_hypothesis": "but ever his gaze returned to that crouched venus on the piano.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "but ever his gaze returned to that crouched venus on the piano",
+ "normalized_hypothesis": "but ever his gaze returned to that crouched venus on the piano",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0374",
+ "category": "phonetic_balance",
+ "text": "But she had become an automaton.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0374.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "but she had become an automaton.",
+ "raw": {
+ "normalized_reference": "but she had become an automaton.",
+ "normalized_hypothesis": "but she had become an automaton.",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "but she had become an automaton",
+ "normalized_hypothesis": "but she had become an automaton",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0375",
+ "category": "phonetic_balance",
+ "text": "I do not blame you for anything; remember that.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0375.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I do not blame you for anything. Remember that.",
+ "raw": {
+ "normalized_reference": "i do not blame you for anything; remember that.",
+ "normalized_hypothesis": "i do not blame you for anything. remember that.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i do not blame you for anything remember that",
+ "normalized_hypothesis": "i do not blame you for anything remember that",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0376",
+ "category": "phonetic_balance",
+ "text": "The next thing to watch out for is bed sores.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0376.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The next thing to watch out for is bed sores.",
+ "raw": {
+ "normalized_reference": "the next thing to watch out for is bed sores.",
+ "normalized_hypothesis": "the next thing to watch out for is bed sores.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the next thing to watch out for is bed sores",
+ "normalized_hypothesis": "the next thing to watch out for is bed sores",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0377",
+ "category": "phonetic_balance",
+ "text": "He will follow us soon.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0377.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He will follow us soon.",
+ "raw": {
+ "normalized_reference": "he will follow us soon.",
+ "normalized_hypothesis": "he will follow us soon.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he will follow us soon",
+ "normalized_hypothesis": "he will follow us soon",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0378",
+ "category": "phonetic_balance",
+ "text": "His beady black eyes saw bargains where other men saw bankruptcy.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0378.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "His beady black eyes saw bargains where other men saw bankruptcy.",
+ "raw": {
+ "normalized_reference": "his beady black eyes saw bargains where other men saw bankruptcy.",
+ "normalized_hypothesis": "his beady black eyes saw bargains where other men saw bankruptcy.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "his beady black eyes saw bargains where other men saw bankruptcy",
+ "normalized_hypothesis": "his beady black eyes saw bargains where other men saw bankruptcy",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0379",
+ "category": "phonetic_balance",
+ "text": "They robbed me a few years later.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0379.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "They robbed me a few years later.",
+ "raw": {
+ "normalized_reference": "they robbed me a few years later.",
+ "normalized_hypothesis": "they robbed me a few years later.",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "they robbed me a few years later",
+ "normalized_hypothesis": "they robbed me a few years later",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0380",
+ "category": "phonetic_balance",
+ "text": "Gregson held a lighted match until it burnt his fingertips.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0380.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Gregson held a lighted match until it burned his fingertips.",
+ "raw": {
+ "normalized_reference": "gregson held a lighted match until it burnt his fingertips.",
+ "normalized_hypothesis": "gregson held a lighted match until it burned his fingertips.",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ },
+ "semantic": {
+ "normalized_reference": "gregson held a lighted match until it burnt his fingertips",
+ "normalized_hypothesis": "gregson held a lighted match until it burned his fingertips",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ }
+ },
+ {
+ "id": "release500_0381",
+ "category": "phonetic_balance",
+ "text": "A burst of laughter was his reward.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0381.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "A burst of laughter was his reward.",
+ "raw": {
+ "normalized_reference": "a burst of laughter was his reward.",
+ "normalized_hypothesis": "a burst of laughter was his reward.",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "a burst of laughter was his reward",
+ "normalized_hypothesis": "a burst of laughter was his reward",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0382",
+ "category": "phonetic_balance",
+ "text": "He cried, and swung the club wildly.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0382.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He cried and swung the club wildly.",
+ "raw": {
+ "normalized_reference": "he cried, and swung the club wildly.",
+ "normalized_hypothesis": "he cried and swung the club wildly.",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he cried and swung the club wildly",
+ "normalized_hypothesis": "he cried and swung the club wildly",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0383",
+ "category": "phonetic_balance",
+ "text": "Sandel would never become a world champion.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0383.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Sandal would never become a world champion.",
+ "raw": {
+ "normalized_reference": "sandel would never become a world champion.",
+ "normalized_hypothesis": "sandal would never become a world champion.",
+ "errors": 1,
+ "words": 7,
+ "wer": 0.14285714285714285
+ },
+ "semantic": {
+ "normalized_reference": "sandel would never become a world champion",
+ "normalized_hypothesis": "sandal would never become a world champion",
+ "errors": 1,
+ "words": 7,
+ "wer": 0.14285714285714285
+ }
+ },
+ {
+ "id": "release500_0384",
+ "category": "phonetic_balance",
+ "text": "And after the bath a shave would not be bad.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0384.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "and after the Bathyshev would not be bad.",
+ "raw": {
+ "normalized_reference": "and after the bath a shave would not be bad.",
+ "normalized_hypothesis": "and after the bathyshev would not be bad.",
+ "errors": 3,
+ "words": 10,
+ "wer": 0.3
+ },
+ "semantic": {
+ "normalized_reference": "and after the bath a shave would not be bad",
+ "normalized_hypothesis": "and after the bathyshev would not be bad",
+ "errors": 3,
+ "words": 10,
+ "wer": 0.3
+ }
+ },
+ {
+ "id": "release500_0385",
+ "category": "phonetic_balance",
+ "text": "They saw each other for the first time in Boston.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0385.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "They saw each other for the first time in Boston.",
+ "raw": {
+ "normalized_reference": "they saw each other for the first time in boston.",
+ "normalized_hypothesis": "they saw each other for the first time in boston.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "they saw each other for the 1st time in boston",
+ "normalized_hypothesis": "they saw each other for the 1st time in boston",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0386",
+ "category": "phonetic_balance",
+ "text": "I had been born with no organic, chemical predisposition toward alcohol.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0386.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I had been born with no organic, chemical predisposition toward alcohol.",
+ "raw": {
+ "normalized_reference": "i had been born with no organic, chemical predisposition toward alcohol.",
+ "normalized_hypothesis": "i had been born with no organic, chemical predisposition toward alcohol.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i had been born with no organic chemical predisposition toward alcohol",
+ "normalized_hypothesis": "i had been born with no organic chemical predisposition toward alcohol",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0387",
+ "category": "phonetic_balance",
+ "text": "They had no fixed values, to be altered by adjectives and adverbs.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0387.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "They had no fixed values to be altered by adjectives and adverbs.",
+ "raw": {
+ "normalized_reference": "they had no fixed values, to be altered by adjectives and adverbs.",
+ "normalized_hypothesis": "they had no fixed values to be altered by adjectives and adverbs.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "they had no fixed values to be altered by adjectives and adverbs",
+ "normalized_hypothesis": "they had no fixed values to be altered by adjectives and adverbs",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0388",
+ "category": "phonetic_balance",
+ "text": "Nor was Elam Harnish an exception.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0388.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "nor was the lamb harness an exception.",
+ "raw": {
+ "normalized_reference": "nor was elam harnish an exception.",
+ "normalized_hypothesis": "nor was the lamb harness an exception.",
+ "errors": 3,
+ "words": 6,
+ "wer": 0.5
+ },
+ "semantic": {
+ "normalized_reference": "nor was elam harnish an exception",
+ "normalized_hypothesis": "nor was the lamb harness an exception",
+ "errors": 3,
+ "words": 6,
+ "wer": 0.5
+ }
+ },
+ {
+ "id": "release500_0389",
+ "category": "phonetic_balance",
+ "text": "He was a wise hyena.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0389.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He was a wise hyena.",
+ "raw": {
+ "normalized_reference": "he was a wise hyena.",
+ "normalized_hypothesis": "he was a wise hyena.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he was a wise hyena",
+ "normalized_hypothesis": "he was a wise hyena",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0390",
+ "category": "phonetic_balance",
+ "text": "And now behold the perversity of things.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0390.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "and now behold the perversity of things.",
+ "raw": {
+ "normalized_reference": "and now behold the perversity of things.",
+ "normalized_hypothesis": "and now behold the perversity of things.",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "and now behold the perversity of things",
+ "normalized_hypothesis": "and now behold the perversity of things",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0391",
+ "category": "phonetic_balance",
+ "text": "The Fire People, like ourselves, lived in caves.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0391.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The fire people, like ourselves, live in caves.",
+ "raw": {
+ "normalized_reference": "the fire people, like ourselves, lived in caves.",
+ "normalized_hypothesis": "the fire people, like ourselves, live in caves.",
+ "errors": 1,
+ "words": 8,
+ "wer": 0.125
+ },
+ "semantic": {
+ "normalized_reference": "the fire people like ourselves lived in caves",
+ "normalized_hypothesis": "the fire people like ourselves live in caves",
+ "errors": 1,
+ "words": 8,
+ "wer": 0.125
+ }
+ },
+ {
+ "id": "release500_0392",
+ "category": "phonetic_balance",
+ "text": "To her the bridge was tambo, which is the native for taboo.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0392.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "to where the bridge was tambo, which is the native for taboo.",
+ "raw": {
+ "normalized_reference": "to her the bridge was tambo, which is the native for taboo.",
+ "normalized_hypothesis": "to where the bridge was tambo, which is the native for taboo.",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "to her the bridge was tambo which is the native for taboo",
+ "normalized_hypothesis": "to where the bridge was tambo which is the native for taboo",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ }
+ },
+ {
+ "id": "release500_0393",
+ "category": "phonetic_balance",
+ "text": "The voyage was our idea of a good time.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0393.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The voyage was our idea of a good time.",
+ "raw": {
+ "normalized_reference": "the voyage was our idea of a good time.",
+ "normalized_hypothesis": "the voyage was our idea of a good time.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the voyage was our idea of a good time",
+ "normalized_hypothesis": "the voyage was our idea of a good time",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0394",
+ "category": "phonetic_balance",
+ "text": "Of course much grumbling went on, and little outbursts were continually occurring.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0394.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Of course much grumbling went on, and little outbursts were continually occurring.",
+ "raw": {
+ "normalized_reference": "of course much grumbling went on, and little outbursts were continually occurring.",
+ "normalized_hypothesis": "of course much grumbling went on, and little outbursts were continually occurring.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "of course much grumbling went on and little outbursts were continually occurring",
+ "normalized_hypothesis": "of course much grumbling went on and little outbursts were continually occurring",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0395",
+ "category": "phonetic_balance",
+ "text": "Get down and dig in.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0395.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Get down and dig in.",
+ "raw": {
+ "normalized_reference": "get down and dig in.",
+ "normalized_hypothesis": "get down and dig in.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "get down and dig in",
+ "normalized_hypothesis": "get down and dig in",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0396",
+ "category": "phonetic_balance",
+ "text": "It was put together in a casual, helter-skelter sort of way.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0396.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It was put together in a casual, helter-skelter sort of way.",
+ "raw": {
+ "normalized_reference": "it was put together in a casual, helter-skelter sort of way.",
+ "normalized_hypothesis": "it was put together in a casual, helter-skelter sort of way.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it was put together in a casual helter skelter sort of way",
+ "normalized_hypothesis": "it was put together in a casual helter skelter sort of way",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0397",
+ "category": "phonetic_balance",
+ "text": "He deluged me, overwhelmed me with argument.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0397.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He deluged me, overwhelmed me with argument.",
+ "raw": {
+ "normalized_reference": "he deluged me, overwhelmed me with argument.",
+ "normalized_hypothesis": "he deluged me, overwhelmed me with argument.",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he deluged me overwhelmed me with argument",
+ "normalized_hypothesis": "he deluged me overwhelmed me with argument",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0398",
+ "category": "phonetic_balance",
+ "text": "The girl faced him, her eyes shining with sudden fear.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0398.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The girl faced him, her eyes shining with sudden fear.",
+ "raw": {
+ "normalized_reference": "the girl faced him, her eyes shining with sudden fear.",
+ "normalized_hypothesis": "the girl faced him, her eyes shining with sudden fear.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the girl faced him her eyes shining with sudden fear",
+ "normalized_hypothesis": "the girl faced him her eyes shining with sudden fear",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0399",
+ "category": "phonetic_balance",
+ "text": "It will break our hearts and our backs to hoist anchor by hand.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0399.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It will break our hearts and our backs to hoist anchor by hand.",
+ "raw": {
+ "normalized_reference": "it will break our hearts and our backs to hoist anchor by hand.",
+ "normalized_hypothesis": "it will break our hearts and our backs to hoist anchor by hand.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it will break our hearts and our backs to hoist anchor by hand",
+ "normalized_hypothesis": "it will break our hearts and our backs to hoist anchor by hand",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0400",
+ "category": "phonetic_balance",
+ "text": "He made no reply as he waited for Whittemore to continue.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0400.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He made no reply as he waited for Whittemore to continue.",
+ "raw": {
+ "normalized_reference": "he made no reply as he waited for whittemore to continue.",
+ "normalized_hypothesis": "he made no reply as he waited for whittemore to continue.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he made no reply as he waited for whittemore to continue",
+ "normalized_hypothesis": "he made no reply as he waited for whittemore to continue",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0401",
+ "category": "phonetic_balance",
+ "text": "Who the devil gave it to you to be judge and jury.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0401.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "who the devil gave it to you to be judge and jury.",
+ "raw": {
+ "normalized_reference": "who the devil gave it to you to be judge and jury.",
+ "normalized_hypothesis": "who the devil gave it to you to be judge and jury.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "who the devil gave it to you to be judge and jury",
+ "normalized_hypothesis": "who the devil gave it to you to be judge and jury",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0402",
+ "category": "phonetic_balance",
+ "text": "Daylight found himself charmed and made curious by the little man.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0402.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Daylight found himself charmed and made curious by the little men.",
+ "raw": {
+ "normalized_reference": "daylight found himself charmed and made curious by the little man.",
+ "normalized_hypothesis": "daylight found himself charmed and made curious by the little men.",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ },
+ "semantic": {
+ "normalized_reference": "daylight found himself charmed and made curious by the little man",
+ "normalized_hypothesis": "daylight found himself charmed and made curious by the little men",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ }
+ },
+ {
+ "id": "release500_0403",
+ "category": "phonetic_balance",
+ "text": "It's a Yankee, Joan cried.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0403.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It's a Yankee, Joan cried.",
+ "raw": {
+ "normalized_reference": "it's a yankee, joan cried.",
+ "normalized_hypothesis": "it's a yankee, joan cried.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it is a yankee joan cried",
+ "normalized_hypothesis": "it is a yankee joan cried",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0404",
+ "category": "phonetic_balance",
+ "text": "But what they want with your toothbrush is more than I can imagine.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0404.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "But what they want with your toothbrush is more than I can imagine.",
+ "raw": {
+ "normalized_reference": "but what they want with your toothbrush is more than i can imagine.",
+ "normalized_hypothesis": "but what they want with your toothbrush is more than i can imagine.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "but what they want with your toothbrush is more than i can imagine",
+ "normalized_hypothesis": "but what they want with your toothbrush is more than i can imagine",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0405",
+ "category": "phonetic_balance",
+ "text": "They were babbling and chattering all together.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0405.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "They were babbling and chattering altogether.",
+ "raw": {
+ "normalized_reference": "they were babbling and chattering all together.",
+ "normalized_hypothesis": "they were babbling and chattering altogether.",
+ "errors": 2,
+ "words": 7,
+ "wer": 0.2857142857142857
+ },
+ "semantic": {
+ "normalized_reference": "they were babbling and chattering all together",
+ "normalized_hypothesis": "they were babbling and chattering altogether",
+ "errors": 2,
+ "words": 7,
+ "wer": 0.2857142857142857
+ }
+ },
+ {
+ "id": "release500_0406",
+ "category": "phonetic_balance",
+ "text": "Men of Selden's stamp don't stop at women and children.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0406.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Men of Seldon's stamp don't stop at women and children.",
+ "raw": {
+ "normalized_reference": "men of selden's stamp don't stop at women and children.",
+ "normalized_hypothesis": "men of seldon's stamp don't stop at women and children.",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ },
+ "semantic": {
+ "normalized_reference": "men of selden is stamp do not stop at women and children",
+ "normalized_hypothesis": "men of seldon is stamp do not stop at women and children",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ }
+ },
+ {
+ "id": "release500_0407",
+ "category": "phonetic_balance",
+ "text": "Cameron looked at his hands with their long, sinewy fingers.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0407.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Cameron looked at his hands with their long sinewy fingers.",
+ "raw": {
+ "normalized_reference": "cameron looked at his hands with their long, sinewy fingers.",
+ "normalized_hypothesis": "cameron looked at his hands with their long sinewy fingers.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "cameron looked at his hands with their long sinewy fingers",
+ "normalized_hypothesis": "cameron looked at his hands with their long sinewy fingers",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0408",
+ "category": "phonetic_balance",
+ "text": "Saxon waited, for she knew a fresh idea had struck Billy.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0408.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Saxon waited, for she knew a fresh idea had struck Billy.",
+ "raw": {
+ "normalized_reference": "saxon waited, for she knew a fresh idea had struck billy.",
+ "normalized_hypothesis": "saxon waited, for she knew a fresh idea had struck billy.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "saxon waited for she knew a fresh idea had struck billy",
+ "normalized_hypothesis": "saxon waited for she knew a fresh idea had struck billy",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0409",
+ "category": "phonetic_balance",
+ "text": "He saw only the effect in a general, sketchy way.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0409.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He saw only the effect in a general, sketchy way.",
+ "raw": {
+ "normalized_reference": "he saw only the effect in a general, sketchy way.",
+ "normalized_hypothesis": "he saw only the effect in a general, sketchy way.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he saw only the effect in a general sketchy way",
+ "normalized_hypothesis": "he saw only the effect in a general sketchy way",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0410",
+ "category": "phonetic_balance",
+ "text": "He began to follow the footprints of the dog.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0410.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He began to follow the footprints of the dog.",
+ "raw": {
+ "normalized_reference": "he began to follow the footprints of the dog.",
+ "normalized_hypothesis": "he began to follow the footprints of the dog.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he began to follow the footprints of the dog",
+ "normalized_hypothesis": "he began to follow the footprints of the dog",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0411",
+ "category": "phonetic_balance",
+ "text": "Do you value your hide.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0411.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Do you value your height?",
+ "raw": {
+ "normalized_reference": "do you value your hide.",
+ "normalized_hypothesis": "do you value your height?",
+ "errors": 1,
+ "words": 5,
+ "wer": 0.2
+ },
+ "semantic": {
+ "normalized_reference": "do you value your hide",
+ "normalized_hypothesis": "do you value your height",
+ "errors": 1,
+ "words": 5,
+ "wer": 0.2
+ }
+ },
+ {
+ "id": "release500_0412",
+ "category": "phonetic_balance",
+ "text": "A rising tide of fat had submerged them.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0412.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "A rising tide of fat had submerged them.",
+ "raw": {
+ "normalized_reference": "a rising tide of fat had submerged them.",
+ "normalized_hypothesis": "a rising tide of fat had submerged them.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "a rising tide of fat had submerged them",
+ "normalized_hypothesis": "a rising tide of fat had submerged them",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0413",
+ "category": "phonetic_balance",
+ "text": "There was nothing on the rock.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0413.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "There was nothing on the rock.",
+ "raw": {
+ "normalized_reference": "there was nothing on the rock.",
+ "normalized_hypothesis": "there was nothing on the rock.",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "there was nothing on the rock",
+ "normalized_hypothesis": "there was nothing on the rock",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0414",
+ "category": "phonetic_balance",
+ "text": "Whatever he guessed he locked away in the taboo room of Naomi.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0414.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Whatever he guessed, he locked away in the taboo room of Naomi.",
+ "raw": {
+ "normalized_reference": "whatever he guessed he locked away in the taboo room of naomi.",
+ "normalized_hypothesis": "whatever he guessed, he locked away in the taboo room of naomi.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "whatever he guessed he locked away in the taboo room of naomi",
+ "normalized_hypothesis": "whatever he guessed he locked away in the taboo room of naomi",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0415",
+ "category": "phonetic_balance",
+ "text": "Will we ever forget it.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0415.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Will we ever forget it?",
+ "raw": {
+ "normalized_reference": "will we ever forget it.",
+ "normalized_hypothesis": "will we ever forget it?",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "will we ever forget it",
+ "normalized_hypothesis": "will we ever forget it",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0416",
+ "category": "phonetic_balance",
+ "text": "Jeanne and Pierre both gazed toward the great rock.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0416.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Jean and Pierre both gazed toward the great rock.",
+ "raw": {
+ "normalized_reference": "jeanne and pierre both gazed toward the great rock.",
+ "normalized_hypothesis": "jean and pierre both gazed toward the great rock.",
+ "errors": 1,
+ "words": 9,
+ "wer": 0.1111111111111111
+ },
+ "semantic": {
+ "normalized_reference": "jeanne and pierre both gazed toward the great rock",
+ "normalized_hypothesis": "jean and pierre both gazed toward the great rock",
+ "errors": 1,
+ "words": 9,
+ "wer": 0.1111111111111111
+ }
+ },
+ {
+ "id": "release500_0417",
+ "category": "phonetic_balance",
+ "text": "Does the old boy often go off at half-cock that way.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0417.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Does the old boy often go off at half-cock that way?",
+ "raw": {
+ "normalized_reference": "does the old boy often go off at half-cock that way.",
+ "normalized_hypothesis": "does the old boy often go off at half-cock that way?",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "does the old boy often go off at half cock that way",
+ "normalized_hypothesis": "does the old boy often go off at half cock that way",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0418",
+ "category": "phonetic_balance",
+ "text": "The temperature dropped to fifty below zero and remained there the whole trip.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0418.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The temperature dropped to 50 below zero and remained there the whole trip.",
+ "raw": {
+ "normalized_reference": "the temperature dropped to fifty below zero and remained there the whole trip.",
+ "normalized_hypothesis": "the temperature dropped to 50 below zero and remained there the whole trip.",
+ "errors": 1,
+ "words": 13,
+ "wer": 0.07692307692307693
+ },
+ "semantic": {
+ "normalized_reference": "the temperature dropped to 50 below 0 and remained there the whole trip",
+ "normalized_hypothesis": "the temperature dropped to 50 below 0 and remained there the whole trip",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0419",
+ "category": "phonetic_balance",
+ "text": "It's worth eight dollars.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0419.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It's worth $8.",
+ "raw": {
+ "normalized_reference": "it's worth eight dollars.",
+ "normalized_hypothesis": "it's worth $8.",
+ "errors": 2,
+ "words": 4,
+ "wer": 0.5
+ },
+ "semantic": {
+ "normalized_reference": "it is worth $8",
+ "normalized_hypothesis": "it is worth $8",
+ "errors": 0,
+ "words": 4,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0420",
+ "category": "phonetic_balance",
+ "text": "But I did not enjoy it long.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0420.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "But I did not enjoy it lol",
+ "raw": {
+ "normalized_reference": "but i did not enjoy it long.",
+ "normalized_hypothesis": "but i did not enjoy it lol",
+ "errors": 1,
+ "words": 7,
+ "wer": 0.14285714285714285
+ },
+ "semantic": {
+ "normalized_reference": "but i did not enjoy it long",
+ "normalized_hypothesis": "but i did not enjoy it lol",
+ "errors": 1,
+ "words": 7,
+ "wer": 0.14285714285714285
+ }
+ },
+ {
+ "id": "release500_0421",
+ "category": "phonetic_balance",
+ "text": "Shorty turned to their employers.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0421.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Shorty turned to their employers.",
+ "raw": {
+ "normalized_reference": "shorty turned to their employers.",
+ "normalized_hypothesis": "shorty turned to their employers.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "shorty turned to their employers",
+ "normalized_hypothesis": "shorty turned to their employers",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0422",
+ "category": "phonetic_balance",
+ "text": "It was the third or fourth time that Philip had heard MacDougall swear.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0422.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It was the third or fourth time that Philip had heard MacDougall swear.",
+ "raw": {
+ "normalized_reference": "it was the third or fourth time that philip had heard macdougall swear.",
+ "normalized_hypothesis": "it was the third or fourth time that philip had heard macdougall swear.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it was the 3rd or 4th time that philip had heard macdougall swear",
+ "normalized_hypothesis": "it was the 3rd or 4th time that philip had heard macdougall swear",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0423",
+ "category": "phonetic_balance",
+ "text": "Ernest saw in the affair the most sinister import.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0423.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Ernest saw in the affair the most sinister import.",
+ "raw": {
+ "normalized_reference": "ernest saw in the affair the most sinister import.",
+ "normalized_hypothesis": "ernest saw in the affair the most sinister import.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "ernest saw in the affair the most sinister import",
+ "normalized_hypothesis": "ernest saw in the affair the most sinister import",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0424",
+ "category": "phonetic_balance",
+ "text": "Also, at regular intervals, he would mutter.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0424.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Also, at regular intervals, he would mutter,",
+ "raw": {
+ "normalized_reference": "also, at regular intervals, he would mutter.",
+ "normalized_hypothesis": "also, at regular intervals, he would mutter,",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "also at regular intervals he would mutter",
+ "normalized_hypothesis": "also at regular intervals he would mutter",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0425",
+ "category": "phonetic_balance",
+ "text": "He may anticipate the day of his death.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0425.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "he may anticipate the day of his death.",
+ "raw": {
+ "normalized_reference": "he may anticipate the day of his death.",
+ "normalized_hypothesis": "he may anticipate the day of his death.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he may anticipate the day of his death",
+ "normalized_hypothesis": "he may anticipate the day of his death",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0426",
+ "category": "phonetic_balance",
+ "text": "I defy any man to get a Solomon Island sore in California.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0426.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I defy any man to get a Solomon Island sore in California.",
+ "raw": {
+ "normalized_reference": "i defy any man to get a solomon island sore in california.",
+ "normalized_hypothesis": "i defy any man to get a solomon island sore in california.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i defy any man to get a solomon island sore in california",
+ "normalized_hypothesis": "i defy any man to get a solomon island sore in california",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0427",
+ "category": "phonetic_balance",
+ "text": "Now our figuring was all right.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0427.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Now our figuring was all right.",
+ "raw": {
+ "normalized_reference": "now our figuring was all right.",
+ "normalized_hypothesis": "now our figuring was all right.",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "now our figuring was all right",
+ "normalized_hypothesis": "now our figuring was all right",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0428",
+ "category": "phonetic_balance",
+ "text": "MacDougall tapped his forehead suspiciously with a stubby forefinger.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0428.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "MacDougall tapped his forehead suspiciously with a stubby forefinger.",
+ "raw": {
+ "normalized_reference": "macdougall tapped his forehead suspiciously with a stubby forefinger.",
+ "normalized_hypothesis": "macdougall tapped his forehead suspiciously with a stubby forefinger.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "macdougall tapped his forehead suspiciously with a stubby forefinger",
+ "normalized_hypothesis": "macdougall tapped his forehead suspiciously with a stubby forefinger",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0429",
+ "category": "phonetic_balance",
+ "text": "O'Brien took off his coat and bared his right arm.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0429.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "O'Brien took off his coat and bared his right arm.",
+ "raw": {
+ "normalized_reference": "o'brien took off his coat and bared his right arm.",
+ "normalized_hypothesis": "o'brien took off his coat and bared his right arm.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "0 brien took off his coat and bared his right arm",
+ "normalized_hypothesis": "0 brien took off his coat and bared his right arm",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0430",
+ "category": "phonetic_balance",
+ "text": "But it won't continue, she said with easy confidence.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0430.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "but it won't continue she said with easy confidence",
+ "raw": {
+ "normalized_reference": "but it won't continue, she said with easy confidence.",
+ "normalized_hypothesis": "but it won't continue she said with easy confidence",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "but it will not continue she said with easy confidence",
+ "normalized_hypothesis": "but it will not continue she said with easy confidence",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0431",
+ "category": "phonetic_balance",
+ "text": "I was still weak from my prolonged immersion.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0431.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I was still weak from my prolonged immersion.",
+ "raw": {
+ "normalized_reference": "i was still weak from my prolonged immersion.",
+ "normalized_hypothesis": "i was still weak from my prolonged immersion.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i was still weak from my prolonged immersion",
+ "normalized_hypothesis": "i was still weak from my prolonged immersion",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0432",
+ "category": "phonetic_balance",
+ "text": "Oolong was two hundred and fifty miles from the nearest land.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0432.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Wulong was 250 miles from the nearest land.",
+ "raw": {
+ "normalized_reference": "oolong was two hundred and fifty miles from the nearest land.",
+ "normalized_hypothesis": "wulong was 250 miles from the nearest land.",
+ "errors": 5,
+ "words": 11,
+ "wer": 0.45454545454545453
+ },
+ "semantic": {
+ "normalized_reference": "oolong was 250 miles from the nearest land",
+ "normalized_hypothesis": "wulong was 250 miles from the nearest land",
+ "errors": 1,
+ "words": 8,
+ "wer": 0.125
+ }
+ },
+ {
+ "id": "release500_0433",
+ "category": "phonetic_balance",
+ "text": "I just do appreciate it without being able to express my feelings.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0433.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I just do appreciate it without being able to express my feelings.",
+ "raw": {
+ "normalized_reference": "i just do appreciate it without being able to express my feelings.",
+ "normalized_hypothesis": "i just do appreciate it without being able to express my feelings.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i just do appreciate it without being able to express my feelings",
+ "normalized_hypothesis": "i just do appreciate it without being able to express my feelings",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0434",
+ "category": "phonetic_balance",
+ "text": "Now just what do you want to know.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0434.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Now just what do you want to know?",
+ "raw": {
+ "normalized_reference": "now just what do you want to know.",
+ "normalized_hypothesis": "now just what do you want to know?",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "now just what do you want to know",
+ "normalized_hypothesis": "now just what do you want to know",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0435",
+ "category": "phonetic_balance",
+ "text": "But she swung obediently on her heel into the wind.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0435.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "but she swung obediently on her heel into the wind.",
+ "raw": {
+ "normalized_reference": "but she swung obediently on her heel into the wind.",
+ "normalized_hypothesis": "but she swung obediently on her heel into the wind.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "but she swung obediently on her heel into the wind",
+ "normalized_hypothesis": "but she swung obediently on her heel into the wind",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0436",
+ "category": "phonetic_balance",
+ "text": "Ask him, I laughed, then turned to Pasquini.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0436.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Ask him. I laughed, then turned to Pasquini.",
+ "raw": {
+ "normalized_reference": "ask him, i laughed, then turned to pasquini.",
+ "normalized_hypothesis": "ask him. i laughed, then turned to pasquini.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "ask him i laughed then turned to pasquini",
+ "normalized_hypothesis": "ask him i laughed then turned to pasquini",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0437",
+ "category": "phonetic_balance",
+ "text": "He did not rush in.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0437.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He did not rush in.",
+ "raw": {
+ "normalized_reference": "he did not rush in.",
+ "normalized_hypothesis": "he did not rush in.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he did not rush in",
+ "normalized_hypothesis": "he did not rush in",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0438",
+ "category": "phonetic_balance",
+ "text": "They obeyed him, and went here and there at his commands.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0438.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "They obeyed him, and went here and there at his commands.",
+ "raw": {
+ "normalized_reference": "they obeyed him, and went here and there at his commands.",
+ "normalized_hypothesis": "they obeyed him, and went here and there at his commands.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "they obeyed him and went here and there at his commands",
+ "normalized_hypothesis": "they obeyed him and went here and there at his commands",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0439",
+ "category": "phonetic_balance",
+ "text": "Tom Spink has a harpoon.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0439.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Tom Spink has a harpoon.",
+ "raw": {
+ "normalized_reference": "tom spink has a harpoon.",
+ "normalized_hypothesis": "tom spink has a harpoon.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "tom spink has a harpoon",
+ "normalized_hypothesis": "tom spink has a harpoon",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0440",
+ "category": "phonetic_balance",
+ "text": "It was my reports from the north which chiefly induced people to buy.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0440.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It was my reports from the north which chiefly induced people to buy.",
+ "raw": {
+ "normalized_reference": "it was my reports from the north which chiefly induced people to buy.",
+ "normalized_hypothesis": "it was my reports from the north which chiefly induced people to buy.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it was my reports from the north which chiefly induced people to buy",
+ "normalized_hypothesis": "it was my reports from the north which chiefly induced people to buy",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0441",
+ "category": "phonetic_balance",
+ "text": "Cherokee identified himself with his instinct.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0441.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Shiroki identified himself with his instinct.",
+ "raw": {
+ "normalized_reference": "cherokee identified himself with his instinct.",
+ "normalized_hypothesis": "shiroki identified himself with his instinct.",
+ "errors": 1,
+ "words": 6,
+ "wer": 0.16666666666666666
+ },
+ "semantic": {
+ "normalized_reference": "cherokee identified himself with his instinct",
+ "normalized_hypothesis": "shiroki identified himself with his instinct",
+ "errors": 1,
+ "words": 6,
+ "wer": 0.16666666666666666
+ }
+ },
+ {
+ "id": "release500_0442",
+ "category": "phonetic_balance",
+ "text": "I graduated last of my class.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0442.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I graduated last of my class.",
+ "raw": {
+ "normalized_reference": "i graduated last of my class.",
+ "normalized_hypothesis": "i graduated last of my class.",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i graduated last of my class",
+ "normalized_hypothesis": "i graduated last of my class",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0443",
+ "category": "phonetic_balance",
+ "text": "It was a gigantic inadequacy.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0443.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It was a gigantic inadequacy.",
+ "raw": {
+ "normalized_reference": "it was a gigantic inadequacy.",
+ "normalized_hypothesis": "it was a gigantic inadequacy.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it was a gigantic inadequacy",
+ "normalized_hypothesis": "it was a gigantic inadequacy",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0444",
+ "category": "phonetic_balance",
+ "text": "Did I possess too much vitality.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0444.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Did I possess too much vitality?",
+ "raw": {
+ "normalized_reference": "did i possess too much vitality.",
+ "normalized_hypothesis": "did i possess too much vitality?",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "did i possess too much vitality",
+ "normalized_hypothesis": "did i possess too much vitality",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0445",
+ "category": "phonetic_balance",
+ "text": "Therefore, hurrah for the game.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0445.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Therefore, hurrah for the game!",
+ "raw": {
+ "normalized_reference": "therefore, hurrah for the game.",
+ "normalized_hypothesis": "therefore, hurrah for the game!",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "therefore hurrah for the game",
+ "normalized_hypothesis": "therefore hurrah for the game",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0446",
+ "category": "phonetic_balance",
+ "text": "He had fulfilled his duty and paid properly.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0446.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He had fulfilled his duty and paid properly.",
+ "raw": {
+ "normalized_reference": "he had fulfilled his duty and paid properly.",
+ "normalized_hypothesis": "he had fulfilled his duty and paid properly.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he had fulfilled his duty and paid properly",
+ "normalized_hypothesis": "he had fulfilled his duty and paid properly",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0447",
+ "category": "phonetic_balance",
+ "text": "How can you manage all alone, Mr Young.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0447.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "How can you manage all alone, Mr. Yum?",
+ "raw": {
+ "normalized_reference": "how can you manage all alone, mr young.",
+ "normalized_hypothesis": "how can you manage all alone, mr. yum?",
+ "errors": 1,
+ "words": 8,
+ "wer": 0.125
+ },
+ "semantic": {
+ "normalized_reference": "how can you manage all alone mister young",
+ "normalized_hypothesis": "how can you manage all alone mister yum",
+ "errors": 1,
+ "words": 8,
+ "wer": 0.125
+ }
+ },
+ {
+ "id": "release500_0448",
+ "category": "phonetic_balance",
+ "text": "Red-Eye never committed a more outrageous deed.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0448.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Red Eye never committed a more outrageous deed.",
+ "raw": {
+ "normalized_reference": "red-eye never committed a more outrageous deed.",
+ "normalized_hypothesis": "red eye never committed a more outrageous deed.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "red eye never committed a more outrageous deed",
+ "normalized_hypothesis": "red eye never committed a more outrageous deed",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0449",
+ "category": "phonetic_balance",
+ "text": "He would keep his faith with Gregson for the promised day or two.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0449.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He would keep his faith with Grixin for the promised day or two.",
+ "raw": {
+ "normalized_reference": "he would keep his faith with gregson for the promised day or two.",
+ "normalized_hypothesis": "he would keep his faith with grixin for the promised day or two.",
+ "errors": 1,
+ "words": 13,
+ "wer": 0.07692307692307693
+ },
+ "semantic": {
+ "normalized_reference": "he would keep his faith with gregson for the promised day or 2",
+ "normalized_hypothesis": "he would keep his faith with grixin for the promised day or 2",
+ "errors": 1,
+ "words": 13,
+ "wer": 0.07692307692307693
+ }
+ },
+ {
+ "id": "release500_0450",
+ "category": "phonetic_balance",
+ "text": "There was one chance, and only one, of saving Jeanne.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0450.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "There was one chance, and only one, of saving Jean.",
+ "raw": {
+ "normalized_reference": "there was one chance, and only one, of saving jeanne.",
+ "normalized_hypothesis": "there was one chance, and only one, of saving jean.",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ },
+ "semantic": {
+ "normalized_reference": "there was one chance and only one of saving jeanne",
+ "normalized_hypothesis": "there was one chance and only one of saving jean",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ }
+ },
+ {
+ "id": "release500_0451",
+ "category": "phonetic_balance",
+ "text": "I was beneath the water, suffocating and drowning.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0451.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I was beneath the water, suffocating and drowning.",
+ "raw": {
+ "normalized_reference": "i was beneath the water, suffocating and drowning.",
+ "normalized_hypothesis": "i was beneath the water, suffocating and drowning.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i was beneath the water suffocating and drowning",
+ "normalized_hypothesis": "i was beneath the water suffocating and drowning",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0452",
+ "category": "phonetic_balance",
+ "text": "Give them their choice between a fine or an official whipping.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0452.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Give them their choice between a fine or an official whipping.",
+ "raw": {
+ "normalized_reference": "give them their choice between a fine or an official whipping.",
+ "normalized_hypothesis": "give them their choice between a fine or an official whipping.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "give them their choice between a fine or an official whipping",
+ "normalized_hypothesis": "give them their choice between a fine or an official whipping",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0453",
+ "category": "phonetic_balance",
+ "text": "They must have been swept away by the chaotic currents.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0453.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "They must have been swept away by the chaotic currents.",
+ "raw": {
+ "normalized_reference": "they must have been swept away by the chaotic currents.",
+ "normalized_hypothesis": "they must have been swept away by the chaotic currents.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "they must have been swept away by the chaotic currents",
+ "normalized_hypothesis": "they must have been swept away by the chaotic currents",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0454",
+ "category": "phonetic_balance",
+ "text": "You yellow giant thing of the frost.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0454.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "You yellow giant thing of the frost.",
+ "raw": {
+ "normalized_reference": "you yellow giant thing of the frost.",
+ "normalized_hypothesis": "you yellow giant thing of the frost.",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "you yellow giant thing of the frost",
+ "normalized_hypothesis": "you yellow giant thing of the frost",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0455",
+ "category": "phonetic_balance",
+ "text": "He had peeled off his shirt and was wildly waving it.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0455.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He had peeled off his shirt and was wildly waving it.",
+ "raw": {
+ "normalized_reference": "he had peeled off his shirt and was wildly waving it.",
+ "normalized_hypothesis": "he had peeled off his shirt and was wildly waving it.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he had peeled off his shirt and was wildly waving it",
+ "normalized_hypothesis": "he had peeled off his shirt and was wildly waving it",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0456",
+ "category": "phonetic_balance",
+ "text": "To these he gave castor oil.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0456.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "To these he gave castor oil.",
+ "raw": {
+ "normalized_reference": "to these he gave castor oil.",
+ "normalized_hypothesis": "to these he gave castor oil.",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "to these he gave castor oil",
+ "normalized_hypothesis": "to these he gave castor oil",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0457",
+ "category": "phonetic_balance",
+ "text": "Tudor surveyed him with withering disgust.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0457.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Tudor surveyed him with withering disgust.",
+ "raw": {
+ "normalized_reference": "tudor surveyed him with withering disgust.",
+ "normalized_hypothesis": "tudor surveyed him with withering disgust.",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "tudor surveyed him with withering disgust",
+ "normalized_hypothesis": "tudor surveyed him with withering disgust",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0458",
+ "category": "phonetic_balance",
+ "text": "After all, it was simply a mistake in judgment.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0458.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "After all, it was simply a mistake in judgment.",
+ "raw": {
+ "normalized_reference": "after all, it was simply a mistake in judgment.",
+ "normalized_hypothesis": "after all, it was simply a mistake in judgment.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "after all it was simply a mistake in judgment",
+ "normalized_hypothesis": "after all it was simply a mistake in judgment",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0459",
+ "category": "phonetic_balance",
+ "text": "That is the strange part of it.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0459.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "That is the strange part of it.",
+ "raw": {
+ "normalized_reference": "that is the strange part of it.",
+ "normalized_hypothesis": "that is the strange part of it.",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "that is the strange part of it",
+ "normalized_hypothesis": "that is the strange part of it",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0460",
+ "category": "phonetic_balance",
+ "text": "Not a wheel moved in his empire.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0460.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Not a wheel moved in his empire.",
+ "raw": {
+ "normalized_reference": "not a wheel moved in his empire.",
+ "normalized_hypothesis": "not a wheel moved in his empire.",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "not a wheel moved in his empire",
+ "normalized_hypothesis": "not a wheel moved in his empire",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0461",
+ "category": "phonetic_balance",
+ "text": "Its diameter was not more than two hundred yards.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0461.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Its diameter was not more than 200 yards.",
+ "raw": {
+ "normalized_reference": "its diameter was not more than two hundred yards.",
+ "normalized_hypothesis": "its diameter was not more than 200 yards.",
+ "errors": 2,
+ "words": 9,
+ "wer": 0.2222222222222222
+ },
+ "semantic": {
+ "normalized_reference": "its diameter was not more than 200 yards",
+ "normalized_hypothesis": "its diameter was not more than 200 yards",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0462",
+ "category": "phonetic_balance",
+ "text": "Nimrod replied, with a slight manifestation of sensitiveness.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0462.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Nimrod replied, with a slight manifestation of sensitiveness.",
+ "raw": {
+ "normalized_reference": "nimrod replied, with a slight manifestation of sensitiveness.",
+ "normalized_hypothesis": "nimrod replied, with a slight manifestation of sensitiveness.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "nimrod replied with a slight manifestation of sensitiveness",
+ "normalized_hypothesis": "nimrod replied with a slight manifestation of sensitiveness",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0463",
+ "category": "phonetic_balance",
+ "text": "I tried to read George Moore last night, and was dreadfully bored.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0463.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I tried to read George Moore last night and was dreadfully bored.",
+ "raw": {
+ "normalized_reference": "i tried to read george moore last night, and was dreadfully bored.",
+ "normalized_hypothesis": "i tried to read george moore last night and was dreadfully bored.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i tried to read george moore last night and was dreadfully bored",
+ "normalized_hypothesis": "i tried to read george moore last night and was dreadfully bored",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0464",
+ "category": "phonetic_balance",
+ "text": "Yes, it was a man who asked, a stranger.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0464.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Yes, it was a man who asked a stranger.",
+ "raw": {
+ "normalized_reference": "yes, it was a man who asked, a stranger.",
+ "normalized_hypothesis": "yes, it was a man who asked a stranger.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "yes it was a man who asked a stranger",
+ "normalized_hypothesis": "yes it was a man who asked a stranger",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0465",
+ "category": "phonetic_balance",
+ "text": "We were now good friends.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0465.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "We were now good friends.",
+ "raw": {
+ "normalized_reference": "we were now good friends.",
+ "normalized_hypothesis": "we were now good friends.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "we were now good friends",
+ "normalized_hypothesis": "we were now good friends",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0466",
+ "category": "phonetic_balance",
+ "text": "Yet, in accordance with Ernest's test of truth, it worked.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0466.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Yet, in accordance with Ernest's test of truth, it worked.",
+ "raw": {
+ "normalized_reference": "yet, in accordance with ernest's test of truth, it worked.",
+ "normalized_hypothesis": "yet, in accordance with ernest's test of truth, it worked.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "yet in accordance with ernest is test of truth it worked",
+ "normalized_hypothesis": "yet in accordance with ernest is test of truth it worked",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0467",
+ "category": "phonetic_balance",
+ "text": "It's the strap hangers that'll keep us from going under.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0467.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It's the strap hangers that'll keep us from going under.",
+ "raw": {
+ "normalized_reference": "it's the strap hangers that'll keep us from going under.",
+ "normalized_hypothesis": "it's the strap hangers that'll keep us from going under.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it is the strap hangers that will keep us from going under",
+ "normalized_hypothesis": "it is the strap hangers that will keep us from going under",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0468",
+ "category": "phonetic_balance",
+ "text": "I never allow what can't be changed to annoy me.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0468.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I never allow what can't be changed to annoy me.",
+ "raw": {
+ "normalized_reference": "i never allow what can't be changed to annoy me.",
+ "normalized_hypothesis": "i never allow what can't be changed to annoy me.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i never allow what can not be changed to annoy me",
+ "normalized_hypothesis": "i never allow what can not be changed to annoy me",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0469",
+ "category": "phonetic_balance",
+ "text": "After all superfluous flesh is gone what is left is stringy and resistant.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0469.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "After all superfluous flesh is gone what is left is stringy and resistant.",
+ "raw": {
+ "normalized_reference": "after all superfluous flesh is gone what is left is stringy and resistant.",
+ "normalized_hypothesis": "after all superfluous flesh is gone what is left is stringy and resistant.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "after all superfluous flesh is gone what is left is stringy and resistant",
+ "normalized_hypothesis": "after all superfluous flesh is gone what is left is stringy and resistant",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0470",
+ "category": "phonetic_balance",
+ "text": "Then he shouted, Shut up.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0470.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Then he shouted, shut up.",
+ "raw": {
+ "normalized_reference": "then he shouted, shut up.",
+ "normalized_hypothesis": "then he shouted, shut up.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "then he shouted shut up",
+ "normalized_hypothesis": "then he shouted shut up",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0471",
+ "category": "phonetic_balance",
+ "text": "He had worshiped her, as Dante might have worshiped Beatrice.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0471.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He had worshipped her, as Dante might have worshipped Beatrice.",
+ "raw": {
+ "normalized_reference": "he had worshiped her, as dante might have worshiped beatrice.",
+ "normalized_hypothesis": "he had worshipped her, as dante might have worshipped beatrice.",
+ "errors": 2,
+ "words": 10,
+ "wer": 0.2
+ },
+ "semantic": {
+ "normalized_reference": "he had worshiped her as dante might have worshiped beatrice",
+ "normalized_hypothesis": "he had worshipped her as dante might have worshipped beatrice",
+ "errors": 2,
+ "words": 10,
+ "wer": 0.2
+ }
+ },
+ {
+ "id": "release500_0472",
+ "category": "phonetic_balance",
+ "text": "He looked like one who had passed through an uncomfortable hour or two.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0472.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He looked like one who had passed through an uncomfortable hour or two.",
+ "raw": {
+ "normalized_reference": "he looked like one who had passed through an uncomfortable hour or two.",
+ "normalized_hypothesis": "he looked like one who had passed through an uncomfortable hour or two.",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he looked like one who had passed through an uncomfortable hour or 2",
+ "normalized_hypothesis": "he looked like one who had passed through an uncomfortable hour or 2",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0473",
+ "category": "phonetic_balance",
+ "text": "The history of our westward-faring race is written in it.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0473.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The history of our westward-faring race is written in it.",
+ "raw": {
+ "normalized_reference": "the history of our westward-faring race is written in it.",
+ "normalized_hypothesis": "the history of our westward-faring race is written in it.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the history of our westward faring race is written in it",
+ "normalized_hypothesis": "the history of our westward faring race is written in it",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0474",
+ "category": "phonetic_balance",
+ "text": "He may desire to escape pain, or to enjoy pleasure.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0474.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He may desire to escape pain, or to enjoy pleasure.",
+ "raw": {
+ "normalized_reference": "he may desire to escape pain, or to enjoy pleasure.",
+ "normalized_hypothesis": "he may desire to escape pain, or to enjoy pleasure.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he may desire to escape pain or to enjoy pleasure",
+ "normalized_hypothesis": "he may desire to escape pain or to enjoy pleasure",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0475",
+ "category": "phonetic_balance",
+ "text": "And the air was growing chilly.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0475.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "and the air was growing chilly.",
+ "raw": {
+ "normalized_reference": "and the air was growing chilly.",
+ "normalized_hypothesis": "and the air was growing chilly.",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "and the air was growing chilly",
+ "normalized_hypothesis": "and the air was growing chilly",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0476",
+ "category": "phonetic_balance",
+ "text": "Dig in; you're sure good, was Daylight's answer.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0476.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Dig in, you're sure good, was Deadlight's answer.",
+ "raw": {
+ "normalized_reference": "dig in; you're sure good, was daylight's answer.",
+ "normalized_hypothesis": "dig in, you're sure good, was deadlight's answer.",
+ "errors": 1,
+ "words": 8,
+ "wer": 0.125
+ },
+ "semantic": {
+ "normalized_reference": "dig in you are sure good was daylight is answer",
+ "normalized_hypothesis": "dig in you are sure good was deadlight is answer",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ }
+ },
+ {
+ "id": "release500_0477",
+ "category": "phonetic_balance",
+ "text": "My idea was that he would have more influence over the natives.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0477.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "My idea was that he would have more influence over the natives.",
+ "raw": {
+ "normalized_reference": "my idea was that he would have more influence over the natives.",
+ "normalized_hypothesis": "my idea was that he would have more influence over the natives.",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "my idea was that he would have more influence over the natives",
+ "normalized_hypothesis": "my idea was that he would have more influence over the natives",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0478",
+ "category": "phonetic_balance",
+ "text": "Now run along, and tell them to hurry.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0478.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Now run along and tell them to hurry.",
+ "raw": {
+ "normalized_reference": "now run along, and tell them to hurry.",
+ "normalized_hypothesis": "now run along and tell them to hurry.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "now run along and tell them to hurry",
+ "normalized_hypothesis": "now run along and tell them to hurry",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0479",
+ "category": "phonetic_balance",
+ "text": "I have no idea, replied Philip.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0479.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I have no idea, replied Philip.",
+ "raw": {
+ "normalized_reference": "i have no idea, replied philip.",
+ "normalized_hypothesis": "i have no idea, replied philip.",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i have no idea replied philip",
+ "normalized_hypothesis": "i have no idea replied philip",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0480",
+ "category": "phonetic_balance",
+ "text": "For the rest, he was a mere automaton.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0480.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "For the rest, he was a mere automaton.",
+ "raw": {
+ "normalized_reference": "for the rest, he was a mere automaton.",
+ "normalized_hypothesis": "for the rest, he was a mere automaton.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "for the rest he was a mere automaton",
+ "normalized_hypothesis": "for the rest he was a mere automaton",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0481",
+ "category": "phonetic_balance",
+ "text": "Why, the average review is more nauseating than cod liver oil.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0481.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Why, the average review is more nauseating than cod liver oil.",
+ "raw": {
+ "normalized_reference": "why, the average review is more nauseating than cod liver oil.",
+ "normalized_hypothesis": "why, the average review is more nauseating than cod liver oil.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "why the average review is more nauseating than cod liver oil",
+ "normalized_hypothesis": "why the average review is more nauseating than cod liver oil",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0482",
+ "category": "phonetic_balance",
+ "text": "I have been robbed, sir, I amended.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0482.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I have been robbed, sir. I amend it.",
+ "raw": {
+ "normalized_reference": "i have been robbed, sir, i amended.",
+ "normalized_hypothesis": "i have been robbed, sir. i amend it.",
+ "errors": 2,
+ "words": 7,
+ "wer": 0.2857142857142857
+ },
+ "semantic": {
+ "normalized_reference": "i have been robbed sir i amended",
+ "normalized_hypothesis": "i have been robbed sir i amend it",
+ "errors": 2,
+ "words": 7,
+ "wer": 0.2857142857142857
+ }
+ },
+ {
+ "id": "release500_0483",
+ "category": "phonetic_balance",
+ "text": "He moved away as quietly as he had come.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0483.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He moved away as quietly as he had come.",
+ "raw": {
+ "normalized_reference": "he moved away as quietly as he had come.",
+ "normalized_hypothesis": "he moved away as quietly as he had come.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he moved away as quietly as he had come",
+ "normalized_hypothesis": "he moved away as quietly as he had come",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0484",
+ "category": "phonetic_balance",
+ "text": "A trickle of fresh blood ran over his face.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0484.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "A trickle of fresh blood ran over his face.",
+ "raw": {
+ "normalized_reference": "a trickle of fresh blood ran over his face.",
+ "normalized_hypothesis": "a trickle of fresh blood ran over his face.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "a trickle of fresh blood ran over his face",
+ "normalized_hypothesis": "a trickle of fresh blood ran over his face",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0485",
+ "category": "phonetic_balance",
+ "text": "Suppose you saw me at work through the window.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0485.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Suppose you saw me at work through the window.",
+ "raw": {
+ "normalized_reference": "suppose you saw me at work through the window.",
+ "normalized_hypothesis": "suppose you saw me at work through the window.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "suppose you saw me at work through the window",
+ "normalized_hypothesis": "suppose you saw me at work through the window",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0486",
+ "category": "phonetic_balance",
+ "text": "But who was Eileen's double.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0486.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "But who was Eileen Stubble?",
+ "raw": {
+ "normalized_reference": "but who was eileen's double.",
+ "normalized_hypothesis": "but who was eileen stubble?",
+ "errors": 2,
+ "words": 5,
+ "wer": 0.4
+ },
+ "semantic": {
+ "normalized_reference": "but who was eileen is double",
+ "normalized_hypothesis": "but who was eileen stubble",
+ "errors": 2,
+ "words": 6,
+ "wer": 0.3333333333333333
+ }
+ },
+ {
+ "id": "release500_0487",
+ "category": "phonetic_balance",
+ "text": "Darkness hid him from Jeanne.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0487.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Darkness heard him from Jean.",
+ "raw": {
+ "normalized_reference": "darkness hid him from jeanne.",
+ "normalized_hypothesis": "darkness heard him from jean.",
+ "errors": 2,
+ "words": 5,
+ "wer": 0.4
+ },
+ "semantic": {
+ "normalized_reference": "darkness hid him from jeanne",
+ "normalized_hypothesis": "darkness heard him from jean",
+ "errors": 2,
+ "words": 5,
+ "wer": 0.4
+ }
+ },
+ {
+ "id": "release500_0488",
+ "category": "phonetic_balance",
+ "text": "By this time Charley was as enraged as the Greek.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0488.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "By this time, Charlie was as enraged as the Greek.",
+ "raw": {
+ "normalized_reference": "by this time charley was as enraged as the greek.",
+ "normalized_hypothesis": "by this time, charlie was as enraged as the greek.",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ },
+ "semantic": {
+ "normalized_reference": "by this time charley was as enraged as the greek",
+ "normalized_hypothesis": "by this time charlie was as enraged as the greek",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ }
+ },
+ {
+ "id": "release500_0489",
+ "category": "phonetic_balance",
+ "text": "I am writing these lines in Honolulu, Hawaii.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0489.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I am writing these lines in Honolulu, Hawaii.",
+ "raw": {
+ "normalized_reference": "i am writing these lines in honolulu, hawaii.",
+ "normalized_hypothesis": "i am writing these lines in honolulu, hawaii.",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i am writing these lines in honolulu hawaii",
+ "normalized_hypothesis": "i am writing these lines in honolulu hawaii",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0490",
+ "category": "phonetic_balance",
+ "text": "Nobody knows how the natives got them.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0490.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Nobody knows how the natives got them.",
+ "raw": {
+ "normalized_reference": "nobody knows how the natives got them.",
+ "normalized_hypothesis": "nobody knows how the natives got them.",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "nobody knows how the natives got them",
+ "normalized_hypothesis": "nobody knows how the natives got them",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0491",
+ "category": "phonetic_balance",
+ "text": "And how would we ever find ourselves.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0491.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "And how would we ever find ourselves?",
+ "raw": {
+ "normalized_reference": "and how would we ever find ourselves.",
+ "normalized_hypothesis": "and how would we ever find ourselves?",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "and how would we ever find ourselves",
+ "normalized_hypothesis": "and how would we ever find ourselves",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0492",
+ "category": "phonetic_balance",
+ "text": "They are coming ashore, whoever they are.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0492.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "They are coming ashore, whoever they are.",
+ "raw": {
+ "normalized_reference": "they are coming ashore, whoever they are.",
+ "normalized_hypothesis": "they are coming ashore, whoever they are.",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "they are coming ashore whoever they are",
+ "normalized_hypothesis": "they are coming ashore whoever they are",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0493",
+ "category": "phonetic_balance",
+ "text": "This is my fifth voyage.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0493.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "This is my fifth voyage.",
+ "raw": {
+ "normalized_reference": "this is my fifth voyage.",
+ "normalized_hypothesis": "this is my fifth voyage.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "this is my 5th voyage",
+ "normalized_hypothesis": "this is my 5th voyage",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0494",
+ "category": "phonetic_balance",
+ "text": "And yet if she came he had no words to say.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0494.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "And yet if she came he had no words to say.",
+ "raw": {
+ "normalized_reference": "and yet if she came he had no words to say.",
+ "normalized_hypothesis": "and yet if she came he had no words to say.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "and yet if she came he had no words to say",
+ "normalized_hypothesis": "and yet if she came he had no words to say",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0495",
+ "category": "phonetic_balance",
+ "text": "You must sleep, he urged.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0495.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "You must sleep, he urged.",
+ "raw": {
+ "normalized_reference": "you must sleep, he urged.",
+ "normalized_hypothesis": "you must sleep, he urged.",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "you must sleep he urged",
+ "normalized_hypothesis": "you must sleep he urged",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0496",
+ "category": "phonetic_balance",
+ "text": "Could the incident have anything to do with Jeanne and Pierre.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0496.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Could the incident have anything to do with Jean and Pierre?",
+ "raw": {
+ "normalized_reference": "could the incident have anything to do with jeanne and pierre.",
+ "normalized_hypothesis": "could the incident have anything to do with jean and pierre?",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ },
+ "semantic": {
+ "normalized_reference": "could the incident have anything to do with jeanne and pierre",
+ "normalized_hypothesis": "could the incident have anything to do with jean and pierre",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ }
+ },
+ {
+ "id": "release500_0497",
+ "category": "phonetic_balance",
+ "text": "Each day she became a more vital part of him.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0497.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Each day she became a more vital part of him.",
+ "raw": {
+ "normalized_reference": "each day she became a more vital part of him.",
+ "normalized_hypothesis": "each day she became a more vital part of him.",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "each day she became a more vital part of him",
+ "normalized_hypothesis": "each day she became a more vital part of him",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0498",
+ "category": "phonetic_balance",
+ "text": "But why continue the tirade, for tirade it was.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0498.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "But why continue the tirade? For tirade it was.",
+ "raw": {
+ "normalized_reference": "but why continue the tirade, for tirade it was.",
+ "normalized_hypothesis": "but why continue the tirade? for tirade it was.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "but why continue the tirade for tirade it was",
+ "normalized_hypothesis": "but why continue the tirade for tirade it was",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0499",
+ "category": "phonetic_balance",
+ "text": "That is what distinguishes all of us from the lower animals.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0499.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "That is what distinguishes all of us from the lower animals.",
+ "raw": {
+ "normalized_reference": "that is what distinguishes all of us from the lower animals.",
+ "normalized_hypothesis": "that is what distinguishes all of us from the lower animals.",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "that is what distinguishes all of us from the lower animals",
+ "normalized_hypothesis": "that is what distinguishes all of us from the lower animals",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0500",
+ "category": "phonetic_balance",
+ "text": "There was something pathetic in the girl's attitude now.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0500.wav",
+ "asr": "whisper-large-v3",
+ "hypothesis": "There was something pathetic in the girl's attitude now.",
+ "raw": {
+ "normalized_reference": "there was something pathetic in the girl's attitude now.",
+ "normalized_hypothesis": "there was something pathetic in the girl's attitude now.",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "there was something pathetic in the girl is attitude now",
+ "normalized_hypothesis": "there was something pathetic in the girl is attitude now",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0001",
+ "category": "conversational",
+ "text": "It wasn't until later that I realized what had actually happened.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0001.wav",
+ "render_seconds": 76.20106935000513,
+ "audio_seconds": 4.31891156462585,
+ "rtf": 17.64358177049325,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "I'm listening to you. I don't know why you're listening. I don't know why you're listening. I don't know why you're listening. I don't know why you're listening. I don't know why you're listening. I don't know why you're listening. I don't know why you're listening. I don't know why you're listening. I don't know why you're listening. I don't know why you're listening. I don't know why you're listening. I don't know why you're listening. I don't know why you're listening. I don't know why you're listening. I don't know why you're listening. I don't know why you're listening. I don't know why you're listening. I don't know why you're listening. I don't know why you're listening. I don't know why you're listening. I don't know why you're listening. I don't know why you're listening. I don't know why you're listening. I don't know why you're listening. I don't know why you're listening. I don't know why you're listening. I don't know why you're listening. I don't know why you're listening. I don't know why you're listening. I don't know why you're listening. I don't know why you're listening. I don't know why you're listening. I don't know why you're listening. I don't know why you're listening. I don't know why you're listening. I don't know why you're listening. I don't know why you're listening. I don't know why you're listening. I don't know why you're listening. I don't know why you're listening. I don't know why you're listening. I don't know why you're listening. I don't know why you're listening. I don't know why you're listening. I don't know why you're listening. I don't know why you're listening. I don't know why you're listening. I don't know why you're listening. I don't know why you",
+ "raw": {
+ "normalized_reference": "it wasn't until later that i realized what had actually happened.",
+ "normalized_hypothesis": "i'm listening to you. i don't know why you're listening. i don't know why you're listening. i don't know why you're listening. i don't know why you're listening. i don't know why you're listening. i don't know why you're listening. i don't know why you're listening. i don't know why you're listening. i don't know why you're listening. i don't know why you're listening. i don't know why you're listening. i don't know why you're listening. i don't know why you're listening. i don't know why you're listening. i don't know why you're listening. i don't know why you're listening. i don't know why you're listening. i don't know why you're listening. i don't know why you're listening. i don't know why you're listening. i don't know why you're listening. i don't know why you're listening. i don't know why you're listening. i don't know why you're listening. i don't know why you're listening. i don't know why you're listening. i don't know why you're listening. i don't know why you're listening. i don't know why you're listening. i don't know why you're listening. i don't know why you're listening. i don't know why you're listening. i don't know why you're listening. i don't know why you're listening. i don't know why you're listening. i don't know why you're listening. i don't know why you're listening. i don't know why you're listening. i don't know why you're listening. i don't know why you're listening. i don't know why you're listening. i don't know why you're listening. i don't know why you're listening. i don't know why you're listening. i don't know why you're listening. i don't know why you're listening. i don't know why you're listening. i don't know why you're listening. i don't know why you",
+ "errors": 296,
+ "words": 11,
+ "wer": 26.90909090909091
+ },
+ "semantic": {
+ "normalized_reference": "it was not until later that i realized what had actually happened",
+ "normalized_hypothesis": "i am listening to you i do not know why you are listening i do not know why you are listening i do not know why you are listening i do not know why you are listening i do not know why you are listening i do not know why you are listening i do not know why you are listening i do not know why you are listening i do not know why you are listening i do not know why you are listening i do not know why you are listening i do not know why you are listening i do not know why you are listening i do not know why you are listening i do not know why you are listening i do not know why you are listening i do not know why you are listening i do not know why you are listening i do not know why you are listening i do not know why you are listening i do not know why you are listening i do not know why you are listening i do not know why you are listening i do not know why you are listening i do not know why you are listening i do not know why you are listening i do not know why you are listening i do not know why you are listening i do not know why you are listening i do not know why you are listening i do not know why you are listening i do not know why you are listening i do not know why you are listening i do not know why you are listening i do not know why you are listening i do not know why you are listening i do not know why you are listening i do not know why you are listening i do not know why you are listening i do not know why you are listening i do not know why you are listening i do not know why you are listening i do not know why you are listening i do not know why you are listening i do not know why you are listening i do not know why you are listening i do not know why you are listening i do not know why you are listening i do not know why you",
+ "errors": 393,
+ "words": 12,
+ "wer": 32.75
+ }
+ },
+ {
+ "id": "release500_0002",
+ "category": "descriptive",
+ "text": "The light at that time of afternoon is perfect for about twenty minutes.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0002.wav",
+ "render_seconds": 89.0924288610695,
+ "audio_seconds": 4.806530612244898,
+ "rtf": 18.535704034444656,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "We are at the Traveller's tomorrow. It's perfect for burnt-in-the-wires.",
+ "raw": {
+ "normalized_reference": "the light at that time of afternoon is perfect for about twenty minutes.",
+ "normalized_hypothesis": "we are at the traveller's tomorrow. it's perfect for burnt-in-the-wires.",
+ "errors": 11,
+ "words": 13,
+ "wer": 0.8461538461538461
+ },
+ "semantic": {
+ "normalized_reference": "the light at that time of afternoon is perfect for about 20 minutes",
+ "normalized_hypothesis": "we are at the traveller is tomorrow it is perfect for burnt in the wires",
+ "errors": 11,
+ "words": 13,
+ "wer": 0.8461538461538461
+ }
+ },
+ {
+ "id": "release500_0003",
+ "category": "emotional",
+ "text": "That was the scariest thing I've ever done, and also maybe the best.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0003.wav",
+ "render_seconds": 76.90116830705665,
+ "audio_seconds": 4.527891156462585,
+ "rtf": 16.983881822622184,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "There was this girl I saw in the water. I knew she was the best.",
+ "raw": {
+ "normalized_reference": "that was the scariest thing i've ever done, and also maybe the best.",
+ "normalized_hypothesis": "there was this girl i saw in the water. i knew she was the best.",
+ "errors": 12,
+ "words": 13,
+ "wer": 0.9230769230769231
+ },
+ "semantic": {
+ "normalized_reference": "that was the scariest thing i have ever done and also maybe the best",
+ "normalized_hypothesis": "there was this girl i saw in the water i knew she was the best",
+ "errors": 12,
+ "words": 14,
+ "wer": 0.8571428571428571
+ }
+ },
+ {
+ "id": "release500_0004",
+ "category": "homographs",
+ "text": "After I record the record, please present the present to the bass player.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0004.wav",
+ "render_seconds": 78.39508487901185,
+ "audio_seconds": 5.015510204081632,
+ "rtf": 15.63053043242017,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "I've never heard of that book. That's why I sent the question to the best guy.",
+ "raw": {
+ "normalized_reference": "after i record the record, please present the present to the bass player.",
+ "normalized_hypothesis": "i've never heard of that book. that's why i sent the question to the best guy.",
+ "errors": 13,
+ "words": 13,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "after i record the record please present the present to the bass player",
+ "normalized_hypothesis": "i have never heard of that book that is why i sent the question to the best guy",
+ "errors": 15,
+ "words": 13,
+ "wer": 1.1538461538461537
+ }
+ },
+ {
+ "id": "release500_0005",
+ "category": "long",
+ "text": "If I'm being completely honest, I had no idea what I was doing for most of it, and I think that might be the most normal thing about the whole experience.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0005.wav",
+ "render_seconds": 108.40687325003091,
+ "audio_seconds": 9.47374149659864,
+ "rtf": 11.442878538426688,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "I'd like to thank you for your interest. I'd like to have your interest around for the rest of it. I'd like to thank you for your interest around for the rest of it. I'd like to thank you for your interest around for the rest of it. I'd like to thank you for your interest around for the rest of it. I'd like to thank you for your interest around for the rest of it. I'd like to thank you for your interest around for the rest of it. I'd like to thank you for your interest around for the rest of it. I'd like to thank you for your interest around for the rest of it. I'd like to thank you for your interest around for the rest of it. I'd like to thank you for your interest around for the rest of it. I'd like to thank you for your interest around for the rest of it. I'd like to thank you for your interest around for the rest of it. I'd like to thank you for your interest around for the rest of it. I'd like to thank you for your interest around for the rest of it. I'd like to thank you for your interest around for the rest of it. I'd like to thank you for your interest around for the rest of it. I'd like to thank you for your interest around for the rest of it. I'd like to thank you for your interest around for the rest of it. I'd like to thank you for your interest around for the rest of it. I'd like to thank you for your interest around for the rest of it. I'd like to thank you for your interest around for the rest of it. I'd like to thank you for your interest around for the rest of it. I'd like to thank you for your interest around for the rest of it. I'd like to thank you for your interest around for the rest of it. I'd like to thank you for your interest around for the rest of it. I'd like to thank you for your interest around for the rest of it. I'd like to thank you for your interest around for the rest of it. I'd like to",
+ "raw": {
+ "normalized_reference": "if i'm being completely honest, i had no idea what i was doing for most of it, and i think that might be the most normal thing about the whole experience.",
+ "normalized_hypothesis": "i'd like to thank you for your interest. i'd like to have your interest around for the rest of it. i'd like to thank you for your interest around for the rest of it. i'd like to thank you for your interest around for the rest of it. i'd like to thank you for your interest around for the rest of it. i'd like to thank you for your interest around for the rest of it. i'd like to thank you for your interest around for the rest of it. i'd like to thank you for your interest around for the rest of it. i'd like to thank you for your interest around for the rest of it. i'd like to thank you for your interest around for the rest of it. i'd like to thank you for your interest around for the rest of it. i'd like to thank you for your interest around for the rest of it. i'd like to thank you for your interest around for the rest of it. i'd like to thank you for your interest around for the rest of it. i'd like to thank you for your interest around for the rest of it. i'd like to thank you for your interest around for the rest of it. i'd like to thank you for your interest around for the rest of it. i'd like to thank you for your interest around for the rest of it. i'd like to thank you for your interest around for the rest of it. i'd like to thank you for your interest around for the rest of it. i'd like to thank you for your interest around for the rest of it. i'd like to thank you for your interest around for the rest of it. i'd like to thank you for your interest around for the rest of it. i'd like to thank you for your interest around for the rest of it. i'd like to thank you for your interest around for the rest of it. i'd like to thank you for your interest around for the rest of it. i'd like to thank you for your interest around for the rest of it. i'd like to thank you for your interest around for the rest of it. i'd like to",
+ "errors": 382,
+ "words": 31,
+ "wer": 12.32258064516129
+ },
+ "semantic": {
+ "normalized_reference": "if i am being completely honest i had no idea what i was doing for most of it and i think that might be the most normal thing about the whole experience",
+ "normalized_hypothesis": "i would like to thank you for your interest i would like to have your interest around for the rest of it i would like to thank you for your interest around for the rest of it i would like to thank you for your interest around for the rest of it i would like to thank you for your interest around for the rest of it i would like to thank you for your interest around for the rest of it i would like to thank you for your interest around for the rest of it i would like to thank you for your interest around for the rest of it i would like to thank you for your interest around for the rest of it i would like to thank you for your interest around for the rest of it i would like to thank you for your interest around for the rest of it i would like to thank you for your interest around for the rest of it i would like to thank you for your interest around for the rest of it i would like to thank you for your interest around for the rest of it i would like to thank you for your interest around for the rest of it i would like to thank you for your interest around for the rest of it i would like to thank you for your interest around for the rest of it i would like to thank you for your interest around for the rest of it i would like to thank you for your interest around for the rest of it i would like to thank you for your interest around for the rest of it i would like to thank you for your interest around for the rest of it i would like to thank you for your interest around for the rest of it i would like to thank you for your interest around for the rest of it i would like to thank you for your interest around for the rest of it i would like to thank you for your interest around for the rest of it i would like to thank you for your interest around for the rest of it i would like to thank you for your interest around for the rest of it i would like to thank you for your interest around for the rest of it i would like to",
+ "errors": 407,
+ "words": 32,
+ "wer": 12.71875
+ }
+ },
+ {
+ "id": "release500_0006",
+ "category": "long_range",
+ "text": "The committee postponed its decision after discovering that three pages of the report had been printed in the wrong order.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0006.wav",
+ "render_seconds": 122.28961312398314,
+ "audio_seconds": 7.9412244897959186,
+ "rtf": 15.399339646060788,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "The club at this moment's discussion is going to serve one of the top matches in Europe, Europe, and quite a bit in the world, Europe.",
+ "raw": {
+ "normalized_reference": "the committee postponed its decision after discovering that three pages of the report had been printed in the wrong order.",
+ "normalized_hypothesis": "the club at this moment's discussion is going to serve one of the top matches in europe, europe, and quite a bit in the world, europe.",
+ "errors": 21,
+ "words": 20,
+ "wer": 1.05
+ },
+ "semantic": {
+ "normalized_reference": "the committee postponed its decision after discovering that 3 pages of the report had been printed in the wrong order",
+ "normalized_hypothesis": "the club at this moment is discussion is going to serve one of the top matches in europe europe and quite a bit in the world europe",
+ "errors": 22,
+ "words": 20,
+ "wer": 1.1
+ }
+ },
+ {
+ "id": "release500_0007",
+ "category": "names_places",
+ "text": "Gwendolyn photographed the eucalyptus trees outside Ljubljana.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0007.wav",
+ "render_seconds": 75.72085215803236,
+ "audio_seconds": 4.527891156462585,
+ "rtf": 16.723205028892362,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "Let me answer your questions.",
+ "raw": {
+ "normalized_reference": "gwendolyn photographed the eucalyptus trees outside ljubljana.",
+ "normalized_hypothesis": "let me answer your questions.",
+ "errors": 7,
+ "words": 7,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "gwendolyn photographed the eucalyptus trees outside ljubljana",
+ "normalized_hypothesis": "let me answer your questions",
+ "errors": 7,
+ "words": 7,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0008",
+ "category": "narrative",
+ "text": "He had this theory, and the more time passes, the more I think he was right.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0008.wav",
+ "render_seconds": 92.60325879510492,
+ "audio_seconds": 5.294149659863946,
+ "rtf": 17.491620891859096,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "I like this town. What do you talk about, sis? Do you think you're a slut?",
+ "raw": {
+ "normalized_reference": "he had this theory, and the more time passes, the more i think he was right.",
+ "normalized_hypothesis": "i like this town. what do you talk about, sis? do you think you're a slut?",
+ "errors": 14,
+ "words": 16,
+ "wer": 0.875
+ },
+ "semantic": {
+ "normalized_reference": "he had this theory and the more time passes the more i think he was right",
+ "normalized_hypothesis": "i like this town what do you talk about sis do you think you are a slut",
+ "errors": 15,
+ "words": 16,
+ "wer": 0.9375
+ }
+ },
+ {
+ "id": "release500_0009",
+ "category": "numbers",
+ "text": "The package weighs twelve point six kilograms and arrived on July twenty-first.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0009.wav",
+ "render_seconds": 94.99756314198021,
+ "audio_seconds": 5.7817687074829935,
+ "rtf": 16.430536735070465,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "The perfect glass glazed like such colors. Like a world in a clear 24th.",
+ "raw": {
+ "normalized_reference": "the package weighs twelve point six kilograms and arrived on july twenty-first.",
+ "normalized_hypothesis": "the perfect glass glazed like such colors. like a world in a clear 24th.",
+ "errors": 13,
+ "words": 13,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "the package weighs 12.6 kilograms and arrived on july 21st",
+ "normalized_hypothesis": "the perfect glass glazed like such colors like a world in a clear 24th",
+ "errors": 13,
+ "words": 11,
+ "wer": 1.1818181818181819
+ }
+ },
+ {
+ "id": "release500_0010",
+ "category": "punctuation",
+ "text": "First, close the window; second, turn off the lamp; finally, lock the door.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0010.wav",
+ "render_seconds": 90.09958808403462,
+ "audio_seconds": 5.224489795918367,
+ "rtf": 17.245624281709752,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "First, here's the regular. Second, twice a year. Third, a year and a half.",
+ "raw": {
+ "normalized_reference": "first, close the window; second, turn off the lamp; finally, lock the door.",
+ "normalized_hypothesis": "first, here's the regular. second, twice a year. third, a year and a half.",
+ "errors": 11,
+ "words": 13,
+ "wer": 0.8461538461538461
+ },
+ "semantic": {
+ "normalized_reference": "1st close the window 2nd turn off the lamp finally lock the door",
+ "normalized_hypothesis": "1st here is the regular 2nd twice a year 3rd a year and a half",
+ "errors": 12,
+ "words": 13,
+ "wer": 0.9230769230769231
+ }
+ },
+ {
+ "id": "release500_0011",
+ "category": "question",
+ "text": "When did you first realize something was wrong?",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0011.wav",
+ "render_seconds": 31.208350338041782,
+ "audio_seconds": 3.1346938775510202,
+ "rtf": 9.955788844818017,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "I'm not going to fall asleep. I'm not going to fall asleep. I'm not going to fall asleep. I'm not going to fall asleep. I'm not going to fall asleep. I'm not going to fall asleep. I'm not going to fall asleep. I'm not going to fall asleep. I'm not going to fall asleep. I'm not going to fall asleep. I'm not going to fall asleep. I'm not going to fall asleep. I'm not going to fall asleep. I'm not going to fall asleep. I'm not going to fall asleep. I'm not going to fall asleep. I'm not going to fall asleep. I'm not going to fall asleep. I'm not going to fall asleep. I'm not going to fall asleep. I'm not going to fall asleep. I'm not going to fall asleep. I'm not going to fall asleep. I'm not going to fall asleep. I'm not going to fall asleep. I'm not going to fall asleep. I'm not going to fall asleep. I'm not going to fall asleep. I'm not going to fall asleep. I'm not going to fall asleep. I'm not going to fall asleep. I'm not going to fall asleep. I'm not going to fall asleep. I'm not going to fall asleep. I'm not going to fall asleep. I'm not going to fall asleep. I'm not going to fall asleep. I'm not going to fall asleep. I'm not going to fall asleep. I'm not going to fall asleep. I'm not going to fall asleep. I'm not going to fall asleep. I'm not going to fall asleep. I'm not going to fall asleep. I'm not going to fall asleep. I'm not going to fall asleep. I'm not going to fall asleep. I'm not going to fall asleep. I'm not going to fall asleep. I'm not going to fall asleep. I'm not going to fall asleep. I'm not going to fall asleep. I'm not going to fall asleep. I'm not going to fall asleep. I'm not going to fall asleep. I'm not going",
+ "raw": {
+ "normalized_reference": "when did you first realize something was wrong?",
+ "normalized_hypothesis": "i'm not going to fall asleep. i'm not going to fall asleep. i'm not going to fall asleep. i'm not going to fall asleep. i'm not going to fall asleep. i'm not going to fall asleep. i'm not going to fall asleep. i'm not going to fall asleep. i'm not going to fall asleep. i'm not going to fall asleep. i'm not going to fall asleep. i'm not going to fall asleep. i'm not going to fall asleep. i'm not going to fall asleep. i'm not going to fall asleep. i'm not going to fall asleep. i'm not going to fall asleep. i'm not going to fall asleep. i'm not going to fall asleep. i'm not going to fall asleep. i'm not going to fall asleep. i'm not going to fall asleep. i'm not going to fall asleep. i'm not going to fall asleep. i'm not going to fall asleep. i'm not going to fall asleep. i'm not going to fall asleep. i'm not going to fall asleep. i'm not going to fall asleep. i'm not going to fall asleep. i'm not going to fall asleep. i'm not going to fall asleep. i'm not going to fall asleep. i'm not going to fall asleep. i'm not going to fall asleep. i'm not going to fall asleep. i'm not going to fall asleep. i'm not going to fall asleep. i'm not going to fall asleep. i'm not going to fall asleep. i'm not going to fall asleep. i'm not going to fall asleep. i'm not going to fall asleep. i'm not going to fall asleep. i'm not going to fall asleep. i'm not going to fall asleep. i'm not going to fall asleep. i'm not going to fall asleep. i'm not going to fall asleep. i'm not going to fall asleep. i'm not going to fall asleep. i'm not going to fall asleep. i'm not going to fall asleep. i'm not going to fall asleep. i'm not going to fall asleep. i'm not going",
+ "errors": 333,
+ "words": 8,
+ "wer": 41.625
+ },
+ "semantic": {
+ "normalized_reference": "when did you 1st realize something was wrong",
+ "normalized_hypothesis": "i am not going to fall asleep i am not going to fall asleep i am not going to fall asleep i am not going to fall asleep i am not going to fall asleep i am not going to fall asleep i am not going to fall asleep i am not going to fall asleep i am not going to fall asleep i am not going to fall asleep i am not going to fall asleep i am not going to fall asleep i am not going to fall asleep i am not going to fall asleep i am not going to fall asleep i am not going to fall asleep i am not going to fall asleep i am not going to fall asleep i am not going to fall asleep i am not going to fall asleep i am not going to fall asleep i am not going to fall asleep i am not going to fall asleep i am not going to fall asleep i am not going to fall asleep i am not going to fall asleep i am not going to fall asleep i am not going to fall asleep i am not going to fall asleep i am not going to fall asleep i am not going to fall asleep i am not going to fall asleep i am not going to fall asleep i am not going to fall asleep i am not going to fall asleep i am not going to fall asleep i am not going to fall asleep i am not going to fall asleep i am not going to fall asleep i am not going to fall asleep i am not going to fall asleep i am not going to fall asleep i am not going to fall asleep i am not going to fall asleep i am not going to fall asleep i am not going to fall asleep i am not going to fall asleep i am not going to fall asleep i am not going to fall asleep i am not going to fall asleep i am not going to fall asleep i am not going to fall asleep i am not going to fall asleep i am not going to fall asleep i am not going to fall asleep i am not going",
+ "errors": 389,
+ "words": 8,
+ "wer": 48.625
+ }
+ },
+ {
+ "id": "release500_0012",
+ "category": "rare_words",
+ "text": "The archaeologist catalogued a Byzantine reliquary and several anachronistic fragments.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0012.wav",
+ "render_seconds": 103.6040112740593,
+ "audio_seconds": 6.478367346938776,
+ "rtf": 15.992302647520495,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you",
+ "raw": {
+ "normalized_reference": "the archaeologist catalogued a byzantine reliquary and several anachronistic fragments.",
+ "normalized_hypothesis": "i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you",
+ "errors": 381,
+ "words": 10,
+ "wer": 38.1
+ },
+ "semantic": {
+ "normalized_reference": "the archaeologist catalogued a byzantine reliquary and several anachronistic fragments",
+ "normalized_hypothesis": "i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you",
+ "errors": 381,
+ "words": 10,
+ "wer": 38.1
+ }
+ },
+ {
+ "id": "release500_0013",
+ "category": "short",
+ "text": "I have no idea.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0013.wav",
+ "render_seconds": 20.591476502013393,
+ "audio_seconds": 1.5325170068027212,
+ "rtf": 13.436377156409662,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "Staffordshire Apple.",
+ "raw": {
+ "normalized_reference": "i have no idea.",
+ "normalized_hypothesis": "staffordshire apple.",
+ "errors": 4,
+ "words": 4,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "i have no idea",
+ "normalized_hypothesis": "staffordshire apple",
+ "errors": 4,
+ "words": 4,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0014",
+ "category": "technical",
+ "text": "The system runs on three core components that all have to stay in sync.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0014.wav",
+ "render_seconds": 94.01606339204591,
+ "audio_seconds": 4.9458503401360545,
+ "rtf": 19.00907971862724,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "The sister was in a twink. Look at this. I have to start a song.",
+ "raw": {
+ "normalized_reference": "the system runs on three core components that all have to stay in sync.",
+ "normalized_hypothesis": "the sister was in a twink. look at this. i have to start a song.",
+ "errors": 12,
+ "words": 14,
+ "wer": 0.8571428571428571
+ },
+ "semantic": {
+ "normalized_reference": "the system runs on 3 core components that all have to stay in sync",
+ "normalized_hypothesis": "the sister was in a twink look at this i have to start a song",
+ "errors": 12,
+ "words": 14,
+ "wer": 0.8571428571428571
+ }
+ },
+ {
+ "id": "release500_0015",
+ "category": "conversational",
+ "text": "We improvised the entire second half and somehow nobody noticed.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0015.wav",
+ "render_seconds": 81.10544405097608,
+ "audio_seconds": 4.597551020408163,
+ "rtf": 17.641010114070614,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "You are for the last time to ask a question so I will do my best.",
+ "raw": {
+ "normalized_reference": "we improvised the entire second half and somehow nobody noticed.",
+ "normalized_hypothesis": "you are for the last time to ask a question so i will do my best.",
+ "errors": 15,
+ "words": 10,
+ "wer": 1.5
+ },
+ "semantic": {
+ "normalized_reference": "we improvised the entire 2nd half and somehow nobody noticed",
+ "normalized_hypothesis": "you are for the last time to ask a question so i will do my best",
+ "errors": 15,
+ "words": 10,
+ "wer": 1.5
+ }
+ },
+ {
+ "id": "release500_0016",
+ "category": "descriptive",
+ "text": "There's a smell in old buildings that's impossible to describe but immediately familiar.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0016.wav",
+ "render_seconds": 92.49169414199423,
+ "audio_seconds": 5.851428571428571,
+ "rtf": 15.806686010594717,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "I'm just trying to make my own special place in this world. I don't know why I'm doing this all the time.",
+ "raw": {
+ "normalized_reference": "there's a smell in old buildings that's impossible to describe but immediately familiar.",
+ "normalized_hypothesis": "i'm just trying to make my own special place in this world. i don't know why i'm doing this all the time.",
+ "errors": 21,
+ "words": 13,
+ "wer": 1.6153846153846154
+ },
+ "semantic": {
+ "normalized_reference": "there is a smell in old buildings that is impossible to describe but immediately familiar",
+ "normalized_hypothesis": "i am just trying to make my own special place in this world i do not know why i am doing this all the time",
+ "errors": 24,
+ "words": 15,
+ "wer": 1.6
+ }
+ },
+ {
+ "id": "release500_0017",
+ "category": "emotional",
+ "text": "I specifically asked for one thing. One thing. And it still didn't happen.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0017.wav",
+ "render_seconds": 99.40190737403464,
+ "audio_seconds": 5.712108843537415,
+ "rtf": 17.401963109735963,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "Splash, Splash, Splash, Splash, Splash, Splash, Splash, Splash, Splash, Splash, Splash, Splash, Splash, Splash, Splash, Splash, Splash, Splash, Splash, Splash, Splash, Splash, Splash, Splash, Splash, Splash, Splash, Splash, Splash, Splash, Splash, Splash, Splash, Splash, Splash, Splash, Splash, Splash, Splash, Splash, Splash, Splash, Splash, Splash, Splash, Splash, Splash, Splash, Splash, Splash, Splash, Splash, Splash, Splash, Splash, Splash, Splash, Splash, Splash, Splash, Splash, Splash, Splash, Splash, Splash, Splash, Splash, Splash, Splash, Splash, Splash, Splash, Splash, Splash, Splash, Splash, Splash, Splash, Splash, Splash, Splash, Splash, Splash, Splash, Splash, Splash, Splash, Splash, Splash, Splash, Splash, Splash, Splash, Splash, Splash, Splash, Splash, Splash, Splash, Splash, Splash, Splash, Splash, Splash, Splash, Splash, Splash, Splash, Splash, Splash, Splash, Splash, Splash, Splash, Splash, Splash, Splash, Splash, Splash, Splash, Splash, Splash, Splash, Splash, Splash, Splash, Splash, Splash, Splash, Splash, Splash, Splash, Splash, Splash, Splash, Splash, Splash, Splash, Splash, Splash, Splash, Splash, Splash, Splash, Splash, Splash, Splash, Splash,",
+ "raw": {
+ "normalized_reference": "i specifically asked for one thing. one thing. and it still didn't happen.",
+ "normalized_hypothesis": "splash, splash, splash, splash, splash, splash, splash, splash, splash, splash, splash, splash, splash, splash, splash, splash, splash, splash, splash, splash, splash, splash, splash, splash, splash, splash, splash, splash, splash, splash, splash, splash, splash, splash, splash, splash, splash, splash, splash, splash, splash, splash, splash, splash, splash, splash, splash, splash, splash, splash, splash, splash, splash, splash, splash, splash, splash, splash, splash, splash, splash, splash, splash, splash, splash, splash, splash, splash, splash, splash, splash, splash, splash, splash, splash, splash, splash, splash, splash, splash, splash, splash, splash, splash, splash, splash, splash, splash, splash, splash, splash, splash, splash, splash, splash, splash, splash, splash, splash, splash, splash, splash, splash, splash, splash, splash, splash, splash, splash, splash, splash, splash, splash, splash, splash, splash, splash, splash, splash, splash, splash, splash, splash, splash, splash, splash, splash, splash, splash, splash, splash, splash, splash, splash, splash, splash, splash, splash, splash, splash, splash, splash, splash, splash, splash, splash, splash, splash,",
+ "errors": 148,
+ "words": 13,
+ "wer": 11.384615384615385
+ },
+ "semantic": {
+ "normalized_reference": "i specifically asked for one thing one thing and it still did not happen",
+ "normalized_hypothesis": "splash splash splash splash splash splash splash splash splash splash splash splash splash splash splash splash splash splash splash splash splash splash splash splash splash splash splash splash splash splash splash splash splash splash splash splash splash splash splash splash splash splash splash splash splash splash splash splash splash splash splash splash splash splash splash splash splash splash splash splash splash splash splash splash splash splash splash splash splash splash splash splash splash splash splash splash splash splash splash splash splash splash splash splash splash splash splash splash splash splash splash splash splash splash splash splash splash splash splash splash splash splash splash splash splash splash splash splash splash splash splash splash splash splash splash splash splash splash splash splash splash splash splash splash splash splash splash splash splash splash splash splash splash splash splash splash splash splash splash splash splash splash splash splash splash splash splash splash",
+ "errors": 148,
+ "words": 14,
+ "wer": 10.571428571428571
+ }
+ },
+ {
+ "id": "release500_0018",
+ "category": "homographs",
+ "text": "They refuse to place the refuse beside the produce aisle.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0018.wav",
+ "render_seconds": 74.38826269307174,
+ "audio_seconds": 3.9706122448979593,
+ "rtf": 18.734708429073372,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "I will have to guess the word for this. I sure hope to answer it.",
+ "raw": {
+ "normalized_reference": "they refuse to place the refuse beside the produce aisle.",
+ "normalized_hypothesis": "i will have to guess the word for this. i sure hope to answer it.",
+ "errors": 13,
+ "words": 10,
+ "wer": 1.3
+ },
+ "semantic": {
+ "normalized_reference": "they refuse to place the refuse beside the produce aisle",
+ "normalized_hypothesis": "i will have to guess the word for this i sure hope to answer it",
+ "errors": 13,
+ "words": 10,
+ "wer": 1.3
+ }
+ },
+ {
+ "id": "release500_0019",
+ "category": "long",
+ "text": "There were days last year when I genuinely didn't know how I was going to get through it, and somehow I did, and I still don't fully understand how.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0019.wav",
+ "render_seconds": 99.78391260211356,
+ "audio_seconds": 8.916462585034013,
+ "rtf": 11.19097530556541,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "Now what does this show? What a channel I didn't know how was going to get on it. And so how I did. And I still don't feel like I still know.",
+ "raw": {
+ "normalized_reference": "there were days last year when i genuinely didn't know how i was going to get through it, and somehow i did, and i still don't fully understand how.",
+ "normalized_hypothesis": "now what does this show? what a channel i didn't know how was going to get on it. and so how i did. and i still don't feel like i still know.",
+ "errors": 18,
+ "words": 29,
+ "wer": 0.6206896551724138
+ },
+ "semantic": {
+ "normalized_reference": "there were days last year when i genuinely did not know how i was going to get through it and somehow i did and i still do not fully understand how",
+ "normalized_hypothesis": "now what does this show what a channel i did not know how was going to get on it and so how i did and i still do not feel like i still know",
+ "errors": 18,
+ "words": 31,
+ "wer": 0.5806451612903226
+ }
+ },
+ {
+ "id": "release500_0020",
+ "category": "long_range",
+ "text": "Before announcing the winner, the judge thanked the volunteers who had worked through the storm to repair the damaged stage.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0020.wav",
+ "render_seconds": 125.08898339990992,
+ "audio_seconds": 7.801904761904762,
+ "rtf": 16.033133858631032,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "A few hours ago, a church by a brilliant house was locked or disclosed to the world and the rest of the world.",
+ "raw": {
+ "normalized_reference": "before announcing the winner, the judge thanked the volunteers who had worked through the storm to repair the damaged stage.",
+ "normalized_hypothesis": "a few hours ago, a church by a brilliant house was locked or disclosed to the world and the rest of the world.",
+ "errors": 22,
+ "words": 20,
+ "wer": 1.1
+ },
+ "semantic": {
+ "normalized_reference": "before announcing the winner the judge thanked the volunteers who had worked through the storm to repair the damaged stage",
+ "normalized_hypothesis": "a few hours ago a church by a brilliant house was locked or disclosed to the world and the rest of the world",
+ "errors": 22,
+ "words": 20,
+ "wer": 1.1
+ }
+ },
+ {
+ "id": "release500_0021",
+ "category": "names_places",
+ "text": "Professor Nguyen mailed the fluorescent poster to Reykjavik on Thursday.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0021.wav",
+ "render_seconds": 90.39663011499215,
+ "audio_seconds": 5.154829931972789,
+ "rtf": 17.536297280014576,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "The fact that I'm living by this election is so correct that I will just die.",
+ "raw": {
+ "normalized_reference": "professor nguyen mailed the fluorescent poster to reykjavik on thursday.",
+ "normalized_hypothesis": "the fact that i'm living by this election is so correct that i will just die.",
+ "errors": 16,
+ "words": 10,
+ "wer": 1.6
+ },
+ "semantic": {
+ "normalized_reference": "professor nguyen mailed the fluorescent poster to reykjavik on thursday",
+ "normalized_hypothesis": "the fact that i am living by this election is so correct that i will just die",
+ "errors": 17,
+ "words": 10,
+ "wer": 1.7
+ }
+ },
+ {
+ "id": "release500_0022",
+ "category": "narrative",
+ "text": "The plan was simple. The execution was anything but.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0022.wav",
+ "render_seconds": 79.6991829710314,
+ "audio_seconds": 3.9706122448979593,
+ "rtf": 20.072265448090764,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "The crowd was shopping. Lots of cash was on the side of the road.",
+ "raw": {
+ "normalized_reference": "the plan was simple. the execution was anything but.",
+ "normalized_hypothesis": "the crowd was shopping. lots of cash was on the side of the road.",
+ "errors": 11,
+ "words": 9,
+ "wer": 1.2222222222222223
+ },
+ "semantic": {
+ "normalized_reference": "the plan was simple the execution was anything but",
+ "normalized_hypothesis": "the crowd was shopping lots of cash was on the side of the road",
+ "errors": 11,
+ "words": 9,
+ "wer": 1.2222222222222223
+ }
+ },
+ {
+ "id": "release500_0023",
+ "category": "numbers",
+ "text": "Her score improved from eighty-two percent to ninety-seven percent in two weeks.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0023.wav",
+ "render_seconds": 55.29268986696843,
+ "audio_seconds": 5.50312925170068,
+ "rtf": 10.047499765679834,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "Thank you for watching. Please like, share, and subscribe.",
+ "raw": {
+ "normalized_reference": "her score improved from eighty-two percent to ninety-seven percent in two weeks.",
+ "normalized_hypothesis": "thank you for watching. please like, share, and subscribe.",
+ "errors": 14,
+ "words": 14,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "her score improved from 82% to 97% in 2 weeks",
+ "normalized_hypothesis": "thank you for watching please like share and subscribe",
+ "errors": 10,
+ "words": 10,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0024",
+ "category": "punctuation",
+ "text": "Why would anyone pack spoons, batteries, and a violin for one short trip?",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0024.wav",
+ "render_seconds": 88.19932402996346,
+ "audio_seconds": 5.294149659863946,
+ "rtf": 16.659771577411377,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "Thank you for watching. Please subscribe to my channel. Thank you for watching. Please subscribe to my channel. Thank you for watching. Please subscribe to my channel. Thank you for watching. Please subscribe to my channel. Thank you for watching. Please subscribe to my channel. Thank you for watching. Please subscribe to my channel. Thank you for watching. Please subscribe to my channel. Thank you for watching. Please subscribe to my channel. Thank you for watching. Please subscribe to my channel. Thank you for watching. Please subscribe to my channel. Thank you for watching. Please subscribe to my channel. Thank you for watching. Please subscribe to my channel. Thank you for watching. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please",
+ "raw": {
+ "normalized_reference": "why would anyone pack spoons, batteries, and a violin for one short trip?",
+ "normalized_hypothesis": "thank you for watching. please subscribe to my channel. thank you for watching. please subscribe to my channel. thank you for watching. please subscribe to my channel. thank you for watching. please subscribe to my channel. thank you for watching. please subscribe to my channel. thank you for watching. please subscribe to my channel. thank you for watching. please subscribe to my channel. thank you for watching. please subscribe to my channel. thank you for watching. please subscribe to my channel. thank you for watching. please subscribe to my channel. thank you for watching. please subscribe to my channel. thank you for watching. please subscribe to my channel. thank you for watching. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please",
+ "errors": 367,
+ "words": 13,
+ "wer": 28.23076923076923
+ },
+ "semantic": {
+ "normalized_reference": "why would anyone pack spoons batteries and a violin for one short trip",
+ "normalized_hypothesis": "thank you for watching please subscribe to my channel thank you for watching please subscribe to my channel thank you for watching please subscribe to my channel thank you for watching please subscribe to my channel thank you for watching please subscribe to my channel thank you for watching please subscribe to my channel thank you for watching please subscribe to my channel thank you for watching please subscribe to my channel thank you for watching please subscribe to my channel thank you for watching please subscribe to my channel thank you for watching please subscribe to my channel thank you for watching please subscribe to my channel thank you for watching please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please",
+ "errors": 367,
+ "words": 13,
+ "wer": 28.23076923076923
+ }
+ },
+ {
+ "id": "release500_0025",
+ "category": "question",
+ "text": "How long have you been sitting on this?",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0025.wav",
+ "render_seconds": 35.49992396996822,
+ "audio_seconds": 2.5774149659863945,
+ "rtf": 13.77346078860148,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "I love you, Bruce. I love you, Bruce. I love you, Bruce. I love you, Bruce. I love you, Bruce. I love you, Bruce. I love you, Bruce. I love you, Bruce. I love you, Bruce. I love you, Bruce. I love you, Bruce. I love you, Bruce. I love you, Bruce. I love you, Bruce. I love you, Bruce. I love you, Bruce. I love you, Bruce. I love you, Bruce. I love you, Bruce. I love you, Bruce. I love you, Bruce. I love you, Bruce. I love you, Bruce. I love you, Bruce. I love you, Bruce. I love you, Bruce. I love you, Bruce. I love you, Bruce. I love you, Bruce. I love you, Bruce. I love you, Bruce. I love you, Bruce. I love you, Bruce. I love you, Bruce. I love you, Bruce. I love you, Bruce. I love you, Bruce. I love you, Bruce. I love you, Bruce. I love you, Bruce. I love you, Bruce. I love you, Bruce. I love you, Bruce. I love you, Bruce. I love you, Bruce. I love you, Bruce. I love you, Bruce. I love you, Bruce. I love you, Bruce. I love you, Bruce. I love you, Bruce. I love you, Bruce. I love you, Bruce. I love you, Bruce. I love you, Bruce. I love you, Bruce. I love you, Bruce. I love you, Bruce. I love you, Bruce. I love you, Bruce. I love you, Bruce. I love you, Bruce. I love you, Bruce. I love you, Bruce. I love you, Bruce. I love you, Bruce. I love you, Bruce. I love you, Bruce. I love you, Bruce. I love you, Bruce. I love you, Bruce. I love you, Bruce. I love you, Bruce. I love you, Bruce.",
+ "raw": {
+ "normalized_reference": "how long have you been sitting on this?",
+ "normalized_hypothesis": "i love you, bruce. i love you, bruce. i love you, bruce. i love you, bruce. i love you, bruce. i love you, bruce. i love you, bruce. i love you, bruce. i love you, bruce. i love you, bruce. i love you, bruce. i love you, bruce. i love you, bruce. i love you, bruce. i love you, bruce. i love you, bruce. i love you, bruce. i love you, bruce. i love you, bruce. i love you, bruce. i love you, bruce. i love you, bruce. i love you, bruce. i love you, bruce. i love you, bruce. i love you, bruce. i love you, bruce. i love you, bruce. i love you, bruce. i love you, bruce. i love you, bruce. i love you, bruce. i love you, bruce. i love you, bruce. i love you, bruce. i love you, bruce. i love you, bruce. i love you, bruce. i love you, bruce. i love you, bruce. i love you, bruce. i love you, bruce. i love you, bruce. i love you, bruce. i love you, bruce. i love you, bruce. i love you, bruce. i love you, bruce. i love you, bruce. i love you, bruce. i love you, bruce. i love you, bruce. i love you, bruce. i love you, bruce. i love you, bruce. i love you, bruce. i love you, bruce. i love you, bruce. i love you, bruce. i love you, bruce. i love you, bruce. i love you, bruce. i love you, bruce. i love you, bruce. i love you, bruce. i love you, bruce. i love you, bruce. i love you, bruce. i love you, bruce. i love you, bruce. i love you, bruce. i love you, bruce. i love you, bruce. i love you, bruce.",
+ "errors": 295,
+ "words": 8,
+ "wer": 36.875
+ },
+ "semantic": {
+ "normalized_reference": "how long have you been sitting on this",
+ "normalized_hypothesis": "i love you bruce i love you bruce i love you bruce i love you bruce i love you bruce i love you bruce i love you bruce i love you bruce i love you bruce i love you bruce i love you bruce i love you bruce i love you bruce i love you bruce i love you bruce i love you bruce i love you bruce i love you bruce i love you bruce i love you bruce i love you bruce i love you bruce i love you bruce i love you bruce i love you bruce i love you bruce i love you bruce i love you bruce i love you bruce i love you bruce i love you bruce i love you bruce i love you bruce i love you bruce i love you bruce i love you bruce i love you bruce i love you bruce i love you bruce i love you bruce i love you bruce i love you bruce i love you bruce i love you bruce i love you bruce i love you bruce i love you bruce i love you bruce i love you bruce i love you bruce i love you bruce i love you bruce i love you bruce i love you bruce i love you bruce i love you bruce i love you bruce i love you bruce i love you bruce i love you bruce i love you bruce i love you bruce i love you bruce i love you bruce i love you bruce i love you bruce i love you bruce i love you bruce i love you bruce i love you bruce i love you bruce i love you bruce i love you bruce i love you bruce",
+ "errors": 295,
+ "words": 8,
+ "wer": 36.875
+ }
+ },
+ {
+ "id": "release500_0026",
+ "category": "rare_words",
+ "text": "A mischievous chameleon rested beside the chrysanthemum and the bougainvillea.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0026.wav",
+ "render_seconds": 99.68826815101784,
+ "audio_seconds": 5.572789115646258,
+ "rtf": 17.88839772729446,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "This covers how I was to assert the question to you on the brand new day.",
+ "raw": {
+ "normalized_reference": "a mischievous chameleon rested beside the chrysanthemum and the bougainvillea.",
+ "normalized_hypothesis": "this covers how i was to assert the question to you on the brand new day.",
+ "errors": 14,
+ "words": 10,
+ "wer": 1.4
+ },
+ "semantic": {
+ "normalized_reference": "a mischievous chameleon rested beside the chrysanthemum and the bougainvillea",
+ "normalized_hypothesis": "this covers how i was to assert the question to you on the brand new day",
+ "errors": 14,
+ "words": 10,
+ "wer": 1.4
+ }
+ },
+ {
+ "id": "release500_0027",
+ "category": "short",
+ "text": "We don't talk about that.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0027.wav",
+ "render_seconds": 50.900528709054925,
+ "audio_seconds": 1.9504761904761905,
+ "rtf": 26.096462472904136,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "Why they actually have to burn it down?",
+ "raw": {
+ "normalized_reference": "we don't talk about that.",
+ "normalized_hypothesis": "why they actually have to burn it down?",
+ "errors": 8,
+ "words": 5,
+ "wer": 1.6
+ },
+ "semantic": {
+ "normalized_reference": "we do not talk about that",
+ "normalized_hypothesis": "why they actually have to burn it down",
+ "errors": 8,
+ "words": 6,
+ "wer": 1.3333333333333333
+ }
+ },
+ {
+ "id": "release500_0028",
+ "category": "technical",
+ "text": "Hash the UTF eight payload with S H A two fifty-six before opening the checkpoint.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0028.wav",
+ "render_seconds": 98.01105747499969,
+ "audio_seconds": 5.7817687074829935,
+ "rtf": 16.95174304502183,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you",
+ "raw": {
+ "normalized_reference": "hash the utf eight payload with s h a two fifty-six before opening the checkpoint.",
+ "normalized_hypothesis": "i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you",
+ "errors": 381,
+ "words": 16,
+ "wer": 23.8125
+ },
+ "semantic": {
+ "normalized_reference": "hash the utf 8 payload with s h a 256 before opening the checkpoint",
+ "normalized_hypothesis": "i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you",
+ "errors": 381,
+ "words": 14,
+ "wer": 27.214285714285715
+ }
+ },
+ {
+ "id": "release500_0029",
+ "category": "conversational",
+ "text": "I thought I had it all figured out, and then this happened.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0029.wav",
+ "render_seconds": 71.49475576600526,
+ "audio_seconds": 4.179591836734694,
+ "rtf": 17.10567886979618,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "Stop being a fool, you son of a bitch. I know what you're up to.",
+ "raw": {
+ "normalized_reference": "i thought i had it all figured out, and then this happened.",
+ "normalized_hypothesis": "stop being a fool, you son of a bitch. i know what you're up to.",
+ "errors": 15,
+ "words": 12,
+ "wer": 1.25
+ },
+ "semantic": {
+ "normalized_reference": "i thought i had it all figured out and then this happened",
+ "normalized_hypothesis": "stop being a fool you son of a bitch i know what you are up to",
+ "errors": 16,
+ "words": 12,
+ "wer": 1.3333333333333333
+ }
+ },
+ {
+ "id": "release500_0030",
+ "category": "descriptive",
+ "text": "The kind of quiet you only get at three in the morning in a city.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0030.wav",
+ "render_seconds": 74.69945722492412,
+ "audio_seconds": 4.179591836734694,
+ "rtf": 17.872428730572665,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "The plot of the book is a little bit different than the story of the show.",
+ "raw": {
+ "normalized_reference": "the kind of quiet you only get at three in the morning in a city.",
+ "normalized_hypothesis": "the plot of the book is a little bit different than the story of the show.",
+ "errors": 13,
+ "words": 15,
+ "wer": 0.8666666666666667
+ },
+ "semantic": {
+ "normalized_reference": "the kind of quiet you only get at 3 in the morning in a city",
+ "normalized_hypothesis": "the plot of the book is a little bit different than the story of the show",
+ "errors": 13,
+ "words": 15,
+ "wer": 0.8666666666666667
+ }
+ },
+ {
+ "id": "release500_0031",
+ "category": "emotional",
+ "text": "The joy of it was almost overwhelming, like too much good all at once.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0031.wav",
+ "render_seconds": 74.61474312399514,
+ "audio_seconds": 4.876190476190477,
+ "rtf": 15.301851617225564,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "The climate industry has been holding up. It couldn't have been yet this.",
+ "raw": {
+ "normalized_reference": "the joy of it was almost overwhelming, like too much good all at once.",
+ "normalized_hypothesis": "the climate industry has been holding up. it couldn't have been yet this.",
+ "errors": 13,
+ "words": 14,
+ "wer": 0.9285714285714286
+ },
+ "semantic": {
+ "normalized_reference": "the joy of it was almost overwhelming like too much good all at once",
+ "normalized_hypothesis": "the climate industry has been holding up it could not have been yet this",
+ "errors": 13,
+ "words": 14,
+ "wer": 0.9285714285714286
+ }
+ },
+ {
+ "id": "release500_0032",
+ "category": "homographs",
+ "text": "Please close the door while I stand close to the microphone.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0032.wav",
+ "render_seconds": 82.40504369803239,
+ "audio_seconds": 4.1099319727891155,
+ "rtf": 20.050220841516754,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "Just kiss the girl where you're stuck. Kiss to get rid of her.",
+ "raw": {
+ "normalized_reference": "please close the door while i stand close to the microphone.",
+ "normalized_hypothesis": "just kiss the girl where you're stuck. kiss to get rid of her.",
+ "errors": 11,
+ "words": 11,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "please close the door while i stand close to the microphone",
+ "normalized_hypothesis": "just kiss the girl where you are stuck kiss to get rid of her",
+ "errors": 12,
+ "words": 11,
+ "wer": 1.0909090909090908
+ }
+ },
+ {
+ "id": "release500_0033",
+ "category": "long",
+ "text": "I think the most underrated skill in the world is knowing when to stop explaining yourself and just let people draw their own conclusions.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0033.wav",
+ "render_seconds": 132.40289115905762,
+ "audio_seconds": 8.916462585034013,
+ "rtf": 14.849262237839866,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "I'm the best in the world at scale, and you're the best in the world at scale, and you're the best in the world at scale, and you're the best in the world at scale, and you're the best in the world at scale, and you're the best in the world at scale, and you're the best in the world at scale, and you're the best in the world at scale, and you're the best in the world at scale, and you're the best in the world at scale, and you're the best in the world at scale, and you're the best in the world at scale, and you're the best in the world at scale, and you're the best in the world at scale, and you're the best in the world at scale, and you're the best in the world at scale, and you're the best in the world at scale, and you're the best in the world at scale, and you're the best in the world at scale, and you're the best in the world at scale, and you're the best in the world at scale, and you're the best in the world at scale, and you're the best in the world at scale, and you're the best in the world at scale, and you're the best in the world at scale, and you're the best in the world at scale, and you're the best in the world at scale, and you're the best in the world at scale, and you're the best in the world at scale, and you're the best in the world at scale, and you're the best in the world at scale, and you're the best in the world at scale, and you're the best in the world at scale, and you're the best in the world at scale, and you're the best in the world at scale, and you're the best in the world at scale, and you're the best in the world at scale, and you're the best in the world at scale, and you're the best in the world at scale, and you're the best in the world at scale, and you're the best",
+ "raw": {
+ "normalized_reference": "i think the most underrated skill in the world is knowing when to stop explaining yourself and just let people draw their own conclusions.",
+ "normalized_hypothesis": "i'm the best in the world at scale, and you're the best in the world at scale, and you're the best in the world at scale, and you're the best in the world at scale, and you're the best in the world at scale, and you're the best in the world at scale, and you're the best in the world at scale, and you're the best in the world at scale, and you're the best in the world at scale, and you're the best in the world at scale, and you're the best in the world at scale, and you're the best in the world at scale, and you're the best in the world at scale, and you're the best in the world at scale, and you're the best in the world at scale, and you're the best in the world at scale, and you're the best in the world at scale, and you're the best in the world at scale, and you're the best in the world at scale, and you're the best in the world at scale, and you're the best in the world at scale, and you're the best in the world at scale, and you're the best in the world at scale, and you're the best in the world at scale, and you're the best in the world at scale, and you're the best in the world at scale, and you're the best in the world at scale, and you're the best in the world at scale, and you're the best in the world at scale, and you're the best in the world at scale, and you're the best in the world at scale, and you're the best in the world at scale, and you're the best in the world at scale, and you're the best in the world at scale, and you're the best in the world at scale, and you're the best in the world at scale, and you're the best in the world at scale, and you're the best in the world at scale, and you're the best in the world at scale, and you're the best in the world at scale, and you're the best",
+ "errors": 358,
+ "words": 24,
+ "wer": 14.916666666666666
+ },
+ "semantic": {
+ "normalized_reference": "i think the most underrated skill in the world is knowing when to stop explaining yourself and just let people draw their own conclusions",
+ "normalized_hypothesis": "i am the best in the world at scale and you are the best in the world at scale and you are the best in the world at scale and you are the best in the world at scale and you are the best in the world at scale and you are the best in the world at scale and you are the best in the world at scale and you are the best in the world at scale and you are the best in the world at scale and you are the best in the world at scale and you are the best in the world at scale and you are the best in the world at scale and you are the best in the world at scale and you are the best in the world at scale and you are the best in the world at scale and you are the best in the world at scale and you are the best in the world at scale and you are the best in the world at scale and you are the best in the world at scale and you are the best in the world at scale and you are the best in the world at scale and you are the best in the world at scale and you are the best in the world at scale and you are the best in the world at scale and you are the best in the world at scale and you are the best in the world at scale and you are the best in the world at scale and you are the best in the world at scale and you are the best in the world at scale and you are the best in the world at scale and you are the best in the world at scale and you are the best in the world at scale and you are the best in the world at scale and you are the best in the world at scale and you are the best in the world at scale and you are the best in the world at scale and you are the best in the world at scale and you are the best in the world at scale and you are the best in the world at scale and you are the best in the world at scale and you are the best",
+ "errors": 398,
+ "words": 24,
+ "wer": 16.583333333333332
+ }
+ },
+ {
+ "id": "release500_0034",
+ "category": "long_range",
+ "text": "When the final bus disappeared around the corner, Nadia remembered that her umbrella, notebook, and lunch were still inside the quiet office.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0034.wav",
+ "render_seconds": 125.5994538969826,
+ "audio_seconds": 9.404081632653062,
+ "rtf": 13.355844706927401,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "Well, it's certainly best to see how alive you are. Now, I want to remind you that the end will stay in a certain place.",
+ "raw": {
+ "normalized_reference": "when the final bus disappeared around the corner, nadia remembered that her umbrella, notebook, and lunch were still inside the quiet office.",
+ "normalized_hypothesis": "well, it's certainly best to see how alive you are. now, i want to remind you that the end will stay in a certain place.",
+ "errors": 25,
+ "words": 22,
+ "wer": 1.1363636363636365
+ },
+ "semantic": {
+ "normalized_reference": "when the final bus disappeared around the corner nadia remembered that her umbrella notebook and lunch were still inside the quiet office",
+ "normalized_hypothesis": "well it is certainly best to see how alive you are now i want to remind you that the end will stay in a certain place",
+ "errors": 26,
+ "words": 22,
+ "wer": 1.1818181818181819
+ }
+ },
+ {
+ "id": "release500_0035",
+ "category": "names_places",
+ "text": "Aurelia and Joaquin crossed the Schuylkill River before noon.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0035.wav",
+ "render_seconds": 80.80838598497212,
+ "audio_seconds": 4.31891156462585,
+ "rtf": 18.710359028148474,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "I am taking questions I could never before hear.",
+ "raw": {
+ "normalized_reference": "aurelia and joaquin crossed the schuylkill river before noon.",
+ "normalized_hypothesis": "i am taking questions i could never before hear.",
+ "errors": 8,
+ "words": 9,
+ "wer": 0.8888888888888888
+ },
+ "semantic": {
+ "normalized_reference": "aurelia and joaquin crossed the schuylkill river before noon",
+ "normalized_hypothesis": "i am taking questions i could never before hear",
+ "errors": 8,
+ "words": 9,
+ "wer": 0.8888888888888888
+ }
+ },
+ {
+ "id": "release500_0036",
+ "category": "narrative",
+ "text": "The village had a name, but nobody used it. Everyone just called it home.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0036.wav",
+ "render_seconds": 91.90063758695032,
+ "audio_seconds": 5.224489795918367,
+ "rtf": 17.59035641312721,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "Good night, everyone. Good night, everyone. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love",
+ "raw": {
+ "normalized_reference": "the village had a name, but nobody used it. everyone just called it home.",
+ "normalized_hypothesis": "good night, everyone. good night, everyone. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love",
+ "errors": 332,
+ "words": 14,
+ "wer": 23.714285714285715
+ },
+ "semantic": {
+ "normalized_reference": "the village had a name but nobody used it everyone just called it home",
+ "normalized_hypothesis": "good night everyone good night everyone i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love",
+ "errors": 332,
+ "words": 14,
+ "wer": 23.714285714285715
+ }
+ },
+ {
+ "id": "release500_0037",
+ "category": "numbers",
+ "text": "Increase the gain by three decibels, wait five seconds, then reduce it by one.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0037.wav",
+ "render_seconds": 50.48591187398415,
+ "audio_seconds": 5.50312925170068,
+ "rtf": 9.174037091420676,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "You question all those wonders of peace, lots of circles, and wonders of the world.",
+ "raw": {
+ "normalized_reference": "increase the gain by three decibels, wait five seconds, then reduce it by one.",
+ "normalized_hypothesis": "you question all those wonders of peace, lots of circles, and wonders of the world.",
+ "errors": 15,
+ "words": 14,
+ "wer": 1.0714285714285714
+ },
+ "semantic": {
+ "normalized_reference": "increase the gain by 3 decibels wait 5 seconds then reduce it by one",
+ "normalized_hypothesis": "you question all those wonders of peace lots of circles and wonders of the world",
+ "errors": 15,
+ "words": 14,
+ "wer": 1.0714285714285714
+ }
+ },
+ {
+ "id": "release500_0038",
+ "category": "punctuation",
+ "text": "The result was unexpected: clear, stable, and surprisingly natural.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0038.wav",
+ "render_seconds": 86.60410415800288,
+ "audio_seconds": 5.433469387755102,
+ "rtf": 15.939006549512248,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "I will save you as an unexpected child stabbing a shop-losser right now, boy.",
+ "raw": {
+ "normalized_reference": "the result was unexpected: clear, stable, and surprisingly natural.",
+ "normalized_hypothesis": "i will save you as an unexpected child stabbing a shop-losser right now, boy.",
+ "errors": 14,
+ "words": 9,
+ "wer": 1.5555555555555556
+ },
+ "semantic": {
+ "normalized_reference": "the result was unexpected clear stable and surprisingly natural",
+ "normalized_hypothesis": "i will save you as an unexpected child stabbing a shop losser right now boy",
+ "errors": 14,
+ "words": 9,
+ "wer": 1.5555555555555556
+ }
+ },
+ {
+ "id": "release500_0039",
+ "category": "question",
+ "text": "Who told you that was the right way to do it?",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0039.wav",
+ "render_seconds": 63.40366674901452,
+ "audio_seconds": 2.9257142857142857,
+ "rtf": 21.671175158354572,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "the creature that was the one that did that",
+ "raw": {
+ "normalized_reference": "who told you that was the right way to do it?",
+ "normalized_hypothesis": "the creature that was the one that did that",
+ "errors": 8,
+ "words": 11,
+ "wer": 0.7272727272727273
+ },
+ "semantic": {
+ "normalized_reference": "who told you that was the right way to do it",
+ "normalized_hypothesis": "the creature that was the one that did that",
+ "errors": 8,
+ "words": 11,
+ "wer": 0.7272727272727273
+ }
+ },
+ {
+ "id": "release500_0040",
+ "category": "rare_words",
+ "text": "The meticulous cartographer examined an iridescent archipelago through binoculars.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0040.wav",
+ "render_seconds": 89.78799257206265,
+ "audio_seconds": 5.851428571428571,
+ "rtf": 15.344627636827113,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "Thank you for watching. I hope you enjoyed the video.",
+ "raw": {
+ "normalized_reference": "the meticulous cartographer examined an iridescent archipelago through binoculars.",
+ "normalized_hypothesis": "thank you for watching. i hope you enjoyed the video.",
+ "errors": 10,
+ "words": 9,
+ "wer": 1.1111111111111112
+ },
+ "semantic": {
+ "normalized_reference": "the meticulous cartographer examined an iridescent archipelago through binoculars",
+ "normalized_hypothesis": "thank you for watching i hope you enjoyed the video",
+ "errors": 10,
+ "words": 9,
+ "wer": 1.1111111111111112
+ }
+ },
+ {
+ "id": "release500_0041",
+ "category": "short",
+ "text": "I mean. Yeah. Kind of.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0041.wav",
+ "render_seconds": 46.79784265195485,
+ "audio_seconds": 2.2987755102040817,
+ "rtf": 20.357726295683484,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you.",
+ "raw": {
+ "normalized_reference": "i mean. yeah. kind of.",
+ "normalized_hypothesis": "i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you.",
+ "errors": 332,
+ "words": 5,
+ "wer": 66.4
+ },
+ "semantic": {
+ "normalized_reference": "i mean yeah kind of",
+ "normalized_hypothesis": "i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you",
+ "errors": 332,
+ "words": 5,
+ "wer": 66.4
+ }
+ },
+ {
+ "id": "release500_0042",
+ "category": "technical",
+ "text": "The logs are your best friend when something behaves unexpectedly.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0042.wav",
+ "render_seconds": 87.90835851500742,
+ "audio_seconds": 4.527891156462585,
+ "rtf": 19.414856823476697,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "Yes, I hope this client will send you the marks you expected.",
+ "raw": {
+ "normalized_reference": "the logs are your best friend when something behaves unexpectedly.",
+ "normalized_hypothesis": "yes, i hope this client will send you the marks you expected.",
+ "errors": 12,
+ "words": 10,
+ "wer": 1.2
+ },
+ "semantic": {
+ "normalized_reference": "the logs are your best friend when something behaves unexpectedly",
+ "normalized_hypothesis": "yes i hope this client will send you the marks you expected",
+ "errors": 12,
+ "words": 10,
+ "wer": 1.2
+ }
+ },
+ {
+ "id": "release500_0043",
+ "category": "conversational",
+ "text": "He kept talking and I kept nodding and neither of us was listening.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0043.wav",
+ "render_seconds": 50.20885578193702,
+ "audio_seconds": 4.527891156462585,
+ "rtf": 11.088794771551594,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "I can't do it. I can't do it. I can't do it. I can't do it. I can't do it. I can't do it. I can't do it. I can't do it.",
+ "raw": {
+ "normalized_reference": "he kept talking and i kept nodding and neither of us was listening.",
+ "normalized_hypothesis": "i can't do it. i can't do it. i can't do it. i can't do it. i can't do it. i can't do it. i can't do it. i can't do it.",
+ "errors": 31,
+ "words": 13,
+ "wer": 2.3846153846153846
+ },
+ "semantic": {
+ "normalized_reference": "he kept talking and i kept nodding and neither of us was listening",
+ "normalized_hypothesis": "i can not do it i can not do it i can not do it i can not do it i can not do it i can not do it i can not do it i can not do it",
+ "errors": 39,
+ "words": 13,
+ "wer": 3.0
+ }
+ },
+ {
+ "id": "release500_0044",
+ "category": "descriptive",
+ "text": "It's one of those places where everyone seems to know each other already.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0044.wav",
+ "render_seconds": 76.58786594006233,
+ "audio_seconds": 4.876190476190477,
+ "rtf": 15.706495944739343,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "Actually, there's cases wherever you choose to know your channel is at.",
+ "raw": {
+ "normalized_reference": "it's one of those places where everyone seems to know each other already.",
+ "normalized_hypothesis": "actually, there's cases wherever you choose to know your channel is at.",
+ "errors": 12,
+ "words": 13,
+ "wer": 0.9230769230769231
+ },
+ "semantic": {
+ "normalized_reference": "it is one of those places where everyone seems to know each other already",
+ "normalized_hypothesis": "actually there is cases wherever you choose to know your channel is at",
+ "errors": 13,
+ "words": 14,
+ "wer": 0.9285714285714286
+ }
+ },
+ {
+ "id": "release500_0045",
+ "category": "emotional",
+ "text": "He said he was proud of me and nobody had ever said that to me before.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0045.wav",
+ "render_seconds": 90.09963908197824,
+ "audio_seconds": 4.876190476190477,
+ "rtf": 18.47746504610882,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "You should always play with me. I know that I will never shake that toy with you.",
+ "raw": {
+ "normalized_reference": "he said he was proud of me and nobody had ever said that to me before.",
+ "normalized_hypothesis": "you should always play with me. i know that i will never shake that toy with you.",
+ "errors": 16,
+ "words": 16,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "he said he was proud of me and nobody had ever said that to me before",
+ "normalized_hypothesis": "you should always play with me i know that i will never shake that toy with you",
+ "errors": 16,
+ "words": 16,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0046",
+ "category": "homographs",
+ "text": "The minute detail took a minute to explain clearly.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0046.wav",
+ "render_seconds": 64.10040538094472,
+ "audio_seconds": 3.622312925170068,
+ "rtf": 17.695987813624686,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "Thank you.",
+ "raw": {
+ "normalized_reference": "the minute detail took a minute to explain clearly.",
+ "normalized_hypothesis": "thank you.",
+ "errors": 9,
+ "words": 9,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "the minute detail took a minute to explain clearly",
+ "normalized_hypothesis": "thank you",
+ "errors": 9,
+ "words": 9,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0047",
+ "category": "long",
+ "text": "The strange thing about really wanting something is that getting it never quite feels the way you imagined it would, and not getting it never quite feels as bad.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0047.wav",
+ "render_seconds": 159.90445923490915,
+ "audio_seconds": 9.961360544217687,
+ "rtf": 16.052471800925787,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "A strange thing about the way I look at you so far is that, coming up level one says that I have a bad dream of you. And you're coming up level one twice as bad.",
+ "raw": {
+ "normalized_reference": "the strange thing about really wanting something is that getting it never quite feels the way you imagined it would, and not getting it never quite feels as bad.",
+ "normalized_hypothesis": "a strange thing about the way i look at you so far is that, coming up level one says that i have a bad dream of you. and you're coming up level one twice as bad.",
+ "errors": 28,
+ "words": 29,
+ "wer": 0.9655172413793104
+ },
+ "semantic": {
+ "normalized_reference": "the strange thing about really wanting something is that getting it never quite feels the way you imagined it would and not getting it never quite feels as bad",
+ "normalized_hypothesis": "a strange thing about the way i look at you so far is that coming up level one says that i have a bad dream of you and you are coming up level one twice as bad",
+ "errors": 29,
+ "words": 29,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0048",
+ "category": "long_range",
+ "text": "Even though the recipe looked simple at first, preparing the sauce required patience, careful timing, and a much larger pan.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0048.wav",
+ "render_seconds": 120.39180352794938,
+ "audio_seconds": 8.289523809523809,
+ "rtf": 14.523367842870732,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "I don't know if you want to see me eat something at first, but now I'm going to show you how it tastes. I'll sit down and have a little cup of milk.",
+ "raw": {
+ "normalized_reference": "even though the recipe looked simple at first, preparing the sauce required patience, careful timing, and a much larger pan.",
+ "normalized_hypothesis": "i don't know if you want to see me eat something at first, but now i'm going to show you how it tastes. i'll sit down and have a little cup of milk.",
+ "errors": 29,
+ "words": 20,
+ "wer": 1.45
+ },
+ "semantic": {
+ "normalized_reference": "even though the recipe looked simple at 1st preparing the sauce required patience careful timing and a much larger pan",
+ "normalized_hypothesis": "i do not know if you want to see me eat something at 1st but now i am going to show you how it tastes i will sit down and have a little cup of milk",
+ "errors": 32,
+ "words": 20,
+ "wer": 1.6
+ }
+ },
+ {
+ "id": "release500_0049",
+ "category": "names_places",
+ "text": "Xiomara met Siobhan near the old library in Worcestershire.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0049.wav",
+ "render_seconds": 48.40293605194893,
+ "audio_seconds": 4.667210884353741,
+ "rtf": 10.370848297044795,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "I don't know what's happening. I don't know what's happening. I don't know what's happening. I don't know what's happening. I don't know what's happening. I don't know what's happening. I don't know what's happening. I don't know what's happening. I don't know what's happening. I don't know what's happening. I don't know what's happening. I don't know what's happening. I don't know what's happening. I don't know what's happening. I don't know what's happening. I don't know what's happening. I don't know what's happening. I don't know what's happening. I don't know what's happening. I don't know what's happening. I don't know what's happening. I don't know what's happening. I don't know what's happening. I don't know what's happening. I don't know what's happening. I don't know what's happening. I don't know what's happening. I don't know what's happening. I don't know what's happening. I don't know what's happening. I don't know what's happening. I don't know what's happening. I don't know what's happening. I don't know what's happening. I don't know what's happening. I don't know what's happening. I don't know what's happening. I don't know what's happening. I don't know what's happening. I don't know what's happening. I don't know what's happening. I don't know what's happening. I don't know what's happening. I don't know what's happening. I don't know what's happening. I don't know what's happening. I don't know what's happening. I don't know what's happening. I don't know what's happening. I don't know what's happening. I don't know what's happening. I don't know what's happening. I don't know what's happening. I don't know what's happening. I don't know what's happening. I don't know",
+ "raw": {
+ "normalized_reference": "xiomara met siobhan near the old library in worcestershire.",
+ "normalized_hypothesis": "i don't know what's happening. i don't know what's happening. i don't know what's happening. i don't know what's happening. i don't know what's happening. i don't know what's happening. i don't know what's happening. i don't know what's happening. i don't know what's happening. i don't know what's happening. i don't know what's happening. i don't know what's happening. i don't know what's happening. i don't know what's happening. i don't know what's happening. i don't know what's happening. i don't know what's happening. i don't know what's happening. i don't know what's happening. i don't know what's happening. i don't know what's happening. i don't know what's happening. i don't know what's happening. i don't know what's happening. i don't know what's happening. i don't know what's happening. i don't know what's happening. i don't know what's happening. i don't know what's happening. i don't know what's happening. i don't know what's happening. i don't know what's happening. i don't know what's happening. i don't know what's happening. i don't know what's happening. i don't know what's happening. i don't know what's happening. i don't know what's happening. i don't know what's happening. i don't know what's happening. i don't know what's happening. i don't know what's happening. i don't know what's happening. i don't know what's happening. i don't know what's happening. i don't know what's happening. i don't know what's happening. i don't know what's happening. i don't know what's happening. i don't know what's happening. i don't know what's happening. i don't know what's happening. i don't know what's happening. i don't know what's happening. i don't know what's happening. i don't know",
+ "errors": 278,
+ "words": 9,
+ "wer": 30.88888888888889
+ },
+ "semantic": {
+ "normalized_reference": "xiomara met siobhan near the old library in worcestershire",
+ "normalized_hypothesis": "i do not know what is happening i do not know what is happening i do not know what is happening i do not know what is happening i do not know what is happening i do not know what is happening i do not know what is happening i do not know what is happening i do not know what is happening i do not know what is happening i do not know what is happening i do not know what is happening i do not know what is happening i do not know what is happening i do not know what is happening i do not know what is happening i do not know what is happening i do not know what is happening i do not know what is happening i do not know what is happening i do not know what is happening i do not know what is happening i do not know what is happening i do not know what is happening i do not know what is happening i do not know what is happening i do not know what is happening i do not know what is happening i do not know what is happening i do not know what is happening i do not know what is happening i do not know what is happening i do not know what is happening i do not know what is happening i do not know what is happening i do not know what is happening i do not know what is happening i do not know what is happening i do not know what is happening i do not know what is happening i do not know what is happening i do not know what is happening i do not know what is happening i do not know what is happening i do not know what is happening i do not know what is happening i do not know what is happening i do not know what is happening i do not know what is happening i do not know what is happening i do not know what is happening i do not know what is happening i do not know what is happening i do not know what is happening i do not know what is happening i do not know",
+ "errors": 389,
+ "words": 9,
+ "wer": 43.22222222222222
+ }
+ },
+ {
+ "id": "release500_0050",
+ "category": "narrative",
+ "text": "She walked into the room and everyone went quiet, not because she was loud, but because she wasn't.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0050.wav",
+ "render_seconds": 103.90908648003824,
+ "audio_seconds": 6.408707482993197,
+ "rtf": 16.21373525875257,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "She looked into the world and had a long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long,",
+ "raw": {
+ "normalized_reference": "she walked into the room and everyone went quiet, not because she was loud, but because she wasn't.",
+ "normalized_hypothesis": "she looked into the world and had a long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long, long,",
+ "errors": 222,
+ "words": 18,
+ "wer": 12.333333333333334
+ },
+ "semantic": {
+ "normalized_reference": "she walked into the room and everyone went quiet not because she was loud but because she was not",
+ "normalized_hypothesis": "she looked into the world and had a long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long long",
+ "errors": 222,
+ "words": 19,
+ "wer": 11.68421052631579
+ }
+ },
+ {
+ "id": "release500_0051",
+ "category": "numbers",
+ "text": "At exactly seven forty-five, Morgan placed eighteen dollars and fifty cents beside the notebook.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0051.wav",
+ "render_seconds": 110.0988628109917,
+ "audio_seconds": 6.826666666666667,
+ "rtf": 16.127763107078863,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "On March 7th, 45th, we will cast an actual bill of some $50,000 a year in a bid.",
+ "raw": {
+ "normalized_reference": "at exactly seven forty-five, morgan placed eighteen dollars and fifty cents beside the notebook.",
+ "normalized_hypothesis": "on march 7th, 45th, we will cast an actual bill of some $50,000 a year in a bid.",
+ "errors": 19,
+ "words": 15,
+ "wer": 1.2666666666666666
+ },
+ "semantic": {
+ "normalized_reference": "at exactly 745 morgan placed $18.50 beside the notebook",
+ "normalized_hypothesis": "on march 7th 45th we will cast an actual bill of some $50000 a year in a bid",
+ "errors": 18,
+ "words": 10,
+ "wer": 1.8
+ }
+ },
+ {
+ "id": "release500_0052",
+ "category": "punctuation",
+ "text": "Does punctuation work? Yes: commas, semicolons, and periods should create sensible boundaries.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0052.wav",
+ "render_seconds": 105.61094658903312,
+ "audio_seconds": 7.453605442176871,
+ "rtf": 14.169108816979291,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "There's a place you have to work. Yes. Purpose. So that's what it is. I call action. Action should be balanced.",
+ "raw": {
+ "normalized_reference": "does punctuation work? yes: commas, semicolons, and periods should create sensible boundaries.",
+ "normalized_hypothesis": "there's a place you have to work. yes. purpose. so that's what it is. i call action. action should be balanced.",
+ "errors": 19,
+ "words": 12,
+ "wer": 1.5833333333333333
+ },
+ "semantic": {
+ "normalized_reference": "does punctuation work yes commas semicolons and periods should create sensible boundaries",
+ "normalized_hypothesis": "there is a place you have to work yes purpose so that is what it is i call action action should be balanced",
+ "errors": 21,
+ "words": 12,
+ "wer": 1.75
+ }
+ },
+ {
+ "id": "release500_0053",
+ "category": "question",
+ "text": "Have you ever just stopped and thought about how strange all of this is?",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0053.wav",
+ "render_seconds": 90.99906517192721,
+ "audio_seconds": 4.876190476190477,
+ "rtf": 18.661917662211636,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "Oh, I love it. I love it. I love it. I love it. I love it. I love it. I love it. I love it. I love it. I love it. I love it. I love it. I love it. I love it. I love it. I love it. I love it. I love it. I love it. I love it. I love it. I love it. I love it. I love it. I love it. I love it. I love it. I love it. I love it. I love it. I love it. I love it. I love it. I love it. I love it. I love it. I love it. I love it. I love it. I love it. I love it. I love it. I love it. I love it. I love it. I love it. I love it. I love it. I love it. I love it. I love it. I love it. I love it. I love it. I love it. I love it. I love it. I love it. I love it. I love it. I love it. I love it. I love it. I love it. I love it. I love it. I love it. I love it. I love it. I love it. I love it. I love it. I love it. I love it. I love it. I love it. I love it. I love it. I love it. I love it. I love it. I love it. I love it. I love it. I love it. I love it. I love it. I love it. I love it. I love it. I love it. I love it. I love it. I love it. I love it. I love it. I love it. I love it. I love it. I love it. I love it. I love it. I love it. I love it. I love it. I love it. I love it. I love it. I love it. I love it. I love",
+ "raw": {
+ "normalized_reference": "have you ever just stopped and thought about how strange all of this is?",
+ "normalized_hypothesis": "oh, i love it. i love it. i love it. i love it. i love it. i love it. i love it. i love it. i love it. i love it. i love it. i love it. i love it. i love it. i love it. i love it. i love it. i love it. i love it. i love it. i love it. i love it. i love it. i love it. i love it. i love it. i love it. i love it. i love it. i love it. i love it. i love it. i love it. i love it. i love it. i love it. i love it. i love it. i love it. i love it. i love it. i love it. i love it. i love it. i love it. i love it. i love it. i love it. i love it. i love it. i love it. i love it. i love it. i love it. i love it. i love it. i love it. i love it. i love it. i love it. i love it. i love it. i love it. i love it. i love it. i love it. i love it. i love it. i love it. i love it. i love it. i love it. i love it. i love it. i love it. i love it. i love it. i love it. i love it. i love it. i love it. i love it. i love it. i love it. i love it. i love it. i love it. i love it. i love it. i love it. i love it. i love it. i love it. i love it. i love it. i love it. i love it. i love it. i love it. i love it. i love it. i love it. i love it. i love it. i love it. i love it. i love it. i love it. i love it. i love it. i love",
+ "errors": 333,
+ "words": 14,
+ "wer": 23.785714285714285
+ },
+ "semantic": {
+ "normalized_reference": "have you ever just stopped and thought about how strange all of this is",
+ "normalized_hypothesis": "0 i love it i love it i love it i love it i love it i love it i love it i love it i love it i love it i love it i love it i love it i love it i love it i love it i love it i love it i love it i love it i love it i love it i love it i love it i love it i love it i love it i love it i love it i love it i love it i love it i love it i love it i love it i love it i love it i love it i love it i love it i love it i love it i love it i love it i love it i love it i love it i love it i love it i love it i love it i love it i love it i love it i love it i love it i love it i love it i love it i love it i love it i love it i love it i love it i love it i love it i love it i love it i love it i love it i love it i love it i love it i love it i love it i love it i love it i love it i love it i love it i love it i love it i love it i love it i love it i love it i love it i love it i love it i love it i love it i love it i love it i love it i love it i love it i love it i love it i love it i love it i love it i love it i love it i love it i love it i love it i love it i love it i love it i love it i love",
+ "errors": 333,
+ "words": 14,
+ "wer": 23.785714285714285
+ }
+ },
+ {
+ "id": "release500_0054",
+ "category": "rare_words",
+ "text": "Her idiosyncratic explanation sounded plausible but remained fundamentally ambiguous.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0054.wav",
+ "render_seconds": 114.90803999104537,
+ "audio_seconds": 6.339047619047619,
+ "rtf": 18.127019529837384,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "I ask you, Father God, to bless you so that Christ, the Lord, may accept our prayer and bless you.",
+ "raw": {
+ "normalized_reference": "her idiosyncratic explanation sounded plausible but remained fundamentally ambiguous.",
+ "normalized_hypothesis": "i ask you, father god, to bless you so that christ, the lord, may accept our prayer and bless you.",
+ "errors": 20,
+ "words": 9,
+ "wer": 2.2222222222222223
+ },
+ "semantic": {
+ "normalized_reference": "her idiosyncratic explanation sounded plausible but remained fundamentally ambiguous",
+ "normalized_hypothesis": "i ask you father god to bless you so that christ the lord may accept our prayer and bless you",
+ "errors": 20,
+ "words": 9,
+ "wer": 2.2222222222222223
+ }
+ },
+ {
+ "id": "release500_0055",
+ "category": "short",
+ "text": "That tracks.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0055.wav",
+ "render_seconds": 23.48697709396947,
+ "audio_seconds": 1.4628571428571429,
+ "rtf": 16.055550747830694,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "I love you.",
+ "raw": {
+ "normalized_reference": "that tracks.",
+ "normalized_hypothesis": "i love you.",
+ "errors": 3,
+ "words": 2,
+ "wer": 1.5
+ },
+ "semantic": {
+ "normalized_reference": "that tracks",
+ "normalized_hypothesis": "i love you",
+ "errors": 3,
+ "words": 2,
+ "wer": 1.5
+ }
+ },
+ {
+ "id": "release500_0056",
+ "category": "technical",
+ "text": "The key thing to remember is that order matters here more than you'd think.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0056.wav",
+ "render_seconds": 93.39177715405822,
+ "audio_seconds": 4.597551020408163,
+ "rtf": 20.313374824879496,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you.",
+ "raw": {
+ "normalized_reference": "the key thing to remember is that order matters here more than you'd think.",
+ "normalized_hypothesis": "thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you.",
+ "errors": 296,
+ "words": 14,
+ "wer": 21.142857142857142
+ },
+ "semantic": {
+ "normalized_reference": "the key thing to remember is that order matters here more than you would think",
+ "normalized_hypothesis": "thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you",
+ "errors": 295,
+ "words": 15,
+ "wer": 19.666666666666668
+ }
+ },
+ {
+ "id": "release500_0057",
+ "category": "conversational",
+ "text": "I would not say that again in this room if I were you.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0057.wav",
+ "render_seconds": 77.50085372989997,
+ "audio_seconds": 3.7616326530612243,
+ "rtf": 20.602983033664817,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "Thank you for watching. Please subscribe to my channel. Thank you for watching. Please subscribe to my channel. Thank you for watching. Please subscribe to my channel. Thank you for watching. Please subscribe to my channel. Thank you for watching. Please subscribe to my channel. Thank you for watching. Please subscribe to my channel. Thank you for watching. Please subscribe to my channel. Thank you for watching. Please subscribe to my channel. Thank you for watching. Please subscribe to my channel. Thank you for watching. Please subscribe to my channel. Thank you for watching. Please subscribe to my channel. Thank you for watching. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel.",
+ "raw": {
+ "normalized_reference": "i would not say that again in this room if i were you.",
+ "normalized_hypothesis": "thank you for watching. please subscribe to my channel. thank you for watching. please subscribe to my channel. thank you for watching. please subscribe to my channel. thank you for watching. please subscribe to my channel. thank you for watching. please subscribe to my channel. thank you for watching. please subscribe to my channel. thank you for watching. please subscribe to my channel. thank you for watching. please subscribe to my channel. thank you for watching. please subscribe to my channel. thank you for watching. please subscribe to my channel. thank you for watching. please subscribe to my channel. thank you for watching. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel.",
+ "errors": 367,
+ "words": 13,
+ "wer": 28.23076923076923
+ },
+ "semantic": {
+ "normalized_reference": "i would not say that again in this room if i were you",
+ "normalized_hypothesis": "thank you for watching please subscribe to my channel thank you for watching please subscribe to my channel thank you for watching please subscribe to my channel thank you for watching please subscribe to my channel thank you for watching please subscribe to my channel thank you for watching please subscribe to my channel thank you for watching please subscribe to my channel thank you for watching please subscribe to my channel thank you for watching please subscribe to my channel thank you for watching please subscribe to my channel thank you for watching please subscribe to my channel thank you for watching please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel",
+ "errors": 367,
+ "words": 13,
+ "wer": 28.23076923076923
+ }
+ },
+ {
+ "id": "release500_0058",
+ "category": "descriptive",
+ "text": "The room was filled with the quiet that means something is about to change.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0058.wav",
+ "render_seconds": 88.89165852591395,
+ "audio_seconds": 4.73687074829932,
+ "rtf": 18.76590332506513,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "I will show you what's about to come.",
+ "raw": {
+ "normalized_reference": "the room was filled with the quiet that means something is about to change.",
+ "normalized_hypothesis": "i will show you what's about to come.",
+ "errors": 12,
+ "words": 14,
+ "wer": 0.8571428571428571
+ },
+ "semantic": {
+ "normalized_reference": "the room was filled with the quiet that means something is about to change",
+ "normalized_hypothesis": "i will show you what is about to come",
+ "errors": 11,
+ "words": 14,
+ "wer": 0.7857142857142857
+ }
+ },
+ {
+ "id": "release500_0059",
+ "category": "emotional",
+ "text": "I didn't expect to feel that way about it, but I did.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0059.wav",
+ "render_seconds": 32.98780106299091,
+ "audio_seconds": 3.4829931972789114,
+ "rtf": 9.471106945819656,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "I'm not scared to say goodbye, baby. I love you.",
+ "raw": {
+ "normalized_reference": "i didn't expect to feel that way about it, but i did.",
+ "normalized_hypothesis": "i'm not scared to say goodbye, baby. i love you.",
+ "errors": 11,
+ "words": 12,
+ "wer": 0.9166666666666666
+ },
+ "semantic": {
+ "normalized_reference": "i did not expect to feel that way about it but i did",
+ "normalized_hypothesis": "i am not scared to say goodbye baby i love you",
+ "errors": 10,
+ "words": 13,
+ "wer": 0.7692307692307693
+ }
+ },
+ {
+ "id": "release500_0060",
+ "category": "homographs",
+ "text": "I object to the strange object that was left near the entrance.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0060.wav",
+ "render_seconds": 75.39093819109257,
+ "audio_seconds": 4.1099319727891155,
+ "rtf": 18.343597580261203,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "Welcome to the Skyhooker. I'm your host, Ron Harkins.",
+ "raw": {
+ "normalized_reference": "i object to the strange object that was left near the entrance.",
+ "normalized_hypothesis": "welcome to the skyhooker. i'm your host, ron harkins.",
+ "errors": 10,
+ "words": 12,
+ "wer": 0.8333333333333334
+ },
+ "semantic": {
+ "normalized_reference": "i object to the strange object that was left near the entrance",
+ "normalized_hypothesis": "welcome to the skyhooker i am your host ron harkins",
+ "errors": 10,
+ "words": 12,
+ "wer": 0.8333333333333334
+ }
+ },
+ {
+ "id": "release500_0061",
+ "category": "long",
+ "text": "I've learned more from the things that went wrong than from the things that worked, which is a frustrating way to have to learn things but apparently the only way that stuck.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0061.wav",
+ "render_seconds": 106.90791040903423,
+ "audio_seconds": 10.65795918367347,
+ "rtf": 10.030805013102553,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "If you want to know some of the things that work for you, then some of the things that work for you should first come out of the abyss to your face. With the power of your eye, you will understand.",
+ "raw": {
+ "normalized_reference": "i've learned more from the things that went wrong than from the things that worked, which is a frustrating way to have to learn things but apparently the only way that stuck.",
+ "normalized_hypothesis": "if you want to know some of the things that work for you, then some of the things that work for you should first come out of the abyss to your face. with the power of your eye, you will understand.",
+ "errors": 34,
+ "words": 32,
+ "wer": 1.0625
+ },
+ "semantic": {
+ "normalized_reference": "i have learned more from the things that went wrong than from the things that worked which is a frustrating way to have to learn things but apparently the only way that stuck",
+ "normalized_hypothesis": "if you want to know some of the things that work for you then some of the things that work for you should 1st come out of the abyss to your face with the power of your eye you will understand",
+ "errors": 34,
+ "words": 33,
+ "wer": 1.0303030303030303
+ }
+ },
+ {
+ "id": "release500_0062",
+ "category": "long_range",
+ "text": "Because every nearby hotel was full, the exhausted hikers followed the narrow road until they found a farmhouse with one empty room.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0062.wav",
+ "render_seconds": 118.20163273497019,
+ "audio_seconds": 8.916462585034013,
+ "rtf": 13.256561288483137,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "It's a hard one, Robert, to say this to you. You first exactly see it in how we work. We take our first steps with the empty words.",
+ "raw": {
+ "normalized_reference": "because every nearby hotel was full, the exhausted hikers followed the narrow road until they found a farmhouse with one empty room.",
+ "normalized_hypothesis": "it's a hard one, robert, to say this to you. you first exactly see it in how we work. we take our first steps with the empty words.",
+ "errors": 26,
+ "words": 22,
+ "wer": 1.1818181818181819
+ },
+ "semantic": {
+ "normalized_reference": "because every nearby hotel was full the exhausted hikers followed the narrow road until they found a farmhouse with one empty room",
+ "normalized_hypothesis": "it is a hard one robert to say this to you you 1st exactly see it in how we work we take our 1st steps with the empty words",
+ "errors": 27,
+ "words": 22,
+ "wer": 1.2272727272727273
+ }
+ },
+ {
+ "id": "release500_0063",
+ "category": "names_places",
+ "text": "The itinerary sends Beatrice from Albuquerque to Des Moines and then to Boise.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0063.wav",
+ "render_seconds": 92.99552207894158,
+ "audio_seconds": 5.363809523809524,
+ "rtf": 17.337588455768582,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "I don't know what I should do. I don't know what I should do. I don't know what I should do. I don't know what I should do. I don't know what I should do. I don't know what I should do. I don't know what I should do. I don't know what I should do. I don't know what I should do. I don't know what I should do. I don't know what I should do. I don't know what I should do. I don't know what I should do. I don't know what I should do. I don't know what I should do. I don't know what I should do. I don't know what I should do. I don't know what I should do. I don't know what I should do. I don't know what I should do. I don't know what I should do. I don't know what I should do. I don't know what I should do. I don't know what I should do. I don't know what I should do. I don't know what I should do. I don't know what I should do. I don't know what I should do. I don't know what I should do. I don't know what I should do. I don't know what I should do. I don't know what I should do. I don't know what I should do. I don't know what I should do. I don't know what I should do. I don't know what I should do. I don't know what I should do. I don't know what I should do. I don't know what I should do. I don't know what I should do. I don't know what I should do. I don't know what I should do. I don't know what I should do. I don't know what I should do. I don't know what I should do. I don't know what I should do. I don't know what I should do. I don't know what I should do. I don't know what I should do. I don't",
+ "raw": {
+ "normalized_reference": "the itinerary sends beatrice from albuquerque to des moines and then to boise.",
+ "normalized_hypothesis": "i don't know what i should do. i don't know what i should do. i don't know what i should do. i don't know what i should do. i don't know what i should do. i don't know what i should do. i don't know what i should do. i don't know what i should do. i don't know what i should do. i don't know what i should do. i don't know what i should do. i don't know what i should do. i don't know what i should do. i don't know what i should do. i don't know what i should do. i don't know what i should do. i don't know what i should do. i don't know what i should do. i don't know what i should do. i don't know what i should do. i don't know what i should do. i don't know what i should do. i don't know what i should do. i don't know what i should do. i don't know what i should do. i don't know what i should do. i don't know what i should do. i don't know what i should do. i don't know what i should do. i don't know what i should do. i don't know what i should do. i don't know what i should do. i don't know what i should do. i don't know what i should do. i don't know what i should do. i don't know what i should do. i don't know what i should do. i don't know what i should do. i don't know what i should do. i don't know what i should do. i don't know what i should do. i don't know what i should do. i don't know what i should do. i don't know what i should do. i don't know what i should do. i don't know what i should do. i don't know what i should do. i don't know what i should do. i don't know what i should do. i don't",
+ "errors": 345,
+ "words": 13,
+ "wer": 26.53846153846154
+ },
+ "semantic": {
+ "normalized_reference": "the itinerary sends beatrice from albuquerque to des moines and then to boise",
+ "normalized_hypothesis": "i do not know what i should do i do not know what i should do i do not know what i should do i do not know what i should do i do not know what i should do i do not know what i should do i do not know what i should do i do not know what i should do i do not know what i should do i do not know what i should do i do not know what i should do i do not know what i should do i do not know what i should do i do not know what i should do i do not know what i should do i do not know what i should do i do not know what i should do i do not know what i should do i do not know what i should do i do not know what i should do i do not know what i should do i do not know what i should do i do not know what i should do i do not know what i should do i do not know what i should do i do not know what i should do i do not know what i should do i do not know what i should do i do not know what i should do i do not know what i should do i do not know what i should do i do not know what i should do i do not know what i should do i do not know what i should do i do not know what i should do i do not know what i should do i do not know what i should do i do not know what i should do i do not know what i should do i do not know what i should do i do not know what i should do i do not know what i should do i do not know what i should do i do not know what i should do i do not know what i should do i do not know what i should do i do not know what i should do i do not know what i should do i do not know what i should do i do not",
+ "errors": 395,
+ "words": 13,
+ "wer": 30.384615384615383
+ }
+ },
+ {
+ "id": "release500_0064",
+ "category": "narrative",
+ "text": "We had been driving for six hours when we finally saw the sign for the town.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0064.wav",
+ "render_seconds": 83.19974137598183,
+ "audio_seconds": 5.085170068027211,
+ "rtf": 16.361250511383414,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "It's time to call it a day. I'm going to call it a day. I'm going to call it a day. I'm going to call it a day. I'm going to call it a day. I'm going to call it a day. I'm going to call it a day. I'm going to call it a day. I'm going to call it a day. I'm going to call it a day. I'm going to call it a day. I'm going to call it a day. I'm going to call it a day. I'm going to call it a day. I'm going to call it a day. I'm going to call it a day. I'm going to call it a day. I'm going to call it a day. I'm going to call it a day. I'm going to call it a day. I'm going to call it a day. I'm going to call it a day. I'm going to call it a day. I'm going to call it a day. I'm going to call it a day. I'm going to call it a day. I'm going to call it a day. I'm going to call it a day. I'm going to call it a day. I'm going to call it a day. I'm going to call it a day. I'm going to call it a day. I'm going to call it a day. I'm going to call it a day. I'm going to call it a day. I'm going to call it a day. I'm going to call it a day. I'm going to call it a day. I'm going to call it a day. I'm going to call it a day. I'm going to call it a day. I'm going to call it a day. I'm going to call it a day. I'm going to call it a day. I'm going to call it a day. I'm going to call it a day. I'm going to call it a day. I'm going to call it a day. I'm going to call it a day. I'm going",
+ "raw": {
+ "normalized_reference": "we had been driving for six hours when we finally saw the sign for the town.",
+ "normalized_hypothesis": "it's time to call it a day. i'm going to call it a day. i'm going to call it a day. i'm going to call it a day. i'm going to call it a day. i'm going to call it a day. i'm going to call it a day. i'm going to call it a day. i'm going to call it a day. i'm going to call it a day. i'm going to call it a day. i'm going to call it a day. i'm going to call it a day. i'm going to call it a day. i'm going to call it a day. i'm going to call it a day. i'm going to call it a day. i'm going to call it a day. i'm going to call it a day. i'm going to call it a day. i'm going to call it a day. i'm going to call it a day. i'm going to call it a day. i'm going to call it a day. i'm going to call it a day. i'm going to call it a day. i'm going to call it a day. i'm going to call it a day. i'm going to call it a day. i'm going to call it a day. i'm going to call it a day. i'm going to call it a day. i'm going to call it a day. i'm going to call it a day. i'm going to call it a day. i'm going to call it a day. i'm going to call it a day. i'm going to call it a day. i'm going to call it a day. i'm going to call it a day. i'm going to call it a day. i'm going to call it a day. i'm going to call it a day. i'm going to call it a day. i'm going to call it a day. i'm going to call it a day. i'm going to call it a day. i'm going to call it a day. i'm going to call it a day. i'm going",
+ "errors": 345,
+ "words": 16,
+ "wer": 21.5625
+ },
+ "semantic": {
+ "normalized_reference": "we had been driving for 6 hours when we finally saw the sign for the town",
+ "normalized_hypothesis": "it is time to call it a day i am going to call it a day i am going to call it a day i am going to call it a day i am going to call it a day i am going to call it a day i am going to call it a day i am going to call it a day i am going to call it a day i am going to call it a day i am going to call it a day i am going to call it a day i am going to call it a day i am going to call it a day i am going to call it a day i am going to call it a day i am going to call it a day i am going to call it a day i am going to call it a day i am going to call it a day i am going to call it a day i am going to call it a day i am going to call it a day i am going to call it a day i am going to call it a day i am going to call it a day i am going to call it a day i am going to call it a day i am going to call it a day i am going to call it a day i am going to call it a day i am going to call it a day i am going to call it a day i am going to call it a day i am going to call it a day i am going to call it a day i am going to call it a day i am going to call it a day i am going to call it a day i am going to call it a day i am going to call it a day i am going to call it a day i am going to call it a day i am going to call it a day i am going to call it a day i am going to call it a day i am going to call it a day i am going to call it a day i am going to call it a day i am going",
+ "errors": 395,
+ "words": 16,
+ "wer": 24.6875
+ }
+ },
+ {
+ "id": "release500_0065",
+ "category": "numbers",
+ "text": "Call extension zero four nine after eight thirty in the morning.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0065.wav",
+ "render_seconds": 62.2007240439998,
+ "audio_seconds": 4.458231292517007,
+ "rtf": 13.951883597515824,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "The extension sounds too loud. I have to add a new one.",
+ "raw": {
+ "normalized_reference": "call extension zero four nine after eight thirty in the morning.",
+ "normalized_hypothesis": "the extension sounds too loud. i have to add a new one.",
+ "errors": 11,
+ "words": 11,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "call extension 049 after 830 in the morning",
+ "normalized_hypothesis": "the extension sounds too loud i have to add a new one",
+ "errors": 11,
+ "words": 8,
+ "wer": 1.375
+ }
+ },
+ {
+ "id": "release500_0066",
+ "category": "punctuation",
+ "text": "Stop - listen carefully - then answer the question without guessing.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0066.wav",
+ "render_seconds": 83.50207139598206,
+ "audio_seconds": 4.597551020408163,
+ "rtf": 18.162293583110444,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "Stop, Russell Kelsey. I'm not trying to be nice to you, Russell.",
+ "raw": {
+ "normalized_reference": "stop - listen carefully - then answer the question without guessing.",
+ "normalized_hypothesis": "stop, russell kelsey. i'm not trying to be nice to you, russell.",
+ "errors": 11,
+ "words": 9,
+ "wer": 1.2222222222222223
+ },
+ "semantic": {
+ "normalized_reference": "stop listen carefully then answer the question without guessing",
+ "normalized_hypothesis": "stop russell kelsey i am not trying to be nice to you russell",
+ "errors": 12,
+ "words": 9,
+ "wer": 1.3333333333333333
+ }
+ },
+ {
+ "id": "release500_0067",
+ "category": "question",
+ "text": "Does it matter at this point, honestly?",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0067.wav",
+ "render_seconds": 25.000298880971968,
+ "audio_seconds": 2.716734693877551,
+ "rtf": 9.202333572473155,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "This video was made possible in part by...",
+ "raw": {
+ "normalized_reference": "does it matter at this point, honestly?",
+ "normalized_hypothesis": "this video was made possible in part by...",
+ "errors": 8,
+ "words": 7,
+ "wer": 1.1428571428571428
+ },
+ "semantic": {
+ "normalized_reference": "does it matter at this point honestly",
+ "normalized_hypothesis": "this video was made possible in part by .",
+ "errors": 8,
+ "words": 7,
+ "wer": 1.1428571428571428
+ }
+ },
+ {
+ "id": "release500_0068",
+ "category": "rare_words",
+ "text": "An entrepreneurial meteorologist described the unprecedented atmospheric phenomenon.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0068.wav",
+ "render_seconds": 86.80405418702867,
+ "audio_seconds": 5.851428571428571,
+ "rtf": 14.834677229228532,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you",
+ "raw": {
+ "normalized_reference": "an entrepreneurial meteorologist described the unprecedented atmospheric phenomenon.",
+ "normalized_hypothesis": "i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you",
+ "errors": 381,
+ "words": 8,
+ "wer": 47.625
+ },
+ "semantic": {
+ "normalized_reference": "an entrepreneurial meteorologist described the unprecedented atmospheric phenomenon",
+ "normalized_hypothesis": "i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you",
+ "errors": 381,
+ "words": 8,
+ "wer": 47.625
+ }
+ },
+ {
+ "id": "release500_0069",
+ "category": "short",
+ "text": "That's not what I expected.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0069.wav",
+ "render_seconds": 42.80378484702669,
+ "audio_seconds": 2.089795918367347,
+ "rtf": 20.482279858440503,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "That's what I'm really expecting.",
+ "raw": {
+ "normalized_reference": "that's not what i expected.",
+ "normalized_hypothesis": "that's what i'm really expecting.",
+ "errors": 4,
+ "words": 5,
+ "wer": 0.8
+ },
+ "semantic": {
+ "normalized_reference": "that is not what i expected",
+ "normalized_hypothesis": "that is what i am really expecting",
+ "errors": 4,
+ "words": 6,
+ "wer": 0.6666666666666666
+ }
+ },
+ {
+ "id": "release500_0070",
+ "category": "technical",
+ "text": "There's a known issue with that version that causes failures under load.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0070.wav",
+ "render_seconds": 83.70731660805177,
+ "audio_seconds": 4.806530612244898,
+ "rtf": 17.415329966858597,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "I'll see you in the next video.",
+ "raw": {
+ "normalized_reference": "there's a known issue with that version that causes failures under load.",
+ "normalized_hypothesis": "i'll see you in the next video.",
+ "errors": 12,
+ "words": 12,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "there is a known issue with that version that causes failures under load",
+ "normalized_hypothesis": "i will see you in the next video",
+ "errors": 13,
+ "words": 13,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0071",
+ "category": "conversational",
+ "text": "Don't look now but I think that's him over there.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0071.wav",
+ "render_seconds": 71.40638228901662,
+ "audio_seconds": 2.8560544217687074,
+ "rtf": 25.00175828050077,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "Don't look now, but I feel cancer over here.",
+ "raw": {
+ "normalized_reference": "don't look now but i think that's him over there.",
+ "normalized_hypothesis": "don't look now, but i feel cancer over here.",
+ "errors": 4,
+ "words": 10,
+ "wer": 0.4
+ },
+ "semantic": {
+ "normalized_reference": "do not look now but i think that is him over there",
+ "normalized_hypothesis": "do not look now but i feel cancer over here",
+ "errors": 5,
+ "words": 12,
+ "wer": 0.4166666666666667
+ }
+ },
+ {
+ "id": "release500_0072",
+ "category": "descriptive",
+ "text": "The city at night looks completely different from how it feels during the day.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0072.wav",
+ "render_seconds": 76.19589207996614,
+ "audio_seconds": 5.154829931972789,
+ "rtf": 14.781456049085493,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "Special thanks to all of you for watching this video.",
+ "raw": {
+ "normalized_reference": "the city at night looks completely different from how it feels during the day.",
+ "normalized_hypothesis": "special thanks to all of you for watching this video.",
+ "errors": 14,
+ "words": 14,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "the city at night looks completely different from how it feels during the day",
+ "normalized_hypothesis": "special thanks to all of you for watching this video",
+ "errors": 14,
+ "words": 14,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0073",
+ "category": "emotional",
+ "text": "It's the kind of happy that makes you nervous because you know it can't last.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0073.wav",
+ "render_seconds": 43.68924387602601,
+ "audio_seconds": 4.876190476190477,
+ "rtf": 8.959708216763147,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "That's the end of the video. Thank you for watching. I hope you enjoyed it. I'll see you in the next video.",
+ "raw": {
+ "normalized_reference": "it's the kind of happy that makes you nervous because you know it can't last.",
+ "normalized_hypothesis": "that's the end of the video. thank you for watching. i hope you enjoyed it. i'll see you in the next video.",
+ "errors": 17,
+ "words": 15,
+ "wer": 1.1333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "it is the kind of happy that makes you nervous because you know it can not last",
+ "normalized_hypothesis": "that is the end of the video thank you for watching i hope you enjoyed it i will see you in the next video",
+ "errors": 18,
+ "words": 17,
+ "wer": 1.0588235294117647
+ }
+ },
+ {
+ "id": "release500_0074",
+ "category": "homographs",
+ "text": "The dove dove beneath the branch while the wind began to wind the loose cable.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0074.wav",
+ "render_seconds": 98.30428628996015,
+ "audio_seconds": 5.085170068027211,
+ "rtf": 19.331563148309264,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "Stickers go beneath the box. Make the one at the top of one of your scraps.",
+ "raw": {
+ "normalized_reference": "the dove dove beneath the branch while the wind began to wind the loose cable.",
+ "normalized_hypothesis": "stickers go beneath the box. make the one at the top of one of your scraps.",
+ "errors": 14,
+ "words": 15,
+ "wer": 0.9333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "the dove dove beneath the branch while the wind began to wind the loose cable",
+ "normalized_hypothesis": "stickers go beneath the box make the one at the top of one of your scraps",
+ "errors": 14,
+ "words": 15,
+ "wer": 0.9333333333333333
+ }
+ },
+ {
+ "id": "release500_0075",
+ "category": "long",
+ "text": "I've thought a lot about what I would do differently if I could go back, and the honest answer is probably not as much as I'd like to think.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0075.wav",
+ "render_seconds": 117.79970723902807,
+ "audio_seconds": 8.70748299319728,
+ "rtf": 13.528560128232128,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "That's where I live, my friend. Really, I want to tell you something, so I can tell you about it. I love you, my friend, so much. Thank you, my friend, as much as I'd like to say it.",
+ "raw": {
+ "normalized_reference": "i've thought a lot about what i would do differently if i could go back, and the honest answer is probably not as much as i'd like to think.",
+ "normalized_hypothesis": "that's where i live, my friend. really, i want to tell you something, so i can tell you about it. i love you, my friend, so much. thank you, my friend, as much as i'd like to say it.",
+ "errors": 31,
+ "words": 29,
+ "wer": 1.0689655172413792
+ },
+ "semantic": {
+ "normalized_reference": "i have thought a lot about what i would do differently if i could go back and the honest answer is probably not as much as i would like to think",
+ "normalized_hypothesis": "that is where i live my friend really i want to tell you something so i can tell you about it i love you my friend so much thank you my friend as much as i would like to say it",
+ "errors": 31,
+ "words": 31,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0076",
+ "category": "long_range",
+ "text": "Although the weather changed without warning, the musicians finished packing before anyone opened the enormous glass door.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0076.wav",
+ "render_seconds": 114.79581492196303,
+ "audio_seconds": 7.801904761904762,
+ "rtf": 14.713819051277,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "It is the light of truth that I am looking for. I shall see you for this walk. As long as you are with me, I will always bless the Lord.",
+ "raw": {
+ "normalized_reference": "although the weather changed without warning, the musicians finished packing before anyone opened the enormous glass door.",
+ "normalized_hypothesis": "it is the light of truth that i am looking for. i shall see you for this walk. as long as you are with me, i will always bless the lord.",
+ "errors": 30,
+ "words": 17,
+ "wer": 1.7647058823529411
+ },
+ "semantic": {
+ "normalized_reference": "although the weather changed without warning the musicians finished packing before anyone opened the enormous glass door",
+ "normalized_hypothesis": "it is the light of truth that i am looking for i shall see you for this walk as long as you are with me i will always bless the lord",
+ "errors": 30,
+ "words": 17,
+ "wer": 1.7647058823529411
+ }
+ },
+ {
+ "id": "release500_0077",
+ "category": "names_places",
+ "text": "Saskatchewan looked unusually bright beneath the pale February sunrise.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0077.wav",
+ "render_seconds": 88.89546964899637,
+ "audio_seconds": 5.154829931972789,
+ "rtf": 17.245082926523526,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "Suspect may be in a slumber, but not in a simple slumber.",
+ "raw": {
+ "normalized_reference": "saskatchewan looked unusually bright beneath the pale february sunrise.",
+ "normalized_hypothesis": "suspect may be in a slumber, but not in a simple slumber.",
+ "errors": 12,
+ "words": 9,
+ "wer": 1.3333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "saskatchewan looked unusually bright beneath the pale february sunrise",
+ "normalized_hypothesis": "suspect may be in a slumber but not in a simple slumber",
+ "errors": 12,
+ "words": 9,
+ "wer": 1.3333333333333333
+ }
+ },
+ {
+ "id": "release500_0078",
+ "category": "narrative",
+ "text": "She handled it with a grace I genuinely did not expect.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0078.wav",
+ "render_seconds": 65.6038538919529,
+ "audio_seconds": 3.900952380952381,
+ "rtf": 16.817394186169565,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "I'm happy that you're back home, sir. I'm trying to work on a couple of books, sir.",
+ "raw": {
+ "normalized_reference": "she handled it with a grace i genuinely did not expect.",
+ "normalized_hypothesis": "i'm happy that you're back home, sir. i'm trying to work on a couple of books, sir.",
+ "errors": 17,
+ "words": 11,
+ "wer": 1.5454545454545454
+ },
+ "semantic": {
+ "normalized_reference": "she handled it with a grace i genuinely did not expect",
+ "normalized_hypothesis": "i am happy that you are back home sir i am trying to work on a couple of books sir",
+ "errors": 19,
+ "words": 11,
+ "wer": 1.7272727272727273
+ }
+ },
+ {
+ "id": "release500_0079",
+ "category": "numbers",
+ "text": "Apartment four hundred seven is across from room twenty-nine on the third floor.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0079.wav",
+ "render_seconds": 74.11673066602089,
+ "audio_seconds": 5.642448979591837,
+ "rtf": 13.135560628743574,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "I can't wait to see your little shivers and tears for a while. Tell me I'll have a good time with you.",
+ "raw": {
+ "normalized_reference": "apartment four hundred seven is across from room twenty-nine on the third floor.",
+ "normalized_hypothesis": "i can't wait to see your little shivers and tears for a while. tell me i'll have a good time with you.",
+ "errors": 22,
+ "words": 14,
+ "wer": 1.5714285714285714
+ },
+ "semantic": {
+ "normalized_reference": "apartment 407 is across from room 29 on the 3rd floor",
+ "normalized_hypothesis": "i can not wait to see your little shivers and tears for a while tell me i will have a good time with you",
+ "errors": 24,
+ "words": 11,
+ "wer": 2.1818181818181817
+ }
+ },
+ {
+ "id": "release500_0080",
+ "category": "punctuation",
+ "text": "\"Not yet,\" Rowan whispered. \"Give me another minute... please.\"",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0080.wav",
+ "render_seconds": 90.39526544196997,
+ "audio_seconds": 5.50312925170068,
+ "rtf": 16.426157065824743,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you",
+ "raw": {
+ "normalized_reference": "\"not yet,\" rowan whispered. \"give me another minute... please.\"",
+ "normalized_hypothesis": "i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you",
+ "errors": 381,
+ "words": 9,
+ "wer": 42.333333333333336
+ },
+ "semantic": {
+ "normalized_reference": "not yet rowan whispered give me another minute please",
+ "normalized_hypothesis": "i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you",
+ "errors": 381,
+ "words": 9,
+ "wer": 42.333333333333336
+ }
+ },
+ {
+ "id": "release500_0081",
+ "category": "question",
+ "text": "What are we actually trying to accomplish here?",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0081.wav",
+ "render_seconds": 52.39817513502203,
+ "audio_seconds": 3.274013605442177,
+ "rtf": 16.004263100166717,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "Thank you for watching this video.",
+ "raw": {
+ "normalized_reference": "what are we actually trying to accomplish here?",
+ "normalized_hypothesis": "thank you for watching this video.",
+ "errors": 8,
+ "words": 8,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "what are we actually trying to accomplish here",
+ "normalized_hypothesis": "thank you for watching this video",
+ "errors": 8,
+ "words": 8,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0082",
+ "category": "rare_words",
+ "text": "The physician recommended a noninvasive procedure for the recurring neurological symptoms.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0082.wav",
+ "render_seconds": 98.7946310529951,
+ "audio_seconds": 6.408707482993197,
+ "rtf": 15.415687377707075,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "This is the first time I've learned that an open-vacation procedure for the coronary intubation test.",
+ "raw": {
+ "normalized_reference": "the physician recommended a noninvasive procedure for the recurring neurological symptoms.",
+ "normalized_hypothesis": "this is the first time i've learned that an open-vacation procedure for the coronary intubation test.",
+ "errors": 13,
+ "words": 11,
+ "wer": 1.1818181818181819
+ },
+ "semantic": {
+ "normalized_reference": "the physician recommended a noninvasive procedure for the recurring neurological symptoms",
+ "normalized_hypothesis": "this is the 1st time i have learned that an open vacation procedure for the coronary intubation test",
+ "errors": 14,
+ "words": 11,
+ "wer": 1.2727272727272727
+ }
+ },
+ {
+ "id": "release500_0083",
+ "category": "short",
+ "text": "Are you serious right now?",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0083.wav",
+ "render_seconds": 44.70363724103663,
+ "audio_seconds": 2.020136054421769,
+ "rtf": 22.129023014656468,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "I'm gonna show you what I know.",
+ "raw": {
+ "normalized_reference": "are you serious right now?",
+ "normalized_hypothesis": "i'm gonna show you what i know.",
+ "errors": 6,
+ "words": 5,
+ "wer": 1.2
+ },
+ "semantic": {
+ "normalized_reference": "are you serious right now",
+ "normalized_hypothesis": "i am going to show you what i know",
+ "errors": 8,
+ "words": 5,
+ "wer": 1.6
+ }
+ },
+ {
+ "id": "release500_0084",
+ "category": "technical",
+ "text": "The performance improvement is real, but it comes with tradeoffs.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0084.wav",
+ "render_seconds": 93.99984763294924,
+ "audio_seconds": 4.597551020408163,
+ "rtf": 20.44563447272067,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "People feel that chocolate is warm, but it tastes like chocolate.",
+ "raw": {
+ "normalized_reference": "the performance improvement is real, but it comes with tradeoffs.",
+ "normalized_hypothesis": "people feel that chocolate is warm, but it tastes like chocolate.",
+ "errors": 8,
+ "words": 10,
+ "wer": 0.8
+ },
+ "semantic": {
+ "normalized_reference": "the performance improvement is real but it comes with tradeoffs",
+ "normalized_hypothesis": "people feel that chocolate is warm but it tastes like chocolate",
+ "errors": 8,
+ "words": 10,
+ "wer": 0.8
+ }
+ },
+ {
+ "id": "release500_0085",
+ "category": "conversational",
+ "text": "I don't know. It's just been one of those weeks, you know?",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0085.wav",
+ "render_seconds": 34.8873905390501,
+ "audio_seconds": 3.900952380952381,
+ "rtf": 8.943300797363918,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "Welcome home. It's first day of the new summer.",
+ "raw": {
+ "normalized_reference": "i don't know. it's just been one of those weeks, you know?",
+ "normalized_hypothesis": "welcome home. it's first day of the new summer.",
+ "errors": 10,
+ "words": 12,
+ "wer": 0.8333333333333334
+ },
+ "semantic": {
+ "normalized_reference": "i do not know it is just been one of those weeks you know",
+ "normalized_hypothesis": "welcome home it is 1st day of the new summer",
+ "errors": 11,
+ "words": 14,
+ "wer": 0.7857142857142857
+ }
+ },
+ {
+ "id": "release500_0086",
+ "category": "descriptive",
+ "text": "The moment you stop expecting perfection, things get a lot more interesting.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0086.wav",
+ "render_seconds": 87.6976898870198,
+ "audio_seconds": 5.154829931972789,
+ "rtf": 17.01272225162573,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "You know that you still construct your perfection, as every year you are requesting.",
+ "raw": {
+ "normalized_reference": "the moment you stop expecting perfection, things get a lot more interesting.",
+ "normalized_hypothesis": "you know that you still construct your perfection, as every year you are requesting.",
+ "errors": 12,
+ "words": 12,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "the moment you stop expecting perfection things get a lot more interesting",
+ "normalized_hypothesis": "you know that you still construct your perfection as every year you are requesting",
+ "errors": 12,
+ "words": 12,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0087",
+ "category": "emotional",
+ "text": "It's the kind of thing that stays with you even years later.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0087.wav",
+ "render_seconds": 75.40016950899735,
+ "audio_seconds": 3.8312925170068026,
+ "rtf": 19.680086856929353,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "It's the turn of favorite stars, what you haven't noticed yet.",
+ "raw": {
+ "normalized_reference": "it's the kind of thing that stays with you even years later.",
+ "normalized_hypothesis": "it's the turn of favorite stars, what you haven't noticed yet.",
+ "errors": 8,
+ "words": 12,
+ "wer": 0.6666666666666666
+ },
+ "semantic": {
+ "normalized_reference": "it is the kind of thing that stays with you even years later",
+ "normalized_hypothesis": "it is the turn of favorite stars what you have not noticed yet",
+ "errors": 9,
+ "words": 13,
+ "wer": 0.6923076923076923
+ }
+ },
+ {
+ "id": "release500_0088",
+ "category": "long",
+ "text": "She told me once that the most important skill she'd developed was learning to disagree with someone without making them feel they were wrong.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0088.wav",
+ "render_seconds": 117.30900525301695,
+ "audio_seconds": 8.70748299319728,
+ "rtf": 13.472206072026164,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "She gave me a yes that there is still a clue in the sky. She told me that there is no end in this world. What's so weird? The fact that there is no end in the world.",
+ "raw": {
+ "normalized_reference": "she told me once that the most important skill she'd developed was learning to disagree with someone without making them feel they were wrong.",
+ "normalized_hypothesis": "she gave me a yes that there is still a clue in the sky. she told me that there is no end in this world. what's so weird? the fact that there is no end in the world.",
+ "errors": 34,
+ "words": 24,
+ "wer": 1.4166666666666667
+ },
+ "semantic": {
+ "normalized_reference": "she told me once that the most important skill she would developed was learning to disagree with someone without making them feel they were wrong",
+ "normalized_hypothesis": "she gave me a yes that there is still a clue in the sky she told me that there is no end in this world what is so weird the fact that there is no end in the world",
+ "errors": 35,
+ "words": 25,
+ "wer": 1.4
+ }
+ },
+ {
+ "id": "release500_0089",
+ "category": "narrative",
+ "text": "He stood at the edge of the cliff and wondered, not for the first time, how he'd gotten here.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0089.wav",
+ "render_seconds": 77.19590870197862,
+ "audio_seconds": 5.7817687074829935,
+ "rtf": 13.351607891555508,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "I still don't know how I got the clothes from your bag. I looked for the first time. I don't know how I got them back.",
+ "raw": {
+ "normalized_reference": "he stood at the edge of the cliff and wondered, not for the first time, how he'd gotten here.",
+ "normalized_hypothesis": "i still don't know how i got the clothes from your bag. i looked for the first time. i don't know how i got them back.",
+ "errors": 20,
+ "words": 19,
+ "wer": 1.0526315789473684
+ },
+ "semantic": {
+ "normalized_reference": "he stood at the edge of the cliff and wondered not for the 1st time how he would gotten here",
+ "normalized_hypothesis": "i still do not know how i got the clothes from your bag i looked for the 1st time i do not know how i got them back",
+ "errors": 22,
+ "words": 20,
+ "wer": 1.1
+ }
+ },
+ {
+ "id": "release500_0090",
+ "category": "question",
+ "text": "Why does asking for help feel so hard when you actually need it?",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0090.wav",
+ "render_seconds": 79.99640146596357,
+ "audio_seconds": 4.249251700680272,
+ "rtf": 18.825997399296625,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "Let us ask of the light they see here, what you want, what you need.",
+ "raw": {
+ "normalized_reference": "why does asking for help feel so hard when you actually need it?",
+ "normalized_hypothesis": "let us ask of the light they see here, what you want, what you need.",
+ "errors": 14,
+ "words": 13,
+ "wer": 1.0769230769230769
+ },
+ "semantic": {
+ "normalized_reference": "why does asking for help feel so hard when you actually need it",
+ "normalized_hypothesis": "let us ask of the light they see here what you want what you need",
+ "errors": 14,
+ "words": 13,
+ "wer": 1.0769230769230769
+ }
+ },
+ {
+ "id": "release500_0091",
+ "category": "short",
+ "text": "That's fair.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0091.wav",
+ "render_seconds": 18.09566029999405,
+ "audio_seconds": 1.4628571428571429,
+ "rtf": 12.370080283199059,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "Hatchimals.",
+ "raw": {
+ "normalized_reference": "that's fair.",
+ "normalized_hypothesis": "hatchimals.",
+ "errors": 2,
+ "words": 2,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "that is fair",
+ "normalized_hypothesis": "hatchimals",
+ "errors": 3,
+ "words": 3,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0092",
+ "category": "technical",
+ "text": "If you run into that error, the fix is usually just restarting the service.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0092.wav",
+ "render_seconds": 90.703427462955,
+ "audio_seconds": 5.363809523809524,
+ "rtf": 16.91026257743302,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "I want to go home. I want to go home. I want to go home. I want to go home. I want to go home. I want to go home. I want to go home. I want to go home. I want to go home. I want to go home. I want to go home. I want to go home. I want to go home. I want to go home. I want to go home. I want to go home. I want to go home. I want to go home. I want to go home. I want to go home. I want to go home. I want to go home. I want to go home. I want to go home. I want to go home. I want to go home. I want to go home. I want to go home. I want to go home. I want to go home. I want to go home. I want to go home. I want to go home. I want to go home. I want to go home. I want to go home. I want to go home. I want to go home. I want to go home. I want to go home. I want to go home. I want to go home. I want to go home. I want to go home. I want to go home. I want to go home. I want to go home. I want to go home. I want to go home. I want to go home. I want to go home. I want to go home. I want to go home. I want to go home. I want to go home. I want to go home. I want to go home. I want to go home. I want to go home. I want to go home. I want to go home. I want to go home. I want to go home. I want to go home. I want to go home. I want to go home. I want to go home. I want to go home. I want to go home. I want to go home. I want to go home. I want to go home. I want to go home. I want to go home.",
+ "raw": {
+ "normalized_reference": "if you run into that error, the fix is usually just restarting the service.",
+ "normalized_hypothesis": "i want to go home. i want to go home. i want to go home. i want to go home. i want to go home. i want to go home. i want to go home. i want to go home. i want to go home. i want to go home. i want to go home. i want to go home. i want to go home. i want to go home. i want to go home. i want to go home. i want to go home. i want to go home. i want to go home. i want to go home. i want to go home. i want to go home. i want to go home. i want to go home. i want to go home. i want to go home. i want to go home. i want to go home. i want to go home. i want to go home. i want to go home. i want to go home. i want to go home. i want to go home. i want to go home. i want to go home. i want to go home. i want to go home. i want to go home. i want to go home. i want to go home. i want to go home. i want to go home. i want to go home. i want to go home. i want to go home. i want to go home. i want to go home. i want to go home. i want to go home. i want to go home. i want to go home. i want to go home. i want to go home. i want to go home. i want to go home. i want to go home. i want to go home. i want to go home. i want to go home. i want to go home. i want to go home. i want to go home. i want to go home. i want to go home. i want to go home. i want to go home. i want to go home. i want to go home. i want to go home. i want to go home. i want to go home. i want to go home. i want to go home.",
+ "errors": 370,
+ "words": 14,
+ "wer": 26.428571428571427
+ },
+ "semantic": {
+ "normalized_reference": "if you run into that error the fix is usually just restarting the service",
+ "normalized_hypothesis": "i want to go home i want to go home i want to go home i want to go home i want to go home i want to go home i want to go home i want to go home i want to go home i want to go home i want to go home i want to go home i want to go home i want to go home i want to go home i want to go home i want to go home i want to go home i want to go home i want to go home i want to go home i want to go home i want to go home i want to go home i want to go home i want to go home i want to go home i want to go home i want to go home i want to go home i want to go home i want to go home i want to go home i want to go home i want to go home i want to go home i want to go home i want to go home i want to go home i want to go home i want to go home i want to go home i want to go home i want to go home i want to go home i want to go home i want to go home i want to go home i want to go home i want to go home i want to go home i want to go home i want to go home i want to go home i want to go home i want to go home i want to go home i want to go home i want to go home i want to go home i want to go home i want to go home i want to go home i want to go home i want to go home i want to go home i want to go home i want to go home i want to go home i want to go home i want to go home i want to go home i want to go home i want to go home",
+ "errors": 370,
+ "words": 14,
+ "wer": 26.428571428571427
+ }
+ },
+ {
+ "id": "release500_0093",
+ "category": "conversational",
+ "text": "It was a disaster in the best possible way.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0093.wav",
+ "render_seconds": 65.69613418600056,
+ "audio_seconds": 3.2043537414965986,
+ "rtf": 20.502147854411692,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "Alessandra says try the best kiss ever.",
+ "raw": {
+ "normalized_reference": "it was a disaster in the best possible way.",
+ "normalized_hypothesis": "alessandra says try the best kiss ever.",
+ "errors": 7,
+ "words": 9,
+ "wer": 0.7777777777777778
+ },
+ "semantic": {
+ "normalized_reference": "it was a disaster in the best possible way",
+ "normalized_hypothesis": "alessandra says try the best kiss ever",
+ "errors": 7,
+ "words": 9,
+ "wer": 0.7777777777777778
+ }
+ },
+ {
+ "id": "release500_0094",
+ "category": "descriptive",
+ "text": "It had the look of a place that used to mean something to someone.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0094.wav",
+ "render_seconds": 75.99187482800335,
+ "audio_seconds": 4.458231292517007,
+ "rtf": 17.045296630426776,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "I want to look at the glass of the sky so I can see.",
+ "raw": {
+ "normalized_reference": "it had the look of a place that used to mean something to someone.",
+ "normalized_hypothesis": "i want to look at the glass of the sky so i can see.",
+ "errors": 13,
+ "words": 14,
+ "wer": 0.9285714285714286
+ },
+ "semantic": {
+ "normalized_reference": "it had the look of a place that used to mean something to someone",
+ "normalized_hypothesis": "i want to look at the glass of the sky so i can see",
+ "errors": 13,
+ "words": 14,
+ "wer": 0.9285714285714286
+ }
+ },
+ {
+ "id": "release500_0095",
+ "category": "emotional",
+ "text": "There's this feeling you get when something finally works after weeks of failure.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0095.wav",
+ "render_seconds": 84.90776875498705,
+ "audio_seconds": 5.015510204081632,
+ "rtf": 16.929039180478377,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "I just want to know how much of this story I want to know how much of this story I want to know how much of this story I want to know how much of this story I want to know how much of this story I want to know how much of this story I want to know how much of this story I want to know how much of this story I want to know how much of this story I want to know how much of this story I want to know how much of this story I want to know how much of this story I want to know how much of this story I want to know how much of this story I want to know how much of this story I want to know how much of this story I want to know how much of this story I want to know how much of this story I want to know how much of this story I want to know how much of this story I want to know how much of this story I want to know how much of this story I want to know how much of this story I want to know how much of this story I want to know how much of this story I want to know how much of this story I want to know how much of this story I want to know how much of this story I want to know how much of this story I want to know how much of this story I want to know how much of this story I want to know how much of this story I want to know how much of this story I want to know how much of this story I want to know how much of this story I want to know how much of this story I want to know how much of this story I want to know how much of this story I want to know how much of this story I want to know how much of this story I want to know how much of this story I want to know how much of this story I want to know how much of this story I want to know how much of this story I want to know how much of this story I want to know how much of this story I want to know how much of this story I want to know how much of this story I want to know how much of this story I want",
+ "raw": {
+ "normalized_reference": "there's this feeling you get when something finally works after weeks of failure.",
+ "normalized_hypothesis": "i just want to know how much of this story i want to know how much of this story i want to know how much of this story i want to know how much of this story i want to know how much of this story i want to know how much of this story i want to know how much of this story i want to know how much of this story i want to know how much of this story i want to know how much of this story i want to know how much of this story i want to know how much of this story i want to know how much of this story i want to know how much of this story i want to know how much of this story i want to know how much of this story i want to know how much of this story i want to know how much of this story i want to know how much of this story i want to know how much of this story i want to know how much of this story i want to know how much of this story i want to know how much of this story i want to know how much of this story i want to know how much of this story i want to know how much of this story i want to know how much of this story i want to know how much of this story i want to know how much of this story i want to know how much of this story i want to know how much of this story i want to know how much of this story i want to know how much of this story i want to know how much of this story i want to know how much of this story i want to know how much of this story i want to know how much of this story i want to know how much of this story i want to know how much of this story i want to know how much of this story i want to know how much of this story i want to know how much of this story i want to know how much of this story i want to know how much of this story i want to know how much of this story i want to know how much of this story i want to know how much of this story i want to know how much of this story i want to know how much of this story i want",
+ "errors": 442,
+ "words": 13,
+ "wer": 34.0
+ },
+ "semantic": {
+ "normalized_reference": "there is this feeling you get when something finally works after weeks of failure",
+ "normalized_hypothesis": "i just want to know how much of this story i want to know how much of this story i want to know how much of this story i want to know how much of this story i want to know how much of this story i want to know how much of this story i want to know how much of this story i want to know how much of this story i want to know how much of this story i want to know how much of this story i want to know how much of this story i want to know how much of this story i want to know how much of this story i want to know how much of this story i want to know how much of this story i want to know how much of this story i want to know how much of this story i want to know how much of this story i want to know how much of this story i want to know how much of this story i want to know how much of this story i want to know how much of this story i want to know how much of this story i want to know how much of this story i want to know how much of this story i want to know how much of this story i want to know how much of this story i want to know how much of this story i want to know how much of this story i want to know how much of this story i want to know how much of this story i want to know how much of this story i want to know how much of this story i want to know how much of this story i want to know how much of this story i want to know how much of this story i want to know how much of this story i want to know how much of this story i want to know how much of this story i want to know how much of this story i want to know how much of this story i want to know how much of this story i want to know how much of this story i want to know how much of this story i want to know how much of this story i want to know how much of this story i want to know how much of this story i want to know how much of this story i want to know how much of this story i want",
+ "errors": 442,
+ "words": 14,
+ "wer": 31.571428571428573
+ }
+ },
+ {
+ "id": "release500_0096",
+ "category": "long",
+ "text": "He had a theory that most arguments aren't actually about what they seem to be about, and the longer I live the more I think he was onto something.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0096.wav",
+ "render_seconds": 129.39934604801238,
+ "audio_seconds": 9.125442176870749,
+ "rtf": 14.180063118035706,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "I never thought I'd be pleased to walk around your house and have a little show tomorrow. I knew all I'd have to do was think I was a little stupid.",
+ "raw": {
+ "normalized_reference": "he had a theory that most arguments aren't actually about what they seem to be about, and the longer i live the more i think he was onto something.",
+ "normalized_hypothesis": "i never thought i'd be pleased to walk around your house and have a little show tomorrow. i knew all i'd have to do was think i was a little stupid.",
+ "errors": 29,
+ "words": 29,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "he had a theory that most arguments are not actually about what they seem to be about and the longer i live the more i think he was onto something",
+ "normalized_hypothesis": "i never thought i would be pleased to walk around your house and have a little show tomorrow i knew all i would have to do was think i was a little stupid",
+ "errors": 30,
+ "words": 30,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0097",
+ "category": "narrative",
+ "text": "The thing nobody tells you about starting over is how ordinary most of it feels.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0097.wav",
+ "render_seconds": 57.30049965600483,
+ "audio_seconds": 5.433469387755102,
+ "rtf": 10.545840156040551,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "With that, I'd like to thank you for watching this video, and I'll see you in the next one.",
+ "raw": {
+ "normalized_reference": "the thing nobody tells you about starting over is how ordinary most of it feels.",
+ "normalized_hypothesis": "with that, i'd like to thank you for watching this video, and i'll see you in the next one.",
+ "errors": 18,
+ "words": 15,
+ "wer": 1.2
+ },
+ "semantic": {
+ "normalized_reference": "the thing nobody tells you about starting over is how ordinary most of it feels",
+ "normalized_hypothesis": "with that i would like to thank you for watching this video and i will see you in the next one",
+ "errors": 20,
+ "words": 15,
+ "wer": 1.3333333333333333
+ }
+ },
+ {
+ "id": "release500_0098",
+ "category": "question",
+ "text": "What exactly are we waiting for at this point?",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0098.wav",
+ "render_seconds": 66.59413642797153,
+ "audio_seconds": 3.2043537414965986,
+ "rtf": 20.782392270108303,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "I'm ready for this game.",
+ "raw": {
+ "normalized_reference": "what exactly are we waiting for at this point?",
+ "normalized_hypothesis": "i'm ready for this game.",
+ "errors": 7,
+ "words": 9,
+ "wer": 0.7777777777777778
+ },
+ "semantic": {
+ "normalized_reference": "what exactly are we waiting for at this point",
+ "normalized_hypothesis": "i am ready for this game",
+ "errors": 7,
+ "words": 9,
+ "wer": 0.7777777777777778
+ }
+ },
+ {
+ "id": "release500_0099",
+ "category": "short",
+ "text": "I stand by what I said.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0099.wav",
+ "render_seconds": 58.70346640900243,
+ "audio_seconds": 2.089795918367347,
+ "rtf": 28.0905259183703,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "I'll stay there for my sake.",
+ "raw": {
+ "normalized_reference": "i stand by what i said.",
+ "normalized_hypothesis": "i'll stay there for my sake.",
+ "errors": 6,
+ "words": 6,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "i stand by what i said",
+ "normalized_hypothesis": "i will stay there for my sake",
+ "errors": 6,
+ "words": 6,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0100",
+ "category": "technical",
+ "text": "Double-check that value before you push anything to production.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0100.wav",
+ "render_seconds": 70.89639390504453,
+ "audio_seconds": 4.31891156462585,
+ "rtf": 16.41533818050899,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "Have you checked my file? I feel like I've lost my way for a couple hours.",
+ "raw": {
+ "normalized_reference": "double-check that value before you push anything to production.",
+ "normalized_hypothesis": "have you checked my file? i feel like i've lost my way for a couple hours.",
+ "errors": 16,
+ "words": 10,
+ "wer": 1.6
+ },
+ "semantic": {
+ "normalized_reference": "double check that value before you push anything to production",
+ "normalized_hypothesis": "have you checked my file i feel like i have lost my way for a couple hours",
+ "errors": 17,
+ "words": 10,
+ "wer": 1.7
+ }
+ },
+ {
+ "id": "release500_0101",
+ "category": "conversational",
+ "text": "I understand your point; I just do not agree with the conclusion.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0101.wav",
+ "render_seconds": 81.69200813397765,
+ "audio_seconds": 4.458231292517007,
+ "rtf": 18.32386046706347,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "Thank you for watching. I'll see you in the next video.",
+ "raw": {
+ "normalized_reference": "i understand your point; i just do not agree with the conclusion.",
+ "normalized_hypothesis": "thank you for watching. i'll see you in the next video.",
+ "errors": 12,
+ "words": 12,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "i understand your point i just do not agree with the conclusion",
+ "normalized_hypothesis": "thank you for watching i will see you in the next video",
+ "errors": 11,
+ "words": 12,
+ "wer": 0.9166666666666666
+ }
+ },
+ {
+ "id": "release500_0102",
+ "category": "descriptive",
+ "text": "The sky that day was a strange shade of orange I'd never seen before.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0102.wav",
+ "render_seconds": 88.40734693396371,
+ "audio_seconds": 4.597551020408163,
+ "rtf": 19.229225851226126,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "The score I wrote down was a score that you are going to have to share with me.",
+ "raw": {
+ "normalized_reference": "the sky that day was a strange shade of orange i'd never seen before.",
+ "normalized_hypothesis": "the score i wrote down was a score that you are going to have to share with me.",
+ "errors": 15,
+ "words": 14,
+ "wer": 1.0714285714285714
+ },
+ "semantic": {
+ "normalized_reference": "the sky that day was a strange shade of orange i would never seen before",
+ "normalized_hypothesis": "the score i wrote down was a score that you are going to have to share with me",
+ "errors": 15,
+ "words": 15,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0103",
+ "category": "emotional",
+ "text": "It hurt more than I expected, and I thought I was prepared.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0103.wav",
+ "render_seconds": 58.3985950849019,
+ "audio_seconds": 4.388571428571429,
+ "rtf": 13.306971536273219,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "And yet, we're not in the next chapter. We're in the first part.",
+ "raw": {
+ "normalized_reference": "it hurt more than i expected, and i thought i was prepared.",
+ "normalized_hypothesis": "and yet, we're not in the next chapter. we're in the first part.",
+ "errors": 13,
+ "words": 12,
+ "wer": 1.0833333333333333
+ },
+ "semantic": {
+ "normalized_reference": "it hurt more than i expected and i thought i was prepared",
+ "normalized_hypothesis": "and yet we are not in the next chapter we are in the 1st part",
+ "errors": 15,
+ "words": 12,
+ "wer": 1.25
+ }
+ },
+ {
+ "id": "release500_0104",
+ "category": "long",
+ "text": "The hardest part wasn't making the decision, it was sitting with the uncertainty of not knowing whether it was the right one for months afterward.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0104.wav",
+ "render_seconds": 139.29743686399888,
+ "audio_seconds": 8.986122448979591,
+ "rtf": 15.50139536322662,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "The law of school used to want you to be successful. I was sick and let down for a week. I learned later I used to work in foreign jobs to work.",
+ "raw": {
+ "normalized_reference": "the hardest part wasn't making the decision, it was sitting with the uncertainty of not knowing whether it was the right one for months afterward.",
+ "normalized_hypothesis": "the law of school used to want you to be successful. i was sick and let down for a week. i learned later i used to work in foreign jobs to work.",
+ "errors": 30,
+ "words": 25,
+ "wer": 1.2
+ },
+ "semantic": {
+ "normalized_reference": "the hardest part was not making the decision it was sitting with the uncertainty of not knowing whether it was the right one for months afterward",
+ "normalized_hypothesis": "the law of school used to want you to be successful i was sick and let down for a week i learned later i used to work in foreign jobs to work",
+ "errors": 30,
+ "words": 26,
+ "wer": 1.1538461538461537
+ }
+ },
+ {
+ "id": "release500_0105",
+ "category": "narrative",
+ "text": "There was this one day last summer when everything just clicked into place.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0105.wav",
+ "render_seconds": 79.4943403560901,
+ "audio_seconds": 5.154829931972789,
+ "rtf": 15.421331334915074,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "I wish this were a real show, but I guess you can't expect it to be.",
+ "raw": {
+ "normalized_reference": "there was this one day last summer when everything just clicked into place.",
+ "normalized_hypothesis": "i wish this were a real show, but i guess you can't expect it to be.",
+ "errors": 15,
+ "words": 13,
+ "wer": 1.1538461538461537
+ },
+ "semantic": {
+ "normalized_reference": "there was this one day last summer when everything just clicked into place",
+ "normalized_hypothesis": "i wish this were a real show but i guess you can not expect it to be",
+ "errors": 16,
+ "words": 13,
+ "wer": 1.2307692307692308
+ }
+ },
+ {
+ "id": "release500_0106",
+ "category": "question",
+ "text": "Have you ever thought about why we keep doing things the same way when they stop working?",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0106.wav",
+ "render_seconds": 89.80292898905464,
+ "audio_seconds": 5.433469387755102,
+ "rtf": 16.52773257385696,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "And I love that thing about you, I like that. The one thing that shows why you're a stupid.",
+ "raw": {
+ "normalized_reference": "have you ever thought about why we keep doing things the same way when they stop working?",
+ "normalized_hypothesis": "and i love that thing about you, i like that. the one thing that shows why you're a stupid.",
+ "errors": 18,
+ "words": 17,
+ "wer": 1.0588235294117647
+ },
+ "semantic": {
+ "normalized_reference": "have you ever thought about why we keep doing things the same way when they stop working",
+ "normalized_hypothesis": "and i love that thing about you i like that the one thing that shows why you are a stupid",
+ "errors": 19,
+ "words": 17,
+ "wer": 1.1176470588235294
+ }
+ },
+ {
+ "id": "release500_0107",
+ "category": "short",
+ "text": "I can't believe you just said that.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0107.wav",
+ "render_seconds": 62.10698268702254,
+ "audio_seconds": 2.4380952380952383,
+ "rtf": 25.473567117724084,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "What the hell is this shit?",
+ "raw": {
+ "normalized_reference": "i can't believe you just said that.",
+ "normalized_hypothesis": "what the hell is this shit?",
+ "errors": 7,
+ "words": 7,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "i can not believe you just said that",
+ "normalized_hypothesis": "what the hell is this shit",
+ "errors": 8,
+ "words": 8,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0108",
+ "category": "technical",
+ "text": "You'll want to isolate that variable before drawing any conclusions.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0108.wav",
+ "render_seconds": 91.70395563798957,
+ "audio_seconds": 4.597551020408163,
+ "rtf": 19.94626165776584,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "Hello and welcome back to my channel. I hope you are enjoying this video.",
+ "raw": {
+ "normalized_reference": "you'll want to isolate that variable before drawing any conclusions.",
+ "normalized_hypothesis": "hello and welcome back to my channel. i hope you are enjoying this video.",
+ "errors": 13,
+ "words": 10,
+ "wer": 1.3
+ },
+ "semantic": {
+ "normalized_reference": "you will want to isolate that variable before drawing any conclusions",
+ "normalized_hypothesis": "hello and welcome back to my channel i hope you are enjoying this video",
+ "errors": 13,
+ "words": 11,
+ "wer": 1.1818181818181819
+ }
+ },
+ {
+ "id": "release500_0109",
+ "category": "conversational",
+ "text": "The meeting that could have been an email lasted two hours.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0109.wav",
+ "render_seconds": 80.20279090700205,
+ "audio_seconds": 3.9706122448979593,
+ "rtf": 20.199099272425475,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "Thank you for watching.",
+ "raw": {
+ "normalized_reference": "the meeting that could have been an email lasted two hours.",
+ "normalized_hypothesis": "thank you for watching.",
+ "errors": 11,
+ "words": 11,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "the meeting that could have been an email lasted 2 hours",
+ "normalized_hypothesis": "thank you for watching",
+ "errors": 11,
+ "words": 11,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0110",
+ "category": "descriptive",
+ "text": "There's a specific kind of tired that sleep doesn't fix.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0110.wav",
+ "render_seconds": 68.00529682496563,
+ "audio_seconds": 3.622312925170068,
+ "rtf": 18.773998334716698,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "Thank you for watching.",
+ "raw": {
+ "normalized_reference": "there's a specific kind of tired that sleep doesn't fix.",
+ "normalized_hypothesis": "thank you for watching.",
+ "errors": 10,
+ "words": 10,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "there is a specific kind of tired that sleep does not fix",
+ "normalized_hypothesis": "thank you for watching",
+ "errors": 12,
+ "words": 12,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0111",
+ "category": "emotional",
+ "text": "Something about that moment broke something open in me that I hadn't realized was closed.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0111.wav",
+ "render_seconds": 96.19787661207374,
+ "audio_seconds": 5.642448979591837,
+ "rtf": 17.048958166925683,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "Thank you for watching. Please subscribe to my channel for more videos like this. Thank you for watching. Please subscribe to my channel for more videos like this. Thank you for watching. Please subscribe to my channel for more videos like this. Thank you for watching. Please subscribe to my channel for more videos like this. Thank you for watching. Please subscribe to my channel for more videos like this. Thank you for watching. Please subscribe to my channel for more videos like this. Thank you for watching. Please subscribe to my channel for more videos like this. Thank you for watching. Please subscribe to my channel for more videos like this. Thank you for watching. Please subscribe to my channel for more videos like this. Thank you for watching. Please subscribe to my channel for more videos like this. Thank you for watching. Please subscribe to my channel for more videos like this. Thank you for watching. Please subscribe to my channel for more videos like this. Thank you for watching. Please subscribe to my channel for more videos like this. Thank you for watching. Please subscribe to my channel for more videos like this. Thank you for watching. Please subscribe to my channel for more videos like this. Thank you for watching. Please subscribe to my channel for more videos like this. Thank you for watching. Please subscribe to my channel for more videos like this. Thank you for watching. Please subscribe to my channel for more videos like this. Thank you for watching. Please subscribe to my channel for more videos like this. Thank you for watching. Please subscribe to my channel for more videos like this. Thank you for watching. Please subscribe to my channel for more videos like this. Thank you for watching. Please subscribe to my channel for more videos like this. Thank you for watching. Please subscribe to my channel for watching. Please subscribe to my channel for more videos like this. Thank you for watching. Please subscribe to my channel for watching. Please subscribe to my channel for watching. Please subscribe to my channel for watching. Please subscribe to my channel for watching. Please subscribe to my channel for watching. Please subscribe to my channel for watching. Please subscribe to my channel for watching. Please subscribe to my channel for watching",
+ "raw": {
+ "normalized_reference": "something about that moment broke something open in me that i hadn't realized was closed.",
+ "normalized_hypothesis": "thank you for watching. please subscribe to my channel for more videos like this. thank you for watching. please subscribe to my channel for more videos like this. thank you for watching. please subscribe to my channel for more videos like this. thank you for watching. please subscribe to my channel for more videos like this. thank you for watching. please subscribe to my channel for more videos like this. thank you for watching. please subscribe to my channel for more videos like this. thank you for watching. please subscribe to my channel for more videos like this. thank you for watching. please subscribe to my channel for more videos like this. thank you for watching. please subscribe to my channel for more videos like this. thank you for watching. please subscribe to my channel for more videos like this. thank you for watching. please subscribe to my channel for more videos like this. thank you for watching. please subscribe to my channel for more videos like this. thank you for watching. please subscribe to my channel for more videos like this. thank you for watching. please subscribe to my channel for more videos like this. thank you for watching. please subscribe to my channel for more videos like this. thank you for watching. please subscribe to my channel for more videos like this. thank you for watching. please subscribe to my channel for more videos like this. thank you for watching. please subscribe to my channel for more videos like this. thank you for watching. please subscribe to my channel for more videos like this. thank you for watching. please subscribe to my channel for more videos like this. thank you for watching. please subscribe to my channel for more videos like this. thank you for watching. please subscribe to my channel for more videos like this. thank you for watching. please subscribe to my channel for watching. please subscribe to my channel for more videos like this. thank you for watching. please subscribe to my channel for watching. please subscribe to my channel for watching. please subscribe to my channel for watching. please subscribe to my channel for watching. please subscribe to my channel for watching. please subscribe to my channel for watching. please subscribe to my channel for watching. please subscribe to my channel for watching",
+ "errors": 389,
+ "words": 15,
+ "wer": 25.933333333333334
+ },
+ "semantic": {
+ "normalized_reference": "something about that moment broke something open in me that i had not realized was closed",
+ "normalized_hypothesis": "thank you for watching please subscribe to my channel for more videos like this thank you for watching please subscribe to my channel for more videos like this thank you for watching please subscribe to my channel for more videos like this thank you for watching please subscribe to my channel for more videos like this thank you for watching please subscribe to my channel for more videos like this thank you for watching please subscribe to my channel for more videos like this thank you for watching please subscribe to my channel for more videos like this thank you for watching please subscribe to my channel for more videos like this thank you for watching please subscribe to my channel for more videos like this thank you for watching please subscribe to my channel for more videos like this thank you for watching please subscribe to my channel for more videos like this thank you for watching please subscribe to my channel for more videos like this thank you for watching please subscribe to my channel for more videos like this thank you for watching please subscribe to my channel for more videos like this thank you for watching please subscribe to my channel for more videos like this thank you for watching please subscribe to my channel for more videos like this thank you for watching please subscribe to my channel for more videos like this thank you for watching please subscribe to my channel for more videos like this thank you for watching please subscribe to my channel for more videos like this thank you for watching please subscribe to my channel for more videos like this thank you for watching please subscribe to my channel for more videos like this thank you for watching please subscribe to my channel for more videos like this thank you for watching please subscribe to my channel for watching please subscribe to my channel for more videos like this thank you for watching please subscribe to my channel for watching please subscribe to my channel for watching please subscribe to my channel for watching please subscribe to my channel for watching please subscribe to my channel for watching please subscribe to my channel for watching please subscribe to my channel for watching please subscribe to my channel for watching",
+ "errors": 389,
+ "words": 16,
+ "wer": 24.3125
+ }
+ },
+ {
+ "id": "release500_0112",
+ "category": "long",
+ "text": "The thing about making something good is that you don't usually know it's good while you're making it. You just know you're not done yet.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0112.wav",
+ "render_seconds": 124.59800055297092,
+ "audio_seconds": 7.662585034013605,
+ "rtf": 16.260570029551424,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "I think about what it feels like to do this, and I don't know if I know what to do or not to do. I just want to do it.",
+ "raw": {
+ "normalized_reference": "the thing about making something good is that you don't usually know it's good while you're making it. you just know you're not done yet.",
+ "normalized_hypothesis": "i think about what it feels like to do this, and i don't know if i know what to do or not to do. i just want to do it.",
+ "errors": 27,
+ "words": 25,
+ "wer": 1.08
+ },
+ "semantic": {
+ "normalized_reference": "the thing about making something good is that you do not usually know it is good while you are making it you just know you are not done yet",
+ "normalized_hypothesis": "i think about what it feels like to do this and i do not know if i know what to do or not to do i just want to do it",
+ "errors": 28,
+ "words": 29,
+ "wer": 0.9655172413793104
+ }
+ },
+ {
+ "id": "release500_0113",
+ "category": "narrative",
+ "text": "She opened the door slowly, as if she expected to find something she wasn't ready for.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0113.wav",
+ "render_seconds": 88.18888535699807,
+ "audio_seconds": 5.921088435374149,
+ "rtf": 14.894032798114335,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "I thought that I would see you. I thought that you expected to say something. I guess I won't be seeing you.",
+ "raw": {
+ "normalized_reference": "she opened the door slowly, as if she expected to find something she wasn't ready for.",
+ "normalized_hypothesis": "i thought that i would see you. i thought that you expected to say something. i guess i won't be seeing you.",
+ "errors": 19,
+ "words": 16,
+ "wer": 1.1875
+ },
+ "semantic": {
+ "normalized_reference": "she opened the door slowly as if she expected to find something she was not ready for",
+ "normalized_hypothesis": "i thought that i would see you i thought that you expected to say something i guess i will not be seeing you",
+ "errors": 19,
+ "words": 17,
+ "wer": 1.1176470588235294
+ }
+ },
+ {
+ "id": "release500_0114",
+ "category": "question",
+ "text": "What would you do differently if you could go back?",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0114.wav",
+ "render_seconds": 66.09143147303257,
+ "audio_seconds": 3.3436734693877552,
+ "rtf": 19.766114149039282,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "We're not going to do that. We're not going to do that. We're not going to do that. We're not going to do that. We're not going to do that. We're not going to do that. We're not going to do that. We're not going to do that. We're not going to do that. We're not going to do that. We're not going to do that. We're not going to do that. We're not going to do that. We're not going to do that. We're not going to do that. We're not going to do that. We're not going to do that. We're not going to do that. We're not going to do that. We're not going to do that. We're not going to do that. We're not going to do that. We're not going to do that. We're not going to do that. We're not going to do that. We're not going to do that. We're not going to do that. We're not going to do that. We're not going to do that. We're not going to do that. We're not going to do that. We're not going to do that. We're not going to do that. We're not going to do that. We're not going to do that. We're not going to do that. We're not going to do that. We're not going to do that. We're not going to do that. We're not going to do that. We're not going to do that. We're not going to do that. We're not going to do that. We're not going to do that. We're not going to do that. We're not going to do that. We're not going to do that. We're not going to do that. We're not going to do that. We're not going to do that. We're not going to do that. We're not going to do that. We're not going to do that. We're not going to do that. We're not going to do that. We're not going",
+ "raw": {
+ "normalized_reference": "what would you do differently if you could go back?",
+ "normalized_hypothesis": "we're not going to do that. we're not going to do that. we're not going to do that. we're not going to do that. we're not going to do that. we're not going to do that. we're not going to do that. we're not going to do that. we're not going to do that. we're not going to do that. we're not going to do that. we're not going to do that. we're not going to do that. we're not going to do that. we're not going to do that. we're not going to do that. we're not going to do that. we're not going to do that. we're not going to do that. we're not going to do that. we're not going to do that. we're not going to do that. we're not going to do that. we're not going to do that. we're not going to do that. we're not going to do that. we're not going to do that. we're not going to do that. we're not going to do that. we're not going to do that. we're not going to do that. we're not going to do that. we're not going to do that. we're not going to do that. we're not going to do that. we're not going to do that. we're not going to do that. we're not going to do that. we're not going to do that. we're not going to do that. we're not going to do that. we're not going to do that. we're not going to do that. we're not going to do that. we're not going to do that. we're not going to do that. we're not going to do that. we're not going to do that. we're not going to do that. we're not going to do that. we're not going to do that. we're not going to do that. we're not going to do that. we're not going to do that. we're not going to do that. we're not going",
+ "errors": 332,
+ "words": 10,
+ "wer": 33.2
+ },
+ "semantic": {
+ "normalized_reference": "what would you do differently if you could go back",
+ "normalized_hypothesis": "we are not going to do that we are not going to do that we are not going to do that we are not going to do that we are not going to do that we are not going to do that we are not going to do that we are not going to do that we are not going to do that we are not going to do that we are not going to do that we are not going to do that we are not going to do that we are not going to do that we are not going to do that we are not going to do that we are not going to do that we are not going to do that we are not going to do that we are not going to do that we are not going to do that we are not going to do that we are not going to do that we are not going to do that we are not going to do that we are not going to do that we are not going to do that we are not going to do that we are not going to do that we are not going to do that we are not going to do that we are not going to do that we are not going to do that we are not going to do that we are not going to do that we are not going to do that we are not going to do that we are not going to do that we are not going to do that we are not going to do that we are not going to do that we are not going to do that we are not going to do that we are not going to do that we are not going to do that we are not going to do that we are not going to do that we are not going to do that we are not going to do that we are not going to do that we are not going to do that we are not going to do that we are not going to do that we are not going to do that we are not going to do that we are not going",
+ "errors": 388,
+ "words": 10,
+ "wer": 38.8
+ }
+ },
+ {
+ "id": "release500_0115",
+ "category": "short",
+ "text": "Okay. Okay okay okay.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0115.wav",
+ "render_seconds": 56.49792234203778,
+ "audio_seconds": 2.4380952380952383,
+ "rtf": 23.17297596060143,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "I can't. I can't. I can't.",
+ "raw": {
+ "normalized_reference": "okay. okay okay okay.",
+ "normalized_hypothesis": "i can't. i can't. i can't.",
+ "errors": 6,
+ "words": 4,
+ "wer": 1.5
+ },
+ "semantic": {
+ "normalized_reference": "okay okay okay okay",
+ "normalized_hypothesis": "i can not i can not i can not",
+ "errors": 9,
+ "words": 4,
+ "wer": 2.25
+ }
+ },
+ {
+ "id": "release500_0116",
+ "category": "technical",
+ "text": "The latency goes up significantly at scale, which is expected.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0116.wav",
+ "render_seconds": 85.29182029305957,
+ "audio_seconds": 4.597551020408163,
+ "rtf": 18.551576679509584,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much.",
+ "raw": {
+ "normalized_reference": "the latency goes up significantly at scale, which is expected.",
+ "normalized_hypothesis": "i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much.",
+ "errors": 370,
+ "words": 10,
+ "wer": 37.0
+ },
+ "semantic": {
+ "normalized_reference": "the latency goes up significantly at scale which is expected",
+ "normalized_hypothesis": "i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much",
+ "errors": 370,
+ "words": 10,
+ "wer": 37.0
+ }
+ },
+ {
+ "id": "release500_0117",
+ "category": "conversational",
+ "text": "We ended up staying until two in the morning talking about nothing.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0117.wav",
+ "render_seconds": 80.60033227689564,
+ "audio_seconds": 4.458231292517007,
+ "rtf": 18.078992988134246,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "My next story will be totally, totally about myself.",
+ "raw": {
+ "normalized_reference": "we ended up staying until two in the morning talking about nothing.",
+ "normalized_hypothesis": "my next story will be totally, totally about myself.",
+ "errors": 11,
+ "words": 12,
+ "wer": 0.9166666666666666
+ },
+ "semantic": {
+ "normalized_reference": "we ended up staying until 2 in the morning talking about nothing",
+ "normalized_hypothesis": "my next story will be totally totally about myself",
+ "errors": 11,
+ "words": 12,
+ "wer": 0.9166666666666666
+ }
+ },
+ {
+ "id": "release500_0118",
+ "category": "descriptive",
+ "text": "The rain started slowly, and then all at once, the way it usually does.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0118.wav",
+ "render_seconds": 83.00951866898686,
+ "audio_seconds": 5.154829931972789,
+ "rtf": 16.103250691961925,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "I want to go back to sleep. I really want to go back to that place.",
+ "raw": {
+ "normalized_reference": "the rain started slowly, and then all at once, the way it usually does.",
+ "normalized_hypothesis": "i want to go back to sleep. i really want to go back to that place.",
+ "errors": 16,
+ "words": 14,
+ "wer": 1.1428571428571428
+ },
+ "semantic": {
+ "normalized_reference": "the rain started slowly and then all at once the way it usually does",
+ "normalized_hypothesis": "i want to go back to sleep i really want to go back to that place",
+ "errors": 16,
+ "words": 14,
+ "wer": 1.1428571428571428
+ }
+ },
+ {
+ "id": "release500_0119",
+ "category": "emotional",
+ "text": "I honestly did not think I was going to make it through that year.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0119.wav",
+ "render_seconds": 75.00318047497422,
+ "audio_seconds": 4.31891156462585,
+ "rtf": 17.36622279772746,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "I love you. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love",
+ "raw": {
+ "normalized_reference": "i honestly did not think i was going to make it through that year.",
+ "normalized_hypothesis": "i love you. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love",
+ "errors": 368,
+ "words": 14,
+ "wer": 26.285714285714285
+ },
+ "semantic": {
+ "normalized_reference": "i honestly did not think i was going to make it through that year",
+ "normalized_hypothesis": "i love you i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love",
+ "errors": 368,
+ "words": 14,
+ "wer": 26.285714285714285
+ }
+ },
+ {
+ "id": "release500_0120",
+ "category": "long",
+ "text": "There's an assumption that people who seem confident all the time actually feel confident, and in my experience that's almost never true.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0120.wav",
+ "render_seconds": 135.2956613280112,
+ "audio_seconds": 8.986122448979591,
+ "rtf": 15.05606696282828,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "My name is Samson, and today we will see a classic movie that turns you into a perfect human. The next part is not Jewish, not a crime.",
+ "raw": {
+ "normalized_reference": "there's an assumption that people who seem confident all the time actually feel confident, and in my experience that's almost never true.",
+ "normalized_hypothesis": "my name is samson, and today we will see a classic movie that turns you into a perfect human. the next part is not jewish, not a crime.",
+ "errors": 28,
+ "words": 22,
+ "wer": 1.2727272727272727
+ },
+ "semantic": {
+ "normalized_reference": "there is an assumption that people who seem confident all the time actually feel confident and in my experience that is almost never true",
+ "normalized_hypothesis": "my name is samson and today we will see a classic movie that turns you into a perfect human the next part is not jewish not a crime",
+ "errors": 26,
+ "words": 24,
+ "wer": 1.0833333333333333
+ }
+ },
+ {
+ "id": "release500_0121",
+ "category": "narrative",
+ "text": "The moment I opened the door I knew something was different, even if I couldn't say what.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0121.wav",
+ "render_seconds": 61.21020917000715,
+ "audio_seconds": 5.572789115646258,
+ "rtf": 10.983765561512513,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "Hello everyone, welcome back to my channel. I hope you are all doing well. I hope you are all doing well. I hope you are all doing well. I hope you are all doing well. I hope you are all doing well. I hope you are all doing well. I hope you are all doing well. I hope you are all doing well. I hope you are all doing well. I hope you are all doing well. I hope you are all doing well. I hope you are all doing well. I hope you are all doing well. I hope you are all doing well. I hope you are all doing well. I hope you are all doing well. I hope you are all doing well. I hope you are all doing well. I hope you are all doing well. I hope you are all doing well. I hope you are all doing well. I hope you are all doing well. I hope you are all doing well. I hope you are all doing well. I hope you are all doing well. I hope you are all doing well. I hope you are all doing well. I hope you are all doing well. I hope you are all doing well. I hope you are all doing well. I hope you are all doing well. I hope you are all doing well. I hope you are all doing well. I hope you are all doing well. I hope you are all doing well. I hope you are all doing well. I hope you are all doing well. I hope you are all doing well. I hope you are all doing well. I hope you are all doing well. I hope you are all doing well. I hope you are all doing well. I hope you are all doing well. I hope you are all doing well. I hope you are all doing well. I hope you are all doing well. I hope you are all doing well. I hope you are all doing well. I hope you are all doing well. I hope you are all doing well. I hope you are all doing well. I hope you are all doing well. I hope you are all doing well. I hope you are all doing well. I hope you",
+ "raw": {
+ "normalized_reference": "the moment i opened the door i knew something was different, even if i couldn't say what.",
+ "normalized_hypothesis": "hello everyone, welcome back to my channel. i hope you are all doing well. i hope you are all doing well. i hope you are all doing well. i hope you are all doing well. i hope you are all doing well. i hope you are all doing well. i hope you are all doing well. i hope you are all doing well. i hope you are all doing well. i hope you are all doing well. i hope you are all doing well. i hope you are all doing well. i hope you are all doing well. i hope you are all doing well. i hope you are all doing well. i hope you are all doing well. i hope you are all doing well. i hope you are all doing well. i hope you are all doing well. i hope you are all doing well. i hope you are all doing well. i hope you are all doing well. i hope you are all doing well. i hope you are all doing well. i hope you are all doing well. i hope you are all doing well. i hope you are all doing well. i hope you are all doing well. i hope you are all doing well. i hope you are all doing well. i hope you are all doing well. i hope you are all doing well. i hope you are all doing well. i hope you are all doing well. i hope you are all doing well. i hope you are all doing well. i hope you are all doing well. i hope you are all doing well. i hope you are all doing well. i hope you are all doing well. i hope you are all doing well. i hope you are all doing well. i hope you are all doing well. i hope you are all doing well. i hope you are all doing well. i hope you are all doing well. i hope you are all doing well. i hope you are all doing well. i hope you are all doing well. i hope you are all doing well. i hope you are all doing well. i hope you are all doing well. i hope you are all doing well. i hope you are all doing well. i hope you",
+ "errors": 385,
+ "words": 17,
+ "wer": 22.647058823529413
+ },
+ "semantic": {
+ "normalized_reference": "the moment i opened the door i knew something was different even if i could not say what",
+ "normalized_hypothesis": "hello everyone welcome back to my channel i hope you are all doing well i hope you are all doing well i hope you are all doing well i hope you are all doing well i hope you are all doing well i hope you are all doing well i hope you are all doing well i hope you are all doing well i hope you are all doing well i hope you are all doing well i hope you are all doing well i hope you are all doing well i hope you are all doing well i hope you are all doing well i hope you are all doing well i hope you are all doing well i hope you are all doing well i hope you are all doing well i hope you are all doing well i hope you are all doing well i hope you are all doing well i hope you are all doing well i hope you are all doing well i hope you are all doing well i hope you are all doing well i hope you are all doing well i hope you are all doing well i hope you are all doing well i hope you are all doing well i hope you are all doing well i hope you are all doing well i hope you are all doing well i hope you are all doing well i hope you are all doing well i hope you are all doing well i hope you are all doing well i hope you are all doing well i hope you are all doing well i hope you are all doing well i hope you are all doing well i hope you are all doing well i hope you are all doing well i hope you are all doing well i hope you are all doing well i hope you are all doing well i hope you are all doing well i hope you are all doing well i hope you are all doing well i hope you are all doing well i hope you are all doing well i hope you are all doing well i hope you are all doing well i hope you are all doing well i hope you are all doing well i hope you",
+ "errors": 385,
+ "words": 18,
+ "wer": 21.38888888888889
+ }
+ },
+ {
+ "id": "release500_0122",
+ "category": "question",
+ "text": "How do you explain something that doesn't make sense even to you?",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0122.wav",
+ "render_seconds": 76.20779749599751,
+ "audio_seconds": 4.040272108843538,
+ "rtf": 18.862045783795182,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "I hope you enjoyed this video, and I'll see you in the next one.",
+ "raw": {
+ "normalized_reference": "how do you explain something that doesn't make sense even to you?",
+ "normalized_hypothesis": "i hope you enjoyed this video, and i'll see you in the next one.",
+ "errors": 13,
+ "words": 12,
+ "wer": 1.0833333333333333
+ },
+ "semantic": {
+ "normalized_reference": "how do you explain something that does not make sense even to you",
+ "normalized_hypothesis": "i hope you enjoyed this video and i will see you in the next one",
+ "errors": 14,
+ "words": 13,
+ "wer": 1.0769230769230769
+ }
+ },
+ {
+ "id": "release500_0123",
+ "category": "short",
+ "text": "Hold on a second.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0123.wav",
+ "render_seconds": 54.59877931093797,
+ "audio_seconds": 1.6718367346938776,
+ "rtf": 32.65796125776319,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "Thank you for watching.",
+ "raw": {
+ "normalized_reference": "hold on a second.",
+ "normalized_hypothesis": "thank you for watching.",
+ "errors": 4,
+ "words": 4,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "hold on a 2nd",
+ "normalized_hypothesis": "thank you for watching",
+ "errors": 4,
+ "words": 4,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0124",
+ "category": "technical",
+ "text": "A deterministic split prevents duplicate normalized text from leaking into evaluation.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0124.wav",
+ "render_seconds": 96.39785736100748,
+ "audio_seconds": 6.339047619047619,
+ "rtf": 15.206993724317586,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "The tutorial is fixed yet, but I still can't remember the last text. I can't remember the last thing.",
+ "raw": {
+ "normalized_reference": "a deterministic split prevents duplicate normalized text from leaking into evaluation.",
+ "normalized_hypothesis": "the tutorial is fixed yet, but i still can't remember the last text. i can't remember the last thing.",
+ "errors": 18,
+ "words": 11,
+ "wer": 1.6363636363636365
+ },
+ "semantic": {
+ "normalized_reference": "a deterministic split prevents duplicate normalized text from leaking into evaluation",
+ "normalized_hypothesis": "the tutorial is fixed yet but i still can not remember the last text i can not remember the last thing",
+ "errors": 20,
+ "words": 11,
+ "wer": 1.8181818181818181
+ }
+ },
+ {
+ "id": "release500_0125",
+ "category": "conversational",
+ "text": "Honestly, that was much better than I expected, but the ending still felt rushed.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0125.wav",
+ "render_seconds": 103.99647059303243,
+ "audio_seconds": 5.642448979591837,
+ "rtf": 18.431087453192234,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "Yes, I was a little bit over-expecting, but now I can stay safe like that.",
+ "raw": {
+ "normalized_reference": "honestly, that was much better than i expected, but the ending still felt rushed.",
+ "normalized_hypothesis": "yes, i was a little bit over-expecting, but now i can stay safe like that.",
+ "errors": 14,
+ "words": 14,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "honestly that was much better than i expected but the ending still felt rushed",
+ "normalized_hypothesis": "yes i was a little bit over expecting but now i can stay safe like that",
+ "errors": 14,
+ "words": 14,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0126",
+ "category": "descriptive",
+ "text": "She had a habit of tilting her head slightly when thinking hard about something.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0126.wav",
+ "render_seconds": 85.90281561494339,
+ "audio_seconds": 5.224489795918367,
+ "rtf": 16.44233580129776,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "She's looking at her because Clayton will hurt you. I'm thinking you're a great saint.",
+ "raw": {
+ "normalized_reference": "she had a habit of tilting her head slightly when thinking hard about something.",
+ "normalized_hypothesis": "she's looking at her because clayton will hurt you. i'm thinking you're a great saint.",
+ "errors": 14,
+ "words": 14,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "she had a habit of tilting her head slightly when thinking hard about something",
+ "normalized_hypothesis": "she is looking at her because clayton will hurt you i am thinking you are a great saint",
+ "errors": 16,
+ "words": 14,
+ "wer": 1.1428571428571428
+ }
+ },
+ {
+ "id": "release500_0127",
+ "category": "emotional",
+ "text": "I cannot stress enough how much I did not want to be right about this.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0127.wav",
+ "render_seconds": 42.48695929395035,
+ "audio_seconds": 4.667210884353741,
+ "rtf": 9.103286812340691,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "I got an excursion. Let's look at your book. I want to buy this.",
+ "raw": {
+ "normalized_reference": "i cannot stress enough how much i did not want to be right about this.",
+ "normalized_hypothesis": "i got an excursion. let's look at your book. i want to buy this.",
+ "errors": 12,
+ "words": 15,
+ "wer": 0.8
+ },
+ "semantic": {
+ "normalized_reference": "i cannot stress enough how much i did not want to be right about this",
+ "normalized_hypothesis": "i got an excursion let us look at your book i want to buy this",
+ "errors": 13,
+ "words": 15,
+ "wer": 0.8666666666666667
+ }
+ },
+ {
+ "id": "release500_0128",
+ "category": "long",
+ "text": "I used to think the goal was to get to a point where nothing was hard anymore, and now I think that point doesn't exist and I'm not sure I'd want it to.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0128.wav",
+ "render_seconds": 130.00142709596548,
+ "audio_seconds": 9.264761904761905,
+ "rtf": 14.031815219034103,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "I must admit that the previous chapter looked like more or less like this for me. I know one thing about what does exist, and that's what you're going to see.",
+ "raw": {
+ "normalized_reference": "i used to think the goal was to get to a point where nothing was hard anymore, and now i think that point doesn't exist and i'm not sure i'd want it to.",
+ "normalized_hypothesis": "i must admit that the previous chapter looked like more or less like this for me. i know one thing about what does exist, and that's what you're going to see.",
+ "errors": 29,
+ "words": 33,
+ "wer": 0.8787878787878788
+ },
+ "semantic": {
+ "normalized_reference": "i used to think the goal was to get to a point where nothing was hard anymore and now i think that point does not exist and i am not sure i would want it to",
+ "normalized_hypothesis": "i must admit that the previous chapter looked like more or less like this for me i know one thing about what does exist and that is what you are going to see",
+ "errors": 31,
+ "words": 36,
+ "wer": 0.8611111111111112
+ }
+ },
+ {
+ "id": "release500_0129",
+ "category": "narrative",
+ "text": "So this is the part of the story where everything goes completely sideways.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0129.wav",
+ "render_seconds": 89.60730500996578,
+ "audio_seconds": 5.224489795918367,
+ "rtf": 17.151398224563764,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "First, look at the score. What are some clues to get the short ones?",
+ "raw": {
+ "normalized_reference": "so this is the part of the story where everything goes completely sideways.",
+ "normalized_hypothesis": "first, look at the score. what are some clues to get the short ones?",
+ "errors": 13,
+ "words": 13,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "so this is the part of the story where everything goes completely sideways",
+ "normalized_hypothesis": "1st look at the score what are some clues to get the short ones",
+ "errors": 13,
+ "words": 13,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0130",
+ "category": "question",
+ "text": "Is there a version of this where it actually works out okay?",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0130.wav",
+ "render_seconds": 74.49139826593455,
+ "audio_seconds": 4.179591836734694,
+ "rtf": 17.822648999173794,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "It's time for us to get this. We're about to have a look-see, okay?",
+ "raw": {
+ "normalized_reference": "is there a version of this where it actually works out okay?",
+ "normalized_hypothesis": "it's time for us to get this. we're about to have a look-see, okay?",
+ "errors": 13,
+ "words": 12,
+ "wer": 1.0833333333333333
+ },
+ "semantic": {
+ "normalized_reference": "is there a version of this where it actually works out okay",
+ "normalized_hypothesis": "it is time for us to get this we are about to have a look see okay",
+ "errors": 14,
+ "words": 12,
+ "wer": 1.1666666666666667
+ }
+ },
+ {
+ "id": "release500_0131",
+ "category": "short",
+ "text": "Oh no. Oh no no no.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0131.wav",
+ "render_seconds": 46.206697111018,
+ "audio_seconds": 2.089795918367347,
+ "rtf": 22.11062654726447,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "Okay. We'll be right back.",
+ "raw": {
+ "normalized_reference": "oh no. oh no no no.",
+ "normalized_hypothesis": "okay. we'll be right back.",
+ "errors": 6,
+ "words": 6,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "0 no 0 no no no",
+ "normalized_hypothesis": "okay we will be right back",
+ "errors": 6,
+ "words": 6,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0132",
+ "category": "technical",
+ "text": "Always test on a copy before modifying the original.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0132.wav",
+ "render_seconds": 73.90473467798438,
+ "audio_seconds": 3.7616326530612243,
+ "rtf": 19.646983502719372,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "We must test the equipment of your medicine.",
+ "raw": {
+ "normalized_reference": "always test on a copy before modifying the original.",
+ "normalized_hypothesis": "we must test the equipment of your medicine.",
+ "errors": 9,
+ "words": 9,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "always test on a copy before modifying the original",
+ "normalized_hypothesis": "we must test the equipment of your medicine",
+ "errors": 9,
+ "words": 9,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0133",
+ "category": "conversational",
+ "text": "I wasn't even supposed to be there. That's the wild part.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0133.wav",
+ "render_seconds": 43.90362838306464,
+ "audio_seconds": 3.9706122448979593,
+ "rtf": 11.05714325025785,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "And this one's supposed to be blue. It's the green one.",
+ "raw": {
+ "normalized_reference": "i wasn't even supposed to be there. that's the wild part.",
+ "normalized_hypothesis": "and this one's supposed to be blue. it's the green one.",
+ "errors": 7,
+ "words": 11,
+ "wer": 0.6363636363636364
+ },
+ "semantic": {
+ "normalized_reference": "i was not even supposed to be there that is the wild part",
+ "normalized_hypothesis": "and this one is supposed to be blue it is the green one",
+ "errors": 8,
+ "words": 13,
+ "wer": 0.6153846153846154
+ }
+ },
+ {
+ "id": "release500_0134",
+ "category": "descriptive",
+ "text": "It's a small town. Not the kind you leave. The kind you leave and come back to.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0134.wav",
+ "render_seconds": 97.39793498499785,
+ "audio_seconds": 5.7817687074829935,
+ "rtf": 16.845698939658654,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "It's a sweet day. I hope that you're having a laugh. That you're having a laugh until the end of the day.",
+ "raw": {
+ "normalized_reference": "it's a small town. not the kind you leave. the kind you leave and come back to.",
+ "normalized_hypothesis": "it's a sweet day. i hope that you're having a laugh. that you're having a laugh until the end of the day.",
+ "errors": 20,
+ "words": 17,
+ "wer": 1.1764705882352942
+ },
+ "semantic": {
+ "normalized_reference": "it is a small town not the kind you leave the kind you leave and come back to",
+ "normalized_hypothesis": "it is a sweet day i hope that you are having a laugh that you are having a laugh until the end of the day",
+ "errors": 20,
+ "words": 18,
+ "wer": 1.1111111111111112
+ }
+ },
+ {
+ "id": "release500_0135",
+ "category": "emotional",
+ "text": "She cried. I cried. We all just sort of stood there crying.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0135.wav",
+ "render_seconds": 79.69221897004172,
+ "audio_seconds": 4.806530612244898,
+ "rtf": 16.579987812211467,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "Succeed. Acquire. Make a certain step forward.",
+ "raw": {
+ "normalized_reference": "she cried. i cried. we all just sort of stood there crying.",
+ "normalized_hypothesis": "succeed. acquire. make a certain step forward.",
+ "errors": 12,
+ "words": 12,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "she cried i cried we all just sort of stood there crying",
+ "normalized_hypothesis": "succeed acquire make a certain step forward",
+ "errors": 12,
+ "words": 12,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0136",
+ "category": "long",
+ "text": "There's a version of being busy that's actually just a very elaborate way of avoiding the one thing you know you should be doing.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0136.wav",
+ "render_seconds": 101.89901606296189,
+ "audio_seconds": 8.150204081632653,
+ "rtf": 12.502633681603397,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "As a version of one person, it's actually just as horrible as a video game in any way. I don't know what's in that video.",
+ "raw": {
+ "normalized_reference": "there's a version of being busy that's actually just a very elaborate way of avoiding the one thing you know you should be doing.",
+ "normalized_hypothesis": "as a version of one person, it's actually just as horrible as a video game in any way. i don't know what's in that video.",
+ "errors": 19,
+ "words": 24,
+ "wer": 0.7916666666666666
+ },
+ "semantic": {
+ "normalized_reference": "there is a version of being busy that is actually just a very elaborate way of avoiding the one thing you know you should be doing",
+ "normalized_hypothesis": "as a version of one person it is actually just as horrible as a video game in any way i do not know what is in that video",
+ "errors": 22,
+ "words": 26,
+ "wer": 0.8461538461538461
+ }
+ },
+ {
+ "id": "release500_0137",
+ "category": "narrative",
+ "text": "The letter arrived on a Tuesday, which felt appropriate somehow.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0137.wav",
+ "render_seconds": 100.79218495893292,
+ "audio_seconds": 4.667210884353741,
+ "rtf": 21.595806886898234,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "The other will be interesting. What's a departure?",
+ "raw": {
+ "normalized_reference": "the letter arrived on a tuesday, which felt appropriate somehow.",
+ "normalized_hypothesis": "the other will be interesting. what's a departure?",
+ "errors": 9,
+ "words": 10,
+ "wer": 0.9
+ },
+ "semantic": {
+ "normalized_reference": "the letter arrived on a tuesday which felt appropriate somehow",
+ "normalized_hypothesis": "the other will be interesting what is a departure",
+ "errors": 9,
+ "words": 10,
+ "wer": 0.9
+ }
+ },
+ {
+ "id": "release500_0138",
+ "category": "question",
+ "text": "What exactly did she say when you told her?",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0138.wav",
+ "render_seconds": 64.09573250508402,
+ "audio_seconds": 2.9257142857142857,
+ "rtf": 21.907721071073638,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "That's not what you should say when you're talking.",
+ "raw": {
+ "normalized_reference": "what exactly did she say when you told her?",
+ "normalized_hypothesis": "that's not what you should say when you're talking.",
+ "errors": 8,
+ "words": 9,
+ "wer": 0.8888888888888888
+ },
+ "semantic": {
+ "normalized_reference": "what exactly did she say when you told her",
+ "normalized_hypothesis": "that is not what you should say when you are talking",
+ "errors": 8,
+ "words": 9,
+ "wer": 0.8888888888888888
+ }
+ },
+ {
+ "id": "release500_0139",
+ "category": "short",
+ "text": "No way. Absolutely not.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0139.wav",
+ "render_seconds": 23.296498080948368,
+ "audio_seconds": 2.5774149659863945,
+ "rtf": 9.03870676176998,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "Okay. I love you.",
+ "raw": {
+ "normalized_reference": "no way. absolutely not.",
+ "normalized_hypothesis": "okay. i love you.",
+ "errors": 4,
+ "words": 4,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "no way absolutely not",
+ "normalized_hypothesis": "okay i love you",
+ "errors": 4,
+ "words": 4,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0140",
+ "category": "technical",
+ "text": "The first thing you want to do is make sure everything is backed up.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0140.wav",
+ "render_seconds": 76.69429815898184,
+ "audio_seconds": 4.597551020408163,
+ "rtf": 16.6815545534007,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "The first thing I look into is the hole I went through as a doctor.",
+ "raw": {
+ "normalized_reference": "the first thing you want to do is make sure everything is backed up.",
+ "normalized_hypothesis": "the first thing i look into is the hole i went through as a doctor.",
+ "errors": 12,
+ "words": 14,
+ "wer": 0.8571428571428571
+ },
+ "semantic": {
+ "normalized_reference": "the 1st thing you want to do is make sure everything is backed up",
+ "normalized_hypothesis": "the 1st thing i look into is the hole i went through as a doctor",
+ "errors": 12,
+ "words": 14,
+ "wer": 0.8571428571428571
+ }
+ },
+ {
+ "id": "release500_0141",
+ "category": "conversational",
+ "text": "I have the most insane idea and you need to just hear me out.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0141.wav",
+ "render_seconds": 74.30298421508633,
+ "audio_seconds": 4.179591836734694,
+ "rtf": 17.777569465523584,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "Sparrow, the rooster is so bad now. I need to cast a spell on him.",
+ "raw": {
+ "normalized_reference": "i have the most insane idea and you need to just hear me out.",
+ "normalized_hypothesis": "sparrow, the rooster is so bad now. i need to cast a spell on him.",
+ "errors": 13,
+ "words": 14,
+ "wer": 0.9285714285714286
+ },
+ "semantic": {
+ "normalized_reference": "i have the most insane idea and you need to just hear me out",
+ "normalized_hypothesis": "sparrow the rooster is so bad now i need to cast a spell on him",
+ "errors": 13,
+ "words": 14,
+ "wer": 0.9285714285714286
+ }
+ },
+ {
+ "id": "release500_0142",
+ "category": "descriptive",
+ "text": "There's a coffee shop on the corner that's been there longer than anyone can remember.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0142.wav",
+ "render_seconds": 83.41060081892647,
+ "audio_seconds": 5.085170068027211,
+ "rtf": 16.402716074997578,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "I don't know. I don't know. I don't know. I don't know. I don't know. I don't know. I don't know. I don't know. I don't know. I don't know. I don't know. I don't know. I don't know. I don't know. I don't know. I don't know. I don't know. I don't know. I don't know. I don't know. I don't know. I don't know. I don't know. I don't know. I don't know. I don't know. I don't know. I don't know. I don't know. I don't know. I don't know. I don't know. I don't know. I don't know. I don't know. I don't know. I don't know. I don't know. I don't know. I don't know. I don't know. I don't know. I don't know. I don't know. I don't know. I don't know. I don't know. I don't know. I don't know. I don't know. I don't know. I don't know. I don't know. I don't know. I don't know. I don't know. I don't know. I don't know. I don't know. I don't know. I don't know. I don't know. I don't know. I don't know. I don't know. I don't know. I don't know. I don't know. I don't know. I don't know. I don't know. I don't know. I don't know. I don't know. I don't know. I don't know. I don't know. I don't know. I don't know. I don't know. I don't know. I don't know. I don't know. I don't know. I don't know. I don't know. I don't know. I don't know. I don't know",
+ "raw": {
+ "normalized_reference": "there's a coffee shop on the corner that's been there longer than anyone can remember.",
+ "normalized_hypothesis": "i don't know. i don't know. i don't know. i don't know. i don't know. i don't know. i don't know. i don't know. i don't know. i don't know. i don't know. i don't know. i don't know. i don't know. i don't know. i don't know. i don't know. i don't know. i don't know. i don't know. i don't know. i don't know. i don't know. i don't know. i don't know. i don't know. i don't know. i don't know. i don't know. i don't know. i don't know. i don't know. i don't know. i don't know. i don't know. i don't know. i don't know. i don't know. i don't know. i don't know. i don't know. i don't know. i don't know. i don't know. i don't know. i don't know. i don't know. i don't know. i don't know. i don't know. i don't know. i don't know. i don't know. i don't know. i don't know. i don't know. i don't know. i don't know. i don't know. i don't know. i don't know. i don't know. i don't know. i don't know. i don't know. i don't know. i don't know. i don't know. i don't know. i don't know. i don't know. i don't know. i don't know. i don't know. i don't know. i don't know. i don't know. i don't know. i don't know. i don't know. i don't know. i don't know. i don't know. i don't know. i don't know. i don't know. i don't know. i don't know. i don't know",
+ "errors": 267,
+ "words": 15,
+ "wer": 17.8
+ },
+ "semantic": {
+ "normalized_reference": "there is a coffee shop on the corner that has been there longer than anyone can remember",
+ "normalized_hypothesis": "i do not know i do not know i do not know i do not know i do not know i do not know i do not know i do not know i do not know i do not know i do not know i do not know i do not know i do not know i do not know i do not know i do not know i do not know i do not know i do not know i do not know i do not know i do not know i do not know i do not know i do not know i do not know i do not know i do not know i do not know i do not know i do not know i do not know i do not know i do not know i do not know i do not know i do not know i do not know i do not know i do not know i do not know i do not know i do not know i do not know i do not know i do not know i do not know i do not know i do not know i do not know i do not know i do not know i do not know i do not know i do not know i do not know i do not know i do not know i do not know i do not know i do not know i do not know i do not know i do not know i do not know i do not know i do not know i do not know i do not know i do not know i do not know i do not know i do not know i do not know i do not know i do not know i do not know i do not know i do not know i do not know i do not know i do not know i do not know i do not know i do not know i do not know i do not know i do not know",
+ "errors": 356,
+ "words": 17,
+ "wer": 20.941176470588236
+ }
+ },
+ {
+ "id": "release500_0143",
+ "category": "emotional",
+ "text": "I kept it together until I got to the car, and then I completely fell apart.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0143.wav",
+ "render_seconds": 96.30345823802054,
+ "audio_seconds": 5.294149659863946,
+ "rtf": 18.190543227010973,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "I thought that you had a hard time getting to the pool. I thought I could be on your side a bit.",
+ "raw": {
+ "normalized_reference": "i kept it together until i got to the car, and then i completely fell apart.",
+ "normalized_hypothesis": "i thought that you had a hard time getting to the pool. i thought i could be on your side a bit.",
+ "errors": 18,
+ "words": 16,
+ "wer": 1.125
+ },
+ "semantic": {
+ "normalized_reference": "i kept it together until i got to the car and then i completely fell apart",
+ "normalized_hypothesis": "i thought that you had a hard time getting to the pool i thought i could be on your side a bit",
+ "errors": 18,
+ "words": 16,
+ "wer": 1.125
+ }
+ },
+ {
+ "id": "release500_0144",
+ "category": "long",
+ "text": "The older I get, the more I think that most of what we call wisdom is just pattern recognition built from having been wrong enough times to start noticing.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0144.wav",
+ "render_seconds": 148.19997806602623,
+ "audio_seconds": 9.682721088435374,
+ "rtf": 15.305612617823922,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "Hi, I'm Matt. Do you want to take a risk to look to your stories? Just tell me what you feel like. I'm hoping that awareness tells the story I want to share.",
+ "raw": {
+ "normalized_reference": "the older i get, the more i think that most of what we call wisdom is just pattern recognition built from having been wrong enough times to start noticing.",
+ "normalized_hypothesis": "hi, i'm matt. do you want to take a risk to look to your stories? just tell me what you feel like. i'm hoping that awareness tells the story i want to share.",
+ "errors": 33,
+ "words": 29,
+ "wer": 1.1379310344827587
+ },
+ "semantic": {
+ "normalized_reference": "the older i get the more i think that most of what we call wisdom is just pattern recognition built from having been wrong enough times to start noticing",
+ "normalized_hypothesis": "hi i am matt do you want to take a risk to look to your stories just tell me what you feel like i am hoping that awareness tells the story i want to share",
+ "errors": 34,
+ "words": 29,
+ "wer": 1.1724137931034482
+ }
+ },
+ {
+ "id": "release500_0145",
+ "category": "narrative",
+ "text": "I remember the exact moment I realized I had made a huge mistake.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0145.wav",
+ "render_seconds": 42.1013664790662,
+ "audio_seconds": 4.527891156462585,
+ "rtf": 9.298228474192804,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "I hope that I actually be one of the writers that had made a thought mistake.",
+ "raw": {
+ "normalized_reference": "i remember the exact moment i realized i had made a huge mistake.",
+ "normalized_hypothesis": "i hope that i actually be one of the writers that had made a thought mistake.",
+ "errors": 11,
+ "words": 13,
+ "wer": 0.8461538461538461
+ },
+ "semantic": {
+ "normalized_reference": "i remember the exact moment i realized i had made a huge mistake",
+ "normalized_hypothesis": "i hope that i actually be one of the writers that had made a thought mistake",
+ "errors": 11,
+ "words": 13,
+ "wer": 0.8461538461538461
+ }
+ },
+ {
+ "id": "release500_0146",
+ "category": "question",
+ "text": "Do you think it would have gone differently if we'd left earlier?",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0146.wav",
+ "render_seconds": 79.8013374650618,
+ "audio_seconds": 3.9706122448979593,
+ "rtf": 20.09799309101577,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "Thank you for watching. Please like, share, and subscribe.",
+ "raw": {
+ "normalized_reference": "do you think it would have gone differently if we'd left earlier?",
+ "normalized_hypothesis": "thank you for watching. please like, share, and subscribe.",
+ "errors": 11,
+ "words": 12,
+ "wer": 0.9166666666666666
+ },
+ "semantic": {
+ "normalized_reference": "do you think it would have gone differently if we would left earlier",
+ "normalized_hypothesis": "thank you for watching please like share and subscribe",
+ "errors": 12,
+ "words": 13,
+ "wer": 0.9230769230769231
+ }
+ },
+ {
+ "id": "release500_0147",
+ "category": "short",
+ "text": "You have to be kidding me.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0147.wav",
+ "render_seconds": 50.69574046693742,
+ "audio_seconds": 1.9504761904761905,
+ "rtf": 25.991468501115378,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "I want to go back to heaven.",
+ "raw": {
+ "normalized_reference": "you have to be kidding me.",
+ "normalized_hypothesis": "i want to go back to heaven.",
+ "errors": 6,
+ "words": 6,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "you have to be kidding me",
+ "normalized_hypothesis": "i want to go back to heaven",
+ "errors": 6,
+ "words": 6,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0148",
+ "category": "technical",
+ "text": "If the checksum fails, restart the worker process without deleting the cached tensors.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0148.wav",
+ "render_seconds": 106.50020606000908,
+ "audio_seconds": 5.851428571428571,
+ "rtf": 18.200718809083583,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "At the exact surface, we start the work of pieces without the help of the card spaces.",
+ "raw": {
+ "normalized_reference": "if the checksum fails, restart the worker process without deleting the cached tensors.",
+ "normalized_hypothesis": "at the exact surface, we start the work of pieces without the help of the card spaces.",
+ "errors": 13,
+ "words": 13,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "if the checksum fails restart the worker process without deleting the cached tensors",
+ "normalized_hypothesis": "at the exact surface we start the work of pieces without the help of the card spaces",
+ "errors": 13,
+ "words": 13,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0149",
+ "category": "conversational",
+ "text": "I need you to understand how close that was to not working.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0149.wav",
+ "render_seconds": 68.99980113690253,
+ "audio_seconds": 3.900952380952381,
+ "rtf": 17.68793730316105,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "Thank you for watching. Please subscribe, like, and share.",
+ "raw": {
+ "normalized_reference": "i need you to understand how close that was to not working.",
+ "normalized_hypothesis": "thank you for watching. please subscribe, like, and share.",
+ "errors": 11,
+ "words": 12,
+ "wer": 0.9166666666666666
+ },
+ "semantic": {
+ "normalized_reference": "i need you to understand how close that was to not working",
+ "normalized_hypothesis": "thank you for watching please subscribe like and share",
+ "errors": 11,
+ "words": 12,
+ "wer": 0.9166666666666666
+ }
+ },
+ {
+ "id": "release500_0150",
+ "category": "descriptive",
+ "text": "It was the sort of morning where the whole world looks rinsed clean.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0150.wav",
+ "render_seconds": 87.79981729900464,
+ "audio_seconds": 4.527891156462585,
+ "rtf": 19.390885130639546,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "I wish the sun and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the",
+ "raw": {
+ "normalized_reference": "it was the sort of morning where the whole world looks rinsed clean.",
+ "normalized_hypothesis": "i wish the sun and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the",
+ "errors": 442,
+ "words": 13,
+ "wer": 34.0
+ },
+ "semantic": {
+ "normalized_reference": "it was the sort of morning where the whole world looks rinsed clean",
+ "normalized_hypothesis": "i wish the sun and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the moon and the",
+ "errors": 442,
+ "words": 13,
+ "wer": 34.0
+ }
+ },
+ {
+ "id": "release500_0151",
+ "category": "emotional",
+ "text": "She started laughing and I started laughing and we couldn't stop for a full minute.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0151.wav",
+ "render_seconds": 49.10540115297772,
+ "audio_seconds": 5.433469387755102,
+ "rtf": 9.037577585997251,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "Stop it, you asshole. Stop it, you asshole. I can't stand for fear.",
+ "raw": {
+ "normalized_reference": "she started laughing and i started laughing and we couldn't stop for a full minute.",
+ "normalized_hypothesis": "stop it, you asshole. stop it, you asshole. i can't stand for fear.",
+ "errors": 14,
+ "words": 15,
+ "wer": 0.9333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "she started laughing and i started laughing and we could not stop for a full minute",
+ "normalized_hypothesis": "stop it you asshole stop it you asshole i can not stand for fear",
+ "errors": 14,
+ "words": 16,
+ "wer": 0.875
+ }
+ },
+ {
+ "id": "release500_0152",
+ "category": "long",
+ "text": "We spent three years building something we were genuinely proud of, and then the market shifted and none of it mattered the way we thought it would.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0152.wav",
+ "render_seconds": 136.8979239440523,
+ "audio_seconds": 9.543401360544218,
+ "rtf": 14.344772767289925,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "Yes, I'm trying to ask you to show me why you're trying to rip me apart. And I know that I'm frustrated. I know that I'm not the one who wants to rip you apart.",
+ "raw": {
+ "normalized_reference": "we spent three years building something we were genuinely proud of, and then the market shifted and none of it mattered the way we thought it would.",
+ "normalized_hypothesis": "yes, i'm trying to ask you to show me why you're trying to rip me apart. and i know that i'm frustrated. i know that i'm not the one who wants to rip you apart.",
+ "errors": 33,
+ "words": 27,
+ "wer": 1.2222222222222223
+ },
+ "semantic": {
+ "normalized_reference": "we spent 3 years building something we were genuinely proud of and then the market shifted and none of it mattered the way we thought it would",
+ "normalized_hypothesis": "yes i am trying to ask you to show me why you are trying to rip me apart and i know that i am frustrated i know that i am not the one who wants to rip you apart",
+ "errors": 37,
+ "words": 27,
+ "wer": 1.3703703703703705
+ }
+ },
+ {
+ "id": "release500_0153",
+ "category": "narrative",
+ "text": "And somehow, despite all of that, it actually turned out okay.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0153.wav",
+ "render_seconds": 75.70023390301503,
+ "audio_seconds": 4.527891156462585,
+ "rtf": 16.718651417883425,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "It's so high. It's so high. It's so high. It's so high. It's so high. It's so high. It's so high. It's so high. It's so high. It's so high. It's so high. It's so high. It's so high. It's so high. It's so high. It's so high. It's so high. It's so high. It's so high. It's so high. It's so high. It's so high. It's so high. It's so high. It's so high. It's so high. It's so high. It's so high. It's so high. It's so high. It's so high. It's so high. It's so high. It's so high. It's so high. It's so high. It's so high. It's so high. It's so high. It's so high. It's so high. It's so high. It's so high. It's so high. It's so high. It's so high. It's so high. It's so high. It's so high. It's so high. It's so high. It's so high. It's so high. It's so high. It's so high. It's so high. It's so high. It's so high. It's so high. It's so high. It's so high. It's so high. It's so high. It's so high. It's so high. It's so high. It's so high. It's so high. It's so high. It's so high. It's so high. It's so high. It's so high. It's so high. It's so high. It's so high. It's so high. It's so high. It's so high. It's so high. It's so high. It's so high. It's so high. It's so high. It's so high. It's so high. It's so high. It's so high. It's so high",
+ "raw": {
+ "normalized_reference": "and somehow, despite all of that, it actually turned out okay.",
+ "normalized_hypothesis": "it's so high. it's so high. it's so high. it's so high. it's so high. it's so high. it's so high. it's so high. it's so high. it's so high. it's so high. it's so high. it's so high. it's so high. it's so high. it's so high. it's so high. it's so high. it's so high. it's so high. it's so high. it's so high. it's so high. it's so high. it's so high. it's so high. it's so high. it's so high. it's so high. it's so high. it's so high. it's so high. it's so high. it's so high. it's so high. it's so high. it's so high. it's so high. it's so high. it's so high. it's so high. it's so high. it's so high. it's so high. it's so high. it's so high. it's so high. it's so high. it's so high. it's so high. it's so high. it's so high. it's so high. it's so high. it's so high. it's so high. it's so high. it's so high. it's so high. it's so high. it's so high. it's so high. it's so high. it's so high. it's so high. it's so high. it's so high. it's so high. it's so high. it's so high. it's so high. it's so high. it's so high. it's so high. it's so high. it's so high. it's so high. it's so high. it's so high. it's so high. it's so high. it's so high. it's so high. it's so high. it's so high. it's so high. it's so high. it's so high. it's so high",
+ "errors": 267,
+ "words": 11,
+ "wer": 24.272727272727273
+ },
+ "semantic": {
+ "normalized_reference": "and somehow despite all of that it actually turned out okay",
+ "normalized_hypothesis": "it is so high it is so high it is so high it is so high it is so high it is so high it is so high it is so high it is so high it is so high it is so high it is so high it is so high it is so high it is so high it is so high it is so high it is so high it is so high it is so high it is so high it is so high it is so high it is so high it is so high it is so high it is so high it is so high it is so high it is so high it is so high it is so high it is so high it is so high it is so high it is so high it is so high it is so high it is so high it is so high it is so high it is so high it is so high it is so high it is so high it is so high it is so high it is so high it is so high it is so high it is so high it is so high it is so high it is so high it is so high it is so high it is so high it is so high it is so high it is so high it is so high it is so high it is so high it is so high it is so high it is so high it is so high it is so high it is so high it is so high it is so high it is so high it is so high it is so high it is so high it is so high it is so high it is so high it is so high it is so high it is so high it is so high it is so high it is so high it is so high it is so high it is so high it is so high it is so high",
+ "errors": 355,
+ "words": 11,
+ "wer": 32.27272727272727
+ }
+ },
+ {
+ "id": "release500_0154",
+ "category": "question",
+ "text": "Is that something you actually want, or something you think you should want?",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0154.wav",
+ "render_seconds": 92.69187824998517,
+ "audio_seconds": 4.876190476190477,
+ "rtf": 19.00907659423524,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "I said certain rocks that work. You're certain I say you should work.",
+ "raw": {
+ "normalized_reference": "is that something you actually want, or something you think you should want?",
+ "normalized_hypothesis": "i said certain rocks that work. you're certain i say you should work.",
+ "errors": 11,
+ "words": 13,
+ "wer": 0.8461538461538461
+ },
+ "semantic": {
+ "normalized_reference": "is that something you actually want or something you think you should want",
+ "normalized_hypothesis": "i said certain rocks that work you are certain i say you should work",
+ "errors": 12,
+ "words": 13,
+ "wer": 0.9230769230769231
+ }
+ },
+ {
+ "id": "release500_0155",
+ "category": "short",
+ "text": "This is fine.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0155.wav",
+ "render_seconds": 28.795869291061535,
+ "audio_seconds": 1.5325170068027212,
+ "rtf": 18.789918260769024,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "As long as I'm alive.",
+ "raw": {
+ "normalized_reference": "this is fine.",
+ "normalized_hypothesis": "as long as i'm alive.",
+ "errors": 5,
+ "words": 3,
+ "wer": 1.6666666666666667
+ },
+ "semantic": {
+ "normalized_reference": "this is fine",
+ "normalized_hypothesis": "as long as i am alive",
+ "errors": 6,
+ "words": 3,
+ "wer": 2.0
+ }
+ },
+ {
+ "id": "release500_0156",
+ "category": "technical",
+ "text": "The tricky part isn't the setup, it's making sure all dependencies line up.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0156.wav",
+ "render_seconds": 89.29555551894009,
+ "audio_seconds": 4.9458503401360545,
+ "rtf": 18.054641644592035,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "The talking boomers of the summit are tracking how we eat the onions in this unit.",
+ "raw": {
+ "normalized_reference": "the tricky part isn't the setup, it's making sure all dependencies line up.",
+ "normalized_hypothesis": "the talking boomers of the summit are tracking how we eat the onions in this unit.",
+ "errors": 14,
+ "words": 13,
+ "wer": 1.0769230769230769
+ },
+ "semantic": {
+ "normalized_reference": "the tricky part is not the setup it is making sure all dependencies line up",
+ "normalized_hypothesis": "the talking boomers of the summit are tracking how we eat the onions in this unit",
+ "errors": 15,
+ "words": 15,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0157",
+ "category": "conversational",
+ "text": "He called it a minor setback. I called it a disaster.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0157.wav",
+ "render_seconds": 61.599523954093456,
+ "audio_seconds": 4.1099319727891155,
+ "rtf": 14.987966798946866,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "I came out of the ocean, Mark. I came in a disaster.",
+ "raw": {
+ "normalized_reference": "he called it a minor setback. i called it a disaster.",
+ "normalized_hypothesis": "i came out of the ocean, mark. i came in a disaster.",
+ "errors": 9,
+ "words": 11,
+ "wer": 0.8181818181818182
+ },
+ "semantic": {
+ "normalized_reference": "he called it a minor setback i called it a disaster",
+ "normalized_hypothesis": "i came out of the ocean mark i came in a disaster",
+ "errors": 9,
+ "words": 11,
+ "wer": 0.8181818181818182
+ }
+ },
+ {
+ "id": "release500_0158",
+ "category": "descriptive",
+ "text": "The whole place smelled like pine and old wood and cold air.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0158.wav",
+ "render_seconds": 81.99821299104951,
+ "audio_seconds": 4.31891156462585,
+ "rtf": 18.985851357239603,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "Please subscribe, like, share, and leave a comment. Thank you.",
+ "raw": {
+ "normalized_reference": "the whole place smelled like pine and old wood and cold air.",
+ "normalized_hypothesis": "please subscribe, like, share, and leave a comment. thank you.",
+ "errors": 10,
+ "words": 12,
+ "wer": 0.8333333333333334
+ },
+ "semantic": {
+ "normalized_reference": "the whole place smelled like pine and old wood and cold air",
+ "normalized_hypothesis": "please subscribe like share and leave a comment thank you",
+ "errors": 10,
+ "words": 12,
+ "wer": 0.8333333333333334
+ }
+ },
+ {
+ "id": "release500_0159",
+ "category": "emotional",
+ "text": "I've never been so relieved in my entire life.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0159.wav",
+ "render_seconds": 58.11096977791749,
+ "audio_seconds": 3.2043537414965986,
+ "rtf": 18.135004579980194,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "I've never been so loud in my entire life.",
+ "raw": {
+ "normalized_reference": "i've never been so relieved in my entire life.",
+ "normalized_hypothesis": "i've never been so loud in my entire life.",
+ "errors": 1,
+ "words": 9,
+ "wer": 0.1111111111111111
+ },
+ "semantic": {
+ "normalized_reference": "i have never been so relieved in my entire life",
+ "normalized_hypothesis": "i have never been so loud in my entire life",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ }
+ },
+ {
+ "id": "release500_0160",
+ "category": "narrative",
+ "text": "I wrote it all down that night so I wouldn't forget, and I still haven't looked at it.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0160.wav",
+ "render_seconds": 99.69554700597655,
+ "audio_seconds": 5.363809523809524,
+ "rtf": 18.58670531894094,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "Stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy,",
+ "raw": {
+ "normalized_reference": "i wrote it all down that night so i wouldn't forget, and i still haven't looked at it.",
+ "normalized_hypothesis": "stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy, stay happy,",
+ "errors": 296,
+ "words": 18,
+ "wer": 16.444444444444443
+ },
+ "semantic": {
+ "normalized_reference": "i wrote it all down that night so i would not forget and i still have not looked at it",
+ "normalized_hypothesis": "stay happy stay happy stay happy stay happy stay happy stay happy stay happy stay happy stay happy stay happy stay happy stay happy stay happy stay happy stay happy stay happy stay happy stay happy stay happy stay happy stay happy stay happy stay happy stay happy stay happy stay happy stay happy stay happy stay happy stay happy stay happy stay happy stay happy stay happy stay happy stay happy stay happy stay happy stay happy stay happy stay happy stay happy stay happy stay happy stay happy stay happy stay happy stay happy stay happy stay happy stay happy stay happy stay happy stay happy stay happy stay happy stay happy stay happy stay happy stay happy stay happy stay happy stay happy stay happy stay happy stay happy stay happy stay happy stay happy stay happy stay happy stay happy stay happy stay happy stay happy stay happy stay happy stay happy stay happy stay happy stay happy stay happy stay happy stay happy stay happy stay happy stay happy stay happy stay happy stay happy stay happy stay happy stay happy stay happy stay happy stay happy stay happy stay happy stay happy stay happy stay happy stay happy stay happy stay happy stay happy stay happy stay happy stay happy stay happy stay happy stay happy stay happy stay happy stay happy stay happy stay happy stay happy stay happy stay happy stay happy stay happy stay happy stay happy stay happy stay happy stay happy stay happy stay happy stay happy stay happy stay happy stay happy stay happy stay happy stay happy stay happy stay happy stay happy stay happy stay happy stay happy stay happy stay happy stay happy stay happy stay happy stay happy stay happy",
+ "errors": 296,
+ "words": 20,
+ "wer": 14.8
+ }
+ },
+ {
+ "id": "release500_0161",
+ "category": "short",
+ "text": "That's kind of wild actually.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0161.wav",
+ "render_seconds": 44.50766643101815,
+ "audio_seconds": 2.2291156462585033,
+ "rtf": 19.966512955809534,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "Thanks for watching.",
+ "raw": {
+ "normalized_reference": "that's kind of wild actually.",
+ "normalized_hypothesis": "thanks for watching.",
+ "errors": 5,
+ "words": 5,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "that is kind of wild actually",
+ "normalized_hypothesis": "thanks for watching",
+ "errors": 6,
+ "words": 6,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0162",
+ "category": "technical",
+ "text": "Version pinning will save you a lot of pain later down the line.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0162.wav",
+ "render_seconds": 76.50037862698082,
+ "audio_seconds": 4.31891156462585,
+ "rtf": 17.71288378617405,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you.",
+ "raw": {
+ "normalized_reference": "version pinning will save you a lot of pain later down the line.",
+ "normalized_hypothesis": "i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you.",
+ "errors": 332,
+ "words": 13,
+ "wer": 25.53846153846154
+ },
+ "semantic": {
+ "normalized_reference": "version pinning will save you a lot of pain later down the line",
+ "normalized_hypothesis": "i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you",
+ "errors": 332,
+ "words": 13,
+ "wer": 25.53846153846154
+ }
+ },
+ {
+ "id": "release500_0163",
+ "category": "conversational",
+ "text": "At some point you just have to commit and see what happens.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0163.wav",
+ "render_seconds": 46.49331007502042,
+ "audio_seconds": 4.040272108843538,
+ "rtf": 11.507469997689926,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "That's so good. I'm just going to have to go now. I'll show you how it works.",
+ "raw": {
+ "normalized_reference": "at some point you just have to commit and see what happens.",
+ "normalized_hypothesis": "that's so good. i'm just going to have to go now. i'll show you how it works.",
+ "errors": 14,
+ "words": 12,
+ "wer": 1.1666666666666667
+ },
+ "semantic": {
+ "normalized_reference": "at some point you just have to commit and see what happens",
+ "normalized_hypothesis": "that is so good i am just going to have to go now i will show you how it works",
+ "errors": 17,
+ "words": 12,
+ "wer": 1.4166666666666667
+ }
+ },
+ {
+ "id": "release500_0164",
+ "category": "emotional",
+ "text": "Nobody told me it was going to be this hard, and honestly, I'm grateful.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0164.wav",
+ "render_seconds": 77.01046454103198,
+ "audio_seconds": 5.085170068027211,
+ "rtf": 15.144127632079009,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "I don't see why I was going to have a stroke. I know I should. I don't want to.",
+ "raw": {
+ "normalized_reference": "nobody told me it was going to be this hard, and honestly, i'm grateful.",
+ "normalized_hypothesis": "i don't see why i was going to have a stroke. i know i should. i don't want to.",
+ "errors": 16,
+ "words": 14,
+ "wer": 1.1428571428571428
+ },
+ "semantic": {
+ "normalized_reference": "nobody told me it was going to be this hard and honestly i am grateful",
+ "normalized_hypothesis": "i do not see why i was going to have a stroke i know i should i do not want to",
+ "errors": 17,
+ "words": 15,
+ "wer": 1.1333333333333333
+ }
+ },
+ {
+ "id": "release500_0165",
+ "category": "narrative",
+ "text": "Chapter seven. In which everything that could go wrong did, and a few things that couldn't.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0165.wav",
+ "render_seconds": 97.00171974499244,
+ "audio_seconds": 6.060408163265306,
+ "rtf": 16.005806396499967,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "Crap or something? I will let you know one thing that can do you good, and that's nothing scientific.",
+ "raw": {
+ "normalized_reference": "chapter seven. in which everything that could go wrong did, and a few things that couldn't.",
+ "normalized_hypothesis": "crap or something? i will let you know one thing that can do you good, and that's nothing scientific.",
+ "errors": 19,
+ "words": 16,
+ "wer": 1.1875
+ },
+ "semantic": {
+ "normalized_reference": "chapter 7 in which everything that could go wrong did and a few things that could not",
+ "normalized_hypothesis": "crap or something i will let you know one thing that can do you good and that is nothing scientific",
+ "errors": 19,
+ "words": 17,
+ "wer": 1.1176470588235294
+ }
+ },
+ {
+ "id": "release500_0166",
+ "category": "short",
+ "text": "Wait, what?",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0166.wav",
+ "render_seconds": 36.80749314709101,
+ "audio_seconds": 1.3931972789115645,
+ "rtf": 26.41944088194521,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "Blah, bleh.",
+ "raw": {
+ "normalized_reference": "wait, what?",
+ "normalized_hypothesis": "blah, bleh.",
+ "errors": 2,
+ "words": 2,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "wait what",
+ "normalized_hypothesis": "blah bleh",
+ "errors": 2,
+ "words": 2,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0167",
+ "category": "technical",
+ "text": "The graphics processor remained idle because the data loader exhausted every central processing core.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0167.wav",
+ "render_seconds": 116.69261429400649,
+ "audio_seconds": 7.174965986394557,
+ "rtf": 16.26385609566421,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "It was a special round of life. I kissed the cat a little and she was stuck. I just said three words to show her to me.",
+ "raw": {
+ "normalized_reference": "the graphics processor remained idle because the data loader exhausted every central processing core.",
+ "normalized_hypothesis": "it was a special round of life. i kissed the cat a little and she was stuck. i just said three words to show her to me.",
+ "errors": 26,
+ "words": 14,
+ "wer": 1.8571428571428572
+ },
+ "semantic": {
+ "normalized_reference": "the graphics processor remained idle because the data loader exhausted every central processing core",
+ "normalized_hypothesis": "it was a special round of life i kissed the cat a little and she was stuck i just said 3 words to show her to me",
+ "errors": 26,
+ "words": 14,
+ "wer": 1.8571428571428572
+ }
+ },
+ {
+ "id": "release500_0168",
+ "category": "conversational",
+ "text": "Could you speak a little more slowly? I am trying to write every detail down.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0168.wav",
+ "render_seconds": 114.00417210010346,
+ "audio_seconds": 5.712108843537415,
+ "rtf": 19.958333292105575,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "I'll try to win. I want to try hard.",
+ "raw": {
+ "normalized_reference": "could you speak a little more slowly? i am trying to write every detail down.",
+ "normalized_hypothesis": "i'll try to win. i want to try hard.",
+ "errors": 13,
+ "words": 15,
+ "wer": 0.8666666666666667
+ },
+ "semantic": {
+ "normalized_reference": "could you speak a little more slowly i am trying to write every detail down",
+ "normalized_hypothesis": "i will try to win i want to try hard",
+ "errors": 13,
+ "words": 15,
+ "wer": 0.8666666666666667
+ }
+ },
+ {
+ "id": "release500_0169",
+ "category": "emotional",
+ "text": "I was so excited I could barely form a complete sentence.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0169.wav",
+ "render_seconds": 33.42832995997742,
+ "audio_seconds": 4.31891156462585,
+ "rtf": 7.739989453308784,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry.",
+ "raw": {
+ "normalized_reference": "i was so excited i could barely form a complete sentence.",
+ "normalized_hypothesis": "i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry.",
+ "errors": 222,
+ "words": 11,
+ "wer": 20.181818181818183
+ },
+ "semantic": {
+ "normalized_reference": "i was so excited i could barely form a complete sentence",
+ "normalized_hypothesis": "i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry",
+ "errors": 331,
+ "words": 11,
+ "wer": 30.09090909090909
+ }
+ },
+ {
+ "id": "release500_0170",
+ "category": "narrative",
+ "text": "He looked at me like I had said something in a language he didn't speak.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0170.wav",
+ "render_seconds": 78.58920265594497,
+ "audio_seconds": 4.667210884353741,
+ "rtf": 16.838579743505,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "I looked at my record and said something like a year and a half and back and speak.",
+ "raw": {
+ "normalized_reference": "he looked at me like i had said something in a language he didn't speak.",
+ "normalized_hypothesis": "i looked at my record and said something like a year and a half and back and speak.",
+ "errors": 13,
+ "words": 15,
+ "wer": 0.8666666666666667
+ },
+ "semantic": {
+ "normalized_reference": "he looked at me like i had said something in a language he did not speak",
+ "normalized_hypothesis": "i looked at my record and said something like a year and a half and back and speak",
+ "errors": 13,
+ "words": 16,
+ "wer": 0.8125
+ }
+ },
+ {
+ "id": "release500_0171",
+ "category": "short",
+ "text": "That's actually hilarious.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0171.wav",
+ "render_seconds": 55.39334730489645,
+ "audio_seconds": 2.159455782312925,
+ "rtf": 25.651531167527025,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "That's actually very harsh.",
+ "raw": {
+ "normalized_reference": "that's actually hilarious.",
+ "normalized_hypothesis": "that's actually very harsh.",
+ "errors": 2,
+ "words": 3,
+ "wer": 0.6666666666666666
+ },
+ "semantic": {
+ "normalized_reference": "that is actually hilarious",
+ "normalized_hypothesis": "that is actually very harsh",
+ "errors": 2,
+ "words": 4,
+ "wer": 0.5
+ }
+ },
+ {
+ "id": "release500_0172",
+ "category": "technical",
+ "text": "The encoder maps phonemes, stress, and phrase boundaries into a shared hidden representation.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0172.wav",
+ "render_seconds": 96.20001360599417,
+ "audio_seconds": 6.617687074829932,
+ "rtf": 14.536803042846572,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "Thank you for watching this video. If you have any questions, please feel free to ask them in the comments below.",
+ "raw": {
+ "normalized_reference": "the encoder maps phonemes, stress, and phrase boundaries into a shared hidden representation.",
+ "normalized_hypothesis": "thank you for watching this video. if you have any questions, please feel free to ask them in the comments below.",
+ "errors": 21,
+ "words": 13,
+ "wer": 1.6153846153846154
+ },
+ "semantic": {
+ "normalized_reference": "the encoder maps phonemes stress and phrase boundaries into a shared hidden representation",
+ "normalized_hypothesis": "thank you for watching this video if you have any questions please feel free to ask them in the comments below",
+ "errors": 21,
+ "words": 13,
+ "wer": 1.6153846153846154
+ }
+ },
+ {
+ "id": "release500_0173",
+ "category": "conversational",
+ "text": "She showed up three hours late and acted like nothing had happened.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0173.wav",
+ "render_seconds": 78.99925330001861,
+ "audio_seconds": 4.597551020408163,
+ "rtf": 17.182898666996234,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "She fell out of her house, yeah. About ten years ago, I think it happened.",
+ "raw": {
+ "normalized_reference": "she showed up three hours late and acted like nothing had happened.",
+ "normalized_hypothesis": "she fell out of her house, yeah. about ten years ago, i think it happened.",
+ "errors": 13,
+ "words": 12,
+ "wer": 1.0833333333333333
+ },
+ "semantic": {
+ "normalized_reference": "she showed up 3 hours late and acted like nothing had happened",
+ "normalized_hypothesis": "she fell out of her house yeah about 10 years ago i think it happened",
+ "errors": 13,
+ "words": 12,
+ "wer": 1.0833333333333333
+ }
+ },
+ {
+ "id": "release500_0174",
+ "category": "emotional",
+ "text": "Oh my god. This is the best thing I've ever tasted.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0174.wav",
+ "render_seconds": 78.99571508297231,
+ "audio_seconds": 3.9706122448979593,
+ "rtf": 19.89509682907917,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "This is the best thing I've ever tasted.",
+ "raw": {
+ "normalized_reference": "oh my god. this is the best thing i've ever tasted.",
+ "normalized_hypothesis": "this is the best thing i've ever tasted.",
+ "errors": 3,
+ "words": 11,
+ "wer": 0.2727272727272727
+ },
+ "semantic": {
+ "normalized_reference": "0 my god this is the best thing i have ever tasted",
+ "normalized_hypothesis": "this is the best thing i have ever tasted",
+ "errors": 3,
+ "words": 12,
+ "wer": 0.25
+ }
+ },
+ {
+ "id": "release500_0175",
+ "category": "narrative",
+ "text": "The note was short, which was somehow more frightening than a long one would have been.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0175.wav",
+ "render_seconds": 46.504088663030416,
+ "audio_seconds": 5.572789115646258,
+ "rtf": 8.34484989436703,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "I love the shark. What was so great about the shark that you didn't like it at all?",
+ "raw": {
+ "normalized_reference": "the note was short, which was somehow more frightening than a long one would have been.",
+ "normalized_hypothesis": "i love the shark. what was so great about the shark that you didn't like it at all?",
+ "errors": 17,
+ "words": 16,
+ "wer": 1.0625
+ },
+ "semantic": {
+ "normalized_reference": "the note was short which was somehow more frightening than a long one would have been",
+ "normalized_hypothesis": "i love the shark what was so great about the shark that you did not like it at all",
+ "errors": 17,
+ "words": 16,
+ "wer": 1.0625
+ }
+ },
+ {
+ "id": "release500_0176",
+ "category": "short",
+ "text": "Noted.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0176.wav",
+ "render_seconds": 38.10715349006932,
+ "audio_seconds": 1.3235374149659864,
+ "rtf": 28.791897425165452,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "Thank you.",
+ "raw": {
+ "normalized_reference": "noted.",
+ "normalized_hypothesis": "thank you.",
+ "errors": 2,
+ "words": 1,
+ "wer": 2.0
+ },
+ "semantic": {
+ "normalized_reference": "noted",
+ "normalized_hypothesis": "thank you",
+ "errors": 2,
+ "words": 1,
+ "wer": 2.0
+ }
+ },
+ {
+ "id": "release500_0177",
+ "category": "technical",
+ "text": "It's a simple process once you understand the underlying structure.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0177.wav",
+ "render_seconds": 78.19265920203179,
+ "audio_seconds": 4.597551020408163,
+ "rtf": 17.007458722032837,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "It's a certain person's wish for us all to live our lives at home.",
+ "raw": {
+ "normalized_reference": "it's a simple process once you understand the underlying structure.",
+ "normalized_hypothesis": "it's a certain person's wish for us all to live our lives at home.",
+ "errors": 12,
+ "words": 10,
+ "wer": 1.2
+ },
+ "semantic": {
+ "normalized_reference": "it is a simple process once you understand the underlying structure",
+ "normalized_hypothesis": "it is a certain person is wish for us all to live our lives at home",
+ "errors": 13,
+ "words": 11,
+ "wer": 1.1818181818181819
+ }
+ },
+ {
+ "id": "release500_0178",
+ "category": "conversational",
+ "text": "You have to hear this story because it's almost too good to be true.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0178.wav",
+ "render_seconds": 80.99190462101251,
+ "audio_seconds": 4.458231292517007,
+ "rtf": 18.16682430921759,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "I have to go to school. I can't do this. I have to go to school.",
+ "raw": {
+ "normalized_reference": "you have to hear this story because it's almost too good to be true.",
+ "normalized_hypothesis": "i have to go to school. i can't do this. i have to go to school.",
+ "errors": 13,
+ "words": 14,
+ "wer": 0.9285714285714286
+ },
+ "semantic": {
+ "normalized_reference": "you have to hear this story because it is almost too good to be true",
+ "normalized_hypothesis": "i have to go to school i can not do this i have to go to school",
+ "errors": 14,
+ "words": 15,
+ "wer": 0.9333333333333333
+ }
+ },
+ {
+ "id": "release500_0179",
+ "category": "emotional",
+ "text": "I have never felt more seen in my entire life than in that moment.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0179.wav",
+ "render_seconds": 91.00180944101885,
+ "audio_seconds": 4.458231292517007,
+ "rtf": 20.412087994125017,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "Stand up and say your share of what I tell you. Stand up and say your share of what I tell you. Stand up and say your share of what I tell you. Stand up and say your share of what I tell you. Stand up and say your share of what I tell you. Stand up and say your share of what I tell you. Stand up and say your share of what I tell you. Stand up and say your share of what I tell you. Stand up and say your share of what I tell you. Stand up and say your share of what I tell you. Stand up and say your share of what I tell you. Stand up and say your share of what I tell you. Stand up and say your share of what I tell you. Stand up and say your share of what I tell you. Stand up and say your share of what I tell you. Stand up and say your share of what I tell you. Stand up and say your share of what I tell you. Stand up and say your share of what I tell you. Stand up and say your share of what I tell you. Stand up and say your share of what I tell you. Stand up and say your share of what I tell you. Stand up and say your share of what I tell you. Stand up and say your share of what I tell you. Stand up and say your share of what I tell you. Stand up and say your share of what I tell you. Stand up and say your share of what I tell you. Stand up and say your share of what I tell you. Stand up and say your share of what I tell you. Stand up and say your share of what I tell you. Stand up and say your share of what I tell you. Stand up and say your share of what I tell you. Stand up and say your share of what I tell you. Stand up and say your share of what I tell you. Stand up and say your share of what I tell you. Stand up and say your share of what I tell you. Stand up and say your share of what I tell you. Stand up and say your share of what I tell you.",
+ "raw": {
+ "normalized_reference": "i have never felt more seen in my entire life than in that moment.",
+ "normalized_hypothesis": "stand up and say your share of what i tell you. stand up and say your share of what i tell you. stand up and say your share of what i tell you. stand up and say your share of what i tell you. stand up and say your share of what i tell you. stand up and say your share of what i tell you. stand up and say your share of what i tell you. stand up and say your share of what i tell you. stand up and say your share of what i tell you. stand up and say your share of what i tell you. stand up and say your share of what i tell you. stand up and say your share of what i tell you. stand up and say your share of what i tell you. stand up and say your share of what i tell you. stand up and say your share of what i tell you. stand up and say your share of what i tell you. stand up and say your share of what i tell you. stand up and say your share of what i tell you. stand up and say your share of what i tell you. stand up and say your share of what i tell you. stand up and say your share of what i tell you. stand up and say your share of what i tell you. stand up and say your share of what i tell you. stand up and say your share of what i tell you. stand up and say your share of what i tell you. stand up and say your share of what i tell you. stand up and say your share of what i tell you. stand up and say your share of what i tell you. stand up and say your share of what i tell you. stand up and say your share of what i tell you. stand up and say your share of what i tell you. stand up and say your share of what i tell you. stand up and say your share of what i tell you. stand up and say your share of what i tell you. stand up and say your share of what i tell you. stand up and say your share of what i tell you. stand up and say your share of what i tell you.",
+ "errors": 406,
+ "words": 14,
+ "wer": 29.0
+ },
+ "semantic": {
+ "normalized_reference": "i have never felt more seen in my entire life than in that moment",
+ "normalized_hypothesis": "stand up and say your share of what i tell you stand up and say your share of what i tell you stand up and say your share of what i tell you stand up and say your share of what i tell you stand up and say your share of what i tell you stand up and say your share of what i tell you stand up and say your share of what i tell you stand up and say your share of what i tell you stand up and say your share of what i tell you stand up and say your share of what i tell you stand up and say your share of what i tell you stand up and say your share of what i tell you stand up and say your share of what i tell you stand up and say your share of what i tell you stand up and say your share of what i tell you stand up and say your share of what i tell you stand up and say your share of what i tell you stand up and say your share of what i tell you stand up and say your share of what i tell you stand up and say your share of what i tell you stand up and say your share of what i tell you stand up and say your share of what i tell you stand up and say your share of what i tell you stand up and say your share of what i tell you stand up and say your share of what i tell you stand up and say your share of what i tell you stand up and say your share of what i tell you stand up and say your share of what i tell you stand up and say your share of what i tell you stand up and say your share of what i tell you stand up and say your share of what i tell you stand up and say your share of what i tell you stand up and say your share of what i tell you stand up and say your share of what i tell you stand up and say your share of what i tell you stand up and say your share of what i tell you stand up and say your share of what i tell you",
+ "errors": 406,
+ "words": 14,
+ "wer": 29.0
+ }
+ },
+ {
+ "id": "release500_0180",
+ "category": "narrative",
+ "text": "She knew. She had always known. She just waited for the right moment.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0180.wav",
+ "render_seconds": 82.90699441102333,
+ "audio_seconds": 5.224489795918367,
+ "rtf": 15.868916898984937,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you.",
+ "raw": {
+ "normalized_reference": "she knew. she had always known. she just waited for the right moment.",
+ "normalized_hypothesis": "i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you.",
+ "errors": 333,
+ "words": 13,
+ "wer": 25.615384615384617
+ },
+ "semantic": {
+ "normalized_reference": "she knew she had always known she just waited for the right moment",
+ "normalized_hypothesis": "i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you",
+ "errors": 333,
+ "words": 13,
+ "wer": 25.615384615384617
+ }
+ },
+ {
+ "id": "release500_0181",
+ "category": "short",
+ "text": "Yeah, no. Hard pass.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0181.wav",
+ "render_seconds": 18.877496225060895,
+ "audio_seconds": 1.9504761904761905,
+ "rtf": 9.678403826325166,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "I'm a hot place.",
+ "raw": {
+ "normalized_reference": "yeah, no. hard pass.",
+ "normalized_hypothesis": "i'm a hot place.",
+ "errors": 4,
+ "words": 4,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "yeah no hard pass",
+ "normalized_hypothesis": "i am a hot place",
+ "errors": 5,
+ "words": 4,
+ "wer": 1.25
+ }
+ },
+ {
+ "id": "release500_0182",
+ "category": "technical",
+ "text": "Please verify the sample rate, hop length, and mel filter configuration before inference.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0182.wav",
+ "render_seconds": 104.69275171903428,
+ "audio_seconds": 6.199727891156463,
+ "rtf": 16.886668827574216,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "My old friend shall be back. But yet, in my fatal consideration, I feel no such.",
+ "raw": {
+ "normalized_reference": "please verify the sample rate, hop length, and mel filter configuration before inference.",
+ "normalized_hypothesis": "my old friend shall be back. but yet, in my fatal consideration, i feel no such.",
+ "errors": 16,
+ "words": 13,
+ "wer": 1.2307692307692308
+ },
+ "semantic": {
+ "normalized_reference": "please verify the sample rate hop length and mel filter configuration before inference",
+ "normalized_hypothesis": "my old friend shall be back but yet in my fatal consideration i feel no such",
+ "errors": 16,
+ "words": 13,
+ "wer": 1.2307692307692308
+ }
+ },
+ {
+ "id": "release500_0183",
+ "category": "conversational",
+ "text": "That's the worst idea I've ever heard and I'm completely in.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0183.wav",
+ "render_seconds": 56.61120985401794,
+ "audio_seconds": 3.8312925170068026,
+ "rtf": 14.776008253800848,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "Thank you for watching.",
+ "raw": {
+ "normalized_reference": "that's the worst idea i've ever heard and i'm completely in.",
+ "normalized_hypothesis": "thank you for watching.",
+ "errors": 11,
+ "words": 11,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "that is the worst idea i have ever heard and i am completely in",
+ "normalized_hypothesis": "thank you for watching",
+ "errors": 14,
+ "words": 14,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0184",
+ "category": "emotional",
+ "text": "I remember thinking: this is it. This is the thing I've been waiting for.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0184.wav",
+ "render_seconds": 92.70035656006075,
+ "audio_seconds": 5.154829931972789,
+ "rtf": 17.98320367178121,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "I have heard a fairy tale. That's a fact. This is the tale of a rabbit soul.",
+ "raw": {
+ "normalized_reference": "i remember thinking: this is it. this is the thing i've been waiting for.",
+ "normalized_hypothesis": "i have heard a fairy tale. that's a fact. this is the tale of a rabbit soul.",
+ "errors": 13,
+ "words": 14,
+ "wer": 0.9285714285714286
+ },
+ "semantic": {
+ "normalized_reference": "i remember thinking this is it this is the thing i have been waiting for",
+ "normalized_hypothesis": "i have heard a fairy tale that is a fact this is the tale of a rabbit soul",
+ "errors": 14,
+ "words": 15,
+ "wer": 0.9333333333333333
+ }
+ },
+ {
+ "id": "release500_0185",
+ "category": "narrative",
+ "text": "We never talked about it again, which I think was the right call.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0185.wav",
+ "render_seconds": 87.39801979297772,
+ "audio_seconds": 4.458231292517007,
+ "rtf": 19.60374284296833,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "You have it here today, Rebecca. What's your favorite story of all time?",
+ "raw": {
+ "normalized_reference": "we never talked about it again, which i think was the right call.",
+ "normalized_hypothesis": "you have it here today, rebecca. what's your favorite story of all time?",
+ "errors": 13,
+ "words": 13,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "we never talked about it again which i think was the right call",
+ "normalized_hypothesis": "you have it here today rebecca what is your favorite story of all time",
+ "errors": 14,
+ "words": 13,
+ "wer": 1.0769230769230769
+ }
+ },
+ {
+ "id": "release500_0186",
+ "category": "short",
+ "text": "In hindsight, obvious.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0186.wav",
+ "render_seconds": 46.29941581108142,
+ "audio_seconds": 1.9504761904761905,
+ "rtf": 23.737493457829828,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "And how true are you guys?",
+ "raw": {
+ "normalized_reference": "in hindsight, obvious.",
+ "normalized_hypothesis": "and how true are you guys?",
+ "errors": 6,
+ "words": 3,
+ "wer": 2.0
+ },
+ "semantic": {
+ "normalized_reference": "in hindsight obvious",
+ "normalized_hypothesis": "and how true are you guys",
+ "errors": 6,
+ "words": 3,
+ "wer": 2.0
+ }
+ },
+ {
+ "id": "release500_0187",
+ "category": "technical",
+ "text": "Once it's configured correctly, it basically runs itself.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0187.wav",
+ "render_seconds": 30.989758559037,
+ "audio_seconds": 4.179591836734694,
+ "rtf": 7.414541842738344,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "Let's go travel around the world and pass through the world of change.",
+ "raw": {
+ "normalized_reference": "once it's configured correctly, it basically runs itself.",
+ "normalized_hypothesis": "let's go travel around the world and pass through the world of change.",
+ "errors": 13,
+ "words": 8,
+ "wer": 1.625
+ },
+ "semantic": {
+ "normalized_reference": "once it is configured correctly it basically runs itself",
+ "normalized_hypothesis": "let us go travel around the world and pass through the world of change",
+ "errors": 14,
+ "words": 9,
+ "wer": 1.5555555555555556
+ }
+ },
+ {
+ "id": "release500_0188",
+ "category": "conversational",
+ "text": "You got engaged? When? How? Tell me everything right now.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0188.wav",
+ "render_seconds": 73.0189430139726,
+ "audio_seconds": 4.597551020408163,
+ "rtf": 15.882138705986584,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "I want a test. I want to know what's on your mind.",
+ "raw": {
+ "normalized_reference": "you got engaged? when? how? tell me everything right now.",
+ "normalized_hypothesis": "i want a test. i want to know what's on your mind.",
+ "errors": 12,
+ "words": 10,
+ "wer": 1.2
+ },
+ "semantic": {
+ "normalized_reference": "you got engaged when how tell me everything right now",
+ "normalized_hypothesis": "i want a test i want to know what is on your mind",
+ "errors": 13,
+ "words": 10,
+ "wer": 1.3
+ }
+ },
+ {
+ "id": "release500_0189",
+ "category": "emotional",
+ "text": "It's hard to explain how much that moment meant to me.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0189.wav",
+ "render_seconds": 66.28985990094952,
+ "audio_seconds": 3.4829931972789114,
+ "rtf": 19.03244024499918,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "It's a vortex going home, it's not me, it's my toy.",
+ "raw": {
+ "normalized_reference": "it's hard to explain how much that moment meant to me.",
+ "normalized_hypothesis": "it's a vortex going home, it's not me, it's my toy.",
+ "errors": 10,
+ "words": 11,
+ "wer": 0.9090909090909091
+ },
+ "semantic": {
+ "normalized_reference": "it is hard to explain how much that moment meant to me",
+ "normalized_hypothesis": "it is a vortex going home it is not me it is my toy",
+ "errors": 12,
+ "words": 12,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0190",
+ "category": "narrative",
+ "text": "He told the story three different times that night and somehow it got better each time.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0190.wav",
+ "render_seconds": 107.10718399297912,
+ "audio_seconds": 5.642448979591837,
+ "rtf": 18.982392996440886,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "You take the story three different times at night, and share it with an animal each time.",
+ "raw": {
+ "normalized_reference": "he told the story three different times that night and somehow it got better each time.",
+ "normalized_hypothesis": "you take the story three different times at night, and share it with an animal each time.",
+ "errors": 7,
+ "words": 16,
+ "wer": 0.4375
+ },
+ "semantic": {
+ "normalized_reference": "he told the story 3 different times that night and somehow it got better each time",
+ "normalized_hypothesis": "you take the story 3 different times at night and share it with an animal each time",
+ "errors": 7,
+ "words": 16,
+ "wer": 0.4375
+ }
+ },
+ {
+ "id": "release500_0191",
+ "category": "short",
+ "text": "Right. Moving on.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0191.wav",
+ "render_seconds": 34.410585061996244,
+ "audio_seconds": 1.811156462585034,
+ "rtf": 18.99923378948861,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "Thank you for watching.",
+ "raw": {
+ "normalized_reference": "right. moving on.",
+ "normalized_hypothesis": "thank you for watching.",
+ "errors": 4,
+ "words": 3,
+ "wer": 1.3333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "right moving on",
+ "normalized_hypothesis": "thank you for watching",
+ "errors": 4,
+ "words": 3,
+ "wer": 1.3333333333333333
+ }
+ },
+ {
+ "id": "release500_0192",
+ "category": "conversational",
+ "text": "No, I did not move your keys; check beside the small ceramic bowl.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0192.wav",
+ "render_seconds": 80.69222360302228,
+ "audio_seconds": 4.667210884353741,
+ "rtf": 17.28917454180894,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "Thank you for watching. I hope you enjoyed this video.",
+ "raw": {
+ "normalized_reference": "no, i did not move your keys; check beside the small ceramic bowl.",
+ "normalized_hypothesis": "thank you for watching. i hope you enjoyed this video.",
+ "errors": 13,
+ "words": 13,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "no i did not move your keys check beside the small ceramic bowl",
+ "normalized_hypothesis": "thank you for watching i hope you enjoyed this video",
+ "errors": 13,
+ "words": 13,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0193",
+ "category": "emotional",
+ "text": "I was devastated. I'm not going to pretend otherwise.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0193.wav",
+ "render_seconds": 26.12720021198038,
+ "audio_seconds": 4.1099319727891155,
+ "rtf": 6.35708824013691,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "I must have a stomach. Then I'll be able to tell the stories.",
+ "raw": {
+ "normalized_reference": "i was devastated. i'm not going to pretend otherwise.",
+ "normalized_hypothesis": "i must have a stomach. then i'll be able to tell the stories.",
+ "errors": 11,
+ "words": 9,
+ "wer": 1.2222222222222223
+ },
+ "semantic": {
+ "normalized_reference": "i was devastated i am not going to pretend otherwise",
+ "normalized_hypothesis": "i must have a stomach then i will be able to tell the stories",
+ "errors": 11,
+ "words": 10,
+ "wer": 1.1
+ }
+ },
+ {
+ "id": "release500_0194",
+ "category": "narrative",
+ "text": "That was the last summer before everything changed, though we didn't know it yet.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0194.wav",
+ "render_seconds": 64.37945774104446,
+ "audio_seconds": 5.015510204081632,
+ "rtf": 12.836073524215408,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "I was the last circle before I was so fast. What the hell was that?",
+ "raw": {
+ "normalized_reference": "that was the last summer before everything changed, though we didn't know it yet.",
+ "normalized_hypothesis": "i was the last circle before i was so fast. what the hell was that?",
+ "errors": 11,
+ "words": 14,
+ "wer": 0.7857142857142857
+ },
+ "semantic": {
+ "normalized_reference": "that was the last summer before everything changed though we did not know it yet",
+ "normalized_hypothesis": "i was the last circle before i was so fast what the hell was that",
+ "errors": 11,
+ "words": 15,
+ "wer": 0.7333333333333333
+ }
+ },
+ {
+ "id": "release500_0195",
+ "category": "short",
+ "text": "Bold of you to assume.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0195.wav",
+ "render_seconds": 49.709628745913506,
+ "audio_seconds": 1.8808163265306121,
+ "rtf": 26.429815630965297,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "Be as you please.",
+ "raw": {
+ "normalized_reference": "bold of you to assume.",
+ "normalized_hypothesis": "be as you please.",
+ "errors": 4,
+ "words": 5,
+ "wer": 0.8
+ },
+ "semantic": {
+ "normalized_reference": "bold of you to assume",
+ "normalized_hypothesis": "be as you please",
+ "errors": 4,
+ "words": 5,
+ "wer": 0.8
+ }
+ },
+ {
+ "id": "release500_0196",
+ "category": "conversational",
+ "text": "It's one of those decisions that makes total sense in the moment.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0196.wav",
+ "render_seconds": 79.19231077795848,
+ "audio_seconds": 4.388571428571429,
+ "rtf": 18.04512289862335,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "That's what it is to say how I feel. That's what it is to say how I feel. That's what it is to say how I feel. That's what it is to say how I feel. That's what it is to say how I feel. That's what it is to say how I feel. That's what it is to say how I feel. That's what it is to say how I feel. That's what it is to say how I feel. That's what it is to say how I feel. That's what it is to say how I feel. That's what it is to say how I feel. That's what it is to say how I feel. That's what it is to say how I feel. That's what it is to say how I feel. That's what it is to say how I feel. That's what it is to say how I feel. That's what it is to say how I feel. That's what it is to say how I feel. That's what it is to say how I feel. That's what it is to say how I feel. That's what it is to say how I feel. That's what it is to say how I feel. That's what it is to say how I feel. That's what it is to say how I feel. That's what it is to say how I feel. That's what it is to say how I feel. That's what it is to say how I feel. That's what it is to say how I feel. That's what it is to say how I feel. That's what it is to say how I feel. That's what it is to say how I feel. That's what it is to say how I feel. That's what it is to say how I feel. That's what it is to say how I feel. That's what it is to say how I feel. That's what it is to say how I feel. That's what it is to say how I feel. That's what it is to say how I feel. That's what it is to say how I feel. That's what it",
+ "raw": {
+ "normalized_reference": "it's one of those decisions that makes total sense in the moment.",
+ "normalized_hypothesis": "that's what it is to say how i feel. that's what it is to say how i feel. that's what it is to say how i feel. that's what it is to say how i feel. that's what it is to say how i feel. that's what it is to say how i feel. that's what it is to say how i feel. that's what it is to say how i feel. that's what it is to say how i feel. that's what it is to say how i feel. that's what it is to say how i feel. that's what it is to say how i feel. that's what it is to say how i feel. that's what it is to say how i feel. that's what it is to say how i feel. that's what it is to say how i feel. that's what it is to say how i feel. that's what it is to say how i feel. that's what it is to say how i feel. that's what it is to say how i feel. that's what it is to say how i feel. that's what it is to say how i feel. that's what it is to say how i feel. that's what it is to say how i feel. that's what it is to say how i feel. that's what it is to say how i feel. that's what it is to say how i feel. that's what it is to say how i feel. that's what it is to say how i feel. that's what it is to say how i feel. that's what it is to say how i feel. that's what it is to say how i feel. that's what it is to say how i feel. that's what it is to say how i feel. that's what it is to say how i feel. that's what it is to say how i feel. that's what it is to say how i feel. that's what it is to say how i feel. that's what it is to say how i feel. that's what it is to say how i feel. that's what it",
+ "errors": 363,
+ "words": 12,
+ "wer": 30.25
+ },
+ "semantic": {
+ "normalized_reference": "it is one of those decisions that makes total sense in the moment",
+ "normalized_hypothesis": "that is what it is to say how i feel that is what it is to say how i feel that is what it is to say how i feel that is what it is to say how i feel that is what it is to say how i feel that is what it is to say how i feel that is what it is to say how i feel that is what it is to say how i feel that is what it is to say how i feel that is what it is to say how i feel that is what it is to say how i feel that is what it is to say how i feel that is what it is to say how i feel that is what it is to say how i feel that is what it is to say how i feel that is what it is to say how i feel that is what it is to say how i feel that is what it is to say how i feel that is what it is to say how i feel that is what it is to say how i feel that is what it is to say how i feel that is what it is to say how i feel that is what it is to say how i feel that is what it is to say how i feel that is what it is to say how i feel that is what it is to say how i feel that is what it is to say how i feel that is what it is to say how i feel that is what it is to say how i feel that is what it is to say how i feel that is what it is to say how i feel that is what it is to say how i feel that is what it is to say how i feel that is what it is to say how i feel that is what it is to say how i feel that is what it is to say how i feel that is what it is to say how i feel that is what it is to say how i feel that is what it is to say how i feel that is what it is to say how i feel that is what it",
+ "errors": 401,
+ "words": 13,
+ "wer": 30.846153846153847
+ }
+ },
+ {
+ "id": "release500_0197",
+ "category": "emotional",
+ "text": "I was furious. Not the loud kind. The quiet kind that lasts for days.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0197.wav",
+ "render_seconds": 100.68786248005927,
+ "audio_seconds": 5.363809523809524,
+ "rtf": 18.771707316062187,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "I love you. I love you. I love you. I love you. I love you.",
+ "raw": {
+ "normalized_reference": "i was furious. not the loud kind. the quiet kind that lasts for days.",
+ "normalized_hypothesis": "i love you. i love you. i love you. i love you. i love you.",
+ "errors": 14,
+ "words": 14,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "i was furious not the loud kind the quiet kind that lasts for days",
+ "normalized_hypothesis": "i love you i love you i love you i love you i love you",
+ "errors": 14,
+ "words": 14,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0198",
+ "category": "narrative",
+ "text": "That's the part of the story I usually leave out.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0198.wav",
+ "render_seconds": 51.69975427899044,
+ "audio_seconds": 3.3436734693877552,
+ "rtf": 15.46196264447346,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "That's the end of the story. I'll see you in the next one.",
+ "raw": {
+ "normalized_reference": "that's the part of the story i usually leave out.",
+ "normalized_hypothesis": "that's the end of the story. i'll see you in the next one.",
+ "errors": 8,
+ "words": 10,
+ "wer": 0.8
+ },
+ "semantic": {
+ "normalized_reference": "that is the part of the story i usually leave out",
+ "normalized_hypothesis": "that is the end of the story i will see you in the next one",
+ "errors": 8,
+ "words": 11,
+ "wer": 0.7272727272727273
+ }
+ },
+ {
+ "id": "release500_0199",
+ "category": "short",
+ "text": "I've said too much.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0199.wav",
+ "render_seconds": 16.68404570000712,
+ "audio_seconds": 1.7414965986394557,
+ "rtf": 9.580291866800962,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "We shall try it.",
+ "raw": {
+ "normalized_reference": "i've said too much.",
+ "normalized_hypothesis": "we shall try it.",
+ "errors": 4,
+ "words": 4,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "i have said too much",
+ "normalized_hypothesis": "we shall try it",
+ "errors": 5,
+ "words": 5,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0200",
+ "category": "conversational",
+ "text": "I've been faking that I understood this the entire time.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0200.wav",
+ "render_seconds": 76.50434159603901,
+ "audio_seconds": 3.622312925170068,
+ "rtf": 21.12030163502429,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "I don't think you and I are supposed to be together.",
+ "raw": {
+ "normalized_reference": "i've been faking that i understood this the entire time.",
+ "normalized_hypothesis": "i don't think you and i are supposed to be together.",
+ "errors": 10,
+ "words": 10,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "i have been faking that i understood this the entire time",
+ "normalized_hypothesis": "i do not think you and i are supposed to be together",
+ "errors": 10,
+ "words": 11,
+ "wer": 0.9090909090909091
+ }
+ },
+ {
+ "id": "release500_0201",
+ "category": "phonetic_balance",
+ "text": "Peace be unto you and grace before the Lord.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0201.wav",
+ "render_seconds": 40.68864465598017,
+ "audio_seconds": 3.0650340136054424,
+ "rtf": 13.275103791790404,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "Yes, I can, of course. I smell the yoke.",
+ "raw": {
+ "normalized_reference": "peace be unto you and grace before the lord.",
+ "normalized_hypothesis": "yes, i can, of course. i smell the yoke.",
+ "errors": 8,
+ "words": 9,
+ "wer": 0.8888888888888888
+ },
+ "semantic": {
+ "normalized_reference": "peace be unto you and grace before the lord",
+ "normalized_hypothesis": "yes i can of course i smell the yoke",
+ "errors": 8,
+ "words": 9,
+ "wer": 0.8888888888888888
+ }
+ },
+ {
+ "id": "release500_0202",
+ "category": "phonetic_balance",
+ "text": "From now on we're pals.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0202.wav",
+ "render_seconds": 46.31092496099882,
+ "audio_seconds": 1.8808163265306121,
+ "rtf": 24.62277911337828,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "Thanks for watching!",
+ "raw": {
+ "normalized_reference": "from now on we're pals.",
+ "normalized_hypothesis": "thanks for watching!",
+ "errors": 5,
+ "words": 5,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "from now on we are pals",
+ "normalized_hypothesis": "thanks for watching",
+ "errors": 6,
+ "words": 6,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0203",
+ "category": "phonetic_balance",
+ "text": "Here we allow our solicitors to look after our legal work.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0203.wav",
+ "render_seconds": 68.31000176095404,
+ "audio_seconds": 4.179591836734694,
+ "rtf": 16.34370159319701,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "All right. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes. Yes",
+ "raw": {
+ "normalized_reference": "here we allow our solicitors to look after our legal work.",
+ "normalized_hypothesis": "all right. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes. yes",
+ "errors": 223,
+ "words": 11,
+ "wer": 20.272727272727273
+ },
+ "semantic": {
+ "normalized_reference": "here we allow our solicitors to look after our legal work",
+ "normalized_hypothesis": "all right yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes yes",
+ "errors": 223,
+ "words": 11,
+ "wer": 20.272727272727273
+ }
+ },
+ {
+ "id": "release500_0204",
+ "category": "phonetic_balance",
+ "text": "Her words sent a strange chill through Philip.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0204.wav",
+ "render_seconds": 59.90024443494622,
+ "audio_seconds": 3.2043537414965986,
+ "rtf": 18.693393197896345,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "For what sakes can I trade your faith?",
+ "raw": {
+ "normalized_reference": "her words sent a strange chill through philip.",
+ "normalized_hypothesis": "for what sakes can i trade your faith?",
+ "errors": 8,
+ "words": 8,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "her words sent a strange chill through philip",
+ "normalized_hypothesis": "for what sakes can i trade your faith",
+ "errors": 8,
+ "words": 8,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0205",
+ "category": "phonetic_balance",
+ "text": "It was introduced by Representative Dick of Ohio.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0205.wav",
+ "render_seconds": 27.99281141499523,
+ "audio_seconds": 3.7616326530612243,
+ "rtf": 7.44166536097421,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "Thank you guys for watching the video.",
+ "raw": {
+ "normalized_reference": "it was introduced by representative dick of ohio.",
+ "normalized_hypothesis": "thank you guys for watching the video.",
+ "errors": 8,
+ "words": 8,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "it was introduced by representative dick of ohio",
+ "normalized_hypothesis": "thank you guys for watching the video",
+ "errors": 8,
+ "words": 8,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0206",
+ "category": "phonetic_balance",
+ "text": "It's the aurora borealis.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0206.wav",
+ "render_seconds": 27.798951274948195,
+ "audio_seconds": 2.020136054421769,
+ "rtf": 13.760930217596258,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "I'm still on the other side.",
+ "raw": {
+ "normalized_reference": "it's the aurora borealis.",
+ "normalized_hypothesis": "i'm still on the other side.",
+ "errors": 5,
+ "words": 4,
+ "wer": 1.25
+ },
+ "semantic": {
+ "normalized_reference": "it is the aurora borealis",
+ "normalized_hypothesis": "i am still on the other side",
+ "errors": 6,
+ "words": 5,
+ "wer": 1.2
+ }
+ },
+ {
+ "id": "release500_0207",
+ "category": "phonetic_balance",
+ "text": "The boy grew and prospered.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0207.wav",
+ "render_seconds": 40.599264263990335,
+ "audio_seconds": 2.159455782312925,
+ "rtf": 18.80069256176468,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "You don't have to ask me.",
+ "raw": {
+ "normalized_reference": "the boy grew and prospered.",
+ "normalized_hypothesis": "you don't have to ask me.",
+ "errors": 6,
+ "words": 5,
+ "wer": 1.2
+ },
+ "semantic": {
+ "normalized_reference": "the boy grew and prospered",
+ "normalized_hypothesis": "you do not have to ask me",
+ "errors": 7,
+ "words": 5,
+ "wer": 1.4
+ }
+ },
+ {
+ "id": "release500_0208",
+ "category": "phonetic_balance",
+ "text": "His voice was passionately rebellious.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0208.wav",
+ "render_seconds": 64.88870784209576,
+ "audio_seconds": 2.995374149659864,
+ "rtf": 21.66297250360664,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "I love this place. I love my house. I love my house.",
+ "raw": {
+ "normalized_reference": "his voice was passionately rebellious.",
+ "normalized_hypothesis": "i love this place. i love my house. i love my house.",
+ "errors": 12,
+ "words": 5,
+ "wer": 2.4
+ },
+ "semantic": {
+ "normalized_reference": "his voice was passionately rebellious",
+ "normalized_hypothesis": "i love this place i love my house i love my house",
+ "errors": 12,
+ "words": 5,
+ "wer": 2.4
+ }
+ },
+ {
+ "id": "release500_0209",
+ "category": "phonetic_balance",
+ "text": "That's where they cut off the Scottish Chiefs and killed all hands.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0209.wav",
+ "render_seconds": 76.79028228507377,
+ "audio_seconds": 4.388571428571429,
+ "rtf": 17.497785676937383,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "I'm trying to close the show. I'm trying to close the show. I'm trying to close the show. I'm trying to close the show. I'm trying to close the show. I'm trying to close the show. I'm trying to close the show. I'm trying to close the show. I'm trying to close the show. I'm trying to close the show. I'm trying to close the show. I'm trying to close the show. I'm trying to close the show. I'm trying to close the show. I'm trying to close the show. I'm trying to close the show. I'm trying to close the show. I'm trying to close the show. I'm trying to close the show. I'm trying to close the show. I'm trying to close the show. I'm trying to close the show. I'm trying to close the show. I'm trying to close the show. I'm trying to close the show. I'm trying to close the show. I'm trying to close the show. I'm trying to close the show. I'm trying to close the show. I'm trying to close the show. I'm trying to close the show. I'm trying to close the show. I'm trying to close the show. I'm trying to close the show. I'm trying to close the show. I'm trying to close the show. I'm trying to close the show. I'm trying to close the show. I'm trying to close the show. I'm trying to close the show. I'm trying to close the show. I'm trying to close the show. I'm trying to close the show. I'm trying to close the show. I'm trying to close the show. I'm trying to close the show. I'm trying to close the show. I'm trying to close the show. I'm trying to close the show. I'm trying to close the show. I'm trying to close the show. I'm trying to close the show. I'm trying to close the show. I'm trying to close the show. I'm trying to close the show. I'm trying to",
+ "raw": {
+ "normalized_reference": "that's where they cut off the scottish chiefs and killed all hands.",
+ "normalized_hypothesis": "i'm trying to close the show. i'm trying to close the show. i'm trying to close the show. i'm trying to close the show. i'm trying to close the show. i'm trying to close the show. i'm trying to close the show. i'm trying to close the show. i'm trying to close the show. i'm trying to close the show. i'm trying to close the show. i'm trying to close the show. i'm trying to close the show. i'm trying to close the show. i'm trying to close the show. i'm trying to close the show. i'm trying to close the show. i'm trying to close the show. i'm trying to close the show. i'm trying to close the show. i'm trying to close the show. i'm trying to close the show. i'm trying to close the show. i'm trying to close the show. i'm trying to close the show. i'm trying to close the show. i'm trying to close the show. i'm trying to close the show. i'm trying to close the show. i'm trying to close the show. i'm trying to close the show. i'm trying to close the show. i'm trying to close the show. i'm trying to close the show. i'm trying to close the show. i'm trying to close the show. i'm trying to close the show. i'm trying to close the show. i'm trying to close the show. i'm trying to close the show. i'm trying to close the show. i'm trying to close the show. i'm trying to close the show. i'm trying to close the show. i'm trying to close the show. i'm trying to close the show. i'm trying to close the show. i'm trying to close the show. i'm trying to close the show. i'm trying to close the show. i'm trying to close the show. i'm trying to close the show. i'm trying to close the show. i'm trying to close the show. i'm trying to close the show. i'm trying to",
+ "errors": 332,
+ "words": 12,
+ "wer": 27.666666666666668
+ },
+ "semantic": {
+ "normalized_reference": "that is where they cut off the scottish chiefs and killed all hands",
+ "normalized_hypothesis": "i am trying to close the show i am trying to close the show i am trying to close the show i am trying to close the show i am trying to close the show i am trying to close the show i am trying to close the show i am trying to close the show i am trying to close the show i am trying to close the show i am trying to close the show i am trying to close the show i am trying to close the show i am trying to close the show i am trying to close the show i am trying to close the show i am trying to close the show i am trying to close the show i am trying to close the show i am trying to close the show i am trying to close the show i am trying to close the show i am trying to close the show i am trying to close the show i am trying to close the show i am trying to close the show i am trying to close the show i am trying to close the show i am trying to close the show i am trying to close the show i am trying to close the show i am trying to close the show i am trying to close the show i am trying to close the show i am trying to close the show i am trying to close the show i am trying to close the show i am trying to close the show i am trying to close the show i am trying to close the show i am trying to close the show i am trying to close the show i am trying to close the show i am trying to close the show i am trying to close the show i am trying to close the show i am trying to close the show i am trying to close the show i am trying to close the show i am trying to close the show i am trying to close the show i am trying to close the show i am trying to close the show i am trying to close the show i am trying to close the show i am trying to",
+ "errors": 388,
+ "words": 13,
+ "wer": 29.846153846153847
+ }
+ },
+ {
+ "id": "release500_0210",
+ "category": "phonetic_balance",
+ "text": "And Tom King patiently endured.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0210.wav",
+ "render_seconds": 65.00650330004282,
+ "audio_seconds": 2.4380952380952383,
+ "rtf": 26.662823619158186,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "Thank you. That's it. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you",
+ "raw": {
+ "normalized_reference": "and tom king patiently endured.",
+ "normalized_hypothesis": "thank you. that's it. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you",
+ "errors": 296,
+ "words": 5,
+ "wer": 59.2
+ },
+ "semantic": {
+ "normalized_reference": "and tom king patiently endured",
+ "normalized_hypothesis": "thank you that is it thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you",
+ "errors": 297,
+ "words": 5,
+ "wer": 59.4
+ }
+ },
+ {
+ "id": "release500_0211",
+ "category": "phonetic_balance",
+ "text": "From that moment his friendship for Belize turns to hatred and jealousy.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0211.wav",
+ "render_seconds": 47.60291245498229,
+ "audio_seconds": 5.154829931972789,
+ "rtf": 9.234623272384919,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "I can't believe that this one ship for the last two years came back on the moon, can you see?",
+ "raw": {
+ "normalized_reference": "from that moment his friendship for belize turns to hatred and jealousy.",
+ "normalized_hypothesis": "i can't believe that this one ship for the last two years came back on the moon, can you see?",
+ "errors": 18,
+ "words": 12,
+ "wer": 1.5
+ },
+ "semantic": {
+ "normalized_reference": "from that moment his friendship for belize turns to hatred and jealousy",
+ "normalized_hypothesis": "i can not believe that this one ship for the last 2 years came back on the moon can you see",
+ "errors": 19,
+ "words": 12,
+ "wer": 1.5833333333333333
+ }
+ },
+ {
+ "id": "release500_0212",
+ "category": "phonetic_balance",
+ "text": "Then you don't believe in altruism.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0212.wav",
+ "render_seconds": 57.117999506997876,
+ "audio_seconds": 2.4380952380952383,
+ "rtf": 23.427304485292094,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "Spareable reality TV. Spareable reality TV. Spareable reality TV. Spareable reality TV. Spareable reality TV. Spareable reality TV. Spareable reality TV. Spareable reality TV. Spareable reality TV. Spareable reality TV. Spareable reality TV. Spareable reality TV. Spareable reality TV. Spareable reality TV. Spareable reality TV. Spareable reality TV. Spareable reality TV. Spareable reality TV. Spareable reality TV. Spareable reality TV. Spareable reality TV. Spareable reality TV. Spareable reality TV. Spareable reality TV. Spareable reality TV. Spareable reality TV. Spareable reality TV. Spareable reality TV. Spareable reality TV. Spareable reality TV. Spareable reality TV. Spareable reality TV. Spareable reality TV. Spareable reality TV. Spareable reality TV. Spareable reality TV. Spareable reality TV. Spareable reality TV. Spareable reality TV. Spareable reality TV. Spareable reality TV. Spareable reality TV. Spareable reality TV. Spareable reality TV. Spareable reality TV. Spareable reality TV. Spareable reality TV. Spareable reality TV. Spareable reality TV. Spareable reality TV. Spareable reality TV. Spareable reality TV. Spareable reality TV. Spareable reality TV. Spareable reality TV. Spareable reality TV. Spareable reality TV. Spareable reality TV. Spareable reality TV. Spareal reality TV. Spareal reality TV. Spareal reality TV. Spareal reality TV. Spareal reality TV. Spareal reality TV. Spareal reality TV. Spareal reality TV. Spareal reality TV. Spareal reality TV. Spareal reality TV. Spareal reality TV. Spareal reality TV. Spareal reality TV. Spareal reality TV.",
+ "raw": {
+ "normalized_reference": "then you don't believe in altruism.",
+ "normalized_hypothesis": "spareable reality tv. spareable reality tv. spareable reality tv. spareable reality tv. spareable reality tv. spareable reality tv. spareable reality tv. spareable reality tv. spareable reality tv. spareable reality tv. spareable reality tv. spareable reality tv. spareable reality tv. spareable reality tv. spareable reality tv. spareable reality tv. spareable reality tv. spareable reality tv. spareable reality tv. spareable reality tv. spareable reality tv. spareable reality tv. spareable reality tv. spareable reality tv. spareable reality tv. spareable reality tv. spareable reality tv. spareable reality tv. spareable reality tv. spareable reality tv. spareable reality tv. spareable reality tv. spareable reality tv. spareable reality tv. spareable reality tv. spareable reality tv. spareable reality tv. spareable reality tv. spareable reality tv. spareable reality tv. spareable reality tv. spareable reality tv. spareable reality tv. spareable reality tv. spareable reality tv. spareable reality tv. spareable reality tv. spareable reality tv. spareable reality tv. spareable reality tv. spareable reality tv. spareable reality tv. spareable reality tv. spareable reality tv. spareable reality tv. spareable reality tv. spareable reality tv. spareable reality tv. spareable reality tv. spareal reality tv. spareal reality tv. spareal reality tv. spareal reality tv. spareal reality tv. spareal reality tv. spareal reality tv. spareal reality tv. spareal reality tv. spareal reality tv. spareal reality tv. spareal reality tv. spareal reality tv. spareal reality tv. spareal reality tv.",
+ "errors": 222,
+ "words": 6,
+ "wer": 37.0
+ },
+ "semantic": {
+ "normalized_reference": "then you do not believe in altruism",
+ "normalized_hypothesis": "spareable reality tv spareable reality tv spareable reality tv spareable reality tv spareable reality tv spareable reality tv spareable reality tv spareable reality tv spareable reality tv spareable reality tv spareable reality tv spareable reality tv spareable reality tv spareable reality tv spareable reality tv spareable reality tv spareable reality tv spareable reality tv spareable reality tv spareable reality tv spareable reality tv spareable reality tv spareable reality tv spareable reality tv spareable reality tv spareable reality tv spareable reality tv spareable reality tv spareable reality tv spareable reality tv spareable reality tv spareable reality tv spareable reality tv spareable reality tv spareable reality tv spareable reality tv spareable reality tv spareable reality tv spareable reality tv spareable reality tv spareable reality tv spareable reality tv spareable reality tv spareable reality tv spareable reality tv spareable reality tv spareable reality tv spareable reality tv spareable reality tv spareable reality tv spareable reality tv spareable reality tv spareable reality tv spareable reality tv spareable reality tv spareable reality tv spareable reality tv spareable reality tv spareable reality tv spareal reality tv spareal reality tv spareal reality tv spareal reality tv spareal reality tv spareal reality tv spareal reality tv spareal reality tv spareal reality tv spareal reality tv spareal reality tv spareal reality tv spareal reality tv spareal reality tv spareal reality tv",
+ "errors": 222,
+ "words": 7,
+ "wer": 31.714285714285715
+ }
+ },
+ {
+ "id": "release500_0213",
+ "category": "phonetic_balance",
+ "text": "What Pascal glimpsed with the vision of a seer, I have lived.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0213.wav",
+ "render_seconds": 64.60190822998993,
+ "audio_seconds": 4.31891156462585,
+ "rtf": 14.957914109451425,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "Look, look, what's on the left? What's on the right? What's on the left? What's on the left?",
+ "raw": {
+ "normalized_reference": "what pascal glimpsed with the vision of a seer, i have lived.",
+ "normalized_hypothesis": "look, look, what's on the left? what's on the right? what's on the left? what's on the left?",
+ "errors": 17,
+ "words": 12,
+ "wer": 1.4166666666666667
+ },
+ "semantic": {
+ "normalized_reference": "what pascal glimpsed with the vision of a seer i have lived",
+ "normalized_hypothesis": "look look what is on the left what is on the right what is on the left what is on the left",
+ "errors": 20,
+ "words": 12,
+ "wer": 1.6666666666666667
+ }
+ },
+ {
+ "id": "release500_0214",
+ "category": "phonetic_balance",
+ "text": "You were making them talk shop, Ruth charged him.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0214.wav",
+ "render_seconds": 70.70681391202379,
+ "audio_seconds": 3.274013605442177,
+ "rtf": 21.596371436726017,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "I love you. I love you. I love you. I love you. I love you. I love you.",
+ "raw": {
+ "normalized_reference": "you were making them talk shop, ruth charged him.",
+ "normalized_hypothesis": "i love you. i love you. i love you. i love you. i love you. i love you.",
+ "errors": 17,
+ "words": 9,
+ "wer": 1.8888888888888888
+ },
+ "semantic": {
+ "normalized_reference": "you were making them talk shop ruth charged him",
+ "normalized_hypothesis": "i love you i love you i love you i love you i love you i love you",
+ "errors": 17,
+ "words": 9,
+ "wer": 1.8888888888888888
+ }
+ },
+ {
+ "id": "release500_0215",
+ "category": "phonetic_balance",
+ "text": "Philip snatched at the letter which Gregson held out to him.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0215.wav",
+ "render_seconds": 73.00307296798564,
+ "audio_seconds": 3.9706122448979593,
+ "rtf": 18.385847941155923,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "Say it's hard to believe how much cocks are hiding too.",
+ "raw": {
+ "normalized_reference": "philip snatched at the letter which gregson held out to him.",
+ "normalized_hypothesis": "say it's hard to believe how much cocks are hiding too.",
+ "errors": 11,
+ "words": 11,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "philip snatched at the letter which gregson held out to him",
+ "normalized_hypothesis": "say it is hard to believe how much cocks are hiding too",
+ "errors": 12,
+ "words": 11,
+ "wer": 1.0909090909090908
+ }
+ },
+ {
+ "id": "release500_0216",
+ "category": "phonetic_balance",
+ "text": "For a time the exciting thrill of his adventure was gone.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0216.wav",
+ "render_seconds": 80.90028269705363,
+ "audio_seconds": 4.040272108843538,
+ "rtf": 20.023473795236534,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "Doctor, I'm really sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I",
+ "raw": {
+ "normalized_reference": "for a time the exciting thrill of his adventure was gone.",
+ "normalized_hypothesis": "doctor, i'm really sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i",
+ "errors": 223,
+ "words": 11,
+ "wer": 20.272727272727273
+ },
+ "semantic": {
+ "normalized_reference": "for a time the exciting thrill of his adventure was gone",
+ "normalized_hypothesis": "doctor i am really sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i",
+ "errors": 333,
+ "words": 11,
+ "wer": 30.272727272727273
+ }
+ },
+ {
+ "id": "release500_0217",
+ "category": "phonetic_balance",
+ "text": "Obviously, it was a disease that could be contracted by contact.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0217.wav",
+ "render_seconds": 47.70699213095941,
+ "audio_seconds": 5.085170068027211,
+ "rtf": 9.38159225606142,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you.",
+ "raw": {
+ "normalized_reference": "obviously, it was a disease that could be contracted by contact.",
+ "normalized_hypothesis": "thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you.",
+ "errors": 296,
+ "words": 11,
+ "wer": 26.90909090909091
+ },
+ "semantic": {
+ "normalized_reference": "obviously it was a disease that could be contracted by contact",
+ "normalized_hypothesis": "thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you",
+ "errors": 296,
+ "words": 11,
+ "wer": 26.90909090909091
+ }
+ },
+ {
+ "id": "release500_0218",
+ "category": "phonetic_balance",
+ "text": "Blind with rage, he darted in.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0218.wav",
+ "render_seconds": 44.784837095066905,
+ "audio_seconds": 2.2291156462585033,
+ "rtf": 20.090854043502304,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "Thank you.",
+ "raw": {
+ "normalized_reference": "blind with rage, he darted in.",
+ "normalized_hypothesis": "thank you.",
+ "errors": 6,
+ "words": 6,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "blind with rage he darted in",
+ "normalized_hypothesis": "thank you",
+ "errors": 6,
+ "words": 6,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0219",
+ "category": "phonetic_balance",
+ "text": "You live on an income which your father earned.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0219.wav",
+ "render_seconds": 61.30533464089967,
+ "audio_seconds": 3.1346938775510202,
+ "rtf": 19.55704034768284,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "I love you. I love you. I love you. I love you.",
+ "raw": {
+ "normalized_reference": "you live on an income which your father earned.",
+ "normalized_hypothesis": "i love you. i love you. i love you. i love you.",
+ "errors": 11,
+ "words": 9,
+ "wer": 1.2222222222222223
+ },
+ "semantic": {
+ "normalized_reference": "you live on an income which your father earned",
+ "normalized_hypothesis": "i love you i love you i love you i love you",
+ "errors": 11,
+ "words": 9,
+ "wer": 1.2222222222222223
+ }
+ },
+ {
+ "id": "release500_0220",
+ "category": "phonetic_balance",
+ "text": "Something about it seemed to fascinate him, to challenge his presence.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0220.wav",
+ "render_seconds": 90.90409748698585,
+ "audio_seconds": 4.806530612244898,
+ "rtf": 18.91262218436781,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "Thank you for watching, and I'll see you in the next video.",
+ "raw": {
+ "normalized_reference": "something about it seemed to fascinate him, to challenge his presence.",
+ "normalized_hypothesis": "thank you for watching, and i'll see you in the next video.",
+ "errors": 12,
+ "words": 11,
+ "wer": 1.0909090909090908
+ },
+ "semantic": {
+ "normalized_reference": "something about it seemed to fascinate him to challenge his presence",
+ "normalized_hypothesis": "thank you for watching and i will see you in the next video",
+ "errors": 13,
+ "words": 11,
+ "wer": 1.1818181818181819
+ }
+ },
+ {
+ "id": "release500_0221",
+ "category": "phonetic_balance",
+ "text": "Her beautiful hair was done up in shining coils.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0221.wav",
+ "render_seconds": 48.39576102490537,
+ "audio_seconds": 3.4829931972789114,
+ "rtf": 13.89487670050994,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "I'm going to see how I was doing before the crash.",
+ "raw": {
+ "normalized_reference": "her beautiful hair was done up in shining coils.",
+ "normalized_hypothesis": "i'm going to see how i was doing before the crash.",
+ "errors": 11,
+ "words": 9,
+ "wer": 1.2222222222222223
+ },
+ "semantic": {
+ "normalized_reference": "her beautiful hair was done up in shining coils",
+ "normalized_hypothesis": "i am going to see how i was doing before the crash",
+ "errors": 12,
+ "words": 9,
+ "wer": 1.3333333333333333
+ }
+ },
+ {
+ "id": "release500_0222",
+ "category": "phonetic_balance",
+ "text": "It does, was her audacious answer.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0222.wav",
+ "render_seconds": 64.49648476706352,
+ "audio_seconds": 2.716734693877551,
+ "rtf": 23.740442860472598,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "Thank you for watching.",
+ "raw": {
+ "normalized_reference": "it does, was her audacious answer.",
+ "normalized_hypothesis": "thank you for watching.",
+ "errors": 6,
+ "words": 6,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "it does was her audacious answer",
+ "normalized_hypothesis": "thank you for watching",
+ "errors": 6,
+ "words": 6,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0223",
+ "category": "phonetic_balance",
+ "text": "He caught himself with a jerk.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0223.wav",
+ "render_seconds": 16.895986302988604,
+ "audio_seconds": 2.2291156462585033,
+ "rtf": 7.579681355405654,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "You can't say it's nothing at all.",
+ "raw": {
+ "normalized_reference": "he caught himself with a jerk.",
+ "normalized_hypothesis": "you can't say it's nothing at all.",
+ "errors": 7,
+ "words": 6,
+ "wer": 1.1666666666666667
+ },
+ "semantic": {
+ "normalized_reference": "he caught himself with a jerk",
+ "normalized_hypothesis": "you can not say it is nothing at all",
+ "errors": 9,
+ "words": 6,
+ "wer": 1.5
+ }
+ },
+ {
+ "id": "release500_0224",
+ "category": "phonetic_balance",
+ "text": "He did not believe in the burning of daylight for such a luxury.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0224.wav",
+ "render_seconds": 84.80448635201901,
+ "audio_seconds": 4.388571428571429,
+ "rtf": 19.323938947400165,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "I don't know if we have any people in our town. I'm so sick of you all.",
+ "raw": {
+ "normalized_reference": "he did not believe in the burning of daylight for such a luxury.",
+ "normalized_hypothesis": "i don't know if we have any people in our town. i'm so sick of you all.",
+ "errors": 16,
+ "words": 13,
+ "wer": 1.2307692307692308
+ },
+ "semantic": {
+ "normalized_reference": "he did not believe in the burning of daylight for such a luxury",
+ "normalized_hypothesis": "i do not know if we have any people in our town i am so sick of you all",
+ "errors": 17,
+ "words": 13,
+ "wer": 1.3076923076923077
+ }
+ },
+ {
+ "id": "release500_0225",
+ "category": "phonetic_balance",
+ "text": "He had comparatively no advantages at first.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0225.wav",
+ "render_seconds": 71.69164062803611,
+ "audio_seconds": 3.4133333333333336,
+ "rtf": 21.003410340244955,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "Welcome to another episode of Ghosts Unplugged.",
+ "raw": {
+ "normalized_reference": "he had comparatively no advantages at first.",
+ "normalized_hypothesis": "welcome to another episode of ghosts unplugged.",
+ "errors": 7,
+ "words": 7,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "he had comparatively no advantages at 1st",
+ "normalized_hypothesis": "welcome to another episode of ghosts unplugged",
+ "errors": 7,
+ "words": 7,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0226",
+ "category": "phonetic_balance",
+ "text": "Such men believe, when they come together.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0226.wav",
+ "render_seconds": 55.39808167202864,
+ "audio_seconds": 2.8560544217687074,
+ "rtf": 19.3967178110428,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "So, where are we now? Where are we now? Where are we now? Where are we now?",
+ "raw": {
+ "normalized_reference": "such men believe, when they come together.",
+ "normalized_hypothesis": "so, where are we now? where are we now? where are we now? where are we now?",
+ "errors": 17,
+ "words": 7,
+ "wer": 2.4285714285714284
+ },
+ "semantic": {
+ "normalized_reference": "such men believe when they come together",
+ "normalized_hypothesis": "so where are we now where are we now where are we now where are we now",
+ "errors": 17,
+ "words": 7,
+ "wer": 2.4285714285714284
+ }
+ },
+ {
+ "id": "release500_0227",
+ "category": "phonetic_balance",
+ "text": "The ship should be in within a week or ten days.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0227.wav",
+ "render_seconds": 39.50758587103337,
+ "audio_seconds": 3.274013605442177,
+ "rtf": 12.067019454458746,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "The Super Bowl is a rock and roll show.",
+ "raw": {
+ "normalized_reference": "the ship should be in within a week or ten days.",
+ "normalized_hypothesis": "the super bowl is a rock and roll show.",
+ "errors": 9,
+ "words": 11,
+ "wer": 0.8181818181818182
+ },
+ "semantic": {
+ "normalized_reference": "the ship should be in within a week or 10 days",
+ "normalized_hypothesis": "the super bowl is a rock and roll show",
+ "errors": 9,
+ "words": 11,
+ "wer": 0.8181818181818182
+ }
+ },
+ {
+ "id": "release500_0228",
+ "category": "phonetic_balance",
+ "text": "Such a dog the wise driver kills, or turns loose.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0228.wav",
+ "render_seconds": 72.1157341429498,
+ "audio_seconds": 3.622312925170068,
+ "rtf": 19.908753228315845,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "It's a day of love, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love",
+ "raw": {
+ "normalized_reference": "such a dog the wise driver kills, or turns loose.",
+ "normalized_hypothesis": "it's a day of love, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love at face, love",
+ "errors": 332,
+ "words": 10,
+ "wer": 33.2
+ },
+ "semantic": {
+ "normalized_reference": "such a dog the wise driver kills or turns loose",
+ "normalized_hypothesis": "it is a day of love love at face love at face love at face love at face love at face love at face love at face love at face love at face love at face love at face love at face love at face love at face love at face love at face love at face love at face love at face love at face love at face love at face love at face love at face love at face love at face love at face love at face love at face love at face love at face love at face love at face love at face love at face love at face love at face love at face love at face love at face love at face love at face love at face love at face love at face love at face love at face love at face love at face love at face love at face love at face love at face love at face love at face love at face love at face love at face love at face love at face love at face love at face love at face love at face love at face love at face love at face love at face love at face love at face love at face love at face love at face love at face love at face love at face love at face love at face love at face love at face love at face love at face love at face love at face love at face love at face love at face love at face love at face love at face love at face love at face love at face love at face love at face love at face love at face love at face love at face love at face love at face love at face love at face love at face love at face love at face love at face love at face love at face love",
+ "errors": 333,
+ "words": 10,
+ "wer": 33.3
+ }
+ },
+ {
+ "id": "release500_0229",
+ "category": "phonetic_balance",
+ "text": "You have heard always how he was the lover of the Princess Naomi.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0229.wav",
+ "render_seconds": 32.687658572918735,
+ "audio_seconds": 4.458231292517007,
+ "rtf": 7.331979080534445,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "You'll have to wait until the rest of the year for it to place its value.",
+ "raw": {
+ "normalized_reference": "you have heard always how he was the lover of the princess naomi.",
+ "normalized_hypothesis": "you'll have to wait until the rest of the year for it to place its value.",
+ "errors": 14,
+ "words": 13,
+ "wer": 1.0769230769230769
+ },
+ "semantic": {
+ "normalized_reference": "you have heard always how he was the lover of the princess naomi",
+ "normalized_hypothesis": "you will have to wait until the rest of the year for it to place its value",
+ "errors": 14,
+ "words": 13,
+ "wer": 1.0769230769230769
+ }
+ },
+ {
+ "id": "release500_0230",
+ "category": "phonetic_balance",
+ "text": "I want to know how all this is possible.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0230.wav",
+ "render_seconds": 60.4882631029468,
+ "audio_seconds": 2.9257142857142857,
+ "rtf": 20.67469930276502,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "I want to go home. I want to go home. I want to go home. I want to go home. I want to go home. I want to go home. I want to go home. I want to go home. I want to go home. I want to go home. I want to go home. I want to go home. I want to go home. I want to go home. I want to go home. I want to go home. I want to go home. I want to go home. I want to go home. I want to go home. I want to go home. I want to go home. I want to go home. I want to go home. I want to go home. I want to go home. I want to go home. I want to go home. I want to go home. I want to go home. I want to go home. I want to go home. I want to go home. I want to go home. I want to go home. I want to go home. I want to go home. I want to go home. I want to go home. I want to go home. I want to go home. I want to go home. I want to go home. I want to go home. I want to go home. I want to go home. I want to go home. I want to go home. I want to go home. I want to go home. I want to go home. I want to go home. I want to go home. I want to go home. I want to go home. I want to go home. I want to go home. I want to go home. I want to go home. I want to go home. I want to go home. I want to go home. I want to go home. I want to go home. I want to go home. I want to go home. I want to go home. I want to go home. I want to go home. I want to go home. I want to go home. I want to go home. I want to go home. I want to go home.",
+ "raw": {
+ "normalized_reference": "i want to know how all this is possible.",
+ "normalized_hypothesis": "i want to go home. i want to go home. i want to go home. i want to go home. i want to go home. i want to go home. i want to go home. i want to go home. i want to go home. i want to go home. i want to go home. i want to go home. i want to go home. i want to go home. i want to go home. i want to go home. i want to go home. i want to go home. i want to go home. i want to go home. i want to go home. i want to go home. i want to go home. i want to go home. i want to go home. i want to go home. i want to go home. i want to go home. i want to go home. i want to go home. i want to go home. i want to go home. i want to go home. i want to go home. i want to go home. i want to go home. i want to go home. i want to go home. i want to go home. i want to go home. i want to go home. i want to go home. i want to go home. i want to go home. i want to go home. i want to go home. i want to go home. i want to go home. i want to go home. i want to go home. i want to go home. i want to go home. i want to go home. i want to go home. i want to go home. i want to go home. i want to go home. i want to go home. i want to go home. i want to go home. i want to go home. i want to go home. i want to go home. i want to go home. i want to go home. i want to go home. i want to go home. i want to go home. i want to go home. i want to go home. i want to go home. i want to go home. i want to go home. i want to go home.",
+ "errors": 367,
+ "words": 9,
+ "wer": 40.77777777777778
+ },
+ "semantic": {
+ "normalized_reference": "i want to know how all this is possible",
+ "normalized_hypothesis": "i want to go home i want to go home i want to go home i want to go home i want to go home i want to go home i want to go home i want to go home i want to go home i want to go home i want to go home i want to go home i want to go home i want to go home i want to go home i want to go home i want to go home i want to go home i want to go home i want to go home i want to go home i want to go home i want to go home i want to go home i want to go home i want to go home i want to go home i want to go home i want to go home i want to go home i want to go home i want to go home i want to go home i want to go home i want to go home i want to go home i want to go home i want to go home i want to go home i want to go home i want to go home i want to go home i want to go home i want to go home i want to go home i want to go home i want to go home i want to go home i want to go home i want to go home i want to go home i want to go home i want to go home i want to go home i want to go home i want to go home i want to go home i want to go home i want to go home i want to go home i want to go home i want to go home i want to go home i want to go home i want to go home i want to go home i want to go home i want to go home i want to go home i want to go home i want to go home i want to go home i want to go home i want to go home",
+ "errors": 367,
+ "words": 9,
+ "wer": 40.77777777777778
+ }
+ },
+ {
+ "id": "release500_0231",
+ "category": "phonetic_balance",
+ "text": "It was my idea to a tee.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0231.wav",
+ "render_seconds": 44.600040287012234,
+ "audio_seconds": 2.159455782312925,
+ "rtf": 20.653370470611137,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "And that's what it is about to happen.",
+ "raw": {
+ "normalized_reference": "it was my idea to a tee.",
+ "normalized_hypothesis": "and that's what it is about to happen.",
+ "errors": 8,
+ "words": 7,
+ "wer": 1.1428571428571428
+ },
+ "semantic": {
+ "normalized_reference": "it was my idea to a tee",
+ "normalized_hypothesis": "and that is what it is about to happen",
+ "errors": 9,
+ "words": 7,
+ "wer": 1.2857142857142858
+ }
+ },
+ {
+ "id": "release500_0232",
+ "category": "phonetic_balance",
+ "text": "They likewise are disinclined to being eaten.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0232.wav",
+ "render_seconds": 61.39736158796586,
+ "audio_seconds": 3.2043537414965986,
+ "rtf": 19.16060664366292,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much.",
+ "raw": {
+ "normalized_reference": "they likewise are disinclined to being eaten.",
+ "normalized_hypothesis": "i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much.",
+ "errors": 370,
+ "words": 7,
+ "wer": 52.857142857142854
+ },
+ "semantic": {
+ "normalized_reference": "they likewise are disinclined to being eaten",
+ "normalized_hypothesis": "i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much",
+ "errors": 370,
+ "words": 7,
+ "wer": 52.857142857142854
+ }
+ },
+ {
+ "id": "release500_0233",
+ "category": "phonetic_balance",
+ "text": "This state of mind comes of an undue prominence of the ego.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0233.wav",
+ "render_seconds": 76.19261164800264,
+ "audio_seconds": 4.1099319727891155,
+ "rtf": 18.53865517786081,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "A sky where I could see what the world could have meant to me.",
+ "raw": {
+ "normalized_reference": "this state of mind comes of an undue prominence of the ego.",
+ "normalized_hypothesis": "a sky where i could see what the world could have meant to me.",
+ "errors": 14,
+ "words": 12,
+ "wer": 1.1666666666666667
+ },
+ "semantic": {
+ "normalized_reference": "this state of mind comes of an undue prominence of the ego",
+ "normalized_hypothesis": "a sky where i could see what the world could have meant to me",
+ "errors": 14,
+ "words": 12,
+ "wer": 1.1666666666666667
+ }
+ },
+ {
+ "id": "release500_0234",
+ "category": "phonetic_balance",
+ "text": "Then it is as I said, Womble announced with finality.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0234.wav",
+ "render_seconds": 67.09139321104158,
+ "audio_seconds": 3.8312925170068026,
+ "rtf": 17.511425429728536,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "Get off of your side. You'll be about to quit your job.",
+ "raw": {
+ "normalized_reference": "then it is as i said, womble announced with finality.",
+ "normalized_hypothesis": "get off of your side. you'll be about to quit your job.",
+ "errors": 12,
+ "words": 10,
+ "wer": 1.2
+ },
+ "semantic": {
+ "normalized_reference": "then it is as i said womble announced with finality",
+ "normalized_hypothesis": "get off of your side you will be about to quit your job",
+ "errors": 13,
+ "words": 10,
+ "wer": 1.3
+ }
+ },
+ {
+ "id": "release500_0235",
+ "category": "phonetic_balance",
+ "text": "I can see that knife now.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0235.wav",
+ "render_seconds": 32.19995296106208,
+ "audio_seconds": 2.020136054421769,
+ "rtf": 15.939497189103333,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "Thank you for listening.",
+ "raw": {
+ "normalized_reference": "i can see that knife now.",
+ "normalized_hypothesis": "thank you for listening.",
+ "errors": 6,
+ "words": 6,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "i can see that knife now",
+ "normalized_hypothesis": "thank you for listening",
+ "errors": 6,
+ "words": 6,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0236",
+ "category": "phonetic_balance",
+ "text": "If I meet her again I shall apologize, said Eileen.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0236.wav",
+ "render_seconds": 73.49967968300916,
+ "audio_seconds": 3.7616326530612243,
+ "rtf": 19.539302867119407,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "It's like we've all ever seen the X-ray of Superman.",
+ "raw": {
+ "normalized_reference": "if i meet her again i shall apologize, said eileen.",
+ "normalized_hypothesis": "it's like we've all ever seen the x-ray of superman.",
+ "errors": 11,
+ "words": 10,
+ "wer": 1.1
+ },
+ "semantic": {
+ "normalized_reference": "if i meet her again i shall apologize said eileen",
+ "normalized_hypothesis": "it is like we have all ever seen the x ray of superman",
+ "errors": 13,
+ "words": 10,
+ "wer": 1.3
+ }
+ },
+ {
+ "id": "release500_0237",
+ "category": "phonetic_balance",
+ "text": "A wildly exciting time was his during the week preceding Thursday the eighteenth.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0237.wav",
+ "render_seconds": 76.90761393296998,
+ "audio_seconds": 5.572789115646258,
+ "rtf": 13.800560605647691,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "Good night, Secretary. This is the only way possible for us to return to Earth.",
+ "raw": {
+ "normalized_reference": "a wildly exciting time was his during the week preceding thursday the eighteenth.",
+ "normalized_hypothesis": "good night, secretary. this is the only way possible for us to return to earth.",
+ "errors": 15,
+ "words": 13,
+ "wer": 1.1538461538461537
+ },
+ "semantic": {
+ "normalized_reference": "a wildly exciting time was his during the week preceding thursday the 18th",
+ "normalized_hypothesis": "good night secretary this is the only way possible for us to return to earth",
+ "errors": 15,
+ "words": 13,
+ "wer": 1.1538461538461537
+ }
+ },
+ {
+ "id": "release500_0238",
+ "category": "phonetic_balance",
+ "text": "In what bucolic school of fence he had been taught was beyond imagining.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0238.wav",
+ "render_seconds": 74.69169386499561,
+ "audio_seconds": 4.876190476190477,
+ "rtf": 15.317632530907302,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "I hope that the experience I shared with you today was helpful.",
+ "raw": {
+ "normalized_reference": "in what bucolic school of fence he had been taught was beyond imagining.",
+ "normalized_hypothesis": "i hope that the experience i shared with you today was helpful.",
+ "errors": 12,
+ "words": 13,
+ "wer": 0.9230769230769231
+ },
+ "semantic": {
+ "normalized_reference": "in what bucolic school of fence he had been taught was beyond imagining",
+ "normalized_hypothesis": "i hope that the experience i shared with you today was helpful",
+ "errors": 12,
+ "words": 13,
+ "wer": 0.9230769230769231
+ }
+ },
+ {
+ "id": "release500_0239",
+ "category": "phonetic_balance",
+ "text": "So she said, the irate skipper dashed on.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0239.wav",
+ "render_seconds": 64.50826901104301,
+ "audio_seconds": 3.0650340136054424,
+ "rtf": 21.046509997832302,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "Thank you for watching. I hope you enjoyed the video.",
+ "raw": {
+ "normalized_reference": "so she said, the irate skipper dashed on.",
+ "normalized_hypothesis": "thank you for watching. i hope you enjoyed the video.",
+ "errors": 10,
+ "words": 8,
+ "wer": 1.25
+ },
+ "semantic": {
+ "normalized_reference": "so she said the irate skipper dashed on",
+ "normalized_hypothesis": "thank you for watching i hope you enjoyed the video",
+ "errors": 10,
+ "words": 8,
+ "wer": 1.25
+ }
+ },
+ {
+ "id": "release500_0240",
+ "category": "phonetic_balance",
+ "text": "It is not an attempt to smash the market.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0240.wav",
+ "render_seconds": 66.30265488405712,
+ "audio_seconds": 3.0650340136054424,
+ "rtf": 21.631947505229927,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "I guess I'll let you know just how strong I am.",
+ "raw": {
+ "normalized_reference": "it is not an attempt to smash the market.",
+ "normalized_hypothesis": "i guess i'll let you know just how strong i am.",
+ "errors": 11,
+ "words": 9,
+ "wer": 1.2222222222222223
+ },
+ "semantic": {
+ "normalized_reference": "it is not an attempt to smash the market",
+ "normalized_hypothesis": "i guess i will let you know just how strong i am",
+ "errors": 12,
+ "words": 9,
+ "wer": 1.3333333333333333
+ }
+ },
+ {
+ "id": "release500_0241",
+ "category": "phonetic_balance",
+ "text": "He destroyed everything that had belonged to the woman.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0241.wav",
+ "render_seconds": 42.7055435749935,
+ "audio_seconds": 3.622312925170068,
+ "rtf": 11.789578773895816,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "I'm a squirrel. I'm a squirrel. I'm a squirrel. I'm a squirrel. I'm a squirrel. I'm a squirrel. I'm a squirrel. I'm a squirrel.",
+ "raw": {
+ "normalized_reference": "he destroyed everything that had belonged to the woman.",
+ "normalized_hypothesis": "i'm a squirrel. i'm a squirrel. i'm a squirrel. i'm a squirrel. i'm a squirrel. i'm a squirrel. i'm a squirrel. i'm a squirrel.",
+ "errors": 24,
+ "words": 9,
+ "wer": 2.6666666666666665
+ },
+ "semantic": {
+ "normalized_reference": "he destroyed everything that had belonged to the woman",
+ "normalized_hypothesis": "i am a squirrel i am a squirrel i am a squirrel i am a squirrel i am a squirrel i am a squirrel i am a squirrel i am a squirrel",
+ "errors": 32,
+ "words": 9,
+ "wer": 3.5555555555555554
+ }
+ },
+ {
+ "id": "release500_0242",
+ "category": "phonetic_balance",
+ "text": "Did Thorpe go to see any one in Churchill.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0242.wav",
+ "render_seconds": 53.50194810505491,
+ "audio_seconds": 2.995374149659864,
+ "rtf": 17.86152428107529,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "Thank you for watching.",
+ "raw": {
+ "normalized_reference": "did thorpe go to see any one in churchill.",
+ "normalized_hypothesis": "thank you for watching.",
+ "errors": 9,
+ "words": 9,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "did thorpe go to see any one in churchill",
+ "normalized_hypothesis": "thank you for watching",
+ "errors": 9,
+ "words": 9,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0243",
+ "category": "phonetic_balance",
+ "text": "Many other similar disconcerting omissions will be noticed in the Manuscript.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0243.wav",
+ "render_seconds": 98.99978925101459,
+ "audio_seconds": 5.712108843537415,
+ "rtf": 17.33156562011617,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "I would like to show you a discussion in the lessons. You can always turn on your screen.",
+ "raw": {
+ "normalized_reference": "many other similar disconcerting omissions will be noticed in the manuscript.",
+ "normalized_hypothesis": "i would like to show you a discussion in the lessons. you can always turn on your screen.",
+ "errors": 16,
+ "words": 11,
+ "wer": 1.4545454545454546
+ },
+ "semantic": {
+ "normalized_reference": "many other similar disconcerting omissions will be noticed in the manuscript",
+ "normalized_hypothesis": "i would like to show you a discussion in the lessons you can always turn on your screen",
+ "errors": 16,
+ "words": 11,
+ "wer": 1.4545454545454546
+ }
+ },
+ {
+ "id": "release500_0244",
+ "category": "phonetic_balance",
+ "text": "The ringing of the big bell aroused him.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0244.wav",
+ "render_seconds": 65.50034994003363,
+ "audio_seconds": 2.716734693877551,
+ "rtf": 24.109954530210693,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "Do you want to go to bed by your last?",
+ "raw": {
+ "normalized_reference": "the ringing of the big bell aroused him.",
+ "normalized_hypothesis": "do you want to go to bed by your last?",
+ "errors": 10,
+ "words": 8,
+ "wer": 1.25
+ },
+ "semantic": {
+ "normalized_reference": "the ringing of the big bell aroused him",
+ "normalized_hypothesis": "do you want to go to bed by your last",
+ "errors": 10,
+ "words": 8,
+ "wer": 1.25
+ }
+ },
+ {
+ "id": "release500_0245",
+ "category": "phonetic_balance",
+ "text": "A sinewy hand, dripping with water, was clutching the rail.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0245.wav",
+ "render_seconds": 74.70232620602474,
+ "audio_seconds": 4.249251700680272,
+ "rtf": 17.580113268899904,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry.",
+ "raw": {
+ "normalized_reference": "a sinewy hand, dripping with water, was clutching the rail.",
+ "normalized_hypothesis": "i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry.",
+ "errors": 222,
+ "words": 10,
+ "wer": 22.2
+ },
+ "semantic": {
+ "normalized_reference": "a sinewy hand dripping with water was clutching the rail",
+ "normalized_hypothesis": "i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry",
+ "errors": 333,
+ "words": 10,
+ "wer": 33.3
+ }
+ },
+ {
+ "id": "release500_0246",
+ "category": "phonetic_balance",
+ "text": "Enters now the psychology of the situation.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0246.wav",
+ "render_seconds": 56.38914462400135,
+ "audio_seconds": 3.1346938775510202,
+ "rtf": 17.9887245219796,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "I can't stand the thought of you having sex with me.",
+ "raw": {
+ "normalized_reference": "enters now the psychology of the situation.",
+ "normalized_hypothesis": "i can't stand the thought of you having sex with me.",
+ "errors": 9,
+ "words": 7,
+ "wer": 1.2857142857142858
+ },
+ "semantic": {
+ "normalized_reference": "enters now the psychology of the situation",
+ "normalized_hypothesis": "i can not stand the thought of you having sex with me",
+ "errors": 10,
+ "words": 7,
+ "wer": 1.4285714285714286
+ }
+ },
+ {
+ "id": "release500_0247",
+ "category": "phonetic_balance",
+ "text": "Wash your hands of me.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0247.wav",
+ "render_seconds": 54.69748455809895,
+ "audio_seconds": 1.8808163265306121,
+ "rtf": 29.081778899162856,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "I'll see you next time.",
+ "raw": {
+ "normalized_reference": "wash your hands of me.",
+ "normalized_hypothesis": "i'll see you next time.",
+ "errors": 5,
+ "words": 5,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "wash your hands of me",
+ "normalized_hypothesis": "i will see you next time",
+ "errors": 6,
+ "words": 5,
+ "wer": 1.2
+ }
+ },
+ {
+ "id": "release500_0248",
+ "category": "phonetic_balance",
+ "text": "He confessed that the sketch had startled him.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0248.wav",
+ "render_seconds": 62.39778457500506,
+ "audio_seconds": 3.1346938775510202,
+ "rtf": 19.905543256349272,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "I'm terrified that this can't look scary to you.",
+ "raw": {
+ "normalized_reference": "he confessed that the sketch had startled him.",
+ "normalized_hypothesis": "i'm terrified that this can't look scary to you.",
+ "errors": 8,
+ "words": 8,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "he confessed that the sketch had startled him",
+ "normalized_hypothesis": "i am terrified that this can not look scary to you",
+ "errors": 10,
+ "words": 8,
+ "wer": 1.25
+ }
+ },
+ {
+ "id": "release500_0249",
+ "category": "phonetic_balance",
+ "text": "Her face was against his breast.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0249.wav",
+ "render_seconds": 54.099360143067315,
+ "audio_seconds": 2.507755102040816,
+ "rtf": 21.572824275799956,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "This is a case of death.",
+ "raw": {
+ "normalized_reference": "her face was against his breast.",
+ "normalized_hypothesis": "this is a case of death.",
+ "errors": 6,
+ "words": 6,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "her face was against his breast",
+ "normalized_hypothesis": "this is a case of death",
+ "errors": 6,
+ "words": 6,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0250",
+ "category": "phonetic_balance",
+ "text": "He was a pariah; a wanderer without a friend or a home.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0250.wav",
+ "render_seconds": 60.31028752902057,
+ "audio_seconds": 3.900952380952381,
+ "rtf": 15.460400855828027,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "Hello, my friends. I'm here to welcome you to another lesson.",
+ "raw": {
+ "normalized_reference": "he was a pariah; a wanderer without a friend or a home.",
+ "normalized_hypothesis": "hello, my friends. i'm here to welcome you to another lesson.",
+ "errors": 12,
+ "words": 12,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "he was a pariah a wanderer without a friend or a home",
+ "normalized_hypothesis": "hello my friends i am here to welcome you to another lesson",
+ "errors": 12,
+ "words": 12,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0251",
+ "category": "phonetic_balance",
+ "text": "A cry of joy burst from Philip's lips.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0251.wav",
+ "render_seconds": 52.593778323964216,
+ "audio_seconds": 2.786394557823129,
+ "rtf": 18.875208529352392,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "I can't wrap my head around my neck.",
+ "raw": {
+ "normalized_reference": "a cry of joy burst from philip's lips.",
+ "normalized_hypothesis": "i can't wrap my head around my neck.",
+ "errors": 8,
+ "words": 8,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "a cry of joy burst from philip is lips",
+ "normalized_hypothesis": "i can not wrap my head around my neck",
+ "errors": 9,
+ "words": 9,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0252",
+ "category": "phonetic_balance",
+ "text": "And the big Persian knew of his existence before he did of hers.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0252.wav",
+ "render_seconds": 74.51021319197025,
+ "audio_seconds": 4.31891156462585,
+ "rtf": 17.25208124247043,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "I look back all the same now with my subsistence with you in the back of my mind.",
+ "raw": {
+ "normalized_reference": "and the big persian knew of his existence before he did of hers.",
+ "normalized_hypothesis": "i look back all the same now with my subsistence with you in the back of my mind.",
+ "errors": 16,
+ "words": 13,
+ "wer": 1.2307692307692308
+ },
+ "semantic": {
+ "normalized_reference": "and the big persian knew of his existence before he did of hers",
+ "normalized_hypothesis": "i look back all the same now with my subsistence with you in the back of my mind",
+ "errors": 16,
+ "words": 13,
+ "wer": 1.2307692307692308
+ }
+ },
+ {
+ "id": "release500_0253",
+ "category": "phonetic_balance",
+ "text": "The men stared into each other's face.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0253.wav",
+ "render_seconds": 21.394839835935272,
+ "audio_seconds": 2.5774149659863945,
+ "rtf": 8.300890666919566,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "Let's all enter into the space.",
+ "raw": {
+ "normalized_reference": "the men stared into each other's face.",
+ "normalized_hypothesis": "let's all enter into the space.",
+ "errors": 6,
+ "words": 7,
+ "wer": 0.8571428571428571
+ },
+ "semantic": {
+ "normalized_reference": "the men stared into each other is face",
+ "normalized_hypothesis": "let us all enter into the space",
+ "errors": 8,
+ "words": 8,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0254",
+ "category": "phonetic_balance",
+ "text": "It is the fire, partly, she said.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0254.wav",
+ "render_seconds": 48.41107422008645,
+ "audio_seconds": 2.716734693877551,
+ "rtf": 17.81958110565081,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "I have to fly home now, for fuck's sake.",
+ "raw": {
+ "normalized_reference": "it is the fire, partly, she said.",
+ "normalized_hypothesis": "i have to fly home now, for fuck's sake.",
+ "errors": 9,
+ "words": 7,
+ "wer": 1.2857142857142858
+ },
+ "semantic": {
+ "normalized_reference": "it is the fire partly she said",
+ "normalized_hypothesis": "i have to fly home now for fuck is sake",
+ "errors": 10,
+ "words": 7,
+ "wer": 1.4285714285714286
+ }
+ },
+ {
+ "id": "release500_0255",
+ "category": "phonetic_balance",
+ "text": "It seemed nearer to him since he had seen and talked with Gregson.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0255.wav",
+ "render_seconds": 81.30002614797559,
+ "audio_seconds": 4.597551020408163,
+ "rtf": 17.683333102143127,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "I don't know. I don't know. I don't know. I don't know. I don't know. I don't know. I don't know. I don't know. I don't know.",
+ "raw": {
+ "normalized_reference": "it seemed nearer to him since he had seen and talked with gregson.",
+ "normalized_hypothesis": "i don't know. i don't know. i don't know. i don't know. i don't know. i don't know. i don't know. i don't know. i don't know.",
+ "errors": 27,
+ "words": 13,
+ "wer": 2.076923076923077
+ },
+ "semantic": {
+ "normalized_reference": "it seemed nearer to him since he had seen and talked with gregson",
+ "normalized_hypothesis": "i do not know i do not know i do not know i do not know i do not know i do not know i do not know i do not know i do not know",
+ "errors": 36,
+ "words": 13,
+ "wer": 2.769230769230769
+ }
+ },
+ {
+ "id": "release500_0256",
+ "category": "phonetic_balance",
+ "text": "Among my minor afflictions, I may mention a new and mysterious one.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0256.wav",
+ "render_seconds": 70.59623351204209,
+ "audio_seconds": 4.9458503401360545,
+ "rtf": 14.273831324645395,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "Remember, all creatures, about what you know is true.",
+ "raw": {
+ "normalized_reference": "among my minor afflictions, i may mention a new and mysterious one.",
+ "normalized_hypothesis": "remember, all creatures, about what you know is true.",
+ "errors": 12,
+ "words": 12,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "among my minor afflictions i may mention a new and mysterious one",
+ "normalized_hypothesis": "remember all creatures about what you know is true",
+ "errors": 12,
+ "words": 12,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0257",
+ "category": "phonetic_balance",
+ "text": "Between him and all domestic animals there must be no hostilities.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0257.wav",
+ "render_seconds": 83.69677852105815,
+ "audio_seconds": 4.667210884353741,
+ "rtf": 17.932932664697336,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "Thank you for watching. Please subscribe to my channel.",
+ "raw": {
+ "normalized_reference": "between him and all domestic animals there must be no hostilities.",
+ "normalized_hypothesis": "thank you for watching. please subscribe to my channel.",
+ "errors": 11,
+ "words": 11,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "between him and all domestic animals there must be no hostilities",
+ "normalized_hypothesis": "thank you for watching please subscribe to my channel",
+ "errors": 11,
+ "words": 11,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0258",
+ "category": "phonetic_balance",
+ "text": "With these arguments he convinced himself that he should go on alone.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0258.wav",
+ "render_seconds": 74.49195452709682,
+ "audio_seconds": 4.597551020408163,
+ "rtf": 16.202529171820597,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "Thank you for watching. Please subscribe to my channel.",
+ "raw": {
+ "normalized_reference": "with these arguments he convinced himself that he should go on alone.",
+ "normalized_hypothesis": "thank you for watching. please subscribe to my channel.",
+ "errors": 12,
+ "words": 12,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "with these arguments he convinced himself that he should go on alone",
+ "normalized_hypothesis": "thank you for watching please subscribe to my channel",
+ "errors": 12,
+ "words": 12,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0259",
+ "category": "phonetic_balance",
+ "text": "We are consumed in our own flesh-pots.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0259.wav",
+ "render_seconds": 34.19644083199091,
+ "audio_seconds": 2.8560544217687074,
+ "rtf": 11.973315554265112,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "I can show you all your stress points.",
+ "raw": {
+ "normalized_reference": "we are consumed in our own flesh-pots.",
+ "normalized_hypothesis": "i can show you all your stress points.",
+ "errors": 8,
+ "words": 8,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "we are consumed in our own flesh pots",
+ "normalized_hypothesis": "i can show you all your stress points",
+ "errors": 8,
+ "words": 8,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0260",
+ "category": "phonetic_balance",
+ "text": "Mercedes screamed, cried, laughed, and manifested the chaotic abandonment of hysteria.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0260.wav",
+ "render_seconds": 91.18836701591499,
+ "audio_seconds": 6.199727891156463,
+ "rtf": 14.708446663601105,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "Why are you so loud? Why are you so loud? Why are you so loud? Why are you so loud? Why are you so loud? Why are you so loud? Why are you so loud? Why are you so loud? Why are you so loud? Why are you so loud? Why are you so loud? Why are you so loud? Why are you so loud? Why are you so loud? Why are you so loud? Why are you so loud? Why are you so loud? Why are you so loud? Why are you so loud? Why are you so loud? Why are you so loud? Why are you so loud? Why are you so loud? Why are you so loud? Why are you so loud? Why are you so loud? Why are you so loud? Why are you so loud? Why are you so loud? Why are you so loud? Why are you so loud? Why are you so loud? Why are you so loud? Why are you so loud? Why are you so loud? Why are you so loud? Why are you so loud? Why are you so loud? Why are you so loud? Why are you so loud? Why are you so loud? Why are you so loud? Why are you so loud? Why are you so loud? Why are you so loud? Why are you so loud? Why are you so loud? Why are you so loud? Why are you so loud? Why are you so loud? Why are you so loud? Why are you so loud? Why are you so loud? Why are you so loud? Why are you so loud? Why are you so loud? Why are you so loud? Why are you so loud? Why are you so loud? Why are you so loud? Why are you so loud? Why are you so loud? Why are you so loud? Why are you so loud? Why are you so loud? Why are you so loud? Why are you so loud? Why are you so loud? Why are you so loud? Why are you so loud? Why are you so loud? Why are you so loud? Why are you so loud? Why are you so loud?",
+ "raw": {
+ "normalized_reference": "mercedes screamed, cried, laughed, and manifested the chaotic abandonment of hysteria.",
+ "normalized_hypothesis": "why are you so loud? why are you so loud? why are you so loud? why are you so loud? why are you so loud? why are you so loud? why are you so loud? why are you so loud? why are you so loud? why are you so loud? why are you so loud? why are you so loud? why are you so loud? why are you so loud? why are you so loud? why are you so loud? why are you so loud? why are you so loud? why are you so loud? why are you so loud? why are you so loud? why are you so loud? why are you so loud? why are you so loud? why are you so loud? why are you so loud? why are you so loud? why are you so loud? why are you so loud? why are you so loud? why are you so loud? why are you so loud? why are you so loud? why are you so loud? why are you so loud? why are you so loud? why are you so loud? why are you so loud? why are you so loud? why are you so loud? why are you so loud? why are you so loud? why are you so loud? why are you so loud? why are you so loud? why are you so loud? why are you so loud? why are you so loud? why are you so loud? why are you so loud? why are you so loud? why are you so loud? why are you so loud? why are you so loud? why are you so loud? why are you so loud? why are you so loud? why are you so loud? why are you so loud? why are you so loud? why are you so loud? why are you so loud? why are you so loud? why are you so loud? why are you so loud? why are you so loud? why are you so loud? why are you so loud? why are you so loud? why are you so loud? why are you so loud? why are you so loud? why are you so loud? why are you so loud?",
+ "errors": 370,
+ "words": 11,
+ "wer": 33.63636363636363
+ },
+ "semantic": {
+ "normalized_reference": "mercedes screamed cried laughed and manifested the chaotic abandonment of hysteria",
+ "normalized_hypothesis": "why are you so loud why are you so loud why are you so loud why are you so loud why are you so loud why are you so loud why are you so loud why are you so loud why are you so loud why are you so loud why are you so loud why are you so loud why are you so loud why are you so loud why are you so loud why are you so loud why are you so loud why are you so loud why are you so loud why are you so loud why are you so loud why are you so loud why are you so loud why are you so loud why are you so loud why are you so loud why are you so loud why are you so loud why are you so loud why are you so loud why are you so loud why are you so loud why are you so loud why are you so loud why are you so loud why are you so loud why are you so loud why are you so loud why are you so loud why are you so loud why are you so loud why are you so loud why are you so loud why are you so loud why are you so loud why are you so loud why are you so loud why are you so loud why are you so loud why are you so loud why are you so loud why are you so loud why are you so loud why are you so loud why are you so loud why are you so loud why are you so loud why are you so loud why are you so loud why are you so loud why are you so loud why are you so loud why are you so loud why are you so loud why are you so loud why are you so loud why are you so loud why are you so loud why are you so loud why are you so loud why are you so loud why are you so loud why are you so loud why are you so loud",
+ "errors": 370,
+ "words": 11,
+ "wer": 33.63636363636363
+ }
+ },
+ {
+ "id": "release500_0261",
+ "category": "phonetic_balance",
+ "text": "Philip made no effort to follow.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0261.wav",
+ "render_seconds": 63.59963294793852,
+ "audio_seconds": 2.2987755102040817,
+ "rtf": 27.666743736230355,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "So that way it will lie for too long.",
+ "raw": {
+ "normalized_reference": "philip made no effort to follow.",
+ "normalized_hypothesis": "so that way it will lie for too long.",
+ "errors": 9,
+ "words": 6,
+ "wer": 1.5
+ },
+ "semantic": {
+ "normalized_reference": "philip made no effort to follow",
+ "normalized_hypothesis": "so that way it will lie for too long",
+ "errors": 9,
+ "words": 6,
+ "wer": 1.5
+ }
+ },
+ {
+ "id": "release500_0262",
+ "category": "phonetic_balance",
+ "text": "She said with chattering teeth.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0262.wav",
+ "render_seconds": 51.599916751962155,
+ "audio_seconds": 2.2987755102040817,
+ "rtf": 22.446696740466493,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "She said let Charlie attend.",
+ "raw": {
+ "normalized_reference": "she said with chattering teeth.",
+ "normalized_hypothesis": "she said let charlie attend.",
+ "errors": 3,
+ "words": 5,
+ "wer": 0.6
+ },
+ "semantic": {
+ "normalized_reference": "she said with chattering teeth",
+ "normalized_hypothesis": "she said let charlie attend",
+ "errors": 3,
+ "words": 5,
+ "wer": 0.6
+ }
+ },
+ {
+ "id": "release500_0263",
+ "category": "phonetic_balance",
+ "text": "Halfway around the track one donkey got into an argument with its rider.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0263.wav",
+ "render_seconds": 91.50479103601538,
+ "audio_seconds": 5.015510204081632,
+ "rtf": 18.244363447122208,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "Thank you for watching. Please subscribe to my channel. Thank you for watching. Please subscribe to my channel. Thank you for watching. Please subscribe to my channel. Thank you for watching. Please subscribe to my channel. Thank you for watching. Please subscribe to my channel. Thank you for watching. Please subscribe to my channel. Thank you for watching. Please subscribe to my channel. Thank you for watching. Please subscribe to my channel. Thank you for watching. Please subscribe to my channel. Thank you for watching. Please subscribe to my channel. Thank you for watching. Please subscribe to my channel. Thank you for watching. Please subscribe to my channel. Thank you for watching. Please subscribe to my channel. Thank you for watching. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe",
+ "raw": {
+ "normalized_reference": "halfway around the track one donkey got into an argument with its rider.",
+ "normalized_hypothesis": "thank you for watching. please subscribe to my channel. thank you for watching. please subscribe to my channel. thank you for watching. please subscribe to my channel. thank you for watching. please subscribe to my channel. thank you for watching. please subscribe to my channel. thank you for watching. please subscribe to my channel. thank you for watching. please subscribe to my channel. thank you for watching. please subscribe to my channel. thank you for watching. please subscribe to my channel. thank you for watching. please subscribe to my channel. thank you for watching. please subscribe to my channel. thank you for watching. please subscribe to my channel. thank you for watching. please subscribe to my channel. thank you for watching. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe",
+ "errors": 368,
+ "words": 13,
+ "wer": 28.307692307692307
+ },
+ "semantic": {
+ "normalized_reference": "halfway around the track one donkey got into an argument with its rider",
+ "normalized_hypothesis": "thank you for watching please subscribe to my channel thank you for watching please subscribe to my channel thank you for watching please subscribe to my channel thank you for watching please subscribe to my channel thank you for watching please subscribe to my channel thank you for watching please subscribe to my channel thank you for watching please subscribe to my channel thank you for watching please subscribe to my channel thank you for watching please subscribe to my channel thank you for watching please subscribe to my channel thank you for watching please subscribe to my channel thank you for watching please subscribe to my channel thank you for watching please subscribe to my channel thank you for watching please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe",
+ "errors": 368,
+ "words": 13,
+ "wer": 28.307692307692307
+ }
+ },
+ {
+ "id": "release500_0264",
+ "category": "phonetic_balance",
+ "text": "Earth and gravel seemed to fill the pan.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0264.wav",
+ "render_seconds": 53.298670324031264,
+ "audio_seconds": 2.8560544217687074,
+ "rtf": 18.66164381105325,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "I think I'll be sure to hide the power.",
+ "raw": {
+ "normalized_reference": "earth and gravel seemed to fill the pan.",
+ "normalized_hypothesis": "i think i'll be sure to hide the power.",
+ "errors": 7,
+ "words": 8,
+ "wer": 0.875
+ },
+ "semantic": {
+ "normalized_reference": "earth and gravel seemed to fill the pan",
+ "normalized_hypothesis": "i think i will be sure to hide the power",
+ "errors": 8,
+ "words": 8,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0265",
+ "category": "phonetic_balance",
+ "text": "Her achievements with cocoanuts were a revelation.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0265.wav",
+ "render_seconds": 40.108290101983584,
+ "audio_seconds": 3.5526530612244898,
+ "rtf": 11.289672650489406,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "All travelers must pay clearance for what they have seen.",
+ "raw": {
+ "normalized_reference": "her achievements with cocoanuts were a revelation.",
+ "normalized_hypothesis": "all travelers must pay clearance for what they have seen.",
+ "errors": 10,
+ "words": 7,
+ "wer": 1.4285714285714286
+ },
+ "semantic": {
+ "normalized_reference": "her achievements with cocoanuts were a revelation",
+ "normalized_hypothesis": "all travelers must pay clearance for what they have seen",
+ "errors": 10,
+ "words": 7,
+ "wer": 1.4285714285714286
+ }
+ },
+ {
+ "id": "release500_0266",
+ "category": "phonetic_balance",
+ "text": "For a full minute he crouched and listened.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0266.wav",
+ "render_seconds": 55.399573264992796,
+ "audio_seconds": 3.1346938775510202,
+ "rtf": 17.67304095042088,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "So, see, I think that's all that's going to happen.",
+ "raw": {
+ "normalized_reference": "for a full minute he crouched and listened.",
+ "normalized_hypothesis": "so, see, i think that's all that's going to happen.",
+ "errors": 10,
+ "words": 8,
+ "wer": 1.25
+ },
+ "semantic": {
+ "normalized_reference": "for a full minute he crouched and listened",
+ "normalized_hypothesis": "so see i think that is all that is going to happen",
+ "errors": 12,
+ "words": 8,
+ "wer": 1.5
+ }
+ },
+ {
+ "id": "release500_0267",
+ "category": "phonetic_balance",
+ "text": "It was edged with ice.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0267.wav",
+ "render_seconds": 46.69210641994141,
+ "audio_seconds": 1.9504761904761905,
+ "rtf": 23.938824092255118,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "Thank you for watching!",
+ "raw": {
+ "normalized_reference": "it was edged with ice.",
+ "normalized_hypothesis": "thank you for watching!",
+ "errors": 5,
+ "words": 5,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "it was edged with ice",
+ "normalized_hypothesis": "thank you for watching",
+ "errors": 5,
+ "words": 5,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0268",
+ "category": "phonetic_balance",
+ "text": "The Portuguese boy passed the Hawaiian.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0268.wav",
+ "render_seconds": 52.094459216925316,
+ "audio_seconds": 2.8560544217687074,
+ "rtf": 18.240009300895633,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "The book is left us alone.",
+ "raw": {
+ "normalized_reference": "the portuguese boy passed the hawaiian.",
+ "normalized_hypothesis": "the book is left us alone.",
+ "errors": 5,
+ "words": 6,
+ "wer": 0.8333333333333334
+ },
+ "semantic": {
+ "normalized_reference": "the portuguese boy passed the hawaiian",
+ "normalized_hypothesis": "the book is left us alone",
+ "errors": 5,
+ "words": 6,
+ "wer": 0.8333333333333334
+ }
+ },
+ {
+ "id": "release500_0269",
+ "category": "phonetic_balance",
+ "text": "But he did not broach it, preferring to mature it carefully.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0269.wav",
+ "render_seconds": 64.79198672692291,
+ "audio_seconds": 4.1099319727891155,
+ "rtf": 15.764734588284012,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "Thank you for watching, and I'll see you in the next video.",
+ "raw": {
+ "normalized_reference": "but he did not broach it, preferring to mature it carefully.",
+ "normalized_hypothesis": "thank you for watching, and i'll see you in the next video.",
+ "errors": 12,
+ "words": 11,
+ "wer": 1.0909090909090908
+ },
+ "semantic": {
+ "normalized_reference": "but he did not broach it preferring to mature it carefully",
+ "normalized_hypothesis": "thank you for watching and i will see you in the next video",
+ "errors": 13,
+ "words": 11,
+ "wer": 1.1818181818181819
+ }
+ },
+ {
+ "id": "release500_0270",
+ "category": "phonetic_balance",
+ "text": "Come on, Del Mar challenged.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0270.wav",
+ "render_seconds": 39.909629823989235,
+ "audio_seconds": 2.159455782312925,
+ "rtf": 18.481336895559533,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "Happy Valentine's Day.",
+ "raw": {
+ "normalized_reference": "come on, del mar challenged.",
+ "normalized_hypothesis": "happy valentine's day.",
+ "errors": 5,
+ "words": 5,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "come on del mar challenged",
+ "normalized_hypothesis": "happy valentine is day",
+ "errors": 5,
+ "words": 5,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0271",
+ "category": "phonetic_balance",
+ "text": "For a full half minute Jeanne looked at him without speaking.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0271.wav",
+ "render_seconds": 58.31023634294979,
+ "audio_seconds": 4.179591836734694,
+ "rtf": 13.951179593772165,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "For she has the power to do what she wants.",
+ "raw": {
+ "normalized_reference": "for a full half minute jeanne looked at him without speaking.",
+ "normalized_hypothesis": "for she has the power to do what she wants.",
+ "errors": 10,
+ "words": 11,
+ "wer": 0.9090909090909091
+ },
+ "semantic": {
+ "normalized_reference": "for a full half minute jeanne looked at him without speaking",
+ "normalized_hypothesis": "for she has the power to do what she wants",
+ "errors": 10,
+ "words": 11,
+ "wer": 0.9090909090909091
+ }
+ },
+ {
+ "id": "release500_0272",
+ "category": "phonetic_balance",
+ "text": "He bore no grudges and had few enemies.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0272.wav",
+ "render_seconds": 29.90095866797492,
+ "audio_seconds": 2.786394557823129,
+ "rtf": 10.73105694382889,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "I'll be on my way to show you what's going on.",
+ "raw": {
+ "normalized_reference": "he bore no grudges and had few enemies.",
+ "normalized_hypothesis": "i'll be on my way to show you what's going on.",
+ "errors": 11,
+ "words": 8,
+ "wer": 1.375
+ },
+ "semantic": {
+ "normalized_reference": "he bore no grudges and had few enemies",
+ "normalized_hypothesis": "i will be on my way to show you what is going on",
+ "errors": 13,
+ "words": 8,
+ "wer": 1.625
+ }
+ },
+ {
+ "id": "release500_0273",
+ "category": "phonetic_balance",
+ "text": "I'll go over tomorrow afternoon.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0273.wav",
+ "render_seconds": 53.60737494705245,
+ "audio_seconds": 2.2987755102040817,
+ "rtf": 23.31996957036195,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "I love you, too. I love you, too. I love you, too. I love you, too.",
+ "raw": {
+ "normalized_reference": "i'll go over tomorrow afternoon.",
+ "normalized_hypothesis": "i love you, too. i love you, too. i love you, too. i love you, too.",
+ "errors": 16,
+ "words": 5,
+ "wer": 3.2
+ },
+ "semantic": {
+ "normalized_reference": "i will go over tomorrow afternoon",
+ "normalized_hypothesis": "i love you too i love you too i love you too i love you too",
+ "errors": 15,
+ "words": 6,
+ "wer": 2.5
+ }
+ },
+ {
+ "id": "release500_0274",
+ "category": "phonetic_balance",
+ "text": "Well, there are better men in Hawaii, that's all.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0274.wav",
+ "render_seconds": 56.00135361694265,
+ "audio_seconds": 3.274013605442177,
+ "rtf": 17.104801740547227,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "Stay right there, but don't run away. You are the truth.",
+ "raw": {
+ "normalized_reference": "well, there are better men in hawaii, that's all.",
+ "normalized_hypothesis": "stay right there, but don't run away. you are the truth.",
+ "errors": 10,
+ "words": 9,
+ "wer": 1.1111111111111112
+ },
+ "semantic": {
+ "normalized_reference": "well there are better men in hawaii that is all",
+ "normalized_hypothesis": "stay right there but do not run away you are the truth",
+ "errors": 11,
+ "words": 10,
+ "wer": 1.1
+ }
+ },
+ {
+ "id": "release500_0275",
+ "category": "phonetic_balance",
+ "text": "He was an athlete and a giant.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0275.wav",
+ "render_seconds": 52.600108951097354,
+ "audio_seconds": 2.36843537414966,
+ "rtf": 22.208800596884508,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "And that's how I start to look at life.",
+ "raw": {
+ "normalized_reference": "he was an athlete and a giant.",
+ "normalized_hypothesis": "and that's how i start to look at life.",
+ "errors": 9,
+ "words": 7,
+ "wer": 1.2857142857142858
+ },
+ "semantic": {
+ "normalized_reference": "he was an athlete and a giant",
+ "normalized_hypothesis": "and that is how i start to look at life",
+ "errors": 10,
+ "words": 7,
+ "wer": 1.4285714285714286
+ }
+ },
+ {
+ "id": "release500_0276",
+ "category": "phonetic_balance",
+ "text": "McVeigh when he returns from a trip to Honolulu.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0276.wav",
+ "render_seconds": 55.38810772099532,
+ "audio_seconds": 3.3436734693877552,
+ "rtf": 16.565046864799626,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "It's all right. I'll try and find a trick to find.",
+ "raw": {
+ "normalized_reference": "mcveigh when he returns from a trip to honolulu.",
+ "normalized_hypothesis": "it's all right. i'll try and find a trick to find.",
+ "errors": 9,
+ "words": 9,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "mcveigh when he returns from a trip to honolulu",
+ "normalized_hypothesis": "it is all right i will try and find a trick to find",
+ "errors": 11,
+ "words": 9,
+ "wer": 1.2222222222222223
+ }
+ },
+ {
+ "id": "release500_0277",
+ "category": "phonetic_balance",
+ "text": "You're a devil for fighting, and will surely win.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0277.wav",
+ "render_seconds": 32.88942808797583,
+ "audio_seconds": 3.274013605442177,
+ "rtf": 10.045599087708707,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "Oh, that is a fair game. I wish I had that.",
+ "raw": {
+ "normalized_reference": "you're a devil for fighting, and will surely win.",
+ "normalized_hypothesis": "oh, that is a fair game. i wish i had that.",
+ "errors": 10,
+ "words": 9,
+ "wer": 1.1111111111111112
+ },
+ "semantic": {
+ "normalized_reference": "you are a devil for fighting and will surely win",
+ "normalized_hypothesis": "0 that is a fair game i wish i had that",
+ "errors": 10,
+ "words": 10,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0278",
+ "category": "phonetic_balance",
+ "text": "All an appearance can know is mirage.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0278.wav",
+ "render_seconds": 31.810034321970306,
+ "audio_seconds": 2.8560544217687074,
+ "rtf": 11.137754966962737,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "They have brought you out of the woods.",
+ "raw": {
+ "normalized_reference": "all an appearance can know is mirage.",
+ "normalized_hypothesis": "they have brought you out of the woods.",
+ "errors": 8,
+ "words": 7,
+ "wer": 1.1428571428571428
+ },
+ "semantic": {
+ "normalized_reference": "all an appearance can know is mirage",
+ "normalized_hypothesis": "they have brought you out of the woods",
+ "errors": 8,
+ "words": 7,
+ "wer": 1.1428571428571428
+ }
+ },
+ {
+ "id": "release500_0279",
+ "category": "phonetic_balance",
+ "text": "You fired me out of your house, in short.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0279.wav",
+ "render_seconds": 66.29974531300832,
+ "audio_seconds": 2.9257142857142857,
+ "rtf": 22.66104576128214,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry.",
+ "raw": {
+ "normalized_reference": "you fired me out of your house, in short.",
+ "normalized_hypothesis": "i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry.",
+ "errors": 222,
+ "words": 9,
+ "wer": 24.666666666666668
+ },
+ "semantic": {
+ "normalized_reference": "you fired me out of your house in short",
+ "normalized_hypothesis": "i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry",
+ "errors": 333,
+ "words": 9,
+ "wer": 37.0
+ }
+ },
+ {
+ "id": "release500_0280",
+ "category": "phonetic_balance",
+ "text": "He made sure that the magazine was loaded, and resumed his paddling.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0280.wav",
+ "render_seconds": 70.5980139960302,
+ "audio_seconds": 4.876190476190477,
+ "rtf": 14.47810833902963,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "I'm not sure what the rest of this will look like. I'm not sure what the rest of this will look like. I'm not sure what the rest of this will look like. I'm not sure what the rest of this will look like. I'm not sure what the rest of this will look like. I'm not sure what the rest of this will look like. I'm not sure what the rest of this will look like. I'm not sure what the rest of this will look like. I'm not sure what the rest of this will look like. I'm not sure what the rest of this will look like. I'm not sure what the rest of this will look like. I'm not sure what the rest of this will look like. I'm not sure what the rest of this will look like. I'm not sure what the rest of this will look like. I'm not sure what the rest of this will look like. I'm not sure what the rest of this will look like. I'm not sure what the rest of this will look like. I'm not sure what the rest of this will look like. I'm not sure what the rest of this will look like. I'm not sure what the rest of this will look like. I'm not sure what the rest of this will look like. I'm not sure what the rest of this will look like. I'm not sure what the rest of this will look like. I'm not sure what the rest of this will look like. I'm not sure what the rest of this will look like. I'm not sure what the rest of this will look like. I'm not sure what the rest of this will look like. I'm not sure what the rest of this will look like. I'm not sure what the rest of this will look like. I'm not sure what the rest of this will look like. I'm not sure what the rest of this will look like. I'm not sure what the rest of this will look like. I'm not sure what the rest of this will look like. I'm not sure what the rest of this will look like. I'm",
+ "raw": {
+ "normalized_reference": "he made sure that the magazine was loaded, and resumed his paddling.",
+ "normalized_hypothesis": "i'm not sure what the rest of this will look like. i'm not sure what the rest of this will look like. i'm not sure what the rest of this will look like. i'm not sure what the rest of this will look like. i'm not sure what the rest of this will look like. i'm not sure what the rest of this will look like. i'm not sure what the rest of this will look like. i'm not sure what the rest of this will look like. i'm not sure what the rest of this will look like. i'm not sure what the rest of this will look like. i'm not sure what the rest of this will look like. i'm not sure what the rest of this will look like. i'm not sure what the rest of this will look like. i'm not sure what the rest of this will look like. i'm not sure what the rest of this will look like. i'm not sure what the rest of this will look like. i'm not sure what the rest of this will look like. i'm not sure what the rest of this will look like. i'm not sure what the rest of this will look like. i'm not sure what the rest of this will look like. i'm not sure what the rest of this will look like. i'm not sure what the rest of this will look like. i'm not sure what the rest of this will look like. i'm not sure what the rest of this will look like. i'm not sure what the rest of this will look like. i'm not sure what the rest of this will look like. i'm not sure what the rest of this will look like. i'm not sure what the rest of this will look like. i'm not sure what the rest of this will look like. i'm not sure what the rest of this will look like. i'm not sure what the rest of this will look like. i'm not sure what the rest of this will look like. i'm not sure what the rest of this will look like. i'm not sure what the rest of this will look like. i'm",
+ "errors": 373,
+ "words": 12,
+ "wer": 31.083333333333332
+ },
+ "semantic": {
+ "normalized_reference": "he made sure that the magazine was loaded and resumed his paddling",
+ "normalized_hypothesis": "i am not sure what the rest of this will look like i am not sure what the rest of this will look like i am not sure what the rest of this will look like i am not sure what the rest of this will look like i am not sure what the rest of this will look like i am not sure what the rest of this will look like i am not sure what the rest of this will look like i am not sure what the rest of this will look like i am not sure what the rest of this will look like i am not sure what the rest of this will look like i am not sure what the rest of this will look like i am not sure what the rest of this will look like i am not sure what the rest of this will look like i am not sure what the rest of this will look like i am not sure what the rest of this will look like i am not sure what the rest of this will look like i am not sure what the rest of this will look like i am not sure what the rest of this will look like i am not sure what the rest of this will look like i am not sure what the rest of this will look like i am not sure what the rest of this will look like i am not sure what the rest of this will look like i am not sure what the rest of this will look like i am not sure what the rest of this will look like i am not sure what the rest of this will look like i am not sure what the rest of this will look like i am not sure what the rest of this will look like i am not sure what the rest of this will look like i am not sure what the rest of this will look like i am not sure what the rest of this will look like i am not sure what the rest of this will look like i am not sure what the rest of this will look like i am not sure what the rest of this will look like i am not sure what the rest of this will look like i am",
+ "errors": 408,
+ "words": 12,
+ "wer": 34.0
+ }
+ },
+ {
+ "id": "release500_0281",
+ "category": "phonetic_balance",
+ "text": "Change chairs, Daylight commanded.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0281.wav",
+ "render_seconds": 47.59949245606549,
+ "audio_seconds": 2.507755102040816,
+ "rtf": 18.980917401914137,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "Chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp",
+ "raw": {
+ "normalized_reference": "change chairs, daylight commanded.",
+ "normalized_hypothesis": "chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp, chirp",
+ "errors": 148,
+ "words": 4,
+ "wer": 37.0
+ },
+ "semantic": {
+ "normalized_reference": "change chairs daylight commanded",
+ "normalized_hypothesis": "chirp chirp chirp chirp chirp chirp chirp chirp chirp chirp chirp chirp chirp chirp chirp chirp chirp chirp chirp chirp chirp chirp chirp chirp chirp chirp chirp chirp chirp chirp chirp chirp chirp chirp chirp chirp chirp chirp chirp chirp chirp chirp chirp chirp chirp chirp chirp chirp chirp chirp chirp chirp chirp chirp chirp chirp chirp chirp chirp chirp chirp chirp chirp chirp chirp chirp chirp chirp chirp chirp chirp chirp chirp chirp chirp chirp chirp chirp chirp chirp chirp chirp chirp chirp chirp chirp chirp chirp chirp chirp chirp chirp chirp chirp chirp chirp chirp chirp chirp chirp chirp chirp chirp chirp chirp chirp chirp chirp chirp chirp chirp chirp chirp chirp chirp chirp chirp chirp chirp chirp chirp chirp chirp chirp chirp chirp chirp chirp chirp chirp chirp chirp chirp chirp chirp chirp chirp chirp chirp chirp chirp chirp chirp chirp chirp chirp chirp chirp",
+ "errors": 148,
+ "words": 4,
+ "wer": 37.0
+ }
+ },
+ {
+ "id": "release500_0282",
+ "category": "phonetic_balance",
+ "text": "By that answer my professional medical prestige stood or fell.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0282.wav",
+ "render_seconds": 73.81132563599385,
+ "audio_seconds": 4.667210884353741,
+ "rtf": 15.814868336769907,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "Statual processing of the postage stamp machine.",
+ "raw": {
+ "normalized_reference": "by that answer my professional medical prestige stood or fell.",
+ "normalized_hypothesis": "statual processing of the postage stamp machine.",
+ "errors": 10,
+ "words": 10,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "by that answer my professional medical prestige stood or fell",
+ "normalized_hypothesis": "statual processing of the postage stamp machine",
+ "errors": 10,
+ "words": 10,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0283",
+ "category": "phonetic_balance",
+ "text": "The moon had already begun its westward decline.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0283.wav",
+ "render_seconds": 33.006494734901935,
+ "audio_seconds": 3.4133333333333336,
+ "rtf": 9.669871504365801,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "The world is a work of art. It's a nice work of art.",
+ "raw": {
+ "normalized_reference": "the moon had already begun its westward decline.",
+ "normalized_hypothesis": "the world is a work of art. it's a nice work of art.",
+ "errors": 12,
+ "words": 8,
+ "wer": 1.5
+ },
+ "semantic": {
+ "normalized_reference": "the moon had already begun its westward decline",
+ "normalized_hypothesis": "the world is a work of art it is a nice work of art",
+ "errors": 13,
+ "words": 8,
+ "wer": 1.625
+ }
+ },
+ {
+ "id": "release500_0284",
+ "category": "phonetic_balance",
+ "text": "At sea, Monday, March 16, 1908.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0284.wav",
+ "render_seconds": 41.395895043970086,
+ "audio_seconds": 3.8312925170068026,
+ "rtf": 10.804681412399862,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry.",
+ "raw": {
+ "normalized_reference": "at sea, monday, march 16, 1908.",
+ "normalized_hypothesis": "i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry.",
+ "errors": 14,
+ "words": 6,
+ "wer": 2.3333333333333335
+ },
+ "semantic": {
+ "normalized_reference": "at sea monday march 16 1908",
+ "normalized_hypothesis": "i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry",
+ "errors": 21,
+ "words": 6,
+ "wer": 3.5
+ }
+ },
+ {
+ "id": "release500_0285",
+ "category": "phonetic_balance",
+ "text": "And wherever I ranged, the way lay along alcohol-drenched roads.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0285.wav",
+ "render_seconds": 86.40386472502723,
+ "audio_seconds": 4.527891156462585,
+ "rtf": 19.08258430675932,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you.",
+ "raw": {
+ "normalized_reference": "and wherever i ranged, the way lay along alcohol-drenched roads.",
+ "normalized_hypothesis": "i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you.",
+ "errors": 332,
+ "words": 11,
+ "wer": 30.181818181818183
+ },
+ "semantic": {
+ "normalized_reference": "and wherever i ranged the way lay along alcohol drenched roads",
+ "normalized_hypothesis": "i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you",
+ "errors": 332,
+ "words": 11,
+ "wer": 30.181818181818183
+ }
+ },
+ {
+ "id": "release500_0286",
+ "category": "phonetic_balance",
+ "text": "The date was nearly eighteen years old.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0286.wav",
+ "render_seconds": 45.90761814196594,
+ "audio_seconds": 2.8560544217687074,
+ "rtf": 16.073789698144516,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "And don't rush away until you're fed.",
+ "raw": {
+ "normalized_reference": "the date was nearly eighteen years old.",
+ "normalized_hypothesis": "and don't rush away until you're fed.",
+ "errors": 7,
+ "words": 7,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "the date was nearly 18 years old",
+ "normalized_hypothesis": "and do not rush away until you are fed",
+ "errors": 9,
+ "words": 7,
+ "wer": 1.2857142857142858
+ }
+ },
+ {
+ "id": "release500_0287",
+ "category": "phonetic_balance",
+ "text": "Very early in my life, I separated from my mother.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0287.wav",
+ "render_seconds": 68.00331256096251,
+ "audio_seconds": 3.900952380952381,
+ "rtf": 17.432489792238925,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "I love you. I love you. I love you. I love you. I love you.",
+ "raw": {
+ "normalized_reference": "very early in my life, i separated from my mother.",
+ "normalized_hypothesis": "i love you. i love you. i love you. i love you. i love you.",
+ "errors": 14,
+ "words": 10,
+ "wer": 1.4
+ },
+ "semantic": {
+ "normalized_reference": "very early in my life i separated from my mother",
+ "normalized_hypothesis": "i love you i love you i love you i love you i love you",
+ "errors": 14,
+ "words": 10,
+ "wer": 1.4
+ }
+ },
+ {
+ "id": "release500_0288",
+ "category": "phonetic_balance",
+ "text": "There was pride and strength, the ring of triumph in his voice.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0288.wav",
+ "render_seconds": 58.888243380934,
+ "audio_seconds": 4.458231292517007,
+ "rtf": 13.208880274959256,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "I wish the world ends soon. I want to cry. I miss you.",
+ "raw": {
+ "normalized_reference": "there was pride and strength, the ring of triumph in his voice.",
+ "normalized_hypothesis": "i wish the world ends soon. i want to cry. i miss you.",
+ "errors": 13,
+ "words": 12,
+ "wer": 1.0833333333333333
+ },
+ "semantic": {
+ "normalized_reference": "there was pride and strength the ring of triumph in his voice",
+ "normalized_hypothesis": "i wish the world ends soon i want to cry i miss you",
+ "errors": 13,
+ "words": 12,
+ "wer": 1.0833333333333333
+ }
+ },
+ {
+ "id": "release500_0289",
+ "category": "phonetic_balance",
+ "text": "From my earliest recollection my sleep was a period of terror.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0289.wav",
+ "render_seconds": 42.4981372730108,
+ "audio_seconds": 4.73687074829932,
+ "rtf": 8.971774728763481,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "To be a warrior is to work for your power and strength. Use your power against power.",
+ "raw": {
+ "normalized_reference": "from my earliest recollection my sleep was a period of terror.",
+ "normalized_hypothesis": "to be a warrior is to work for your power and strength. use your power against power.",
+ "errors": 17,
+ "words": 11,
+ "wer": 1.5454545454545454
+ },
+ "semantic": {
+ "normalized_reference": "from my earliest recollection my sleep was a period of terror",
+ "normalized_hypothesis": "to be a warrior is to work for your power and strength use your power against power",
+ "errors": 17,
+ "words": 11,
+ "wer": 1.5454545454545454
+ }
+ },
+ {
+ "id": "release500_0290",
+ "category": "phonetic_balance",
+ "text": "We'll have to watch our chances.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0290.wav",
+ "render_seconds": 34.978232913068496,
+ "audio_seconds": 2.2987755102040817,
+ "rtf": 15.21602816708413,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "It is time to get out of here.",
+ "raw": {
+ "normalized_reference": "we'll have to watch our chances.",
+ "normalized_hypothesis": "it is time to get out of here.",
+ "errors": 7,
+ "words": 6,
+ "wer": 1.1666666666666667
+ },
+ "semantic": {
+ "normalized_reference": "we will have to watch our chances",
+ "normalized_hypothesis": "it is time to get out of here",
+ "errors": 7,
+ "words": 7,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0291",
+ "category": "phonetic_balance",
+ "text": "Close beside him gleamed the white fangs of the wolf-dog.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0291.wav",
+ "render_seconds": 67.38768891594373,
+ "audio_seconds": 3.7616326530612243,
+ "rtf": 17.914478932732436,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "This episode will be on the Netflix for the next time.",
+ "raw": {
+ "normalized_reference": "close beside him gleamed the white fangs of the wolf-dog.",
+ "normalized_hypothesis": "this episode will be on the netflix for the next time.",
+ "errors": 10,
+ "words": 11,
+ "wer": 0.9090909090909091
+ },
+ "semantic": {
+ "normalized_reference": "close beside him gleamed the white fangs of the wolf dog",
+ "normalized_hypothesis": "this episode will be on the netflix for the next time",
+ "errors": 10,
+ "words": 11,
+ "wer": 0.9090909090909091
+ }
+ },
+ {
+ "id": "release500_0292",
+ "category": "phonetic_balance",
+ "text": "Would you be satisfied with that one hundredth part of me.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0292.wav",
+ "render_seconds": 58.39204915298615,
+ "audio_seconds": 3.900952380952381,
+ "rtf": 14.968664944002796,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry.",
+ "raw": {
+ "normalized_reference": "would you be satisfied with that one hundredth part of me.",
+ "normalized_hypothesis": "i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry.",
+ "errors": 222,
+ "words": 11,
+ "wer": 20.181818181818183
+ },
+ "semantic": {
+ "normalized_reference": "would you be satisfied with that 100th part of me",
+ "normalized_hypothesis": "i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry",
+ "errors": 333,
+ "words": 10,
+ "wer": 33.3
+ }
+ },
+ {
+ "id": "release500_0293",
+ "category": "phonetic_balance",
+ "text": "Two gigantic owls were tearing at the carcass.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0293.wav",
+ "render_seconds": 63.297486994066276,
+ "audio_seconds": 3.3436734693877552,
+ "rtf": 18.930522843684372,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "So get there today is what I want you to do.",
+ "raw": {
+ "normalized_reference": "two gigantic owls were tearing at the carcass.",
+ "normalized_hypothesis": "so get there today is what i want you to do.",
+ "errors": 11,
+ "words": 8,
+ "wer": 1.375
+ },
+ "semantic": {
+ "normalized_reference": "2 gigantic owls were tearing at the carcass",
+ "normalized_hypothesis": "so get there today is what i want you to do",
+ "errors": 11,
+ "words": 8,
+ "wer": 1.375
+ }
+ },
+ {
+ "id": "release500_0294",
+ "category": "phonetic_balance",
+ "text": "The lush vegetation of that sheltered spot make a natural shield.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0294.wav",
+ "render_seconds": 68.8004981239792,
+ "audio_seconds": 4.597551020408163,
+ "rtf": 14.964596981866924,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "Yes, I could do that. I can't. I can't do that. I can't do that. I can't do that. I can't do that. I can't do that. I can't do that. I can't do that. I can't do that. I can't do that. I can't do that. I can't do that. I can't do that. I can't do that. I can't do that. I can't do that. I can't do that. I can't do that. I can't do that. I can't do that. I can't do that. I can't do that. I can't do that. I can't do that. I can't do that. I can't do that. I can't do that. I can't do that. I can't do that. I can't do that. I can't do that. I can't do that. I can't do that. I can't do that. I can't do that. I can't do that. I can't do that. I can't do that. I can't do that. I can't do that. I can't do that. I can't do that. I can't do that. I can't do that. I can't do that. I can't do that. I can't do that. I can't do that. I can't do that. I can't do that. I can't do that. I can't do that. I can't do that. I can't do that. I can't do that. I can't do that. I can't do that. I can't do that. I can't do that. I can't do that. I can't do that. I can't do that. I can't do that. I can't do that. I can't do that. I can't do that. I can't do that. I can't do that. I can't do that. I can't do that. I can't do that. I can't do that. I can't do that. I",
+ "raw": {
+ "normalized_reference": "the lush vegetation of that sheltered spot make a natural shield.",
+ "normalized_hypothesis": "yes, i could do that. i can't. i can't do that. i can't do that. i can't do that. i can't do that. i can't do that. i can't do that. i can't do that. i can't do that. i can't do that. i can't do that. i can't do that. i can't do that. i can't do that. i can't do that. i can't do that. i can't do that. i can't do that. i can't do that. i can't do that. i can't do that. i can't do that. i can't do that. i can't do that. i can't do that. i can't do that. i can't do that. i can't do that. i can't do that. i can't do that. i can't do that. i can't do that. i can't do that. i can't do that. i can't do that. i can't do that. i can't do that. i can't do that. i can't do that. i can't do that. i can't do that. i can't do that. i can't do that. i can't do that. i can't do that. i can't do that. i can't do that. i can't do that. i can't do that. i can't do that. i can't do that. i can't do that. i can't do that. i can't do that. i can't do that. i can't do that. i can't do that. i can't do that. i can't do that. i can't do that. i can't do that. i can't do that. i can't do that. i can't do that. i can't do that. i can't do that. i can't do that. i can't do that. i can't do that. i can't do that. i can't do that. i can't do that. i can't do that. i",
+ "errors": 295,
+ "words": 11,
+ "wer": 26.818181818181817
+ },
+ "semantic": {
+ "normalized_reference": "the lush vegetation of that sheltered spot make a natural shield",
+ "normalized_hypothesis": "yes i could do that i can not i can not do that i can not do that i can not do that i can not do that i can not do that i can not do that i can not do that i can not do that i can not do that i can not do that i can not do that i can not do that i can not do that i can not do that i can not do that i can not do that i can not do that i can not do that i can not do that i can not do that i can not do that i can not do that i can not do that i can not do that i can not do that i can not do that i can not do that i can not do that i can not do that i can not do that i can not do that i can not do that i can not do that i can not do that i can not do that i can not do that i can not do that i can not do that i can not do that i can not do that i can not do that i can not do that i can not do that i can not do that i can not do that i can not do that i can not do that i can not do that i can not do that i can not do that i can not do that i can not do that i can not do that i can not do that i can not do that i can not do that i can not do that i can not do that i can not do that i can not do that i can not do that i can not do that i can not do that i can not do that i can not do that i can not do that i can not do that i can not do that i can not do that i can not do that i can not do that i can not do that i",
+ "errors": 368,
+ "words": 11,
+ "wer": 33.45454545454545
+ }
+ },
+ {
+ "id": "release500_0295",
+ "category": "phonetic_balance",
+ "text": "Below him the shadow was broken into a pool of rippling starlight.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0295.wav",
+ "render_seconds": 36.89564866293222,
+ "audio_seconds": 4.527891156462585,
+ "rtf": 8.148528175256965,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "A.S.A.S. I can't hear what you're saying. A.S.A.S. A.S.A.S. A.S.A.S. A.S.A.S. A.S.A.S. A.S.A.S. A.S.A.S. A.S.A.S. A.S.A.S. A.S.A.S. A.S.A.S. A.S.A.S. A.S.A.S. A.S.A.S. A.S.A.S. A.S.A.S. A.S.A.S. A.S.A.S. A.S.A.S. A.S.A.S. A.S.A.S. A.S.A.S. A.S.A.S. A.S.A.S. A.S.A.S. A.S.A.S. A.S.A.S. A.S.A.S. A.S.A.S. A.S.A.S. A.S.A.S. A.S.A.S. A.S.A.S. A.S.A.S. A.S.A.S. A.S.A.S. A.S.A.S. A.S.A.S. A.S.A.S. A.S.A.S. A.S.A.S. A.S.A.S. A.S.A.S. A.S.A.S. A.S.A.S. A.S.A.S. A.S.A.S. A.S.A.S. A.S.A.S. A.S.A.S. A.S.A.S. A.S.A.S. A.S.A.S. A.S",
+ "raw": {
+ "normalized_reference": "below him the shadow was broken into a pool of rippling starlight.",
+ "normalized_hypothesis": "a.s.a.s. i can't hear what you're saying. a.s.a.s. a.s.a.s. a.s.a.s. a.s.a.s. a.s.a.s. a.s.a.s. a.s.a.s. a.s.a.s. a.s.a.s. a.s.a.s. a.s.a.s. a.s.a.s. a.s.a.s. a.s.a.s. a.s.a.s. a.s.a.s. a.s.a.s. a.s.a.s. a.s.a.s. a.s.a.s. a.s.a.s. a.s.a.s. a.s.a.s. a.s.a.s. a.s.a.s. a.s.a.s. a.s.a.s. a.s.a.s. a.s.a.s. a.s.a.s. a.s.a.s. a.s.a.s. a.s.a.s. a.s.a.s. a.s.a.s. a.s.a.s. a.s.a.s. a.s.a.s. a.s.a.s. a.s.a.s. a.s.a.s. a.s.a.s. a.s.a.s. a.s.a.s. a.s.a.s. a.s.a.s. a.s.a.s. a.s.a.s. a.s.a.s. a.s.a.s. a.s.a.s. a.s.a.s. a.s.a.s. a.s",
+ "errors": 223,
+ "words": 12,
+ "wer": 18.583333333333332
+ },
+ "semantic": {
+ "normalized_reference": "below him the shadow was broken into a pool of rippling starlight",
+ "normalized_hypothesis": "a s a s i can not hear what you are saying a s a s a s a s a s a s a s a s a s a s a s a s a s a s a s a s a s a s a s a s a s a s a s a s a s a s a s a s a s a s a s a s a s a s a s a s a s a s a s a s a s a s a s a s a s a s a s a s a s a s a s a s a s a s a s a s a s a s a s a s a s a s a s a s a s a s a s a s a s a s a s a s a s a s a s a s a s a s a s a s a s a s a s a s a s a s a s a s a s a s a s a s a s a s a s a s a s a s a s a s a s a s a s a s a s a s a s",
+ "errors": 225,
+ "words": 12,
+ "wer": 18.75
+ }
+ },
+ {
+ "id": "release500_0296",
+ "category": "phonetic_balance",
+ "text": "Lots of men take women buggy riding.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0296.wav",
+ "render_seconds": 31.415752750006504,
+ "audio_seconds": 2.507755102040816,
+ "rtf": 12.527440468345693,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "I'll try and get you out of here.",
+ "raw": {
+ "normalized_reference": "lots of men take women buggy riding.",
+ "normalized_hypothesis": "i'll try and get you out of here.",
+ "errors": 8,
+ "words": 7,
+ "wer": 1.1428571428571428
+ },
+ "semantic": {
+ "normalized_reference": "lots of men take women buggy riding",
+ "normalized_hypothesis": "i will try and get you out of here",
+ "errors": 9,
+ "words": 7,
+ "wer": 1.2857142857142858
+ }
+ },
+ {
+ "id": "release500_0297",
+ "category": "phonetic_balance",
+ "text": "But it contributed to the smash.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0297.wav",
+ "render_seconds": 44.90772803698201,
+ "audio_seconds": 2.4380952380952383,
+ "rtf": 18.4191853276684,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "I don't think you're going to kill us, Scrooge.",
+ "raw": {
+ "normalized_reference": "but it contributed to the smash.",
+ "normalized_hypothesis": "i don't think you're going to kill us, scrooge.",
+ "errors": 8,
+ "words": 6,
+ "wer": 1.3333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "but it contributed to the smash",
+ "normalized_hypothesis": "i do not think you are going to kill us scrooge",
+ "errors": 10,
+ "words": 6,
+ "wer": 1.6666666666666667
+ }
+ },
+ {
+ "id": "release500_0298",
+ "category": "phonetic_balance",
+ "text": "No man ate of the seal meat or the oil.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0298.wav",
+ "render_seconds": 39.89926950400695,
+ "audio_seconds": 2.786394557823129,
+ "rtf": 14.319317912814997,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "Oh, my God. What the fuck? What the fuck? What the fuck? What the fuck? What the fuck? What the fuck? What the fuck? What the fuck? What the fuck? What the fuck? What the fuck? What the fuck? What the fuck? What the fuck? What the fuck? What the fuck? What the fuck? What the fuck? What the fuck? What the fuck? What the fuck? What the fuck? What the fuck? What the fuck? What the fuck? What the fuck? What the fuck? What the fuck? What the fuck? What the fuck? What the fuck? What the fuck? What the fuck? What the fuck? What the fuck? What the fuck? What the fuck? What the fuck? What the fuck? What the fuck? What the fuck? What the fuck? What the fuck? What the fuck? What the fuck? What the fuck? What the fuck? What the fuck? What the fuck? What the fuck? What the fuck? What the fuck? What the fuck? What the fuck? What the fuck? What the fuck? What the fuck? What the fuck? What the fuck? What the fuck? What the fuck? What the fuck? What the fuck? What the fuck? What the fuck? What the fuck? What the fuck? What the fuck? What the fuck? What the fuck? What the fuck? What the fuck? What the fuck? What the fuck? What the fuck? What the fuck? What the fuck? What the fuck? What the fuck? What the fuck? What the fuck? What the fuck? What the fuck? What the fuck? What the fuck? What the fuck? What the fuck? What the fuck? What the fuck? What the fuck? What the fuck? What the fuck? What the fuck? What the fuck? What the fuck? What the fuck? What the fuck? What the fuck? What the fuck? What the fuck? What the fuck? What the fuck? What the fuck? What the fuck? What the fuck? What the fuck? What the fuck? What the fuck? What the fuck? What the fuck",
+ "raw": {
+ "normalized_reference": "no man ate of the seal meat or the oil.",
+ "normalized_hypothesis": "oh, my god. what the fuck? what the fuck? what the fuck? what the fuck? what the fuck? what the fuck? what the fuck? what the fuck? what the fuck? what the fuck? what the fuck? what the fuck? what the fuck? what the fuck? what the fuck? what the fuck? what the fuck? what the fuck? what the fuck? what the fuck? what the fuck? what the fuck? what the fuck? what the fuck? what the fuck? what the fuck? what the fuck? what the fuck? what the fuck? what the fuck? what the fuck? what the fuck? what the fuck? what the fuck? what the fuck? what the fuck? what the fuck? what the fuck? what the fuck? what the fuck? what the fuck? what the fuck? what the fuck? what the fuck? what the fuck? what the fuck? what the fuck? what the fuck? what the fuck? what the fuck? what the fuck? what the fuck? what the fuck? what the fuck? what the fuck? what the fuck? what the fuck? what the fuck? what the fuck? what the fuck? what the fuck? what the fuck? what the fuck? what the fuck? what the fuck? what the fuck? what the fuck? what the fuck? what the fuck? what the fuck? what the fuck? what the fuck? what the fuck? what the fuck? what the fuck? what the fuck? what the fuck? what the fuck? what the fuck? what the fuck? what the fuck? what the fuck? what the fuck? what the fuck? what the fuck? what the fuck? what the fuck? what the fuck? what the fuck? what the fuck? what the fuck? what the fuck? what the fuck? what the fuck? what the fuck? what the fuck? what the fuck? what the fuck? what the fuck? what the fuck? what the fuck? what the fuck? what the fuck? what the fuck? what the fuck? what the fuck? what the fuck? what the fuck? what the fuck? what the fuck",
+ "errors": 331,
+ "words": 10,
+ "wer": 33.1
+ },
+ "semantic": {
+ "normalized_reference": "no man ate of the seal meat or the oil",
+ "normalized_hypothesis": "0 my god what the fuck what the fuck what the fuck what the fuck what the fuck what the fuck what the fuck what the fuck what the fuck what the fuck what the fuck what the fuck what the fuck what the fuck what the fuck what the fuck what the fuck what the fuck what the fuck what the fuck what the fuck what the fuck what the fuck what the fuck what the fuck what the fuck what the fuck what the fuck what the fuck what the fuck what the fuck what the fuck what the fuck what the fuck what the fuck what the fuck what the fuck what the fuck what the fuck what the fuck what the fuck what the fuck what the fuck what the fuck what the fuck what the fuck what the fuck what the fuck what the fuck what the fuck what the fuck what the fuck what the fuck what the fuck what the fuck what the fuck what the fuck what the fuck what the fuck what the fuck what the fuck what the fuck what the fuck what the fuck what the fuck what the fuck what the fuck what the fuck what the fuck what the fuck what the fuck what the fuck what the fuck what the fuck what the fuck what the fuck what the fuck what the fuck what the fuck what the fuck what the fuck what the fuck what the fuck what the fuck what the fuck what the fuck what the fuck what the fuck what the fuck what the fuck what the fuck what the fuck what the fuck what the fuck what the fuck what the fuck what the fuck what the fuck what the fuck what the fuck what the fuck what the fuck what the fuck what the fuck what the fuck what the fuck what the fuck what the fuck what the fuck what the fuck",
+ "errors": 331,
+ "words": 10,
+ "wer": 33.1
+ }
+ },
+ {
+ "id": "release500_0299",
+ "category": "phonetic_balance",
+ "text": "You're going in for grab sharing.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0299.wav",
+ "render_seconds": 38.899090984952636,
+ "audio_seconds": 2.2291156462585033,
+ "rtf": 17.450458907434196,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "Thank you for watching.",
+ "raw": {
+ "normalized_reference": "you're going in for grab sharing.",
+ "normalized_hypothesis": "thank you for watching.",
+ "errors": 5,
+ "words": 6,
+ "wer": 0.8333333333333334
+ },
+ "semantic": {
+ "normalized_reference": "you are going in for grab sharing",
+ "normalized_hypothesis": "thank you for watching",
+ "errors": 6,
+ "words": 7,
+ "wer": 0.8571428571428571
+ }
+ },
+ {
+ "id": "release500_0300",
+ "category": "phonetic_balance",
+ "text": "He it was that lived to found the family of the Patino.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0300.wav",
+ "render_seconds": 62.90078269294463,
+ "audio_seconds": 3.6919727891156464,
+ "rtf": 17.037173967907687,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "How was that nice to find a family out of the tomb?",
+ "raw": {
+ "normalized_reference": "he it was that lived to found the family of the patino.",
+ "normalized_hypothesis": "how was that nice to find a family out of the tomb?",
+ "errors": 7,
+ "words": 12,
+ "wer": 0.5833333333333334
+ },
+ "semantic": {
+ "normalized_reference": "he it was that lived to found the family of the patino",
+ "normalized_hypothesis": "how was that nice to find a family out of the tomb",
+ "errors": 7,
+ "words": 12,
+ "wer": 0.5833333333333334
+ }
+ },
+ {
+ "id": "release500_0301",
+ "category": "phonetic_balance",
+ "text": "For such countries nothing remained but reorganization.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0301.wav",
+ "render_seconds": 41.90868374996353,
+ "audio_seconds": 3.9706122448979593,
+ "rtf": 10.554715788179548,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "Was it a good question or something like that? Why did I say so?",
+ "raw": {
+ "normalized_reference": "for such countries nothing remained but reorganization.",
+ "normalized_hypothesis": "was it a good question or something like that? why did i say so?",
+ "errors": 14,
+ "words": 7,
+ "wer": 2.0
+ },
+ "semantic": {
+ "normalized_reference": "for such countries nothing remained but reorganization",
+ "normalized_hypothesis": "was it a good question or something like that why did i say so",
+ "errors": 14,
+ "words": 7,
+ "wer": 2.0
+ }
+ },
+ {
+ "id": "release500_0302",
+ "category": "phonetic_balance",
+ "text": "Stand off butcher and baker and all the rest.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0302.wav",
+ "render_seconds": 42.398791291052476,
+ "audio_seconds": 3.3436734693877552,
+ "rtf": 12.680302571176583,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "There is steps on the corridor.",
+ "raw": {
+ "normalized_reference": "stand off butcher and baker and all the rest.",
+ "normalized_hypothesis": "there is steps on the corridor.",
+ "errors": 8,
+ "words": 9,
+ "wer": 0.8888888888888888
+ },
+ "semantic": {
+ "normalized_reference": "stand off butcher and baker and all the rest",
+ "normalized_hypothesis": "there is steps on the corridor",
+ "errors": 8,
+ "words": 9,
+ "wer": 0.8888888888888888
+ }
+ },
+ {
+ "id": "release500_0303",
+ "category": "phonetic_balance",
+ "text": "All eyes, however, were staring at him in certitude of expectancy.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0303.wav",
+ "render_seconds": 67.49231372890063,
+ "audio_seconds": 4.806530612244898,
+ "rtf": 14.04179421160042,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "Yes, I love her. I still love her. I'm so into her. That's what I'm expecting.",
+ "raw": {
+ "normalized_reference": "all eyes, however, were staring at him in certitude of expectancy.",
+ "normalized_hypothesis": "yes, i love her. i still love her. i'm so into her. that's what i'm expecting.",
+ "errors": 16,
+ "words": 11,
+ "wer": 1.4545454545454546
+ },
+ "semantic": {
+ "normalized_reference": "all eyes however were staring at him in certitude of expectancy",
+ "normalized_hypothesis": "yes i love her i still love her i am so into her that is what i am expecting",
+ "errors": 19,
+ "words": 11,
+ "wer": 1.7272727272727273
+ }
+ },
+ {
+ "id": "release500_0304",
+ "category": "phonetic_balance",
+ "text": "Two days ago Jeanne learned where her father's men were hiding.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0304.wav",
+ "render_seconds": 67.70582575898152,
+ "audio_seconds": 3.9706122448979593,
+ "rtf": 17.051734489052706,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "Thank you for watching. I hope you enjoyed it.",
+ "raw": {
+ "normalized_reference": "two days ago jeanne learned where her father's men were hiding.",
+ "normalized_hypothesis": "thank you for watching. i hope you enjoyed it.",
+ "errors": 11,
+ "words": 11,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "2 days ago jeanne learned where her father is men were hiding",
+ "normalized_hypothesis": "thank you for watching i hope you enjoyed it",
+ "errors": 12,
+ "words": 12,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0305",
+ "category": "phonetic_balance",
+ "text": "His previous wives had never lived long enough to bear him children.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0305.wav",
+ "render_seconds": 46.40942315000575,
+ "audio_seconds": 4.597551020408163,
+ "rtf": 10.094379147506578,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "I'll try my best to never let you ever step out of my sight.",
+ "raw": {
+ "normalized_reference": "his previous wives had never lived long enough to bear him children.",
+ "normalized_hypothesis": "i'll try my best to never let you ever step out of my sight.",
+ "errors": 13,
+ "words": 12,
+ "wer": 1.0833333333333333
+ },
+ "semantic": {
+ "normalized_reference": "his previous wives had never lived long enough to bear him children",
+ "normalized_hypothesis": "i will try my best to never let you ever step out of my sight",
+ "errors": 14,
+ "words": 12,
+ "wer": 1.1666666666666667
+ }
+ },
+ {
+ "id": "release500_0306",
+ "category": "phonetic_balance",
+ "text": "But this little defect did not worry him.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0306.wav",
+ "render_seconds": 55.49938709696289,
+ "audio_seconds": 2.8560544217687074,
+ "rtf": 19.432188222307417,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "I'll see you in the factory, my little boy.",
+ "raw": {
+ "normalized_reference": "but this little defect did not worry him.",
+ "normalized_hypothesis": "i'll see you in the factory, my little boy.",
+ "errors": 9,
+ "words": 8,
+ "wer": 1.125
+ },
+ "semantic": {
+ "normalized_reference": "but this little defect did not worry him",
+ "normalized_hypothesis": "i will see you in the factory my little boy",
+ "errors": 10,
+ "words": 8,
+ "wer": 1.25
+ }
+ },
+ {
+ "id": "release500_0307",
+ "category": "phonetic_balance",
+ "text": "Eli Harding asked, as Shunk started to follow.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0307.wav",
+ "render_seconds": 29.992313339957036,
+ "audio_seconds": 3.4829931972789114,
+ "rtf": 8.611074337839227,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "I have loved this thing's character for years.",
+ "raw": {
+ "normalized_reference": "eli harding asked, as shunk started to follow.",
+ "normalized_hypothesis": "i have loved this thing's character for years.",
+ "errors": 8,
+ "words": 8,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "eli harding asked as shunk started to follow",
+ "normalized_hypothesis": "i have loved this thing is character for years",
+ "errors": 9,
+ "words": 8,
+ "wer": 1.125
+ }
+ },
+ {
+ "id": "release500_0308",
+ "category": "phonetic_balance",
+ "text": "And this was their sole conversation throughout the meal.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0308.wav",
+ "render_seconds": 65.61097859800793,
+ "audio_seconds": 3.900952380952381,
+ "rtf": 16.819220587868244,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "I never saw a secret or something like that.",
+ "raw": {
+ "normalized_reference": "and this was their sole conversation throughout the meal.",
+ "normalized_hypothesis": "i never saw a secret or something like that.",
+ "errors": 9,
+ "words": 9,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "and this was their sole conversation throughout the meal",
+ "normalized_hypothesis": "i never saw a secret or something like that",
+ "errors": 9,
+ "words": 9,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0309",
+ "category": "phonetic_balance",
+ "text": "Then, again, Tudor had such an irritating way about him.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0309.wav",
+ "render_seconds": 65.31061014102306,
+ "audio_seconds": 4.179591836734694,
+ "rtf": 15.626073715381493,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you.",
+ "raw": {
+ "normalized_reference": "then, again, tudor had such an irritating way about him.",
+ "normalized_hypothesis": "i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you.",
+ "errors": 333,
+ "words": 10,
+ "wer": 33.3
+ },
+ "semantic": {
+ "normalized_reference": "then again tudor had such an irritating way about him",
+ "normalized_hypothesis": "i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you",
+ "errors": 333,
+ "words": 10,
+ "wer": 33.3
+ }
+ },
+ {
+ "id": "release500_0310",
+ "category": "phonetic_balance",
+ "text": "In it there was something that was almost tragedy.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0310.wav",
+ "render_seconds": 32.71393072709907,
+ "audio_seconds": 3.4829931972789114,
+ "rtf": 9.392476204850709,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "I like how the circle ends in this cartoon.",
+ "raw": {
+ "normalized_reference": "in it there was something that was almost tragedy.",
+ "normalized_hypothesis": "i like how the circle ends in this cartoon.",
+ "errors": 9,
+ "words": 9,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "in it there was something that was almost tragedy",
+ "normalized_hypothesis": "i like how the circle ends in this cartoon",
+ "errors": 9,
+ "words": 9,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0311",
+ "category": "phonetic_balance",
+ "text": "Thirty pounds, said the captain with finality.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0311.wav",
+ "render_seconds": 23.599484464037232,
+ "audio_seconds": 3.274013605442177,
+ "rtf": 7.208120462544617,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "What the fuck did you do, son of a bitch? What the fuck did you do, son of a bitch? What the fuck did you do, son of a bitch? What the fuck did you do, son of a bitch? What the fuck did you do, son of a bitch? What the fuck did you do, son of a bitch? What the fuck did you do, son of a bitch? What the fuck did you do, son of a bitch? What the fuck did you do, son of a bitch? What the fuck did you do, son of a bitch? What the fuck did you do, son of a bitch? What the fuck did you do, son of a bitch? What the fuck did you do, son of a bitch? What the fuck did you do, son of a bitch? What the fuck did you do, son of a bitch? What the fuck did you do, son of a bitch? What the fuck did you do, son of a bitch? What the fuck did you do, son of a bitch? What the fuck did you do, son of a bitch? What the fuck did you do, son of a bitch? What the fuck did you do, son of a bitch? What the fuck did you do, son of a bitch? What the fuck did you do, son of a bitch? What the fuck did you do, son of a bitch? What the fuck did you do, son of a bitch? What the fuck did you do, son of a bitch? What the fuck did you do, son of a bitch? What the fuck did you do, son of a bitch? What the fuck did you do, son of a bitch? What the fuck did you do, son of a bitch? What the fuck did you do, son of a bitch? What the fuck did you do, son of a bitch? What the fuck did you do, son of a bitch? What the fuck did you do, son of a bitch? What the fuck did you do, son of a bitch? What the fuck did you do, son of a bitch? What the fuck did you do, son of a bitch?",
+ "raw": {
+ "normalized_reference": "thirty pounds, said the captain with finality.",
+ "normalized_hypothesis": "what the fuck did you do, son of a bitch? what the fuck did you do, son of a bitch? what the fuck did you do, son of a bitch? what the fuck did you do, son of a bitch? what the fuck did you do, son of a bitch? what the fuck did you do, son of a bitch? what the fuck did you do, son of a bitch? what the fuck did you do, son of a bitch? what the fuck did you do, son of a bitch? what the fuck did you do, son of a bitch? what the fuck did you do, son of a bitch? what the fuck did you do, son of a bitch? what the fuck did you do, son of a bitch? what the fuck did you do, son of a bitch? what the fuck did you do, son of a bitch? what the fuck did you do, son of a bitch? what the fuck did you do, son of a bitch? what the fuck did you do, son of a bitch? what the fuck did you do, son of a bitch? what the fuck did you do, son of a bitch? what the fuck did you do, son of a bitch? what the fuck did you do, son of a bitch? what the fuck did you do, son of a bitch? what the fuck did you do, son of a bitch? what the fuck did you do, son of a bitch? what the fuck did you do, son of a bitch? what the fuck did you do, son of a bitch? what the fuck did you do, son of a bitch? what the fuck did you do, son of a bitch? what the fuck did you do, son of a bitch? what the fuck did you do, son of a bitch? what the fuck did you do, son of a bitch? what the fuck did you do, son of a bitch? what the fuck did you do, son of a bitch? what the fuck did you do, son of a bitch? what the fuck did you do, son of a bitch? what the fuck did you do, son of a bitch?",
+ "errors": 369,
+ "words": 7,
+ "wer": 52.714285714285715
+ },
+ "semantic": {
+ "normalized_reference": "\u00a330 said the captain with finality",
+ "normalized_hypothesis": "what the fuck did you do son of a bitch what the fuck did you do son of a bitch what the fuck did you do son of a bitch what the fuck did you do son of a bitch what the fuck did you do son of a bitch what the fuck did you do son of a bitch what the fuck did you do son of a bitch what the fuck did you do son of a bitch what the fuck did you do son of a bitch what the fuck did you do son of a bitch what the fuck did you do son of a bitch what the fuck did you do son of a bitch what the fuck did you do son of a bitch what the fuck did you do son of a bitch what the fuck did you do son of a bitch what the fuck did you do son of a bitch what the fuck did you do son of a bitch what the fuck did you do son of a bitch what the fuck did you do son of a bitch what the fuck did you do son of a bitch what the fuck did you do son of a bitch what the fuck did you do son of a bitch what the fuck did you do son of a bitch what the fuck did you do son of a bitch what the fuck did you do son of a bitch what the fuck did you do son of a bitch what the fuck did you do son of a bitch what the fuck did you do son of a bitch what the fuck did you do son of a bitch what the fuck did you do son of a bitch what the fuck did you do son of a bitch what the fuck did you do son of a bitch what the fuck did you do son of a bitch what the fuck did you do son of a bitch what the fuck did you do son of a bitch what the fuck did you do son of a bitch what the fuck did you do son of a bitch",
+ "errors": 369,
+ "words": 6,
+ "wer": 61.5
+ }
+ },
+ {
+ "id": "release500_0312",
+ "category": "phonetic_balance",
+ "text": "It was over when he made his way through the ring of spectators.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0312.wav",
+ "render_seconds": 66.0076415190706,
+ "audio_seconds": 4.31891156462585,
+ "rtf": 15.283397340132591,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "I receive your welcome back as well for the one of spectacles.",
+ "raw": {
+ "normalized_reference": "it was over when he made his way through the ring of spectators.",
+ "normalized_hypothesis": "i receive your welcome back as well for the one of spectacles.",
+ "errors": 11,
+ "words": 13,
+ "wer": 0.8461538461538461
+ },
+ "semantic": {
+ "normalized_reference": "it was over when he made his way through the ring of spectators",
+ "normalized_hypothesis": "i receive your welcome back as well for the one of spectacles",
+ "errors": 11,
+ "words": 13,
+ "wer": 0.8461538461538461
+ }
+ },
+ {
+ "id": "release500_0313",
+ "category": "phonetic_balance",
+ "text": "She had died from cold and starvation.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0313.wav",
+ "render_seconds": 31.52514024404809,
+ "audio_seconds": 2.9257142857142857,
+ "rtf": 10.775194419352374,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "Shrug that sluggy ass dog off.",
+ "raw": {
+ "normalized_reference": "she had died from cold and starvation.",
+ "normalized_hypothesis": "shrug that sluggy ass dog off.",
+ "errors": 7,
+ "words": 7,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "she had died from cold and starvation",
+ "normalized_hypothesis": "shrug that sluggy ass dog off",
+ "errors": 7,
+ "words": 7,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0314",
+ "category": "phonetic_balance",
+ "text": "We fished sharks on Niihau together.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0314.wav",
+ "render_seconds": 52.09936827199999,
+ "audio_seconds": 2.647074829931973,
+ "rtf": 19.68186455587993,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "I'm sorry, I'm sorry, I'm sorry, I'm sorry, I'm sorry, I'm sorry, I'm sorry, I'm sorry, I'm sorry, I'm sorry, I'm sorry, I'm sorry, I'm sorry, I'm sorry, I'm sorry, I'm sorry, I'm sorry, I'm sorry, I'm sorry, I'm sorry, I'm sorry, I'm sorry, I'm sorry, I'm sorry, I'm sorry, I'm sorry, I'm sorry, I'm sorry, I'm sorry, I'm sorry, I'm sorry, I'm sorry, I'm sorry, I'm sorry, I'm sorry, I'm sorry, I'm sorry, I'm sorry, I'm sorry, I'm sorry, I'm sorry, I'm sorry, I'm sorry, I'm sorry, I'm sorry, I'm sorry, I'm sorry, I'm sorry, I'm sorry, I'm sorry, I'm sorry, I'm sorry, I'm sorry, I'm sorry, I'm sorry, I'm sorry, I'm sorry, I'm sorry, I'm sorry, I'm sorry, I'm sorry, I'm sorry, I'm sorry, I'm sorry, I'm sorry, I'm sorry, I'm sorry, I'm sorry, I'm sorry, I'm sorry, I'm sorry, I'm sorry, I'm sorry, I'm sorry, I'm sorry, I'm sorry, I'm sorry, I'm sorry, I'm sorry, I'm sorry, I'm sorry, I'm sorry, I'm sorry, I'm sorry, I'm sorry, I'm sorry, I'm sorry, I'm sorry, I'm sorry, I'm sorry, I'm sorry, I'm sorry, I'm sorry, I'm sorry, I'm sorry, I'm sorry, I'm sorry, I'm sorry, I'm sorry, I'm sorry, I'm sorry, I'm sorry, I'm sorry, I'm sorry, I'm sorry, I'm sorry, I'm sorry, I'm sorry, I'm sorry, I'm sorry, I'm sorry,",
+ "raw": {
+ "normalized_reference": "we fished sharks on niihau together.",
+ "normalized_hypothesis": "i'm sorry, i'm sorry, i'm sorry, i'm sorry, i'm sorry, i'm sorry, i'm sorry, i'm sorry, i'm sorry, i'm sorry, i'm sorry, i'm sorry, i'm sorry, i'm sorry, i'm sorry, i'm sorry, i'm sorry, i'm sorry, i'm sorry, i'm sorry, i'm sorry, i'm sorry, i'm sorry, i'm sorry, i'm sorry, i'm sorry, i'm sorry, i'm sorry, i'm sorry, i'm sorry, i'm sorry, i'm sorry, i'm sorry, i'm sorry, i'm sorry, i'm sorry, i'm sorry, i'm sorry, i'm sorry, i'm sorry, i'm sorry, i'm sorry, i'm sorry, i'm sorry, i'm sorry, i'm sorry, i'm sorry, i'm sorry, i'm sorry, i'm sorry, i'm sorry, i'm sorry, i'm sorry, i'm sorry, i'm sorry, i'm sorry, i'm sorry, i'm sorry, i'm sorry, i'm sorry, i'm sorry, i'm sorry, i'm sorry, i'm sorry, i'm sorry, i'm sorry, i'm sorry, i'm sorry, i'm sorry, i'm sorry, i'm sorry, i'm sorry, i'm sorry, i'm sorry, i'm sorry, i'm sorry, i'm sorry, i'm sorry, i'm sorry, i'm sorry, i'm sorry, i'm sorry, i'm sorry, i'm sorry, i'm sorry, i'm sorry, i'm sorry, i'm sorry, i'm sorry, i'm sorry, i'm sorry, i'm sorry, i'm sorry, i'm sorry, i'm sorry, i'm sorry, i'm sorry, i'm sorry, i'm sorry, i'm sorry, i'm sorry, i'm sorry, i'm sorry, i'm sorry, i'm sorry, i'm sorry, i'm sorry, i'm sorry, i'm sorry, i'm sorry, i'm sorry,",
+ "errors": 222,
+ "words": 6,
+ "wer": 37.0
+ },
+ "semantic": {
+ "normalized_reference": "we fished sharks on niihau together",
+ "normalized_hypothesis": "i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry",
+ "errors": 333,
+ "words": 6,
+ "wer": 55.5
+ }
+ },
+ {
+ "id": "release500_0315",
+ "category": "phonetic_balance",
+ "text": "His slim hands gripped the edges of the table.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0315.wav",
+ "render_seconds": 65.1922088349238,
+ "audio_seconds": 3.274013605442177,
+ "rtf": 19.912015248366437,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "A slugger has got the interest of a tabby.",
+ "raw": {
+ "normalized_reference": "his slim hands gripped the edges of the table.",
+ "normalized_hypothesis": "a slugger has got the interest of a tabby.",
+ "errors": 7,
+ "words": 9,
+ "wer": 0.7777777777777778
+ },
+ "semantic": {
+ "normalized_reference": "his slim hands gripped the edges of the table",
+ "normalized_hypothesis": "a slugger has got the interest of a tabby",
+ "errors": 7,
+ "words": 9,
+ "wer": 0.7777777777777778
+ }
+ },
+ {
+ "id": "release500_0316",
+ "category": "phonetic_balance",
+ "text": "We had been chased by them ourselves, more than once.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0316.wav",
+ "render_seconds": 51.9782966269413,
+ "audio_seconds": 3.7616326530612243,
+ "rtf": 13.818015053820117,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this?",
+ "raw": {
+ "normalized_reference": "we had been chased by them ourselves, more than once.",
+ "normalized_hypothesis": "what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this?",
+ "errors": 333,
+ "words": 10,
+ "wer": 33.3
+ },
+ "semantic": {
+ "normalized_reference": "we had been chased by them ourselves more than once",
+ "normalized_hypothesis": "what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this",
+ "errors": 333,
+ "words": 10,
+ "wer": 33.3
+ }
+ },
+ {
+ "id": "release500_0317",
+ "category": "phonetic_balance",
+ "text": "Or have they already devised one.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0317.wav",
+ "render_seconds": 24.19049555505626,
+ "audio_seconds": 2.5774149659863945,
+ "rtf": 9.38556494561146,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "Oh, oh, oh. Oh, oh, oh. Oh, oh, oh. Oh, oh, oh. Oh, oh, oh. Oh, oh, oh. Oh, oh, oh. Oh, oh, oh. Oh, oh, oh. Oh, oh, oh. Oh, oh, oh. Oh, oh, oh. Oh, oh, oh. Oh, oh, oh. Oh, oh, oh. Oh, oh, oh. Oh, oh, oh. Oh, oh, oh. Oh, oh, oh. Oh, oh, oh. Oh, oh, oh. Oh, oh, oh. Oh, oh, oh. Oh, oh, oh. Oh, oh, oh. Oh, oh, oh. Oh, oh, oh. Oh, oh, oh. Oh, oh, oh. Oh, oh, oh. Oh, oh, oh. Oh, oh, oh. Oh, oh, oh. Oh, oh, oh. Oh, oh, oh. Oh, oh, oh. Oh, oh, oh. Oh, oh, oh. Oh, oh, oh. Oh, oh, oh. Oh, oh, oh. Oh, oh, oh. Oh, oh, oh. Oh, oh, oh. Oh, oh, oh. Oh, oh, oh. Oh, oh, oh. Oh, oh, oh. Oh, oh, oh. Oh, oh, oh. Oh, oh, oh. Oh, oh, oh. Oh, oh, oh. Oh, oh, oh. Oh, oh, oh. Oh, oh, oh. Oh, oh, oh. Oh, oh, oh. Oh, oh, oh. Oh, oh, oh. Oh, oh, oh. Oh, oh, oh. Oh, oh, oh. Oh, oh, oh. Oh, oh, oh. Oh, oh, oh. Oh, oh, oh. Oh, oh, oh. Oh, oh, oh. Oh, oh, oh. Oh, oh, oh. Oh, oh, oh. Oh, oh. Oh, oh, oh. Oh,",
+ "raw": {
+ "normalized_reference": "or have they already devised one.",
+ "normalized_hypothesis": "oh, oh, oh. oh, oh, oh. oh, oh, oh. oh, oh, oh. oh, oh, oh. oh, oh, oh. oh, oh, oh. oh, oh, oh. oh, oh, oh. oh, oh, oh. oh, oh, oh. oh, oh, oh. oh, oh, oh. oh, oh, oh. oh, oh, oh. oh, oh, oh. oh, oh, oh. oh, oh, oh. oh, oh, oh. oh, oh, oh. oh, oh, oh. oh, oh, oh. oh, oh, oh. oh, oh, oh. oh, oh, oh. oh, oh, oh. oh, oh, oh. oh, oh, oh. oh, oh, oh. oh, oh, oh. oh, oh, oh. oh, oh, oh. oh, oh, oh. oh, oh, oh. oh, oh, oh. oh, oh, oh. oh, oh, oh. oh, oh, oh. oh, oh, oh. oh, oh, oh. oh, oh, oh. oh, oh, oh. oh, oh, oh. oh, oh, oh. oh, oh, oh. oh, oh, oh. oh, oh, oh. oh, oh, oh. oh, oh, oh. oh, oh, oh. oh, oh, oh. oh, oh, oh. oh, oh, oh. oh, oh, oh. oh, oh, oh. oh, oh, oh. oh, oh, oh. oh, oh, oh. oh, oh, oh. oh, oh, oh. oh, oh, oh. oh, oh, oh. oh, oh, oh. oh, oh, oh. oh, oh, oh. oh, oh, oh. oh, oh, oh. oh, oh, oh. oh, oh, oh. oh, oh, oh. oh, oh, oh. oh, oh, oh. oh, oh. oh, oh, oh. oh,",
+ "errors": 222,
+ "words": 6,
+ "wer": 37.0
+ },
+ "semantic": {
+ "normalized_reference": "or have they already devised one",
+ "normalized_hypothesis": "000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000",
+ "errors": 6,
+ "words": 6,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0318",
+ "category": "phonetic_balance",
+ "text": "I cannot follow you, she said.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0318.wav",
+ "render_seconds": 48.49822791304905,
+ "audio_seconds": 2.2987755102040817,
+ "rtf": 21.097418037459192,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "Thank you for watching.",
+ "raw": {
+ "normalized_reference": "i cannot follow you, she said.",
+ "normalized_hypothesis": "thank you for watching.",
+ "errors": 5,
+ "words": 6,
+ "wer": 0.8333333333333334
+ },
+ "semantic": {
+ "normalized_reference": "i cannot follow you she said",
+ "normalized_hypothesis": "thank you for watching",
+ "errors": 5,
+ "words": 6,
+ "wer": 0.8333333333333334
+ }
+ },
+ {
+ "id": "release500_0319",
+ "category": "phonetic_balance",
+ "text": "He went down in midstream, searching the shadows of both shores.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0319.wav",
+ "render_seconds": 51.67209360306151,
+ "audio_seconds": 4.388571428571429,
+ "rtf": 11.77424007882261,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "I won't tell you what I'm going to tell you. Talk to your servers at these halls.",
+ "raw": {
+ "normalized_reference": "he went down in midstream, searching the shadows of both shores.",
+ "normalized_hypothesis": "i won't tell you what i'm going to tell you. talk to your servers at these halls.",
+ "errors": 17,
+ "words": 11,
+ "wer": 1.5454545454545454
+ },
+ "semantic": {
+ "normalized_reference": "he went down in midstream searching the shadows of both shores",
+ "normalized_hypothesis": "i will not tell you what i am going to tell you talk to your servers at these halls",
+ "errors": 19,
+ "words": 11,
+ "wer": 1.7272727272727273
+ }
+ },
+ {
+ "id": "release500_0320",
+ "category": "phonetic_balance",
+ "text": "I was sick once -- typhoid.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0320.wav",
+ "render_seconds": 30.190354880993254,
+ "audio_seconds": 2.2291156462585033,
+ "rtf": 13.543646751422145,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "I'm a sucker. Trust me.",
+ "raw": {
+ "normalized_reference": "i was sick once -- typhoid.",
+ "normalized_hypothesis": "i'm a sucker. trust me.",
+ "errors": 5,
+ "words": 5,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "i was sick once typhoid",
+ "normalized_hypothesis": "i am a sucker trust me",
+ "errors": 5,
+ "words": 5,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0321",
+ "category": "phonetic_balance",
+ "text": "Why not like any railroad station or ferry depot.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0321.wav",
+ "render_seconds": 59.707711667055264,
+ "audio_seconds": 3.622312925170068,
+ "rtf": 16.483311326354272,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "I have been watching you for a long time.",
+ "raw": {
+ "normalized_reference": "why not like any railroad station or ferry depot.",
+ "normalized_hypothesis": "i have been watching you for a long time.",
+ "errors": 9,
+ "words": 9,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "why not like any railroad station or ferry depot",
+ "normalized_hypothesis": "i have been watching you for a long time",
+ "errors": 9,
+ "words": 9,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0322",
+ "category": "phonetic_balance",
+ "text": "Red-Eye swung back and forth on the branch farther down.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0322.wav",
+ "render_seconds": 62.30721454601735,
+ "audio_seconds": 3.900952380952381,
+ "rtf": 15.972308416337455,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "Well, I'll see you there. I'll see you there. I'll see you there. I'll see you there. I'll see you there. I'll see you there. I'll see you there. I'll see you there. I'll see you there. I'll see you there. I'll see you there. I'll see you there. I'll see you there. I'll see you there. I'll see you there. I'll see you there. I'll see you there. I'll see you there. I'll see you there. I'll see you there. I'll see you there. I'll see you there. I'll see you there. I'll see you there. I'll see you there. I'll see you there. I'll see you there. I'll see you there. I'll see you there. I'll see you there. I'll see you there. I'll see you there. I'll see you there. I'll see you there. I'll see you there. I'll see you there. I'll see you there. I'll see you there. I'll see you there. I'll see you there. I'll see you there. I'll see you there. I'll see you there. I'll see you there. I'll see you there. I'll see you there. I'll see you there. I'll see you there. I'll see you there. I'll see you there. I'll see you there. I'll see you there. I'll see you there. I'll see you there. I'll see you there. I'll see you there. I'll see you there. I'll see you there. I'll see you there. I'll see you there. I'll see you there. I'll see you there. I'll see you there. I'll see you there. I'll see you there. I'll see you there. I'll see you there. I'll see you there. I'll see you there. I'll see you there. I'll see you there. I'll see you there. I'll see you there. I'll see you",
+ "raw": {
+ "normalized_reference": "red-eye swung back and forth on the branch farther down.",
+ "normalized_hypothesis": "well, i'll see you there. i'll see you there. i'll see you there. i'll see you there. i'll see you there. i'll see you there. i'll see you there. i'll see you there. i'll see you there. i'll see you there. i'll see you there. i'll see you there. i'll see you there. i'll see you there. i'll see you there. i'll see you there. i'll see you there. i'll see you there. i'll see you there. i'll see you there. i'll see you there. i'll see you there. i'll see you there. i'll see you there. i'll see you there. i'll see you there. i'll see you there. i'll see you there. i'll see you there. i'll see you there. i'll see you there. i'll see you there. i'll see you there. i'll see you there. i'll see you there. i'll see you there. i'll see you there. i'll see you there. i'll see you there. i'll see you there. i'll see you there. i'll see you there. i'll see you there. i'll see you there. i'll see you there. i'll see you there. i'll see you there. i'll see you there. i'll see you there. i'll see you there. i'll see you there. i'll see you there. i'll see you there. i'll see you there. i'll see you there. i'll see you there. i'll see you there. i'll see you there. i'll see you there. i'll see you there. i'll see you there. i'll see you there. i'll see you there. i'll see you there. i'll see you there. i'll see you there. i'll see you there. i'll see you there. i'll see you there. i'll see you there. i'll see you there. i'll see you there. i'll see you there. i'll see you",
+ "errors": 296,
+ "words": 11,
+ "wer": 26.90909090909091
+ },
+ "semantic": {
+ "normalized_reference": "red eye swung back and forth on the branch farther down",
+ "normalized_hypothesis": "well i will see you there i will see you there i will see you there i will see you there i will see you there i will see you there i will see you there i will see you there i will see you there i will see you there i will see you there i will see you there i will see you there i will see you there i will see you there i will see you there i will see you there i will see you there i will see you there i will see you there i will see you there i will see you there i will see you there i will see you there i will see you there i will see you there i will see you there i will see you there i will see you there i will see you there i will see you there i will see you there i will see you there i will see you there i will see you there i will see you there i will see you there i will see you there i will see you there i will see you there i will see you there i will see you there i will see you there i will see you there i will see you there i will see you there i will see you there i will see you there i will see you there i will see you there i will see you there i will see you there i will see you there i will see you there i will see you there i will see you there i will see you there i will see you there i will see you there i will see you there i will see you there i will see you there i will see you there i will see you there i will see you there i will see you there i will see you there i will see you there i will see you there i will see you there i will see you there i will see you there i will see you there i will see you",
+ "errors": 370,
+ "words": 11,
+ "wer": 33.63636363636363
+ }
+ },
+ {
+ "id": "release500_0323",
+ "category": "phonetic_balance",
+ "text": "Within himself he called it no longer his own.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0323.wav",
+ "render_seconds": 53.198633711086586,
+ "audio_seconds": 3.0650340136054424,
+ "rtf": 17.35662099505,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "What the hell are you saying to me? I don't know what I'm saying.",
+ "raw": {
+ "normalized_reference": "within himself he called it no longer his own.",
+ "normalized_hypothesis": "what the hell are you saying to me? i don't know what i'm saying.",
+ "errors": 14,
+ "words": 9,
+ "wer": 1.5555555555555556
+ },
+ "semantic": {
+ "normalized_reference": "within himself he called it no longer his own",
+ "normalized_hypothesis": "what the hell are you saying to me i do not know what i am saying",
+ "errors": 16,
+ "words": 9,
+ "wer": 1.7777777777777777
+ }
+ },
+ {
+ "id": "release500_0324",
+ "category": "phonetic_balance",
+ "text": "It lasted as a deterrent for two days.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0324.wav",
+ "render_seconds": 40.49952131602913,
+ "audio_seconds": 2.8560544217687074,
+ "rtf": 14.180234454688174,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "I'm going to have to tell him for two days.",
+ "raw": {
+ "normalized_reference": "it lasted as a deterrent for two days.",
+ "normalized_hypothesis": "i'm going to have to tell him for two days.",
+ "errors": 7,
+ "words": 8,
+ "wer": 0.875
+ },
+ "semantic": {
+ "normalized_reference": "it lasted as a deterrent for 2 days",
+ "normalized_hypothesis": "i am going to have to tell him for 2 days",
+ "errors": 8,
+ "words": 8,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0325",
+ "category": "phonetic_balance",
+ "text": "He loved to play Chinese lottery.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0325.wav",
+ "render_seconds": 35.21083924500272,
+ "audio_seconds": 2.4380952380952383,
+ "rtf": 14.441945784083144,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "I have to go up there now, Michael.",
+ "raw": {
+ "normalized_reference": "he loved to play chinese lottery.",
+ "normalized_hypothesis": "i have to go up there now, michael.",
+ "errors": 7,
+ "words": 6,
+ "wer": 1.1666666666666667
+ },
+ "semantic": {
+ "normalized_reference": "he loved to play chinese lottery",
+ "normalized_hypothesis": "i have to go up there now michael",
+ "errors": 7,
+ "words": 6,
+ "wer": 1.1666666666666667
+ }
+ },
+ {
+ "id": "release500_0326",
+ "category": "phonetic_balance",
+ "text": "Lord, but I'm glad to see you again, Phil.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0326.wav",
+ "render_seconds": 51.59785965597257,
+ "audio_seconds": 2.9257142857142857,
+ "rtf": 17.6359871871,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "You're up and down. You have to see what I say.",
+ "raw": {
+ "normalized_reference": "lord, but i'm glad to see you again, phil.",
+ "normalized_hypothesis": "you're up and down. you have to see what i say.",
+ "errors": 9,
+ "words": 9,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "lord but i am glad to see you again phil",
+ "normalized_hypothesis": "you are up and down you have to see what i say",
+ "errors": 10,
+ "words": 10,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0327",
+ "category": "phonetic_balance",
+ "text": "There is not an iota of truth in it, certainly not.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0327.wav",
+ "render_seconds": 66.69544376607519,
+ "audio_seconds": 3.6919727891156464,
+ "rtf": 18.064987900967445,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "I wish my little brother-in-law the best of luck. Shut your mouth.",
+ "raw": {
+ "normalized_reference": "there is not an iota of truth in it, certainly not.",
+ "normalized_hypothesis": "i wish my little brother-in-law the best of luck. shut your mouth.",
+ "errors": 14,
+ "words": 11,
+ "wer": 1.2727272727272727
+ },
+ "semantic": {
+ "normalized_reference": "there is not an iota of truth in it certainly not",
+ "normalized_hypothesis": "i wish my little brother in law the best of luck shut your mouth",
+ "errors": 14,
+ "words": 11,
+ "wer": 1.2727272727272727
+ }
+ },
+ {
+ "id": "release500_0328",
+ "category": "phonetic_balance",
+ "text": "She added, with genuine sympathy in her face and voice.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0328.wav",
+ "render_seconds": 69.40363747405354,
+ "audio_seconds": 3.9706122448979593,
+ "rtf": 17.47932892798429,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "Thank you for watching. I'll see you in the next video.",
+ "raw": {
+ "normalized_reference": "she added, with genuine sympathy in her face and voice.",
+ "normalized_hypothesis": "thank you for watching. i'll see you in the next video.",
+ "errors": 11,
+ "words": 10,
+ "wer": 1.1
+ },
+ "semantic": {
+ "normalized_reference": "she added with genuine sympathy in her face and voice",
+ "normalized_hypothesis": "thank you for watching i will see you in the next video",
+ "errors": 12,
+ "words": 10,
+ "wer": 1.2
+ }
+ },
+ {
+ "id": "release500_0329",
+ "category": "phonetic_balance",
+ "text": "We have plenty of capital ourselves, and yet we want more.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0329.wav",
+ "render_seconds": 61.16601952398196,
+ "audio_seconds": 4.179591836734694,
+ "rtf": 14.634448030640213,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "Why can't I talk to you, Sage? I don't want to hear from you.",
+ "raw": {
+ "normalized_reference": "we have plenty of capital ourselves, and yet we want more.",
+ "normalized_hypothesis": "why can't i talk to you, sage? i don't want to hear from you.",
+ "errors": 13,
+ "words": 11,
+ "wer": 1.1818181818181819
+ },
+ "semantic": {
+ "normalized_reference": "we have plenty of capital ourselves and yet we want more",
+ "normalized_hypothesis": "why can not i talk to you sage i do not want to hear from you",
+ "errors": 15,
+ "words": 11,
+ "wer": 1.3636363636363635
+ }
+ },
+ {
+ "id": "release500_0330",
+ "category": "phonetic_balance",
+ "text": "And now put yourself in my place for a moment.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0330.wav",
+ "render_seconds": 57.69193570991047,
+ "audio_seconds": 3.274013605442177,
+ "rtf": 17.621165536396358,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "Oh, get your face out of that guy's face.",
+ "raw": {
+ "normalized_reference": "and now put yourself in my place for a moment.",
+ "normalized_hypothesis": "oh, get your face out of that guy's face.",
+ "errors": 10,
+ "words": 10,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "and now put yourself in my place for a moment",
+ "normalized_hypothesis": "0 get your face out of that guy is face",
+ "errors": 10,
+ "words": 10,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0331",
+ "category": "phonetic_balance",
+ "text": "Fast, but endure.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0331.wav",
+ "render_seconds": 18.198805535095744,
+ "audio_seconds": 1.6718367346938776,
+ "rtf": 10.88551600609975,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "To the future.",
+ "raw": {
+ "normalized_reference": "fast, but endure.",
+ "normalized_hypothesis": "to the future.",
+ "errors": 3,
+ "words": 3,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "fast but endure",
+ "normalized_hypothesis": "to the future",
+ "errors": 3,
+ "words": 3,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0332",
+ "category": "phonetic_balance",
+ "text": "You were looking squeamish this afternoon, he began.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0332.wav",
+ "render_seconds": 53.79991964297369,
+ "audio_seconds": 3.4829931972789114,
+ "rtf": 15.4464613037444,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "I love you guys. I love you. I love you. I love you.",
+ "raw": {
+ "normalized_reference": "you were looking squeamish this afternoon, he began.",
+ "normalized_hypothesis": "i love you guys. i love you. i love you. i love you.",
+ "errors": 12,
+ "words": 8,
+ "wer": 1.5
+ },
+ "semantic": {
+ "normalized_reference": "you were looking squeamish this afternoon he began",
+ "normalized_hypothesis": "i love you guys i love you i love you i love you",
+ "errors": 12,
+ "words": 8,
+ "wer": 1.5
+ }
+ },
+ {
+ "id": "release500_0333",
+ "category": "phonetic_balance",
+ "text": "It is the nearest refuge.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0333.wav",
+ "render_seconds": 44.69639569998253,
+ "audio_seconds": 2.020136054421769,
+ "rtf": 22.125438334783915,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "It is the last race.",
+ "raw": {
+ "normalized_reference": "it is the nearest refuge.",
+ "normalized_hypothesis": "it is the last race.",
+ "errors": 2,
+ "words": 5,
+ "wer": 0.4
+ },
+ "semantic": {
+ "normalized_reference": "it is the nearest refuge",
+ "normalized_hypothesis": "it is the last race",
+ "errors": 2,
+ "words": 5,
+ "wer": 0.4
+ }
+ },
+ {
+ "id": "release500_0334",
+ "category": "phonetic_balance",
+ "text": "I want to die in it.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0334.wav",
+ "render_seconds": 36.895730805932544,
+ "audio_seconds": 1.7414965986394557,
+ "rtf": 21.186220423719078,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "I want to be human.",
+ "raw": {
+ "normalized_reference": "i want to die in it.",
+ "normalized_hypothesis": "i want to be human.",
+ "errors": 3,
+ "words": 6,
+ "wer": 0.5
+ },
+ "semantic": {
+ "normalized_reference": "i want to die in it",
+ "normalized_hypothesis": "i want to be human",
+ "errors": 3,
+ "words": 6,
+ "wer": 0.5
+ }
+ },
+ {
+ "id": "release500_0335",
+ "category": "phonetic_balance",
+ "text": "He stopped, and Philip nodded at the horrified question in his eyes.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0335.wav",
+ "render_seconds": 64.33425731700845,
+ "audio_seconds": 4.73687074829932,
+ "rtf": 13.581594418658435,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "Desperate and silent melancholy, the loss of past promises.",
+ "raw": {
+ "normalized_reference": "he stopped, and philip nodded at the horrified question in his eyes.",
+ "normalized_hypothesis": "desperate and silent melancholy, the loss of past promises.",
+ "errors": 10,
+ "words": 12,
+ "wer": 0.8333333333333334
+ },
+ "semantic": {
+ "normalized_reference": "he stopped and philip nodded at the horrified question in his eyes",
+ "normalized_hypothesis": "desperate and silent melancholy the loss of past promises",
+ "errors": 10,
+ "words": 12,
+ "wer": 0.8333333333333334
+ }
+ },
+ {
+ "id": "release500_0336",
+ "category": "phonetic_balance",
+ "text": "She turned, fearing that Jacques might see what was in her face.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0336.wav",
+ "render_seconds": 70.60388363199309,
+ "audio_seconds": 4.249251700680272,
+ "rtf": 16.61560401815923,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "I don't feel like talking about something without a voice.",
+ "raw": {
+ "normalized_reference": "she turned, fearing that jacques might see what was in her face.",
+ "normalized_hypothesis": "i don't feel like talking about something without a voice.",
+ "errors": 12,
+ "words": 12,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "she turned fearing that jacques might see what was in her face",
+ "normalized_hypothesis": "i do not feel like talking about something without a voice",
+ "errors": 12,
+ "words": 12,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0337",
+ "category": "phonetic_balance",
+ "text": "It was a curious coincidence.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0337.wav",
+ "render_seconds": 26.40043653792236,
+ "audio_seconds": 2.4380952380952383,
+ "rtf": 10.828304048757216,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "I don't know. I don't know what to do. I don't know what to do. I don't know what to do. I don't know what to do. I don't know what to do. I don't know what to do. I don't know what to do. I don't know what to do. I don't know what to do. I don't know what to do. I don't know what to do. I don't know what to do. I don't know what to do. I don't know what to do. I don't know what to do. I don't know what to do. I don't know what to do. I don't know what to do. I don't know what to do. I don't know what to do. I don't know what to do. I don't know what to do. I don't know what to do. I don't know what to do. I don't know what to do. I don't know what to do. I don't know what to do. I don't know what to do. I don't know what to do. I don't know what to do. I don't know what to do. I don't know what to do. I don't know what to do. I don't know what to do. I don't know what to do. I don't know what to do. I don't know what to do. I don't know what to do. I don't know what to do. I don't know what to do. I don't know what to do. I don't know what to do. I don't know what to do. I don't know what to do. I don't know what to do. I don't know what to do. I don't know what to do. I don't know what to do. I don't know what to do. I don't know what to do. I don't know what to do. I don't know what to do. I don't know what to do. I don't know what to do. I don't know what to do",
+ "raw": {
+ "normalized_reference": "it was a curious coincidence.",
+ "normalized_hypothesis": "i don't know. i don't know what to do. i don't know what to do. i don't know what to do. i don't know what to do. i don't know what to do. i don't know what to do. i don't know what to do. i don't know what to do. i don't know what to do. i don't know what to do. i don't know what to do. i don't know what to do. i don't know what to do. i don't know what to do. i don't know what to do. i don't know what to do. i don't know what to do. i don't know what to do. i don't know what to do. i don't know what to do. i don't know what to do. i don't know what to do. i don't know what to do. i don't know what to do. i don't know what to do. i don't know what to do. i don't know what to do. i don't know what to do. i don't know what to do. i don't know what to do. i don't know what to do. i don't know what to do. i don't know what to do. i don't know what to do. i don't know what to do. i don't know what to do. i don't know what to do. i don't know what to do. i don't know what to do. i don't know what to do. i don't know what to do. i don't know what to do. i don't know what to do. i don't know what to do. i don't know what to do. i don't know what to do. i don't know what to do. i don't know what to do. i don't know what to do. i don't know what to do. i don't know what to do. i don't know what to do. i don't know what to do. i don't know what to do. i don't know what to do",
+ "errors": 333,
+ "words": 5,
+ "wer": 66.6
+ },
+ "semantic": {
+ "normalized_reference": "it was a curious coincidence",
+ "normalized_hypothesis": "i do not know i do not know what to do i do not know what to do i do not know what to do i do not know what to do i do not know what to do i do not know what to do i do not know what to do i do not know what to do i do not know what to do i do not know what to do i do not know what to do i do not know what to do i do not know what to do i do not know what to do i do not know what to do i do not know what to do i do not know what to do i do not know what to do i do not know what to do i do not know what to do i do not know what to do i do not know what to do i do not know what to do i do not know what to do i do not know what to do i do not know what to do i do not know what to do i do not know what to do i do not know what to do i do not know what to do i do not know what to do i do not know what to do i do not know what to do i do not know what to do i do not know what to do i do not know what to do i do not know what to do i do not know what to do i do not know what to do i do not know what to do i do not know what to do i do not know what to do i do not know what to do i do not know what to do i do not know what to do i do not know what to do i do not know what to do i do not know what to do i do not know what to do i do not know what to do i do not know what to do i do not know what to do i do not know what to do i do not know what to do i do not know what to do",
+ "errors": 389,
+ "words": 5,
+ "wer": 77.8
+ }
+ },
+ {
+ "id": "release500_0338",
+ "category": "phonetic_balance",
+ "text": "Daylight made no answer, and the door closed behind him.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0338.wav",
+ "render_seconds": 68.90685510099865,
+ "audio_seconds": 3.9706122448979593,
+ "rtf": 17.354214123915163,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "Star Wars. Star Wars. Star Wars. Star Wars. Star Wars. Star Wars. Star Wars. Star Wars. Star Wars. Star Wars. Star Wars. Star Wars. Star Wars. Star Wars. Star Wars. Star Wars. Star Wars. Star Wars. Star Wars. Star Wars. Star Wars. Star Wars. Star Wars. Star Wars. Star Wars. Star Wars. Star Wars. Star Wars. Star Wars. Star Wars. Star Wars. Star Wars. Star Wars. Star Wars. Star Wars. Star Wars. Star Wars. Star Wars. Star Wars. Star Wars. Star Wars. Star Wars. Star Wars. Star Wars. Star Wars. Star Wars. Star Wars. Star Wars. Star Wars. Star Wars. Star Wars. Star Wars. Star Wars. Star Wars. Star Wars. Star Wars. Star Wars. Star Wars. Star Wars. Star Wars. Star Wars. Star Wars. Star Wars. Star Wars. Star Wars. Star Wars. Star Wars. Star Wars. Star Wars. Star Wars. Star Wars. Star Wars. Star Wars. Star Wars. Star Wars. Star Wars. Star Wars. Star Wars. Star Wars. Star Wars. Star Wars. Star Wars. Star Wars. Star Wars. Star Wars. Star Wars. Star Wars. Star Wars. Star Wars. Star Wars. Star Wars. Star Wars. Star Wars. Star Wars. Star Wars. Star Wars. Star Wars. Star Wars. Star Wars. Star Wars. Star Wars. Star Wars. Star Wars. Star Wars. Star Wars. Star Wars. Star Wars. Star Wars. Star Wars. Star Wars. Star Wars. Star Wars. Star Wars. Star Wars. Star Wars. Star Wars. Star Wars. Star Wars. Star Wars. Star Wars. Star Wars. Star Wars. Star Wars. Star Wars. Star Wars. Star Wars. Star Wars. Star Wars. Star Wars. Star Wars. Star wars. Star wars. Star wars. Star wars. Star wars. Star wars. Star wars. Star wars. Star wars. star wars. star wars. star wars. star wars. star wars. star wars. star wars. star wars. star wars.",
+ "raw": {
+ "normalized_reference": "daylight made no answer, and the door closed behind him.",
+ "normalized_hypothesis": "star wars. star wars. star wars. star wars. star wars. star wars. star wars. star wars. star wars. star wars. star wars. star wars. star wars. star wars. star wars. star wars. star wars. star wars. star wars. star wars. star wars. star wars. star wars. star wars. star wars. star wars. star wars. star wars. star wars. star wars. star wars. star wars. star wars. star wars. star wars. star wars. star wars. star wars. star wars. star wars. star wars. star wars. star wars. star wars. star wars. star wars. star wars. star wars. star wars. star wars. star wars. star wars. star wars. star wars. star wars. star wars. star wars. star wars. star wars. star wars. star wars. star wars. star wars. star wars. star wars. star wars. star wars. star wars. star wars. star wars. star wars. star wars. star wars. star wars. star wars. star wars. star wars. star wars. star wars. star wars. star wars. star wars. star wars. star wars. star wars. star wars. star wars. star wars. star wars. star wars. star wars. star wars. star wars. star wars. star wars. star wars. star wars. star wars. star wars. star wars. star wars. star wars. star wars. star wars. star wars. star wars. star wars. star wars. star wars. star wars. star wars. star wars. star wars. star wars. star wars. star wars. star wars. star wars. star wars. star wars. star wars. star wars. star wars. star wars. star wars. star wars. star wars. star wars. star wars. star wars. star wars. star wars. star wars. star wars. star wars. star wars. star wars. star wars. star wars. star wars. star wars. star wars. star wars. star wars. star wars. star wars. star wars. star wars.",
+ "errors": 296,
+ "words": 10,
+ "wer": 29.6
+ },
+ "semantic": {
+ "normalized_reference": "daylight made no answer and the door closed behind him",
+ "normalized_hypothesis": "star wars star wars star wars star wars star wars star wars star wars star wars star wars star wars star wars star wars star wars star wars star wars star wars star wars star wars star wars star wars star wars star wars star wars star wars star wars star wars star wars star wars star wars star wars star wars star wars star wars star wars star wars star wars star wars star wars star wars star wars star wars star wars star wars star wars star wars star wars star wars star wars star wars star wars star wars star wars star wars star wars star wars star wars star wars star wars star wars star wars star wars star wars star wars star wars star wars star wars star wars star wars star wars star wars star wars star wars star wars star wars star wars star wars star wars star wars star wars star wars star wars star wars star wars star wars star wars star wars star wars star wars star wars star wars star wars star wars star wars star wars star wars star wars star wars star wars star wars star wars star wars star wars star wars star wars star wars star wars star wars star wars star wars star wars star wars star wars star wars star wars star wars star wars star wars star wars star wars star wars star wars star wars star wars star wars star wars star wars star wars star wars star wars star wars star wars star wars star wars star wars star wars star wars star wars star wars star wars star wars star wars star wars star wars star wars star wars star wars star wars star wars",
+ "errors": 296,
+ "words": 10,
+ "wer": 29.6
+ }
+ },
+ {
+ "id": "release500_0339",
+ "category": "phonetic_balance",
+ "text": "And he did hurt my arm.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0339.wav",
+ "render_seconds": 28.416230701026507,
+ "audio_seconds": 1.9504761904761905,
+ "rtf": 14.568868279334879,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "We're not dead yet. We're alive.",
+ "raw": {
+ "normalized_reference": "and he did hurt my arm.",
+ "normalized_hypothesis": "we're not dead yet. we're alive.",
+ "errors": 6,
+ "words": 6,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "and he did hurt my arm",
+ "normalized_hypothesis": "we are not dead yet we are alive",
+ "errors": 8,
+ "words": 6,
+ "wer": 1.3333333333333333
+ }
+ },
+ {
+ "id": "release500_0340",
+ "category": "phonetic_balance",
+ "text": "Dunham, can your boy go along with Jesse.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0340.wav",
+ "render_seconds": 50.09988435602281,
+ "audio_seconds": 2.9257142857142857,
+ "rtf": 17.123983910749985,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "Uh-oh. So I'll be right over here. What should I say?",
+ "raw": {
+ "normalized_reference": "dunham, can your boy go along with jesse.",
+ "normalized_hypothesis": "uh-oh. so i'll be right over here. what should i say?",
+ "errors": 12,
+ "words": 8,
+ "wer": 1.5
+ },
+ "semantic": {
+ "normalized_reference": "dunham can your boy go along with jesse",
+ "normalized_hypothesis": "0 so i will be right over here what should i say",
+ "errors": 12,
+ "words": 8,
+ "wer": 1.5
+ }
+ },
+ {
+ "id": "release500_0341",
+ "category": "phonetic_balance",
+ "text": "The creative joy, I murmured.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0341.wav",
+ "render_seconds": 48.50075435405597,
+ "audio_seconds": 2.36843537414966,
+ "rtf": 20.47797245532579,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "It's all the time.",
+ "raw": {
+ "normalized_reference": "the creative joy, i murmured.",
+ "normalized_hypothesis": "it's all the time.",
+ "errors": 5,
+ "words": 5,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "the creative joy i murmured",
+ "normalized_hypothesis": "it is all the time",
+ "errors": 5,
+ "words": 5,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0342",
+ "category": "phonetic_balance",
+ "text": "He pulled, and the log crashed down to break his back.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0342.wav",
+ "render_seconds": 41.507601993042044,
+ "audio_seconds": 3.8312925170068026,
+ "rtf": 10.833837878155505,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "Thank you for watching.",
+ "raw": {
+ "normalized_reference": "he pulled, and the log crashed down to break his back.",
+ "normalized_hypothesis": "thank you for watching.",
+ "errors": 11,
+ "words": 11,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "he pulled and the log crashed down to break his back",
+ "normalized_hypothesis": "thank you for watching",
+ "errors": 11,
+ "words": 11,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0343",
+ "category": "phonetic_balance",
+ "text": "By golly, the boy wins.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0343.wav",
+ "render_seconds": 18.99812872102484,
+ "audio_seconds": 2.020136054421769,
+ "rtf": 9.404380798729296,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "P.T.O.B.O.B.O.B.O.B.O.B.O.B.O.B.O.B.O.B.O.B.O.B.O.B.O.B.O.B.O.B.O.B.O.B.O.B.O.B.O.B.O.B.O.B.O.B.O.B.O.B.O.B.O.B.O.B.O.B.O.B.O.B.O.B.O.B.O.B.O.B.O.B.O.B.O.B.O.B.O.B.O.B.O.B.O.B.O.B.O.B.O.B.O.B.O.B.O.B.O.B.O.B.O.B.O.B.O.B.O.B.O.B.O.B.O.B.O.B.O.B.O.B.O.B.O.B.O.B.O.B.O.B.O.B.O.B.O.B.O.B.O.B.O.B.O.B.O.B.O.B.O.B.O.B.O.B.O.B.O.B.O.B.O.B.O.B.O.B.O.B.O.B.O.B.O.B.O.B.O.B.O.B.O.B.O.B.O.B.O.B.O.B.O.B.O.B.O.B.O.B.O.B.O.B.O.B.O.B.B.O.B.O.B.O.B.B.O.B.O.B.",
+ "raw": {
+ "normalized_reference": "by golly, the boy wins.",
+ "normalized_hypothesis": "p.t.o.b.o.b.o.b.o.b.o.b.o.b.o.b.o.b.o.b.o.b.o.b.o.b.o.b.o.b.o.b.o.b.o.b.o.b.o.b.o.b.o.b.o.b.o.b.o.b.o.b.o.b.o.b.o.b.o.b.o.b.o.b.o.b.o.b.o.b.o.b.o.b.o.b.o.b.o.b.o.b.o.b.o.b.o.b.o.b.o.b.o.b.o.b.o.b.o.b.o.b.o.b.o.b.o.b.o.b.o.b.o.b.o.b.o.b.o.b.o.b.o.b.o.b.o.b.o.b.o.b.o.b.o.b.o.b.o.b.o.b.o.b.o.b.o.b.o.b.o.b.o.b.o.b.o.b.o.b.o.b.o.b.o.b.o.b.o.b.o.b.o.b.o.b.o.b.o.b.o.b.o.b.o.b.o.b.o.b.o.b.o.b.o.b.o.b.o.b.o.b.o.b.o.b.o.b.o.b.b.o.b.o.b.o.b.b.o.b.o.b.",
+ "errors": 222,
+ "words": 5,
+ "wer": 44.4
+ },
+ "semantic": {
+ "normalized_reference": "by golly the boy wins",
+ "normalized_hypothesis": "p t 0 b 0 b 0 b 0 b 0 b 0 b 0 b 0 b 0 b 0 b 0 b 0 b 0 b 0 b 0 b 0 b 0 b 0 b 0 b 0 b 0 b 0 b 0 b 0 b 0 b 0 b 0 b 0 b 0 b 0 b 0 b 0 b 0 b 0 b 0 b 0 b 0 b 0 b 0 b 0 b 0 b 0 b 0 b 0 b 0 b 0 b 0 b 0 b 0 b 0 b 0 b 0 b 0 b 0 b 0 b 0 b 0 b 0 b 0 b 0 b 0 b 0 b 0 b 0 b 0 b 0 b 0 b 0 b 0 b 0 b 0 b 0 b 0 b 0 b 0 b 0 b 0 b 0 b 0 b 0 b 0 b 0 b 0 b 0 b 0 b 0 b 0 b 0 b 0 b 0 b 0 b 0 b 0 b 0 b 0 b 0 b 0 b 0 b 0 b 0 b 0 b 0 b 0 b 0 b b 0 b 0 b 0 b b 0 b 0 b",
+ "errors": 222,
+ "words": 5,
+ "wer": 44.4
+ }
+ },
+ {
+ "id": "release500_0344",
+ "category": "phonetic_balance",
+ "text": "The boy threw back his head with pride.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0344.wav",
+ "render_seconds": 44.50350289302878,
+ "audio_seconds": 2.716734693877551,
+ "rtf": 16.381247308882287,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "A breath of air comes out of the pit.",
+ "raw": {
+ "normalized_reference": "the boy threw back his head with pride.",
+ "normalized_hypothesis": "a breath of air comes out of the pit.",
+ "errors": 9,
+ "words": 8,
+ "wer": 1.125
+ },
+ "semantic": {
+ "normalized_reference": "the boy threw back his head with pride",
+ "normalized_hypothesis": "a breath of air comes out of the pit",
+ "errors": 9,
+ "words": 8,
+ "wer": 1.125
+ }
+ },
+ {
+ "id": "release500_0345",
+ "category": "phonetic_balance",
+ "text": "The last one I knew was an overseer.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0345.wav",
+ "render_seconds": 52.690776895033196,
+ "audio_seconds": 2.647074829931973,
+ "rtf": 19.905284240259764,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "Yes, we're getting out of this house.",
+ "raw": {
+ "normalized_reference": "the last one i knew was an overseer.",
+ "normalized_hypothesis": "yes, we're getting out of this house.",
+ "errors": 8,
+ "words": 8,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "the last one i knew was an overseer",
+ "normalized_hypothesis": "yes we are getting out of this house",
+ "errors": 8,
+ "words": 8,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0346",
+ "category": "phonetic_balance",
+ "text": "Two years ago I gave up civilization for this.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0346.wav",
+ "render_seconds": 57.59371170005761,
+ "audio_seconds": 3.4133333333333336,
+ "rtf": 16.873157724626253,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "So how should we look at it? So you're there for this.",
+ "raw": {
+ "normalized_reference": "two years ago i gave up civilization for this.",
+ "normalized_hypothesis": "so how should we look at it? so you're there for this.",
+ "errors": 10,
+ "words": 9,
+ "wer": 1.1111111111111112
+ },
+ "semantic": {
+ "normalized_reference": "2 years ago i gave up civilization for this",
+ "normalized_hypothesis": "so how should we look at it so you are there for this",
+ "errors": 11,
+ "words": 9,
+ "wer": 1.2222222222222223
+ }
+ },
+ {
+ "id": "release500_0347",
+ "category": "phonetic_balance",
+ "text": "The river bared its bosom, and snorting steamboats challenged the wilderness.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0347.wav",
+ "render_seconds": 64.40615069901105,
+ "audio_seconds": 5.50312925170068,
+ "rtf": 11.703550426170175,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "The weather now is beautiful, and snow is still it. I like the waders.",
+ "raw": {
+ "normalized_reference": "the river bared its bosom, and snorting steamboats challenged the wilderness.",
+ "normalized_hypothesis": "the weather now is beautiful, and snow is still it. i like the waders.",
+ "errors": 11,
+ "words": 11,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "the river bared its bosom and snorting steamboats challenged the wilderness",
+ "normalized_hypothesis": "the weather now is beautiful and snow is still it i like the waders",
+ "errors": 11,
+ "words": 11,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0348",
+ "category": "phonetic_balance",
+ "text": "Mab, she said.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0348.wav",
+ "render_seconds": 37.49199713708367,
+ "audio_seconds": 1.6021768707482993,
+ "rtf": 23.400660577238874,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "I'm just fine.",
+ "raw": {
+ "normalized_reference": "mab, she said.",
+ "normalized_hypothesis": "i'm just fine.",
+ "errors": 3,
+ "words": 3,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "mab she said",
+ "normalized_hypothesis": "i am just fine",
+ "errors": 4,
+ "words": 3,
+ "wer": 1.3333333333333333
+ }
+ },
+ {
+ "id": "release500_0349",
+ "category": "phonetic_balance",
+ "text": "The regret in his voice was provocative of a second burst of laughter.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0349.wav",
+ "render_seconds": 66.40094997501001,
+ "audio_seconds": 4.876190476190477,
+ "rtf": 13.617382319093847,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "Now I'm going to use this as cover to add the second burst of laughter.",
+ "raw": {
+ "normalized_reference": "the regret in his voice was provocative of a second burst of laughter.",
+ "normalized_hypothesis": "now i'm going to use this as cover to add the second burst of laughter.",
+ "errors": 11,
+ "words": 13,
+ "wer": 0.8461538461538461
+ },
+ "semantic": {
+ "normalized_reference": "the regret in his voice was provocative of a 2nd burst of laughter",
+ "normalized_hypothesis": "now i am going to use this as cover to add the 2nd burst of laughter",
+ "errors": 12,
+ "words": 13,
+ "wer": 0.9230769230769231
+ }
+ },
+ {
+ "id": "release500_0350",
+ "category": "phonetic_balance",
+ "text": "And each year something happened, and I did not go.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0350.wav",
+ "render_seconds": 54.688802125980146,
+ "audio_seconds": 3.622312925170068,
+ "rtf": 15.09775749797003,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "I'm not sure how to help you. I'm not that ready.",
+ "raw": {
+ "normalized_reference": "and each year something happened, and i did not go.",
+ "normalized_hypothesis": "i'm not sure how to help you. i'm not that ready.",
+ "errors": 10,
+ "words": 10,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "and each year something happened and i did not go",
+ "normalized_hypothesis": "i am not sure how to help you i am not that ready",
+ "errors": 11,
+ "words": 10,
+ "wer": 1.1
+ }
+ },
+ {
+ "id": "release500_0351",
+ "category": "phonetic_balance",
+ "text": "Gad, your letter came just in time.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0351.wav",
+ "render_seconds": 42.79335663700476,
+ "audio_seconds": 2.647074829931973,
+ "rtf": 16.16628141868755,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "Stay away from me. Stay away from me. Stay away from me. Stay away from me. Stay away from me. Stay away from me. Stay away from me. Stay away from me. Stay away from me. Stay away from me. Stay away from me. Stay away from me. Stay away from me. Stay away from me. Stay away from me. Stay away from me. Stay away from me. Stay away from me. Stay away from me. Stay away from me. Stay away from me. Stay away from me. Stay away from me. Stay away from me. Stay away from me. Stay away from me. Stay away from me. Stay away from me. Stay away from me. Stay away from me. Stay away from me. Stay away from me. Stay away from me. Stay away from me. Stay away from me. Stay away from me. Stay away from me. Stay away from me. Stay away from me. Stay away from me. Stay away from me. Stay away from me. Stay away from me. Stay away from me. Stay away from me. Stay away from me. Stay away from me. Stay away from me. Stay away from me. Stay away from me. Stay away from me. Stay away from me. Stay away from me. Stay away from me. Stay away from me. Stay away from me. Stay away from me. Stay away from me. Stay away from me. Stay away from me. Stay away from me. Stay away from me. Stay away from me. Stay away from me. Stay away from me. Stay away from me. Stay away from me. Stay away from me. Stay away from me. Stay away from me. Stay away from me. Stay away from me. Stay away from me. Stay away from me. Stay away from me. Stay away from me. Stay away from me. Stay away from me. Stay away from me. Stay away from me. Stay away from me. Stay away from me. Stay away from me. Stay away from me. Stay away from me. Stay away from me. Stay away from me. Stay away from me. Stay away from me",
+ "raw": {
+ "normalized_reference": "gad, your letter came just in time.",
+ "normalized_hypothesis": "stay away from me. stay away from me. stay away from me. stay away from me. stay away from me. stay away from me. stay away from me. stay away from me. stay away from me. stay away from me. stay away from me. stay away from me. stay away from me. stay away from me. stay away from me. stay away from me. stay away from me. stay away from me. stay away from me. stay away from me. stay away from me. stay away from me. stay away from me. stay away from me. stay away from me. stay away from me. stay away from me. stay away from me. stay away from me. stay away from me. stay away from me. stay away from me. stay away from me. stay away from me. stay away from me. stay away from me. stay away from me. stay away from me. stay away from me. stay away from me. stay away from me. stay away from me. stay away from me. stay away from me. stay away from me. stay away from me. stay away from me. stay away from me. stay away from me. stay away from me. stay away from me. stay away from me. stay away from me. stay away from me. stay away from me. stay away from me. stay away from me. stay away from me. stay away from me. stay away from me. stay away from me. stay away from me. stay away from me. stay away from me. stay away from me. stay away from me. stay away from me. stay away from me. stay away from me. stay away from me. stay away from me. stay away from me. stay away from me. stay away from me. stay away from me. stay away from me. stay away from me. stay away from me. stay away from me. stay away from me. stay away from me. stay away from me. stay away from me. stay away from me. stay away from me. stay away from me. stay away from me. stay away from me. stay away from me",
+ "errors": 356,
+ "words": 7,
+ "wer": 50.857142857142854
+ },
+ "semantic": {
+ "normalized_reference": "gad your letter came just in time",
+ "normalized_hypothesis": "stay away from me stay away from me stay away from me stay away from me stay away from me stay away from me stay away from me stay away from me stay away from me stay away from me stay away from me stay away from me stay away from me stay away from me stay away from me stay away from me stay away from me stay away from me stay away from me stay away from me stay away from me stay away from me stay away from me stay away from me stay away from me stay away from me stay away from me stay away from me stay away from me stay away from me stay away from me stay away from me stay away from me stay away from me stay away from me stay away from me stay away from me stay away from me stay away from me stay away from me stay away from me stay away from me stay away from me stay away from me stay away from me stay away from me stay away from me stay away from me stay away from me stay away from me stay away from me stay away from me stay away from me stay away from me stay away from me stay away from me stay away from me stay away from me stay away from me stay away from me stay away from me stay away from me stay away from me stay away from me stay away from me stay away from me stay away from me stay away from me stay away from me stay away from me stay away from me stay away from me stay away from me stay away from me stay away from me stay away from me stay away from me stay away from me stay away from me stay away from me stay away from me stay away from me stay away from me stay away from me stay away from me stay away from me stay away from me stay away from me stay away from me",
+ "errors": 356,
+ "words": 7,
+ "wer": 50.857142857142854
+ }
+ },
+ {
+ "id": "release500_0352",
+ "category": "phonetic_balance",
+ "text": "But if Pierre did not return, until tomorrow.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0352.wav",
+ "render_seconds": 48.40563212300185,
+ "audio_seconds": 3.2043537414965986,
+ "rtf": 15.106207375342375,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "Thank you for watching, and I'll see you in the next video.",
+ "raw": {
+ "normalized_reference": "but if pierre did not return, until tomorrow.",
+ "normalized_hypothesis": "thank you for watching, and i'll see you in the next video.",
+ "errors": 12,
+ "words": 8,
+ "wer": 1.5
+ },
+ "semantic": {
+ "normalized_reference": "but if pierre did not return until tomorrow",
+ "normalized_hypothesis": "thank you for watching and i will see you in the next video",
+ "errors": 13,
+ "words": 8,
+ "wer": 1.625
+ }
+ },
+ {
+ "id": "release500_0353",
+ "category": "phonetic_balance",
+ "text": "Billinger may arrive in time.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0353.wav",
+ "render_seconds": 45.29558758705389,
+ "audio_seconds": 2.2987755102040817,
+ "rtf": 19.70422400360121,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "They don't hurt me when I'm broken.",
+ "raw": {
+ "normalized_reference": "billinger may arrive in time.",
+ "normalized_hypothesis": "they don't hurt me when i'm broken.",
+ "errors": 7,
+ "words": 5,
+ "wer": 1.4
+ },
+ "semantic": {
+ "normalized_reference": "billinger may arrive in time",
+ "normalized_hypothesis": "they do not hurt me when i am broken",
+ "errors": 9,
+ "words": 5,
+ "wer": 1.8
+ }
+ },
+ {
+ "id": "release500_0354",
+ "category": "phonetic_balance",
+ "text": "In such a tumbling of values was no time to sell.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0354.wav",
+ "render_seconds": 67.16993414401077,
+ "audio_seconds": 3.4829931972789114,
+ "rtf": 19.28511781087809,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "I'm searching too long away as there's no time to say.",
+ "raw": {
+ "normalized_reference": "in such a tumbling of values was no time to sell.",
+ "normalized_hypothesis": "i'm searching too long away as there's no time to say.",
+ "errors": 8,
+ "words": 11,
+ "wer": 0.7272727272727273
+ },
+ "semantic": {
+ "normalized_reference": "in such a tumbling of values was no time to sell",
+ "normalized_hypothesis": "i am searching too long away as there is no time to say",
+ "errors": 10,
+ "words": 11,
+ "wer": 0.9090909090909091
+ }
+ },
+ {
+ "id": "release500_0355",
+ "category": "phonetic_balance",
+ "text": "There's not an iota of truth in it.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0355.wav",
+ "render_seconds": 37.18858742003795,
+ "audio_seconds": 2.4380952380952383,
+ "rtf": 15.253131558999938,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "Sparrow, I love you, too. Sparrow, I love you, too. Sparrow, I love you, too. Sparrow, I love you, too. Sparrow, I love you, too. Sparrow, I love you, too. Sparrow, I love you, too. Sparrow, I love you, too. Sparrow, I love you, too. Sparrow, I love you, too. Sparrow, I love you, too. Sparrow, I love you, too. Sparrow, I love you, too. Sparrow, I love you, too. Sparrow, I love you, too. Sparrow, I love you, too. Sparrow, I love you, too. Sparrow, I love you, too. Sparrow, I love you, too. Sparrow, I love you, too. Sparrow, I love you, too. Sparrow, I love you, too. Sparrow, I love you, too. Sparrow, I love you, too. Sparrow, I love you, too. Sparrow, I love you, too. Sparrow, I love you, too. Sparrow, I love you, too. Sparrow, I love you, too. Sparrow, I love you, too. Sparrow, I love you, too. Sparrow, I love you, too. Sparrow, I love you, too. Sparrow, I love you, too. Sparrow, I love you, too. Sparrow, I love you, too. Sparrow, I love you, too. Sparrow, I love you, too. Sparrow, I love you, too. Sparrow, I love you, too. Sparrow, I love you, too. Sparrow, I love you, too. Sparrow, I love you, too. Sparrow, I love you, too. Sparrow,",
+ "raw": {
+ "normalized_reference": "there's not an iota of truth in it.",
+ "normalized_hypothesis": "sparrow, i love you, too. sparrow, i love you, too. sparrow, i love you, too. sparrow, i love you, too. sparrow, i love you, too. sparrow, i love you, too. sparrow, i love you, too. sparrow, i love you, too. sparrow, i love you, too. sparrow, i love you, too. sparrow, i love you, too. sparrow, i love you, too. sparrow, i love you, too. sparrow, i love you, too. sparrow, i love you, too. sparrow, i love you, too. sparrow, i love you, too. sparrow, i love you, too. sparrow, i love you, too. sparrow, i love you, too. sparrow, i love you, too. sparrow, i love you, too. sparrow, i love you, too. sparrow, i love you, too. sparrow, i love you, too. sparrow, i love you, too. sparrow, i love you, too. sparrow, i love you, too. sparrow, i love you, too. sparrow, i love you, too. sparrow, i love you, too. sparrow, i love you, too. sparrow, i love you, too. sparrow, i love you, too. sparrow, i love you, too. sparrow, i love you, too. sparrow, i love you, too. sparrow, i love you, too. sparrow, i love you, too. sparrow, i love you, too. sparrow, i love you, too. sparrow, i love you, too. sparrow, i love you, too. sparrow, i love you, too. sparrow,",
+ "errors": 221,
+ "words": 8,
+ "wer": 27.625
+ },
+ "semantic": {
+ "normalized_reference": "there is not an iota of truth in it",
+ "normalized_hypothesis": "sparrow i love you too sparrow i love you too sparrow i love you too sparrow i love you too sparrow i love you too sparrow i love you too sparrow i love you too sparrow i love you too sparrow i love you too sparrow i love you too sparrow i love you too sparrow i love you too sparrow i love you too sparrow i love you too sparrow i love you too sparrow i love you too sparrow i love you too sparrow i love you too sparrow i love you too sparrow i love you too sparrow i love you too sparrow i love you too sparrow i love you too sparrow i love you too sparrow i love you too sparrow i love you too sparrow i love you too sparrow i love you too sparrow i love you too sparrow i love you too sparrow i love you too sparrow i love you too sparrow i love you too sparrow i love you too sparrow i love you too sparrow i love you too sparrow i love you too sparrow i love you too sparrow i love you too sparrow i love you too sparrow i love you too sparrow i love you too sparrow i love you too sparrow i love you too sparrow",
+ "errors": 221,
+ "words": 9,
+ "wer": 24.555555555555557
+ }
+ },
+ {
+ "id": "release500_0356",
+ "category": "phonetic_balance",
+ "text": "It is dog eat dog, and you ate them up.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0356.wav",
+ "render_seconds": 48.39995520899538,
+ "audio_seconds": 2.8560544217687074,
+ "rtf": 16.9464401098569,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "That's terrific. Thank you. I love that word.",
+ "raw": {
+ "normalized_reference": "it is dog eat dog, and you ate them up.",
+ "normalized_hypothesis": "that's terrific. thank you. i love that word.",
+ "errors": 10,
+ "words": 10,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "it is dog eat dog and you ate them up",
+ "normalized_hypothesis": "that is terrific thank you i love that word",
+ "errors": 9,
+ "words": 10,
+ "wer": 0.9
+ }
+ },
+ {
+ "id": "release500_0357",
+ "category": "phonetic_balance",
+ "text": "You can take a vacation on pay.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0357.wav",
+ "render_seconds": 47.408293746062554,
+ "audio_seconds": 2.4380952380952383,
+ "rtf": 19.44480798178347,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "Thank you for watching this video.",
+ "raw": {
+ "normalized_reference": "you can take a vacation on pay.",
+ "normalized_hypothesis": "thank you for watching this video.",
+ "errors": 7,
+ "words": 7,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "you can take a vacation on pay",
+ "normalized_hypothesis": "thank you for watching this video",
+ "errors": 7,
+ "words": 7,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0358",
+ "category": "phonetic_balance",
+ "text": "I have seen myself that one man contemplated by Pascal's philosophic eye.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0358.wav",
+ "render_seconds": 81.69244368898217,
+ "audio_seconds": 4.876190476190477,
+ "rtf": 16.753333178404546,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "I'll show you a face that didn't look like that at the last place to the surface at all.",
+ "raw": {
+ "normalized_reference": "i have seen myself that one man contemplated by pascal's philosophic eye.",
+ "normalized_hypothesis": "i'll show you a face that didn't look like that at the last place to the surface at all.",
+ "errors": 18,
+ "words": 12,
+ "wer": 1.5
+ },
+ "semantic": {
+ "normalized_reference": "i have seen myself that one man contemplated by pascal is philosophic eye",
+ "normalized_hypothesis": "i will show you a face that did not look like that at the last place to the surface at all",
+ "errors": 19,
+ "words": 13,
+ "wer": 1.4615384615384615
+ }
+ },
+ {
+ "id": "release500_0359",
+ "category": "phonetic_balance",
+ "text": "Ah, I had forgotten, he exclaimed.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0359.wav",
+ "render_seconds": 44.70367567194626,
+ "audio_seconds": 2.5774149659863945,
+ "rtf": 17.34438430050702,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "I have to kill my ex-girlfriend.",
+ "raw": {
+ "normalized_reference": "ah, i had forgotten, he exclaimed.",
+ "normalized_hypothesis": "i have to kill my ex-girlfriend.",
+ "errors": 7,
+ "words": 6,
+ "wer": 1.1666666666666667
+ },
+ "semantic": {
+ "normalized_reference": "ah i had forgotten he exclaimed",
+ "normalized_hypothesis": "i have to kill my ex girlfriend",
+ "errors": 7,
+ "words": 6,
+ "wer": 1.1666666666666667
+ }
+ },
+ {
+ "id": "release500_0360",
+ "category": "phonetic_balance",
+ "text": "Providence had delivered him through the maelstrom.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0360.wav",
+ "render_seconds": 47.60065295291133,
+ "audio_seconds": 3.4133333333333336,
+ "rtf": 13.94550379479824,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "Give and take a little bit of sunrise tea.",
+ "raw": {
+ "normalized_reference": "providence had delivered him through the maelstrom.",
+ "normalized_hypothesis": "give and take a little bit of sunrise tea.",
+ "errors": 9,
+ "words": 7,
+ "wer": 1.2857142857142858
+ },
+ "semantic": {
+ "normalized_reference": "providence had delivered him through the maelstrom",
+ "normalized_hypothesis": "give and take a little bit of sunrise tea",
+ "errors": 9,
+ "words": 7,
+ "wer": 1.2857142857142858
+ }
+ },
+ {
+ "id": "release500_0361",
+ "category": "phonetic_balance",
+ "text": "She was his now, forever.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0361.wav",
+ "render_seconds": 41.50763119303156,
+ "audio_seconds": 2.089795918367347,
+ "rtf": 19.862050082602988,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "Come on. Come on. Come on. Come on. Come on. Come on. Come on. Come on. Come on. Come on. Come on. Come on. Come on. Come on. Come on. Come on. Come on. Come on. Come on. Come on. Come on. Come on. Come on. Come on. Come on. Come on. Come on. Come on. Come on. Come on. Come on. Come on. Come on. Come on. Come on. Come on. Come on. Come on. Come on. Come on. Come on. Come on. Come on. Come on. Come on. Come on. Come on. Come on. Come on. Come on. Come on. Come on. Come on. Come on. Come on. Come on. Come on. Come on. Come on. Come on. Come on. Come on. Come on. Come on. Come on. Come on. Come on. Come on. Come on. Come on. Come on. Come on. Come on. Come on. Come on. Come on. Come on. Come on. Come on. Come on. Come on. Come on. Come on. Come on. Come on. Come on. Come on. Come on. Come on. Come on. Come on. Come on. Come on. Come on. Come on. Come on. Come on. Come on. Come on. Come on. Come on. Come on. Come on. Come on. Come on. Come on. Come on. Come on. Come on. Come on. Come on. Come on. Come on. Come on. Come on. Come on. Come on. Come on. Come on. Come on. Come on. Come on. Come on. Come on. Come on. Come on. Come on. Come on. Come on. Come on. Come on. Come on. Come on. Come on. Come on. Come on. Come on. Come on. Come on. Come on. Come on. Come on. Come on. Come on. Come on. Come on. Come on. Come on.",
+ "raw": {
+ "normalized_reference": "she was his now, forever.",
+ "normalized_hypothesis": "come on. come on. come on. come on. come on. come on. come on. come on. come on. come on. come on. come on. come on. come on. come on. come on. come on. come on. come on. come on. come on. come on. come on. come on. come on. come on. come on. come on. come on. come on. come on. come on. come on. come on. come on. come on. come on. come on. come on. come on. come on. come on. come on. come on. come on. come on. come on. come on. come on. come on. come on. come on. come on. come on. come on. come on. come on. come on. come on. come on. come on. come on. come on. come on. come on. come on. come on. come on. come on. come on. come on. come on. come on. come on. come on. come on. come on. come on. come on. come on. come on. come on. come on. come on. come on. come on. come on. come on. come on. come on. come on. come on. come on. come on. come on. come on. come on. come on. come on. come on. come on. come on. come on. come on. come on. come on. come on. come on. come on. come on. come on. come on. come on. come on. come on. come on. come on. come on. come on. come on. come on. come on. come on. come on. come on. come on. come on. come on. come on. come on. come on. come on. come on. come on. come on. come on. come on. come on. come on. come on. come on. come on. come on. come on. come on. come on. come on. come on.",
+ "errors": 296,
+ "words": 5,
+ "wer": 59.2
+ },
+ "semantic": {
+ "normalized_reference": "she was his now forever",
+ "normalized_hypothesis": "come on come on come on come on come on come on come on come on come on come on come on come on come on come on come on come on come on come on come on come on come on come on come on come on come on come on come on come on come on come on come on come on come on come on come on come on come on come on come on come on come on come on come on come on come on come on come on come on come on come on come on come on come on come on come on come on come on come on come on come on come on come on come on come on come on come on come on come on come on come on come on come on come on come on come on come on come on come on come on come on come on come on come on come on come on come on come on come on come on come on come on come on come on come on come on come on come on come on come on come on come on come on come on come on come on come on come on come on come on come on come on come on come on come on come on come on come on come on come on come on come on come on come on come on come on come on come on come on come on come on come on come on come on come on come on come on come on come on come on come on come on come on come on come on come on come on come on come on",
+ "errors": 296,
+ "words": 5,
+ "wer": 59.2
+ }
+ },
+ {
+ "id": "release500_0362",
+ "category": "phonetic_balance",
+ "text": "It was unobtrusive, yet it was there.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0362.wav",
+ "render_seconds": 52.499631601036526,
+ "audio_seconds": 2.716734693877551,
+ "rtf": 19.324533867569034,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "I'm a person. I'm a person.",
+ "raw": {
+ "normalized_reference": "it was unobtrusive, yet it was there.",
+ "normalized_hypothesis": "i'm a person. i'm a person.",
+ "errors": 7,
+ "words": 7,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "it was unobtrusive yet it was there",
+ "normalized_hypothesis": "i am a person i am a person",
+ "errors": 8,
+ "words": 7,
+ "wer": 1.1428571428571428
+ }
+ },
+ {
+ "id": "release500_0363",
+ "category": "phonetic_balance",
+ "text": "Men like Joe Goose dated existence from drunk to drunk.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0363.wav",
+ "render_seconds": 58.39979052299168,
+ "audio_seconds": 3.9706122448979593,
+ "rtf": 14.708006453672864,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "Thank you for watching.",
+ "raw": {
+ "normalized_reference": "men like joe goose dated existence from drunk to drunk.",
+ "normalized_hypothesis": "thank you for watching.",
+ "errors": 10,
+ "words": 10,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "men like joe goose dated existence from drunk to drunk",
+ "normalized_hypothesis": "thank you for watching",
+ "errors": 10,
+ "words": 10,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0364",
+ "category": "phonetic_balance",
+ "text": "I was completely lost in my work.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0364.wav",
+ "render_seconds": 47.81108432204928,
+ "audio_seconds": 2.5774149659863945,
+ "rtf": 18.55001424023766,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "I must put you on the list of the right.",
+ "raw": {
+ "normalized_reference": "i was completely lost in my work.",
+ "normalized_hypothesis": "i must put you on the list of the right.",
+ "errors": 9,
+ "words": 7,
+ "wer": 1.2857142857142858
+ },
+ "semantic": {
+ "normalized_reference": "i was completely lost in my work",
+ "normalized_hypothesis": "i must put you on the list of the right",
+ "errors": 9,
+ "words": 7,
+ "wer": 1.2857142857142858
+ }
+ },
+ {
+ "id": "release500_0365",
+ "category": "phonetic_balance",
+ "text": "The very thing, Ernest agreed.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0365.wav",
+ "render_seconds": 51.60790924797766,
+ "audio_seconds": 2.2987755102040817,
+ "rtf": 22.450173589763008,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "But I must tell you, I have to go.",
+ "raw": {
+ "normalized_reference": "the very thing, ernest agreed.",
+ "normalized_hypothesis": "but i must tell you, i have to go.",
+ "errors": 9,
+ "words": 5,
+ "wer": 1.8
+ },
+ "semantic": {
+ "normalized_reference": "the very thing ernest agreed",
+ "normalized_hypothesis": "but i must tell you i have to go",
+ "errors": 9,
+ "words": 5,
+ "wer": 1.8
+ }
+ },
+ {
+ "id": "release500_0366",
+ "category": "phonetic_balance",
+ "text": "He was the leader, and Tudor was his lieutenant.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0366.wav",
+ "render_seconds": 58.503436186001636,
+ "audio_seconds": 3.622312925170068,
+ "rtf": 16.150850960303185,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "There was a ladder, and she almost had to get to it.",
+ "raw": {
+ "normalized_reference": "he was the leader, and tudor was his lieutenant.",
+ "normalized_hypothesis": "there was a ladder, and she almost had to get to it.",
+ "errors": 10,
+ "words": 9,
+ "wer": 1.1111111111111112
+ },
+ "semantic": {
+ "normalized_reference": "he was the leader and tudor was his lieutenant",
+ "normalized_hypothesis": "there was a ladder and she almost had to get to it",
+ "errors": 10,
+ "words": 9,
+ "wer": 1.1111111111111112
+ }
+ },
+ {
+ "id": "release500_0367",
+ "category": "phonetic_balance",
+ "text": "The Fire-Men wore animal skins around their waists and across their shoulders.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0367.wav",
+ "render_seconds": 59.00359232502524,
+ "audio_seconds": 5.712108843537415,
+ "rtf": 10.329563728776094,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "If any of your enemies speaks a word, I'll rush to the first of fables.",
+ "raw": {
+ "normalized_reference": "the fire-men wore animal skins around their waists and across their shoulders.",
+ "normalized_hypothesis": "if any of your enemies speaks a word, i'll rush to the first of fables.",
+ "errors": 15,
+ "words": 13,
+ "wer": 1.1538461538461537
+ },
+ "semantic": {
+ "normalized_reference": "the fire men wore animal skins around their waists and across their shoulders",
+ "normalized_hypothesis": "if any of your enemies speaks a word i will rush to the 1st of fables",
+ "errors": 16,
+ "words": 13,
+ "wer": 1.2307692307692308
+ }
+ },
+ {
+ "id": "release500_0368",
+ "category": "phonetic_balance",
+ "text": "We could throw stones with our feet.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0368.wav",
+ "render_seconds": 43.21032852400094,
+ "audio_seconds": 2.5774149659863945,
+ "rtf": 16.764987048744032,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "What's it taste on your dirty feet?",
+ "raw": {
+ "normalized_reference": "we could throw stones with our feet.",
+ "normalized_hypothesis": "what's it taste on your dirty feet?",
+ "errors": 6,
+ "words": 7,
+ "wer": 0.8571428571428571
+ },
+ "semantic": {
+ "normalized_reference": "we could throw stones with our feet",
+ "normalized_hypothesis": "what is it taste on your dirty feet",
+ "errors": 7,
+ "words": 7,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0369",
+ "category": "phonetic_balance",
+ "text": "He turned sharply, and faced Gregson across the table.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0369.wav",
+ "render_seconds": 42.790321383974515,
+ "audio_seconds": 4.249251700680272,
+ "rtf": 10.070083957870539,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "The time has come for you to face what shall approach the table.",
+ "raw": {
+ "normalized_reference": "he turned sharply, and faced gregson across the table.",
+ "normalized_hypothesis": "the time has come for you to face what shall approach the table.",
+ "errors": 11,
+ "words": 9,
+ "wer": 1.2222222222222223
+ },
+ "semantic": {
+ "normalized_reference": "he turned sharply and faced gregson across the table",
+ "normalized_hypothesis": "the time has come for you to face what shall approach the table",
+ "errors": 11,
+ "words": 9,
+ "wer": 1.2222222222222223
+ }
+ },
+ {
+ "id": "release500_0370",
+ "category": "phonetic_balance",
+ "text": "They are to attack your camp tomorrow night.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0370.wav",
+ "render_seconds": 49.991705862106755,
+ "audio_seconds": 3.0650340136054424,
+ "rtf": 16.31032661960603,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "Stay at work, Rick. Rick, you're too old.",
+ "raw": {
+ "normalized_reference": "they are to attack your camp tomorrow night.",
+ "normalized_hypothesis": "stay at work, rick. rick, you're too old.",
+ "errors": 8,
+ "words": 8,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "they are to attack your camp tomorrow night",
+ "normalized_hypothesis": "stay at work rick rick you are too old",
+ "errors": 9,
+ "words": 8,
+ "wer": 1.125
+ }
+ },
+ {
+ "id": "release500_0371",
+ "category": "phonetic_balance",
+ "text": "And watch out for wet feet, was his parting advice.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0371.wav",
+ "render_seconds": 58.3955981049221,
+ "audio_seconds": 3.5526530612244898,
+ "rtf": 16.437180073191538,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "ELECTRONICS ELECTRONICS ELECTRONICS ELECTRONICS ELECTRONICS ELECTRONICS ELECTRONICS ELECTRONICS ELECTRONICS ELECTRONICS ELECTRONICS ELECTRONICS ELECTRONICS ELECTRONICS ELECTRONICS ELECTRONICS ELECTRONICS ELECTRONICS ELECTRONICS ELECTRONICS ELECTRONICS ELECTRONICS ELECTRONICS ELECTRONICS ELECTRONICS ELECTRONICS ELECTRONICS ELECTRONICS ELECTRONICS ELECTRONICS ELECTRONICS ELECTRONICS ELECTRONICS ELECTRONICS ELECTRONICS ELECTRONICS ELECTRONICS ELECTRONICS ELECTRONICS ELECTRONICS ELECTRONICS ELECTRONICS ELECTRONICS ELECTRONICS ELECTRONICS ELECTRONICS ELECTRONICS ELECTRONICS ELECTRONICS ELECTRONICS ELECTRONICS ELECTRONICS ELECTRONICS ELECTRONICS ELECTRONICS ELECTRONICS ELECTRONICS ELECTRONICS ELECTRONICS ELECTRONICS ELECTRONICS ELECTRONICS ELECTRONICS ELECT",
+ "raw": {
+ "normalized_reference": "and watch out for wet feet, was his parting advice.",
+ "normalized_hypothesis": "electronics electronics electronics electronics electronics electronics electronics electronics electronics electronics electronics electronics electronics electronics electronics electronics electronics electronics electronics electronics electronics electronics electronics electronics electronics electronics electronics electronics electronics electronics electronics electronics electronics electronics electronics electronics electronics electronics electronics electronics electronics electronics electronics electronics electronics electronics electronics electronics electronics electronics electronics electronics electronics electronics electronics electronics electronics electronics electronics electronics electronics electronics electronics elect",
+ "errors": 64,
+ "words": 10,
+ "wer": 6.4
+ },
+ "semantic": {
+ "normalized_reference": "and watch out for wet feet was his parting advice",
+ "normalized_hypothesis": "electronics electronics electronics electronics electronics electronics electronics electronics electronics electronics electronics electronics electronics electronics electronics electronics electronics electronics electronics electronics electronics electronics electronics electronics electronics electronics electronics electronics electronics electronics electronics electronics electronics electronics electronics electronics electronics electronics electronics electronics electronics electronics electronics electronics electronics electronics electronics electronics electronics electronics electronics electronics electronics electronics electronics electronics electronics electronics electronics electronics electronics electronics electronics elect",
+ "errors": 64,
+ "words": 10,
+ "wer": 6.4
+ }
+ },
+ {
+ "id": "release500_0372",
+ "category": "phonetic_balance",
+ "text": "Two weeks passed, and in that time Thorpe left camp three times.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0372.wav",
+ "render_seconds": 71.31071821902879,
+ "audio_seconds": 4.527891156462585,
+ "rtf": 15.749212106666512,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "T-R-A-X-K-E-S-T. An election for Mr. T-R-A-X-K-E-S-T.",
+ "raw": {
+ "normalized_reference": "two weeks passed, and in that time thorpe left camp three times.",
+ "normalized_hypothesis": "t-r-a-x-k-e-s-t. an election for mr. t-r-a-x-k-e-s-t.",
+ "errors": 20,
+ "words": 12,
+ "wer": 1.6666666666666667
+ },
+ "semantic": {
+ "normalized_reference": "2 weeks passed and in that time thorpe left camp 3 times",
+ "normalized_hypothesis": "t r a x k e s t an election for mister t r a x k e s t",
+ "errors": 20,
+ "words": 12,
+ "wer": 1.6666666666666667
+ }
+ },
+ {
+ "id": "release500_0373",
+ "category": "phonetic_balance",
+ "text": "But ever his gaze returned to that Crouched Venus on the piano.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0373.wav",
+ "render_seconds": 80.58846646803431,
+ "audio_seconds": 4.388571428571429,
+ "rtf": 18.363257333210942,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "Thank you for watching.",
+ "raw": {
+ "normalized_reference": "but ever his gaze returned to that crouched venus on the piano.",
+ "normalized_hypothesis": "thank you for watching.",
+ "errors": 12,
+ "words": 12,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "but ever his gaze returned to that crouched venus on the piano",
+ "normalized_hypothesis": "thank you for watching",
+ "errors": 12,
+ "words": 12,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0374",
+ "category": "phonetic_balance",
+ "text": "But she had become an automaton.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0374.wav",
+ "render_seconds": 46.59230939403642,
+ "audio_seconds": 2.36843537414966,
+ "rtf": 19.672189455777094,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "I am back to normal.",
+ "raw": {
+ "normalized_reference": "but she had become an automaton.",
+ "normalized_hypothesis": "i am back to normal.",
+ "errors": 6,
+ "words": 6,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "but she had become an automaton",
+ "normalized_hypothesis": "i am back to normal",
+ "errors": 6,
+ "words": 6,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0375",
+ "category": "phonetic_balance",
+ "text": "I do not blame you for anything; remember that.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0375.wav",
+ "render_seconds": 44.501929983031005,
+ "audio_seconds": 3.1346938775510202,
+ "rtf": 14.196579226357548,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "Thank you for watching.",
+ "raw": {
+ "normalized_reference": "i do not blame you for anything; remember that.",
+ "normalized_hypothesis": "thank you for watching.",
+ "errors": 7,
+ "words": 9,
+ "wer": 0.7777777777777778
+ },
+ "semantic": {
+ "normalized_reference": "i do not blame you for anything remember that",
+ "normalized_hypothesis": "thank you for watching",
+ "errors": 7,
+ "words": 9,
+ "wer": 0.7777777777777778
+ }
+ },
+ {
+ "id": "release500_0376",
+ "category": "phonetic_balance",
+ "text": "The next thing to watch out for is bed sores.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0376.wav",
+ "render_seconds": 50.298806952894665,
+ "audio_seconds": 3.0650340136054424,
+ "rtf": 16.410521622149137,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "Thank you for watching.",
+ "raw": {
+ "normalized_reference": "the next thing to watch out for is bed sores.",
+ "normalized_hypothesis": "thank you for watching.",
+ "errors": 9,
+ "words": 10,
+ "wer": 0.9
+ },
+ "semantic": {
+ "normalized_reference": "the next thing to watch out for is bed sores",
+ "normalized_hypothesis": "thank you for watching",
+ "errors": 9,
+ "words": 10,
+ "wer": 0.9
+ }
+ },
+ {
+ "id": "release500_0377",
+ "category": "phonetic_balance",
+ "text": "He will follow us soon.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0377.wav",
+ "render_seconds": 37.39577233802993,
+ "audio_seconds": 1.9504761904761905,
+ "rtf": 19.17263718502511,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "And that's the end of the show.",
+ "raw": {
+ "normalized_reference": "he will follow us soon.",
+ "normalized_hypothesis": "and that's the end of the show.",
+ "errors": 7,
+ "words": 5,
+ "wer": 1.4
+ },
+ "semantic": {
+ "normalized_reference": "he will follow us soon",
+ "normalized_hypothesis": "and that is the end of the show",
+ "errors": 8,
+ "words": 5,
+ "wer": 1.6
+ }
+ },
+ {
+ "id": "release500_0378",
+ "category": "phonetic_balance",
+ "text": "His beady black eyes saw bargains where other men saw bankruptcy.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0378.wav",
+ "render_seconds": 35.784436937072314,
+ "audio_seconds": 4.527891156462585,
+ "rtf": 7.9031133259459585,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "I'm a super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super",
+ "raw": {
+ "normalized_reference": "his beady black eyes saw bargains where other men saw bankruptcy.",
+ "normalized_hypothesis": "i'm a super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super",
+ "errors": 443,
+ "words": 11,
+ "wer": 40.27272727272727
+ },
+ "semantic": {
+ "normalized_reference": "his beady black eyes saw bargains where other men saw bankruptcy",
+ "normalized_hypothesis": "i am a super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super super",
+ "errors": 444,
+ "words": 11,
+ "wer": 40.36363636363637
+ }
+ },
+ {
+ "id": "release500_0379",
+ "category": "phonetic_balance",
+ "text": "They robbed me a few years later.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0379.wav",
+ "render_seconds": 30.30690105794929,
+ "audio_seconds": 2.36843537414966,
+ "rtf": 12.796169736668617,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "Stand up with your hands up.",
+ "raw": {
+ "normalized_reference": "they robbed me a few years later.",
+ "normalized_hypothesis": "stand up with your hands up.",
+ "errors": 7,
+ "words": 7,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "they robbed me a few years later",
+ "normalized_hypothesis": "stand up with your hands up",
+ "errors": 7,
+ "words": 7,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0380",
+ "category": "phonetic_balance",
+ "text": "Gregson held a lighted match until it burnt his fingertips.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0380.wav",
+ "render_seconds": 63.697357670986094,
+ "audio_seconds": 4.249251700680272,
+ "rtf": 14.990252909892027,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "Thanks for having me here tonight. I hope you enjoyed this video.",
+ "raw": {
+ "normalized_reference": "gregson held a lighted match until it burnt his fingertips.",
+ "normalized_hypothesis": "thanks for having me here tonight. i hope you enjoyed this video.",
+ "errors": 12,
+ "words": 10,
+ "wer": 1.2
+ },
+ "semantic": {
+ "normalized_reference": "gregson held a lighted match until it burnt his fingertips",
+ "normalized_hypothesis": "thanks for having me here tonight i hope you enjoyed this video",
+ "errors": 12,
+ "words": 10,
+ "wer": 1.2
+ }
+ },
+ {
+ "id": "release500_0381",
+ "category": "phonetic_balance",
+ "text": "A burst of laughter was his reward.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0381.wav",
+ "render_seconds": 47.196240690071136,
+ "audio_seconds": 2.647074829931973,
+ "rtf": 17.829583114310385,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "I'll tell you what's on your social network.",
+ "raw": {
+ "normalized_reference": "a burst of laughter was his reward.",
+ "normalized_hypothesis": "i'll tell you what's on your social network.",
+ "errors": 8,
+ "words": 7,
+ "wer": 1.1428571428571428
+ },
+ "semantic": {
+ "normalized_reference": "a burst of laughter was his reward",
+ "normalized_hypothesis": "i will tell you what is on your social network",
+ "errors": 10,
+ "words": 7,
+ "wer": 1.4285714285714286
+ }
+ },
+ {
+ "id": "release500_0382",
+ "category": "phonetic_balance",
+ "text": "He cried, and swung the club wildly.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0382.wav",
+ "render_seconds": 45.60102565505076,
+ "audio_seconds": 2.716734693877551,
+ "rtf": 16.785233301513575,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "You can't look at me like that right now, you do.",
+ "raw": {
+ "normalized_reference": "he cried, and swung the club wildly.",
+ "normalized_hypothesis": "you can't look at me like that right now, you do.",
+ "errors": 11,
+ "words": 7,
+ "wer": 1.5714285714285714
+ },
+ "semantic": {
+ "normalized_reference": "he cried and swung the club wildly",
+ "normalized_hypothesis": "you can not look at me like that right now you do",
+ "errors": 12,
+ "words": 7,
+ "wer": 1.7142857142857142
+ }
+ },
+ {
+ "id": "release500_0383",
+ "category": "phonetic_balance",
+ "text": "Sandel would never become a world champion.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0383.wav",
+ "render_seconds": 48.79456333303824,
+ "audio_seconds": 3.2043537414965986,
+ "rtf": 15.227583241246224,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you.",
+ "raw": {
+ "normalized_reference": "sandel would never become a world champion.",
+ "normalized_hypothesis": "i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you.",
+ "errors": 333,
+ "words": 7,
+ "wer": 47.57142857142857
+ },
+ "semantic": {
+ "normalized_reference": "sandel would never become a world champion",
+ "normalized_hypothesis": "i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you",
+ "errors": 333,
+ "words": 7,
+ "wer": 47.57142857142857
+ }
+ },
+ {
+ "id": "release500_0384",
+ "category": "phonetic_balance",
+ "text": "And after the bath a shave would not be bad.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0384.wav",
+ "render_seconds": 41.10922727303114,
+ "audio_seconds": 3.0650340136054424,
+ "rtf": 13.412323351241959,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "And I hope that this has helped you. Thank you.",
+ "raw": {
+ "normalized_reference": "and after the bath a shave would not be bad.",
+ "normalized_hypothesis": "and i hope that this has helped you. thank you.",
+ "errors": 9,
+ "words": 10,
+ "wer": 0.9
+ },
+ "semantic": {
+ "normalized_reference": "and after the bath a shave would not be bad",
+ "normalized_hypothesis": "and i hope that this has helped you thank you",
+ "errors": 9,
+ "words": 10,
+ "wer": 0.9
+ }
+ },
+ {
+ "id": "release500_0385",
+ "category": "phonetic_balance",
+ "text": "They saw each other for the first time in Boston.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0385.wav",
+ "render_seconds": 42.49437020299956,
+ "audio_seconds": 3.3436734693877552,
+ "rtf": 12.70888757291857,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "I sure hope I'm not the first to have a birthday.",
+ "raw": {
+ "normalized_reference": "they saw each other for the first time in boston.",
+ "normalized_hypothesis": "i sure hope i'm not the first to have a birthday.",
+ "errors": 9,
+ "words": 10,
+ "wer": 0.9
+ },
+ "semantic": {
+ "normalized_reference": "they saw each other for the 1st time in boston",
+ "normalized_hypothesis": "i sure hope i am not the 1st to have a birthday",
+ "errors": 10,
+ "words": 10,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0386",
+ "category": "phonetic_balance",
+ "text": "I had been born with no organic, chemical predisposition toward alcohol.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0386.wav",
+ "render_seconds": 70.00925771601032,
+ "audio_seconds": 5.224489795918367,
+ "rtf": 13.400209484705103,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "Thank you for listening to our interview.",
+ "raw": {
+ "normalized_reference": "i had been born with no organic, chemical predisposition toward alcohol.",
+ "normalized_hypothesis": "thank you for listening to our interview.",
+ "errors": 11,
+ "words": 11,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "i had been born with no organic chemical predisposition toward alcohol",
+ "normalized_hypothesis": "thank you for listening to our interview",
+ "errors": 11,
+ "words": 11,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0387",
+ "category": "phonetic_balance",
+ "text": "They had no fixed values, to be altered by adjectives and adverbs.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0387.wav",
+ "render_seconds": 72.50669521500822,
+ "audio_seconds": 4.9458503401360545,
+ "rtf": 14.660107004575,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "I have six friends. Can we talk on a trip to some other place?",
+ "raw": {
+ "normalized_reference": "they had no fixed values, to be altered by adjectives and adverbs.",
+ "normalized_hypothesis": "i have six friends. can we talk on a trip to some other place?",
+ "errors": 14,
+ "words": 12,
+ "wer": 1.1666666666666667
+ },
+ "semantic": {
+ "normalized_reference": "they had no fixed values to be altered by adjectives and adverbs",
+ "normalized_hypothesis": "i have 6 friends can we talk on a trip to some other place",
+ "errors": 14,
+ "words": 12,
+ "wer": 1.1666666666666667
+ }
+ },
+ {
+ "id": "release500_0388",
+ "category": "phonetic_balance",
+ "text": "Nor was Elam Harnish an exception.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0388.wav",
+ "render_seconds": 50.79827866400592,
+ "audio_seconds": 2.716734693877551,
+ "rtf": 18.69828466448535,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "I don't want somebody to ever hear this. It sucks.",
+ "raw": {
+ "normalized_reference": "nor was elam harnish an exception.",
+ "normalized_hypothesis": "i don't want somebody to ever hear this. it sucks.",
+ "errors": 10,
+ "words": 6,
+ "wer": 1.6666666666666667
+ },
+ "semantic": {
+ "normalized_reference": "nor was elam harnish an exception",
+ "normalized_hypothesis": "i do not want somebody to ever hear this it sucks",
+ "errors": 11,
+ "words": 6,
+ "wer": 1.8333333333333333
+ }
+ },
+ {
+ "id": "release500_0389",
+ "category": "phonetic_balance",
+ "text": "He was a wise hyena.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0389.wav",
+ "render_seconds": 21.19734215491917,
+ "audio_seconds": 1.8808163265306121,
+ "rtf": 11.270288255111105,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "I love her. I love her. I love her.",
+ "raw": {
+ "normalized_reference": "he was a wise hyena.",
+ "normalized_hypothesis": "i love her. i love her. i love her.",
+ "errors": 9,
+ "words": 5,
+ "wer": 1.8
+ },
+ "semantic": {
+ "normalized_reference": "he was a wise hyena",
+ "normalized_hypothesis": "i love her i love her i love her",
+ "errors": 9,
+ "words": 5,
+ "wer": 1.8
+ }
+ },
+ {
+ "id": "release500_0390",
+ "category": "phonetic_balance",
+ "text": "And now behold the perversity of things.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0390.wav",
+ "render_seconds": 42.79965169506613,
+ "audio_seconds": 2.8560544217687074,
+ "rtf": 14.985586888278204,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "Thank you for watching.",
+ "raw": {
+ "normalized_reference": "and now behold the perversity of things.",
+ "normalized_hypothesis": "thank you for watching.",
+ "errors": 7,
+ "words": 7,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "and now behold the perversity of things",
+ "normalized_hypothesis": "thank you for watching",
+ "errors": 7,
+ "words": 7,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0391",
+ "category": "phonetic_balance",
+ "text": "The Fire People, like ourselves, lived in caves.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0391.wav",
+ "render_seconds": 68.71404180303216,
+ "audio_seconds": 3.4829931972789114,
+ "rtf": 19.728445595792437,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "So creepy. Look at our face. Look at this.",
+ "raw": {
+ "normalized_reference": "the fire people, like ourselves, lived in caves.",
+ "normalized_hypothesis": "so creepy. look at our face. look at this.",
+ "errors": 9,
+ "words": 8,
+ "wer": 1.125
+ },
+ "semantic": {
+ "normalized_reference": "the fire people like ourselves lived in caves",
+ "normalized_hypothesis": "so creepy look at our face look at this",
+ "errors": 9,
+ "words": 8,
+ "wer": 1.125
+ }
+ },
+ {
+ "id": "release500_0392",
+ "category": "phonetic_balance",
+ "text": "To her the bridge was tambo, which is the native for taboo.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0392.wav",
+ "render_seconds": 60.69530165498145,
+ "audio_seconds": 4.040272108843538,
+ "rtf": 15.022577692757059,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "God bless you all. God bless you all. God bless you all. God bless you all. God bless you all. God bless you all. God bless you all. God bless you all. God bless you all. God bless you all. God bless you all. God bless you all. God bless you all. God bless you all. God bless you all. God bless you all. God bless you all. God bless you all. God bless you all. God bless you all. God bless you all. God bless you all. God bless you all. God bless you all. God bless you all. God bless you all. God bless you all. God bless you all. God bless you all. God bless you all. God bless you all. God bless you all. God bless you all. God bless you all. God bless you all. God bless you all. God bless you all. God bless you all. God bless you all. God bless you all. God bless you all. God bless you all. God bless you all. God bless you all. God bless you all. God bless you all. God bless you all. God bless you all. God bless you all. God bless you all. God bless you all. God bless you all. God bless you all. God bless you all. God bless you all. God bless you all. God bless you all. God bless you all. God bless you all. God bless you all. God bless you all. God bless you all. God bless you all. God bless you all. God bless you all. God bless you all. God bless you all. God bless you all. God bless you all. God bless you all. God bless you all. God bless you all. God bless you all. God bless you all. God bless you all. God bless you all. God bless you all. God bless you all. God bless you all. God bless you all. God bless you all. God bless you all. God bless you all. God bless you all. God bless you all. God bless you all. God bless you all. God bless you all. God bless you all",
+ "raw": {
+ "normalized_reference": "to her the bridge was tambo, which is the native for taboo.",
+ "normalized_hypothesis": "god bless you all. god bless you all. god bless you all. god bless you all. god bless you all. god bless you all. god bless you all. god bless you all. god bless you all. god bless you all. god bless you all. god bless you all. god bless you all. god bless you all. god bless you all. god bless you all. god bless you all. god bless you all. god bless you all. god bless you all. god bless you all. god bless you all. god bless you all. god bless you all. god bless you all. god bless you all. god bless you all. god bless you all. god bless you all. god bless you all. god bless you all. god bless you all. god bless you all. god bless you all. god bless you all. god bless you all. god bless you all. god bless you all. god bless you all. god bless you all. god bless you all. god bless you all. god bless you all. god bless you all. god bless you all. god bless you all. god bless you all. god bless you all. god bless you all. god bless you all. god bless you all. god bless you all. god bless you all. god bless you all. god bless you all. god bless you all. god bless you all. god bless you all. god bless you all. god bless you all. god bless you all. god bless you all. god bless you all. god bless you all. god bless you all. god bless you all. god bless you all. god bless you all. god bless you all. god bless you all. god bless you all. god bless you all. god bless you all. god bless you all. god bless you all. god bless you all. god bless you all. god bless you all. god bless you all. god bless you all. god bless you all. god bless you all. god bless you all. god bless you all. god bless you all. god bless you all. god bless you all. god bless you all. god bless you all",
+ "errors": 356,
+ "words": 12,
+ "wer": 29.666666666666668
+ },
+ "semantic": {
+ "normalized_reference": "to her the bridge was tambo which is the native for taboo",
+ "normalized_hypothesis": "god bless you all god bless you all god bless you all god bless you all god bless you all god bless you all god bless you all god bless you all god bless you all god bless you all god bless you all god bless you all god bless you all god bless you all god bless you all god bless you all god bless you all god bless you all god bless you all god bless you all god bless you all god bless you all god bless you all god bless you all god bless you all god bless you all god bless you all god bless you all god bless you all god bless you all god bless you all god bless you all god bless you all god bless you all god bless you all god bless you all god bless you all god bless you all god bless you all god bless you all god bless you all god bless you all god bless you all god bless you all god bless you all god bless you all god bless you all god bless you all god bless you all god bless you all god bless you all god bless you all god bless you all god bless you all god bless you all god bless you all god bless you all god bless you all god bless you all god bless you all god bless you all god bless you all god bless you all god bless you all god bless you all god bless you all god bless you all god bless you all god bless you all god bless you all god bless you all god bless you all god bless you all god bless you all god bless you all god bless you all god bless you all god bless you all god bless you all god bless you all god bless you all god bless you all god bless you all god bless you all god bless you all god bless you all god bless you all god bless you all god bless you all",
+ "errors": 356,
+ "words": 12,
+ "wer": 29.666666666666668
+ }
+ },
+ {
+ "id": "release500_0393",
+ "category": "phonetic_balance",
+ "text": "The voyage was our idea of a good time.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0393.wav",
+ "render_seconds": 50.50033521710429,
+ "audio_seconds": 2.8560544217687074,
+ "rtf": 17.681853270089395,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "Thank you so much for watching.",
+ "raw": {
+ "normalized_reference": "the voyage was our idea of a good time.",
+ "normalized_hypothesis": "thank you so much for watching.",
+ "errors": 9,
+ "words": 9,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "the voyage was our idea of a good time",
+ "normalized_hypothesis": "thank you so much for watching",
+ "errors": 9,
+ "words": 9,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0394",
+ "category": "phonetic_balance",
+ "text": "Of course much grumbling went on, and little outbursts were continually occurring.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0394.wav",
+ "render_seconds": 95.90916807902977,
+ "audio_seconds": 5.712108843537415,
+ "rtf": 16.79050079508548,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "Let's go stretch with me a little. I'm very out of breath, so I'm going to do that too.",
+ "raw": {
+ "normalized_reference": "of course much grumbling went on, and little outbursts were continually occurring.",
+ "normalized_hypothesis": "let's go stretch with me a little. i'm very out of breath, so i'm going to do that too.",
+ "errors": 19,
+ "words": 12,
+ "wer": 1.5833333333333333
+ },
+ "semantic": {
+ "normalized_reference": "of course much grumbling went on and little outbursts were continually occurring",
+ "normalized_hypothesis": "let us go stretch with me a little i am very out of breath so i am going to do that too",
+ "errors": 21,
+ "words": 12,
+ "wer": 1.75
+ }
+ },
+ {
+ "id": "release500_0395",
+ "category": "phonetic_balance",
+ "text": "Get down and dig in.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0395.wav",
+ "render_seconds": 25.104691768996418,
+ "audio_seconds": 1.811156462585034,
+ "rtf": 13.86113916031578,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "Get down, little brats.",
+ "raw": {
+ "normalized_reference": "get down and dig in.",
+ "normalized_hypothesis": "get down, little brats.",
+ "errors": 3,
+ "words": 5,
+ "wer": 0.6
+ },
+ "semantic": {
+ "normalized_reference": "get down and dig in",
+ "normalized_hypothesis": "get down little brats",
+ "errors": 3,
+ "words": 5,
+ "wer": 0.6
+ }
+ },
+ {
+ "id": "release500_0396",
+ "category": "phonetic_balance",
+ "text": "It was put together in a casual, helter-skelter sort of way.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0396.wav",
+ "render_seconds": 59.49858675501309,
+ "audio_seconds": 4.31891156462585,
+ "rtf": 13.7762919811412,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "I was picked to get all the ghastly. Haters, haters, all but me.",
+ "raw": {
+ "normalized_reference": "it was put together in a casual, helter-skelter sort of way.",
+ "normalized_hypothesis": "i was picked to get all the ghastly. haters, haters, all but me.",
+ "errors": 12,
+ "words": 12,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "it was put together in a casual helter skelter sort of way",
+ "normalized_hypothesis": "i was picked to get all the ghastly haters haters all but me",
+ "errors": 12,
+ "words": 12,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0397",
+ "category": "phonetic_balance",
+ "text": "He deluged me, overwhelmed me with argument.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0397.wav",
+ "render_seconds": 34.39524943998549,
+ "audio_seconds": 3.0650340136054424,
+ "rtf": 11.221816556458334,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "At last, we have arrived at the weapon of truth.",
+ "raw": {
+ "normalized_reference": "he deluged me, overwhelmed me with argument.",
+ "normalized_hypothesis": "at last, we have arrived at the weapon of truth.",
+ "errors": 10,
+ "words": 7,
+ "wer": 1.4285714285714286
+ },
+ "semantic": {
+ "normalized_reference": "he deluged me overwhelmed me with argument",
+ "normalized_hypothesis": "at last we have arrived at the weapon of truth",
+ "errors": 10,
+ "words": 7,
+ "wer": 1.4285714285714286
+ }
+ },
+ {
+ "id": "release500_0398",
+ "category": "phonetic_balance",
+ "text": "The girl faced him, her eyes shining with sudden fear.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0398.wav",
+ "render_seconds": 48.99578789505176,
+ "audio_seconds": 3.6919727891156464,
+ "rtf": 13.270896264321582,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "I'm going fast. I'm going fast. I'm going fast. I'm going fast. I'm going fast. I'm going fast. I'm going fast. I'm going fast. I'm going fast. I'm going fast. I'm going fast. I'm going fast. I'm going fast. I'm going fast. I'm going fast. I'm going fast. I'm going fast. I'm going fast. I'm going fast. I'm going fast. I'm going fast. I'm going fast. I'm going fast. I'm going fast. I'm going fast. I'm going fast. I'm going fast. I'm going fast. I'm going fast. I'm going fast. I'm going fast. I'm going fast. I'm going fast. I'm going fast. I'm going fast. I'm going fast. I'm going fast. I'm going fast. I'm going fast. I'm going fast. I'm going fast. I'm going fast. I'm going fast. I'm going fast. I'm going fast. I'm going fast. I'm going fast. I'm going fast. I'm going fast. I'm going fast. I'm going fast. I'm going fast. I'm going fast. I'm going fast. I'm going fast. I'm going fast. I'm going fast. I'm going fast. I'm going fast. I'm going fast. I'm going fast. I'm going fast. I'm going fast. I'm going fast. I'm going fast. I'm going fast. I'm going fast. I'm going fast. I'm going fast. I'm going fast. I'm going fast. I'm going fast. I'm going fast. I'm going fast. I'm going fast. I'm going fast. I'm going fast. I'm going fast. I'm going fast. I'm going fast. I'm going fast. I'm going fast. I'm going fast. I'm going fast. I'm going fast. I'm going fast. I'm going fast. I'm going fast. I'm going fast",
+ "raw": {
+ "normalized_reference": "the girl faced him, her eyes shining with sudden fear.",
+ "normalized_hypothesis": "i'm going fast. i'm going fast. i'm going fast. i'm going fast. i'm going fast. i'm going fast. i'm going fast. i'm going fast. i'm going fast. i'm going fast. i'm going fast. i'm going fast. i'm going fast. i'm going fast. i'm going fast. i'm going fast. i'm going fast. i'm going fast. i'm going fast. i'm going fast. i'm going fast. i'm going fast. i'm going fast. i'm going fast. i'm going fast. i'm going fast. i'm going fast. i'm going fast. i'm going fast. i'm going fast. i'm going fast. i'm going fast. i'm going fast. i'm going fast. i'm going fast. i'm going fast. i'm going fast. i'm going fast. i'm going fast. i'm going fast. i'm going fast. i'm going fast. i'm going fast. i'm going fast. i'm going fast. i'm going fast. i'm going fast. i'm going fast. i'm going fast. i'm going fast. i'm going fast. i'm going fast. i'm going fast. i'm going fast. i'm going fast. i'm going fast. i'm going fast. i'm going fast. i'm going fast. i'm going fast. i'm going fast. i'm going fast. i'm going fast. i'm going fast. i'm going fast. i'm going fast. i'm going fast. i'm going fast. i'm going fast. i'm going fast. i'm going fast. i'm going fast. i'm going fast. i'm going fast. i'm going fast. i'm going fast. i'm going fast. i'm going fast. i'm going fast. i'm going fast. i'm going fast. i'm going fast. i'm going fast. i'm going fast. i'm going fast. i'm going fast. i'm going fast. i'm going fast. i'm going fast",
+ "errors": 267,
+ "words": 10,
+ "wer": 26.7
+ },
+ "semantic": {
+ "normalized_reference": "the girl faced him her eyes shining with sudden fear",
+ "normalized_hypothesis": "i am going fast i am going fast i am going fast i am going fast i am going fast i am going fast i am going fast i am going fast i am going fast i am going fast i am going fast i am going fast i am going fast i am going fast i am going fast i am going fast i am going fast i am going fast i am going fast i am going fast i am going fast i am going fast i am going fast i am going fast i am going fast i am going fast i am going fast i am going fast i am going fast i am going fast i am going fast i am going fast i am going fast i am going fast i am going fast i am going fast i am going fast i am going fast i am going fast i am going fast i am going fast i am going fast i am going fast i am going fast i am going fast i am going fast i am going fast i am going fast i am going fast i am going fast i am going fast i am going fast i am going fast i am going fast i am going fast i am going fast i am going fast i am going fast i am going fast i am going fast i am going fast i am going fast i am going fast i am going fast i am going fast i am going fast i am going fast i am going fast i am going fast i am going fast i am going fast i am going fast i am going fast i am going fast i am going fast i am going fast i am going fast i am going fast i am going fast i am going fast i am going fast i am going fast i am going fast i am going fast i am going fast i am going fast i am going fast i am going fast i am going fast",
+ "errors": 356,
+ "words": 10,
+ "wer": 35.6
+ }
+ },
+ {
+ "id": "release500_0399",
+ "category": "phonetic_balance",
+ "text": "It will break our hearts and our backs to hoist anchor by hand.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0399.wav",
+ "render_seconds": 67.49203227902763,
+ "audio_seconds": 4.597551020408163,
+ "rtf": 14.679996367508672,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "Thank you for watching, and I'll see you in the next video.",
+ "raw": {
+ "normalized_reference": "it will break our hearts and our backs to hoist anchor by hand.",
+ "normalized_hypothesis": "thank you for watching, and i'll see you in the next video.",
+ "errors": 12,
+ "words": 13,
+ "wer": 0.9230769230769231
+ },
+ "semantic": {
+ "normalized_reference": "it will break our hearts and our backs to hoist anchor by hand",
+ "normalized_hypothesis": "thank you for watching and i will see you in the next video",
+ "errors": 13,
+ "words": 13,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0400",
+ "category": "phonetic_balance",
+ "text": "He made no reply as he waited for Whittemore to continue.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0400.wav",
+ "render_seconds": 73.58779493893962,
+ "audio_seconds": 4.040272108843538,
+ "rtf": 18.213573976333723,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "I love you. I love you so much. I love you. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so",
+ "raw": {
+ "normalized_reference": "he made no reply as he waited for whittemore to continue.",
+ "normalized_hypothesis": "i love you. i love you so much. i love you. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so",
+ "errors": 370,
+ "words": 11,
+ "wer": 33.63636363636363
+ },
+ "semantic": {
+ "normalized_reference": "he made no reply as he waited for whittemore to continue",
+ "normalized_hypothesis": "i love you i love you so much i love you i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so",
+ "errors": 370,
+ "words": 11,
+ "wer": 33.63636363636363
+ }
+ },
+ {
+ "id": "release500_0401",
+ "category": "phonetic_balance",
+ "text": "Who the devil gave it to you to be judge and jury.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0401.wav",
+ "render_seconds": 45.79701575706713,
+ "audio_seconds": 3.3436734693877552,
+ "rtf": 13.69661726132989,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "You can tell me how much you like my character call.",
+ "raw": {
+ "normalized_reference": "who the devil gave it to you to be judge and jury.",
+ "normalized_hypothesis": "you can tell me how much you like my character call.",
+ "errors": 11,
+ "words": 12,
+ "wer": 0.9166666666666666
+ },
+ "semantic": {
+ "normalized_reference": "who the devil gave it to you to be judge and jury",
+ "normalized_hypothesis": "you can tell me how much you like my character call",
+ "errors": 11,
+ "words": 12,
+ "wer": 0.9166666666666666
+ }
+ },
+ {
+ "id": "release500_0402",
+ "category": "phonetic_balance",
+ "text": "Daylight found himself charmed and made curious by the little man.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0402.wav",
+ "render_seconds": 70.59723766800016,
+ "audio_seconds": 4.527891156462585,
+ "rtf": 15.59163752583537,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry.",
+ "raw": {
+ "normalized_reference": "daylight found himself charmed and made curious by the little man.",
+ "normalized_hypothesis": "i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry.",
+ "errors": 222,
+ "words": 11,
+ "wer": 20.181818181818183
+ },
+ "semantic": {
+ "normalized_reference": "daylight found himself charmed and made curious by the little man",
+ "normalized_hypothesis": "i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry",
+ "errors": 333,
+ "words": 11,
+ "wer": 30.272727272727273
+ }
+ },
+ {
+ "id": "release500_0403",
+ "category": "phonetic_balance",
+ "text": "It's a Yankee, Joan cried.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0403.wav",
+ "render_seconds": 23.088385968934745,
+ "audio_seconds": 2.089795918367347,
+ "rtf": 11.048153442166038,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "I'll see you next time.",
+ "raw": {
+ "normalized_reference": "it's a yankee, joan cried.",
+ "normalized_hypothesis": "i'll see you next time.",
+ "errors": 5,
+ "words": 5,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "it is a yankee joan cried",
+ "normalized_hypothesis": "i will see you next time",
+ "errors": 6,
+ "words": 6,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0404",
+ "category": "phonetic_balance",
+ "text": "But what they want with your toothbrush is more than I can imagine.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0404.wav",
+ "render_seconds": 71.50919498398434,
+ "audio_seconds": 4.388571428571429,
+ "rtf": 16.29441291952768,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "So you are aware that your task force is running out of energy.",
+ "raw": {
+ "normalized_reference": "but what they want with your toothbrush is more than i can imagine.",
+ "normalized_hypothesis": "so you are aware that your task force is running out of energy.",
+ "errors": 12,
+ "words": 13,
+ "wer": 0.9230769230769231
+ },
+ "semantic": {
+ "normalized_reference": "but what they want with your toothbrush is more than i can imagine",
+ "normalized_hypothesis": "so you are aware that your task force is running out of energy",
+ "errors": 12,
+ "words": 13,
+ "wer": 0.9230769230769231
+ }
+ },
+ {
+ "id": "release500_0405",
+ "category": "phonetic_balance",
+ "text": "They were babbling and chattering all together.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0405.wav",
+ "render_seconds": 47.711288860067725,
+ "audio_seconds": 3.2043537414965986,
+ "rtf": 14.889519918541856,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "Hello, baby. I'm sorry. It's okay.",
+ "raw": {
+ "normalized_reference": "they were babbling and chattering all together.",
+ "normalized_hypothesis": "hello, baby. i'm sorry. it's okay.",
+ "errors": 7,
+ "words": 7,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "they were babbling and chattering all together",
+ "normalized_hypothesis": "hello baby i am sorry it is okay",
+ "errors": 8,
+ "words": 7,
+ "wer": 1.1428571428571428
+ }
+ },
+ {
+ "id": "release500_0406",
+ "category": "phonetic_balance",
+ "text": "Men of Selden's stamp don't stop at women and children.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0406.wav",
+ "render_seconds": 65.10381821589544,
+ "audio_seconds": 3.4133333333333336,
+ "rtf": 19.07338424293812,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "Well, if I can start, I'll start on the next cycle.",
+ "raw": {
+ "normalized_reference": "men of selden's stamp don't stop at women and children.",
+ "normalized_hypothesis": "well, if i can start, i'll start on the next cycle.",
+ "errors": 11,
+ "words": 10,
+ "wer": 1.1
+ },
+ "semantic": {
+ "normalized_reference": "men of selden is stamp do not stop at women and children",
+ "normalized_hypothesis": "well if i can start i will start on the next cycle",
+ "errors": 12,
+ "words": 12,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0407",
+ "category": "phonetic_balance",
+ "text": "Cameron looked at his hands with their long, sinewy fingers.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0407.wav",
+ "render_seconds": 60.201275527942926,
+ "audio_seconds": 4.179591836734694,
+ "rtf": 14.403625492525405,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "Come on up to the floors and do the lessons.",
+ "raw": {
+ "normalized_reference": "cameron looked at his hands with their long, sinewy fingers.",
+ "normalized_hypothesis": "come on up to the floors and do the lessons.",
+ "errors": 10,
+ "words": 10,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "cameron looked at his hands with their long sinewy fingers",
+ "normalized_hypothesis": "come on up to the floors and do the lessons",
+ "errors": 10,
+ "words": 10,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0408",
+ "category": "phonetic_balance",
+ "text": "Saxon waited, for she knew a fresh idea had struck Billy.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0408.wav",
+ "render_seconds": 67.69975536607672,
+ "audio_seconds": 4.179591836734694,
+ "rtf": 16.19769537567265,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "Thank you for watching.",
+ "raw": {
+ "normalized_reference": "saxon waited, for she knew a fresh idea had struck billy.",
+ "normalized_hypothesis": "thank you for watching.",
+ "errors": 10,
+ "words": 11,
+ "wer": 0.9090909090909091
+ },
+ "semantic": {
+ "normalized_reference": "saxon waited for she knew a fresh idea had struck billy",
+ "normalized_hypothesis": "thank you for watching",
+ "errors": 10,
+ "words": 11,
+ "wer": 0.9090909090909091
+ }
+ },
+ {
+ "id": "release500_0409",
+ "category": "phonetic_balance",
+ "text": "He saw only the effect in a general, sketchy way.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0409.wav",
+ "render_seconds": 30.415352317038924,
+ "audio_seconds": 3.4133333333333336,
+ "rtf": 8.910747749132497,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "I say that I have thought that that family is not true.",
+ "raw": {
+ "normalized_reference": "he saw only the effect in a general, sketchy way.",
+ "normalized_hypothesis": "i say that i have thought that that family is not true.",
+ "errors": 12,
+ "words": 10,
+ "wer": 1.2
+ },
+ "semantic": {
+ "normalized_reference": "he saw only the effect in a general sketchy way",
+ "normalized_hypothesis": "i say that i have thought that that family is not true",
+ "errors": 12,
+ "words": 10,
+ "wer": 1.2
+ }
+ },
+ {
+ "id": "release500_0410",
+ "category": "phonetic_balance",
+ "text": "He began to follow the footprints of the dog.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0410.wav",
+ "render_seconds": 46.28845179104246,
+ "audio_seconds": 2.995374149659864,
+ "rtf": 15.453312166795152,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "It's time to see the flip flops of the earth.",
+ "raw": {
+ "normalized_reference": "he began to follow the footprints of the dog.",
+ "normalized_hypothesis": "it's time to see the flip flops of the earth.",
+ "errors": 6,
+ "words": 9,
+ "wer": 0.6666666666666666
+ },
+ "semantic": {
+ "normalized_reference": "he began to follow the footprints of the dog",
+ "normalized_hypothesis": "it is time to see the flip flops of the earth",
+ "errors": 7,
+ "words": 9,
+ "wer": 0.7777777777777778
+ }
+ },
+ {
+ "id": "release500_0411",
+ "category": "phonetic_balance",
+ "text": "Do you value your hide.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0411.wav",
+ "render_seconds": 38.296080961008556,
+ "audio_seconds": 1.811156462585034,
+ "rtf": 21.144545903201088,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "Telephone in your heart.",
+ "raw": {
+ "normalized_reference": "do you value your hide.",
+ "normalized_hypothesis": "telephone in your heart.",
+ "errors": 4,
+ "words": 5,
+ "wer": 0.8
+ },
+ "semantic": {
+ "normalized_reference": "do you value your hide",
+ "normalized_hypothesis": "telephone in your heart",
+ "errors": 4,
+ "words": 5,
+ "wer": 0.8
+ }
+ },
+ {
+ "id": "release500_0412",
+ "category": "phonetic_balance",
+ "text": "A rising tide of fat had submerged them.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0412.wav",
+ "render_seconds": 50.398992425063625,
+ "audio_seconds": 2.8560544217687074,
+ "rtf": 17.646369775353357,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "It's a wonderful day to say that shit last night.",
+ "raw": {
+ "normalized_reference": "a rising tide of fat had submerged them.",
+ "normalized_hypothesis": "it's a wonderful day to say that shit last night.",
+ "errors": 9,
+ "words": 8,
+ "wer": 1.125
+ },
+ "semantic": {
+ "normalized_reference": "a rising tide of fat had submerged them",
+ "normalized_hypothesis": "it is a wonderful day to say that shit last night",
+ "errors": 10,
+ "words": 8,
+ "wer": 1.25
+ }
+ },
+ {
+ "id": "release500_0413",
+ "category": "phonetic_balance",
+ "text": "There was nothing on the rock.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0413.wav",
+ "render_seconds": 40.90144200401846,
+ "audio_seconds": 2.2291156462585033,
+ "rtf": 18.348730391207013,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "Al-Fatiha.",
+ "raw": {
+ "normalized_reference": "there was nothing on the rock.",
+ "normalized_hypothesis": "al-fatiha.",
+ "errors": 6,
+ "words": 6,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "there was nothing on the rock",
+ "normalized_hypothesis": "al fatiha",
+ "errors": 6,
+ "words": 6,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0414",
+ "category": "phonetic_balance",
+ "text": "Whatever he guessed he locked away in the taboo room of Naomi.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0414.wav",
+ "render_seconds": 67.39272928400896,
+ "audio_seconds": 4.458231292517007,
+ "rtf": 15.116472175215634,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "Well, that will be us trying to come out of the wall this time.",
+ "raw": {
+ "normalized_reference": "whatever he guessed he locked away in the taboo room of naomi.",
+ "normalized_hypothesis": "well, that will be us trying to come out of the wall this time.",
+ "errors": 14,
+ "words": 12,
+ "wer": 1.1666666666666667
+ },
+ "semantic": {
+ "normalized_reference": "whatever he guessed he locked away in the taboo room of naomi",
+ "normalized_hypothesis": "well that will be us trying to come out of the wall this time",
+ "errors": 14,
+ "words": 12,
+ "wer": 1.1666666666666667
+ }
+ },
+ {
+ "id": "release500_0415",
+ "category": "phonetic_balance",
+ "text": "Will we ever forget it.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0415.wav",
+ "render_seconds": 16.89603050693404,
+ "audio_seconds": 1.9504761904761905,
+ "rtf": 8.66251564076208,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "What is a photo?",
+ "raw": {
+ "normalized_reference": "will we ever forget it.",
+ "normalized_hypothesis": "what is a photo?",
+ "errors": 5,
+ "words": 5,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "will we ever forget it",
+ "normalized_hypothesis": "what is a photo",
+ "errors": 5,
+ "words": 5,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0416",
+ "category": "phonetic_balance",
+ "text": "Jeanne and Pierre both gazed toward the great rock.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0416.wav",
+ "render_seconds": 54.9072054940043,
+ "audio_seconds": 3.622312925170068,
+ "rtf": 15.158051396519367,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "How can I be touched with that word?",
+ "raw": {
+ "normalized_reference": "jeanne and pierre both gazed toward the great rock.",
+ "normalized_hypothesis": "how can i be touched with that word?",
+ "errors": 9,
+ "words": 9,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "jeanne and pierre both gazed toward the great rock",
+ "normalized_hypothesis": "how can i be touched with that word",
+ "errors": 9,
+ "words": 9,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0417",
+ "category": "phonetic_balance",
+ "text": "Does the old boy often go off at half-cock that way.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0417.wav",
+ "render_seconds": 47.9935970089864,
+ "audio_seconds": 3.4829931972789114,
+ "rtf": 13.779411641251954,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "That's not it, guys. We'll call it a night. We'll call it a night. We'll call it a night. We'll call it a night. We'll call it a night. We'll call it a night. We'll call it a night. We'll call it a night. We'll call it a night. We'll call it a night. We'll call it a night. We'll call it a night. We'll call it a night. We'll call it a night. We'll call it a night. We'll call it a night. We'll call it a night. We'll call it a night. We'll call it a night. We'll call it a night. We'll call it a night. We'll call it a night. We'll call it a night. We'll call it a night. We'll call it a night. We'll call it a night. We'll call it a night. We'll call it a night. We'll call it a night. We'll call it a night. We'll call it a night. We'll call it a night. We'll call it a night. We'll call it a night. We'll call it a night. We'll call it a night. We'll call it a night. We'll call it a night. We'll call it a night. We'll call it a night. We'll call it a night. We'll call it a night. We'll call it a night. We'll call it a night. We'll call it a night. We'll call it a night. We'll call it a night. We'll call it a night. We'll call it a night. We'll call it a night. We'll call it a night. We'll call it a night. We'll call it a night. We'll call it a night. We'll call it a night. We'll call it a night. We'll call it a night. We'll call it a night. We'll call it a night. We'll call it a night. We'll call it a night. We'll call it a night. We'll call",
+ "raw": {
+ "normalized_reference": "does the old boy often go off at half-cock that way.",
+ "normalized_hypothesis": "that's not it, guys. we'll call it a night. we'll call it a night. we'll call it a night. we'll call it a night. we'll call it a night. we'll call it a night. we'll call it a night. we'll call it a night. we'll call it a night. we'll call it a night. we'll call it a night. we'll call it a night. we'll call it a night. we'll call it a night. we'll call it a night. we'll call it a night. we'll call it a night. we'll call it a night. we'll call it a night. we'll call it a night. we'll call it a night. we'll call it a night. we'll call it a night. we'll call it a night. we'll call it a night. we'll call it a night. we'll call it a night. we'll call it a night. we'll call it a night. we'll call it a night. we'll call it a night. we'll call it a night. we'll call it a night. we'll call it a night. we'll call it a night. we'll call it a night. we'll call it a night. we'll call it a night. we'll call it a night. we'll call it a night. we'll call it a night. we'll call it a night. we'll call it a night. we'll call it a night. we'll call it a night. we'll call it a night. we'll call it a night. we'll call it a night. we'll call it a night. we'll call it a night. we'll call it a night. we'll call it a night. we'll call it a night. we'll call it a night. we'll call it a night. we'll call it a night. we'll call it a night. we'll call it a night. we'll call it a night. we'll call it a night. we'll call it a night. we'll call it a night. we'll call",
+ "errors": 316,
+ "words": 12,
+ "wer": 26.333333333333332
+ },
+ "semantic": {
+ "normalized_reference": "does the old boy often go off at half cock that way",
+ "normalized_hypothesis": "that is not it guys we will call it a night we will call it a night we will call it a night we will call it a night we will call it a night we will call it a night we will call it a night we will call it a night we will call it a night we will call it a night we will call it a night we will call it a night we will call it a night we will call it a night we will call it a night we will call it a night we will call it a night we will call it a night we will call it a night we will call it a night we will call it a night we will call it a night we will call it a night we will call it a night we will call it a night we will call it a night we will call it a night we will call it a night we will call it a night we will call it a night we will call it a night we will call it a night we will call it a night we will call it a night we will call it a night we will call it a night we will call it a night we will call it a night we will call it a night we will call it a night we will call it a night we will call it a night we will call it a night we will call it a night we will call it a night we will call it a night we will call it a night we will call it a night we will call it a night we will call it a night we will call it a night we will call it a night we will call it a night we will call it a night we will call it a night we will call it a night we will call it a night we will call it a night we will call it a night we will call it a night we will call it a night we will call it a night we will call",
+ "errors": 380,
+ "words": 12,
+ "wer": 31.666666666666668
+ }
+ },
+ {
+ "id": "release500_0418",
+ "category": "phonetic_balance",
+ "text": "The temperature dropped to fifty below zero and remained there the whole trip.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0418.wav",
+ "render_seconds": 78.09716012491845,
+ "audio_seconds": 5.363809523809524,
+ "rtf": 14.56001742385731,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "It's time to call it a day. It's time to call it a day. It's time to call it a day. It's time to call it a day. It's time to call it a day. It's time to call it a day. It's time to call it a day. It's time to call it a day. It's time to call it a day. It's time to call it a day. It's time to call it a day. It's time to call it a day. It's time to call it a day. It's time to call it a day. It's time to call it a day. It's time to call it a day. It's time to call it a day. It's time to call it a day. It's time to call it a day. It's time to call it a day. It's time to call it a day. It's time to call it a day. It's time to call it a day. It's time to call it a day. It's time to call it a day. It's time to call it a day. It's time to call it a day. It's time to call it a day. It's time to call it a day. It's time to call it a day. It's time to call it a day. It's time to call it a day. It's time to call it a day. It's time to call it a day. It's time to call it a day. It's time to call it a day. It's time to call it a day. It's time to call it a day. It's time to call it a day. It's time to call it a day. It's time to call it a day. It's time to call it a day. It's time to call it a day. It's time to call it a day. It's time to call it a day. It's time to call it a day. It's time to call it a day. It's time to call it a day. It's time to call it a day. It's time",
+ "raw": {
+ "normalized_reference": "the temperature dropped to fifty below zero and remained there the whole trip.",
+ "normalized_hypothesis": "it's time to call it a day. it's time to call it a day. it's time to call it a day. it's time to call it a day. it's time to call it a day. it's time to call it a day. it's time to call it a day. it's time to call it a day. it's time to call it a day. it's time to call it a day. it's time to call it a day. it's time to call it a day. it's time to call it a day. it's time to call it a day. it's time to call it a day. it's time to call it a day. it's time to call it a day. it's time to call it a day. it's time to call it a day. it's time to call it a day. it's time to call it a day. it's time to call it a day. it's time to call it a day. it's time to call it a day. it's time to call it a day. it's time to call it a day. it's time to call it a day. it's time to call it a day. it's time to call it a day. it's time to call it a day. it's time to call it a day. it's time to call it a day. it's time to call it a day. it's time to call it a day. it's time to call it a day. it's time to call it a day. it's time to call it a day. it's time to call it a day. it's time to call it a day. it's time to call it a day. it's time to call it a day. it's time to call it a day. it's time to call it a day. it's time to call it a day. it's time to call it a day. it's time to call it a day. it's time to call it a day. it's time to call it a day. it's time to call it a day. it's time",
+ "errors": 344,
+ "words": 13,
+ "wer": 26.46153846153846
+ },
+ "semantic": {
+ "normalized_reference": "the temperature dropped to 50 below 0 and remained there the whole trip",
+ "normalized_hypothesis": "it is time to call it a day it is time to call it a day it is time to call it a day it is time to call it a day it is time to call it a day it is time to call it a day it is time to call it a day it is time to call it a day it is time to call it a day it is time to call it a day it is time to call it a day it is time to call it a day it is time to call it a day it is time to call it a day it is time to call it a day it is time to call it a day it is time to call it a day it is time to call it a day it is time to call it a day it is time to call it a day it is time to call it a day it is time to call it a day it is time to call it a day it is time to call it a day it is time to call it a day it is time to call it a day it is time to call it a day it is time to call it a day it is time to call it a day it is time to call it a day it is time to call it a day it is time to call it a day it is time to call it a day it is time to call it a day it is time to call it a day it is time to call it a day it is time to call it a day it is time to call it a day it is time to call it a day it is time to call it a day it is time to call it a day it is time to call it a day it is time to call it a day it is time to call it a day it is time to call it a day it is time to call it a day it is time to call it a day it is time to call it a day it is time to call it a day it is time",
+ "errors": 394,
+ "words": 13,
+ "wer": 30.307692307692307
+ }
+ },
+ {
+ "id": "release500_0419",
+ "category": "phonetic_balance",
+ "text": "It's worth eight dollars.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0419.wav",
+ "render_seconds": 40.10603851394262,
+ "audio_seconds": 2.020136054421769,
+ "rtf": 19.853137330110336,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "Thank you for watching. I love you.",
+ "raw": {
+ "normalized_reference": "it's worth eight dollars.",
+ "normalized_hypothesis": "thank you for watching. i love you.",
+ "errors": 7,
+ "words": 4,
+ "wer": 1.75
+ },
+ "semantic": {
+ "normalized_reference": "it is worth $8",
+ "normalized_hypothesis": "thank you for watching i love you",
+ "errors": 7,
+ "words": 4,
+ "wer": 1.75
+ }
+ },
+ {
+ "id": "release500_0420",
+ "category": "phonetic_balance",
+ "text": "But I did not enjoy it long.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0420.wav",
+ "render_seconds": 30.903302449034527,
+ "audio_seconds": 2.159455782312925,
+ "rtf": 14.310690083190764,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "Well, I bet nothing for that now.",
+ "raw": {
+ "normalized_reference": "but i did not enjoy it long.",
+ "normalized_hypothesis": "well, i bet nothing for that now.",
+ "errors": 6,
+ "words": 7,
+ "wer": 0.8571428571428571
+ },
+ "semantic": {
+ "normalized_reference": "but i did not enjoy it long",
+ "normalized_hypothesis": "well i bet nothing for that now",
+ "errors": 6,
+ "words": 7,
+ "wer": 0.8571428571428571
+ }
+ },
+ {
+ "id": "release500_0421",
+ "category": "phonetic_balance",
+ "text": "Shorty turned to their employers.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0421.wav",
+ "render_seconds": 50.088806048966944,
+ "audio_seconds": 2.4380952380952383,
+ "rtf": 20.544236856021595,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "For a total of years.",
+ "raw": {
+ "normalized_reference": "shorty turned to their employers.",
+ "normalized_hypothesis": "for a total of years.",
+ "errors": 5,
+ "words": 5,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "shorty turned to their employers",
+ "normalized_hypothesis": "for a total of years",
+ "errors": 5,
+ "words": 5,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0422",
+ "category": "phonetic_balance",
+ "text": "It was the third or fourth time that Philip had heard MacDougall swear.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0422.wav",
+ "render_seconds": 75.50115589995403,
+ "audio_seconds": 5.015510204081632,
+ "rtf": 15.053534501537058,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "I guess that's all I have to say. I'm sorry if I'm wrong, but I'll tell you soon.",
+ "raw": {
+ "normalized_reference": "it was the third or fourth time that philip had heard macdougall swear.",
+ "normalized_hypothesis": "i guess that's all i have to say. i'm sorry if i'm wrong, but i'll tell you soon.",
+ "errors": 18,
+ "words": 13,
+ "wer": 1.3846153846153846
+ },
+ "semantic": {
+ "normalized_reference": "it was the 3rd or 4th time that philip had heard macdougall swear",
+ "normalized_hypothesis": "i guess that is all i have to say i am sorry if i am wrong but i will tell you soon",
+ "errors": 22,
+ "words": 13,
+ "wer": 1.6923076923076923
+ }
+ },
+ {
+ "id": "release500_0423",
+ "category": "phonetic_balance",
+ "text": "Ernest saw in the affair the most sinister import.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0423.wav",
+ "render_seconds": 64.29856270097662,
+ "audio_seconds": 3.6919727891156464,
+ "rtf": 17.415773726863875,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "I'll have to find a place to restore you.",
+ "raw": {
+ "normalized_reference": "ernest saw in the affair the most sinister import.",
+ "normalized_hypothesis": "i'll have to find a place to restore you.",
+ "errors": 9,
+ "words": 9,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "ernest saw in the affair the most sinister import",
+ "normalized_hypothesis": "i will have to find a place to restore you",
+ "errors": 10,
+ "words": 9,
+ "wer": 1.1111111111111112
+ }
+ },
+ {
+ "id": "release500_0424",
+ "category": "phonetic_balance",
+ "text": "Also, at regular intervals, he would mutter.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0424.wav",
+ "render_seconds": 45.8996271060314,
+ "audio_seconds": 3.3436734693877552,
+ "rtf": 13.727305469943474,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "Stay safe. We'll be right back.",
+ "raw": {
+ "normalized_reference": "also, at regular intervals, he would mutter.",
+ "normalized_hypothesis": "stay safe. we'll be right back.",
+ "errors": 7,
+ "words": 7,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "also at regular intervals he would mutter",
+ "normalized_hypothesis": "stay safe we will be right back",
+ "errors": 7,
+ "words": 7,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0425",
+ "category": "phonetic_balance",
+ "text": "He may anticipate the day of his death.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0425.wav",
+ "render_seconds": 48.688397492980585,
+ "audio_seconds": 2.9257142857142857,
+ "rtf": 16.641542111858598,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "I want to sleep again for a change.",
+ "raw": {
+ "normalized_reference": "he may anticipate the day of his death.",
+ "normalized_hypothesis": "i want to sleep again for a change.",
+ "errors": 8,
+ "words": 8,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "he may anticipate the day of his death",
+ "normalized_hypothesis": "i want to sleep again for a change",
+ "errors": 8,
+ "words": 8,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0426",
+ "category": "phonetic_balance",
+ "text": "I defy any man to get a Solomon Island sore in California.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0426.wav",
+ "render_seconds": 50.70114224101417,
+ "audio_seconds": 4.31891156462585,
+ "rtf": 11.739333274680385,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "Thank you for watching, and I'll see you in the next one.",
+ "raw": {
+ "normalized_reference": "i defy any man to get a solomon island sore in california.",
+ "normalized_hypothesis": "thank you for watching, and i'll see you in the next one.",
+ "errors": 12,
+ "words": 12,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "i defy any man to get a solomon island sore in california",
+ "normalized_hypothesis": "thank you for watching and i will see you in the next one",
+ "errors": 13,
+ "words": 12,
+ "wer": 1.0833333333333333
+ }
+ },
+ {
+ "id": "release500_0427",
+ "category": "phonetic_balance",
+ "text": "Now our figuring was all right.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0427.wav",
+ "render_seconds": 21.79856337502133,
+ "audio_seconds": 2.4380952380952383,
+ "rtf": 8.940817009286093,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "You're mine. Today you're nothing but me.",
+ "raw": {
+ "normalized_reference": "now our figuring was all right.",
+ "normalized_hypothesis": "you're mine. today you're nothing but me.",
+ "errors": 7,
+ "words": 6,
+ "wer": 1.1666666666666667
+ },
+ "semantic": {
+ "normalized_reference": "now our figuring was all right",
+ "normalized_hypothesis": "you are mine today you are nothing but me",
+ "errors": 9,
+ "words": 6,
+ "wer": 1.5
+ }
+ },
+ {
+ "id": "release500_0428",
+ "category": "phonetic_balance",
+ "text": "MacDougall tapped his forehead suspiciously with a stubby forefinger.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0428.wav",
+ "render_seconds": 71.49118694197387,
+ "audio_seconds": 5.085170068027211,
+ "rtf": 14.058760274601559,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "Thank you for watching this video. I hope you enjoyed it.",
+ "raw": {
+ "normalized_reference": "macdougall tapped his forehead suspiciously with a stubby forefinger.",
+ "normalized_hypothesis": "thank you for watching this video. i hope you enjoyed it.",
+ "errors": 11,
+ "words": 9,
+ "wer": 1.2222222222222223
+ },
+ "semantic": {
+ "normalized_reference": "macdougall tapped his forehead suspiciously with a stubby forefinger",
+ "normalized_hypothesis": "thank you for watching this video i hope you enjoyed it",
+ "errors": 11,
+ "words": 9,
+ "wer": 1.2222222222222223
+ }
+ },
+ {
+ "id": "release500_0429",
+ "category": "phonetic_balance",
+ "text": "O'Brien took off his coat and bared his right arm.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0429.wav",
+ "render_seconds": 60.09927179699298,
+ "audio_seconds": 3.4133333333333336,
+ "rtf": 17.607208534275287,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "The boy I'm talking about is killing me as well.",
+ "raw": {
+ "normalized_reference": "o'brien took off his coat and bared his right arm.",
+ "normalized_hypothesis": "the boy i'm talking about is killing me as well.",
+ "errors": 10,
+ "words": 10,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "0 brien took off his coat and bared his right arm",
+ "normalized_hypothesis": "the boy i am talking about is killing me as well",
+ "errors": 11,
+ "words": 11,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0430",
+ "category": "phonetic_balance",
+ "text": "But it won't continue, she said with easy confidence.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0430.wav",
+ "render_seconds": 57.308990280958824,
+ "audio_seconds": 3.5526530612244898,
+ "rtf": 16.131321942595257,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "Thank you for watching. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this video. I hope you enjoyed this",
+ "raw": {
+ "normalized_reference": "but it won't continue, she said with easy confidence.",
+ "normalized_hypothesis": "thank you for watching. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this video. i hope you enjoyed this",
+ "errors": 381,
+ "words": 9,
+ "wer": 42.333333333333336
+ },
+ "semantic": {
+ "normalized_reference": "but it will not continue she said with easy confidence",
+ "normalized_hypothesis": "thank you for watching i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this video i hope you enjoyed this",
+ "errors": 381,
+ "words": 10,
+ "wer": 38.1
+ }
+ },
+ {
+ "id": "release500_0431",
+ "category": "phonetic_balance",
+ "text": "I was still weak from my prolonged immersion.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0431.wav",
+ "render_seconds": 36.81127409497276,
+ "audio_seconds": 3.3436734693877552,
+ "rtf": 11.009231144126375,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "I'm still waiting for you to open the door for me.",
+ "raw": {
+ "normalized_reference": "i was still weak from my prolonged immersion.",
+ "normalized_hypothesis": "i'm still waiting for you to open the door for me.",
+ "errors": 11,
+ "words": 8,
+ "wer": 1.375
+ },
+ "semantic": {
+ "normalized_reference": "i was still weak from my prolonged immersion",
+ "normalized_hypothesis": "i am still waiting for you to open the door for me",
+ "errors": 10,
+ "words": 8,
+ "wer": 1.25
+ }
+ },
+ {
+ "id": "release500_0432",
+ "category": "phonetic_balance",
+ "text": "Oolong was two hundred and fifty miles from the nearest land.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0432.wav",
+ "render_seconds": 48.21655345603358,
+ "audio_seconds": 4.249251700680272,
+ "rtf": 11.347069284767123,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "You must take him to me after I've shown you the snake.",
+ "raw": {
+ "normalized_reference": "oolong was two hundred and fifty miles from the nearest land.",
+ "normalized_hypothesis": "you must take him to me after i've shown you the snake.",
+ "errors": 12,
+ "words": 11,
+ "wer": 1.0909090909090908
+ },
+ "semantic": {
+ "normalized_reference": "oolong was 250 miles from the nearest land",
+ "normalized_hypothesis": "you must take him to me after i have shown you the snake",
+ "errors": 13,
+ "words": 8,
+ "wer": 1.625
+ }
+ },
+ {
+ "id": "release500_0433",
+ "category": "phonetic_balance",
+ "text": "I just do appreciate it without being able to express my feelings.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0433.wav",
+ "render_seconds": 37.79948809801135,
+ "audio_seconds": 4.806530612244898,
+ "rtf": 7.864193770391289,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "Thank you for watching. I love you. Talk to you soon. Bye.",
+ "raw": {
+ "normalized_reference": "i just do appreciate it without being able to express my feelings.",
+ "normalized_hypothesis": "thank you for watching. i love you. talk to you soon. bye.",
+ "errors": 11,
+ "words": 12,
+ "wer": 0.9166666666666666
+ },
+ "semantic": {
+ "normalized_reference": "i just do appreciate it without being able to express my feelings",
+ "normalized_hypothesis": "thank you for watching i love you talk to you soon bye",
+ "errors": 11,
+ "words": 12,
+ "wer": 0.9166666666666666
+ }
+ },
+ {
+ "id": "release500_0434",
+ "category": "phonetic_balance",
+ "text": "Now just what do you want to know.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0434.wav",
+ "render_seconds": 51.29997921502218,
+ "audio_seconds": 2.36843537414966,
+ "rtf": 21.659860249908835,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "Thank you so much for watching.",
+ "raw": {
+ "normalized_reference": "now just what do you want to know.",
+ "normalized_hypothesis": "thank you so much for watching.",
+ "errors": 8,
+ "words": 8,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "now just what do you want to know",
+ "normalized_hypothesis": "thank you so much for watching",
+ "errors": 8,
+ "words": 8,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0435",
+ "category": "phonetic_balance",
+ "text": "But she swung obediently on her heel into the wind.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0435.wav",
+ "render_seconds": 60.00744204397779,
+ "audio_seconds": 3.4133333333333336,
+ "rtf": 17.580305286321618,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "Thank you so much for watching. I'll see you in the next video.",
+ "raw": {
+ "normalized_reference": "but she swung obediently on her heel into the wind.",
+ "normalized_hypothesis": "thank you so much for watching. i'll see you in the next video.",
+ "errors": 12,
+ "words": 10,
+ "wer": 1.2
+ },
+ "semantic": {
+ "normalized_reference": "but she swung obediently on her heel into the wind",
+ "normalized_hypothesis": "thank you so much for watching i will see you in the next video",
+ "errors": 13,
+ "words": 10,
+ "wer": 1.3
+ }
+ },
+ {
+ "id": "release500_0436",
+ "category": "phonetic_balance",
+ "text": "Ask him, I laughed, then turned to Pasquini.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0436.wav",
+ "render_seconds": 40.29292599798646,
+ "audio_seconds": 3.0650340136054424,
+ "rtf": 13.145996363867207,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "Ask your master to ask me.",
+ "raw": {
+ "normalized_reference": "ask him, i laughed, then turned to pasquini.",
+ "normalized_hypothesis": "ask your master to ask me.",
+ "errors": 7,
+ "words": 8,
+ "wer": 0.875
+ },
+ "semantic": {
+ "normalized_reference": "ask him i laughed then turned to pasquini",
+ "normalized_hypothesis": "ask your master to ask me",
+ "errors": 7,
+ "words": 8,
+ "wer": 0.875
+ }
+ },
+ {
+ "id": "release500_0437",
+ "category": "phonetic_balance",
+ "text": "He did not rush in.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0437.wav",
+ "render_seconds": 38.699202345917,
+ "audio_seconds": 1.7414965986394557,
+ "rtf": 22.221807597069528,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "I don't have one.",
+ "raw": {
+ "normalized_reference": "he did not rush in.",
+ "normalized_hypothesis": "i don't have one.",
+ "errors": 5,
+ "words": 5,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "he did not rush in",
+ "normalized_hypothesis": "i do not have one",
+ "errors": 4,
+ "words": 5,
+ "wer": 0.8
+ }
+ },
+ {
+ "id": "release500_0438",
+ "category": "phonetic_balance",
+ "text": "They obeyed him, and went here and there at his commands.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0438.wav",
+ "render_seconds": 48.70280516904313,
+ "audio_seconds": 3.900952380952381,
+ "rtf": 12.484849957884592,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "Bye-bye. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you.",
+ "raw": {
+ "normalized_reference": "they obeyed him, and went here and there at his commands.",
+ "normalized_hypothesis": "bye-bye. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you.",
+ "errors": 26,
+ "words": 11,
+ "wer": 2.3636363636363638
+ },
+ "semantic": {
+ "normalized_reference": "they obeyed him and went here and there at his commands",
+ "normalized_hypothesis": "bye bye i love you i love you i love you i love you i love you i love you i love you i love you",
+ "errors": 26,
+ "words": 11,
+ "wer": 2.3636363636363638
+ }
+ },
+ {
+ "id": "release500_0439",
+ "category": "phonetic_balance",
+ "text": "Tom Spink has a harpoon.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0439.wav",
+ "render_seconds": 38.82369073200971,
+ "audio_seconds": 2.159455782312925,
+ "rtf": 17.978460614936452,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "Yes, it doesn't have to.",
+ "raw": {
+ "normalized_reference": "tom spink has a harpoon.",
+ "normalized_hypothesis": "yes, it doesn't have to.",
+ "errors": 5,
+ "words": 5,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "tom spink has a harpoon",
+ "normalized_hypothesis": "yes it does not have to",
+ "errors": 6,
+ "words": 5,
+ "wer": 1.2
+ }
+ },
+ {
+ "id": "release500_0440",
+ "category": "phonetic_balance",
+ "text": "It was my reports from the north which chiefly induced people to buy.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0440.wav",
+ "render_seconds": 56.10623948997818,
+ "audio_seconds": 4.876190476190477,
+ "rtf": 11.50616239540568,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "I lost a lot of cool action in my life. What should I do now, guys? I can't wait to go.",
+ "raw": {
+ "normalized_reference": "it was my reports from the north which chiefly induced people to buy.",
+ "normalized_hypothesis": "i lost a lot of cool action in my life. what should i do now, guys? i can't wait to go.",
+ "errors": 19,
+ "words": 13,
+ "wer": 1.4615384615384615
+ },
+ "semantic": {
+ "normalized_reference": "it was my reports from the north which chiefly induced people to buy",
+ "normalized_hypothesis": "i lost a lot of cool action in my life what should i do now guys i can not wait to go",
+ "errors": 20,
+ "words": 13,
+ "wer": 1.5384615384615385
+ }
+ },
+ {
+ "id": "release500_0441",
+ "category": "phonetic_balance",
+ "text": "Cherokee identified himself with his instinct.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0441.wav",
+ "render_seconds": 59.99933752510697,
+ "audio_seconds": 3.2043537414965986,
+ "rtf": 18.724317714399465,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "Thank you for listening. We'll see you next time.",
+ "raw": {
+ "normalized_reference": "cherokee identified himself with his instinct.",
+ "normalized_hypothesis": "thank you for listening. we'll see you next time.",
+ "errors": 9,
+ "words": 6,
+ "wer": 1.5
+ },
+ "semantic": {
+ "normalized_reference": "cherokee identified himself with his instinct",
+ "normalized_hypothesis": "thank you for listening we will see you next time",
+ "errors": 10,
+ "words": 6,
+ "wer": 1.6666666666666667
+ }
+ },
+ {
+ "id": "release500_0442",
+ "category": "phonetic_balance",
+ "text": "I graduated last of my class.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0442.wav",
+ "render_seconds": 35.50398636702448,
+ "audio_seconds": 2.4380952380952383,
+ "rtf": 14.562181908349883,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "The author of Lost Reckless.",
+ "raw": {
+ "normalized_reference": "i graduated last of my class.",
+ "normalized_hypothesis": "the author of lost reckless.",
+ "errors": 5,
+ "words": 6,
+ "wer": 0.8333333333333334
+ },
+ "semantic": {
+ "normalized_reference": "i graduated last of my class",
+ "normalized_hypothesis": "the author of lost reckless",
+ "errors": 5,
+ "words": 6,
+ "wer": 0.8333333333333334
+ }
+ },
+ {
+ "id": "release500_0443",
+ "category": "phonetic_balance",
+ "text": "It was a gigantic inadequacy.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0443.wav",
+ "render_seconds": 33.90033615392167,
+ "audio_seconds": 2.2987755102040817,
+ "rtf": 14.74712776582175,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "I was a good boy back when I was a little boy.",
+ "raw": {
+ "normalized_reference": "it was a gigantic inadequacy.",
+ "normalized_hypothesis": "i was a good boy back when i was a little boy.",
+ "errors": 10,
+ "words": 5,
+ "wer": 2.0
+ },
+ "semantic": {
+ "normalized_reference": "it was a gigantic inadequacy",
+ "normalized_hypothesis": "i was a good boy back when i was a little boy",
+ "errors": 10,
+ "words": 5,
+ "wer": 2.0
+ }
+ },
+ {
+ "id": "release500_0444",
+ "category": "phonetic_balance",
+ "text": "Did I possess too much vitality.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0444.wav",
+ "render_seconds": 41.29556576604955,
+ "audio_seconds": 2.507755102040816,
+ "rtf": 16.467144551891504,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "To the past, to the future.",
+ "raw": {
+ "normalized_reference": "did i possess too much vitality.",
+ "normalized_hypothesis": "to the past, to the future.",
+ "errors": 6,
+ "words": 6,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "did i possess too much vitality",
+ "normalized_hypothesis": "to the past to the future",
+ "errors": 6,
+ "words": 6,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0445",
+ "category": "phonetic_balance",
+ "text": "Therefore, hurrah for the game.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0445.wav",
+ "render_seconds": 21.084058632026426,
+ "audio_seconds": 2.2291156462585033,
+ "rtf": 9.458485775475722,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "I'll sign your ass like a girl.",
+ "raw": {
+ "normalized_reference": "therefore, hurrah for the game.",
+ "normalized_hypothesis": "i'll sign your ass like a girl.",
+ "errors": 7,
+ "words": 5,
+ "wer": 1.4
+ },
+ "semantic": {
+ "normalized_reference": "therefore hurrah for the game",
+ "normalized_hypothesis": "i will sign your ass like a girl",
+ "errors": 8,
+ "words": 5,
+ "wer": 1.6
+ }
+ },
+ {
+ "id": "release500_0446",
+ "category": "phonetic_balance",
+ "text": "He had fulfilled his duty and paid properly.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0446.wav",
+ "render_seconds": 47.69303379300982,
+ "audio_seconds": 3.1346938775510202,
+ "rtf": 15.214574582405476,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "That's the saddest time of the year for me.",
+ "raw": {
+ "normalized_reference": "he had fulfilled his duty and paid properly.",
+ "normalized_hypothesis": "that's the saddest time of the year for me.",
+ "errors": 9,
+ "words": 8,
+ "wer": 1.125
+ },
+ "semantic": {
+ "normalized_reference": "he had fulfilled his duty and paid properly",
+ "normalized_hypothesis": "that is the saddest time of the year for me",
+ "errors": 10,
+ "words": 8,
+ "wer": 1.25
+ }
+ },
+ {
+ "id": "release500_0447",
+ "category": "phonetic_balance",
+ "text": "How can you manage all alone, Mr Young.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0447.wav",
+ "render_seconds": 45.49568775296211,
+ "audio_seconds": 2.8560544217687074,
+ "rtf": 15.929559117009887,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "Happy, happy, happy birthday.",
+ "raw": {
+ "normalized_reference": "how can you manage all alone, mr young.",
+ "normalized_hypothesis": "happy, happy, happy birthday.",
+ "errors": 8,
+ "words": 8,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "how can you manage all alone mister young",
+ "normalized_hypothesis": "happy happy happy birthday",
+ "errors": 8,
+ "words": 8,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0448",
+ "category": "phonetic_balance",
+ "text": "Red-Eye never committed a more outrageous deed.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0448.wav",
+ "render_seconds": 43.70079360692762,
+ "audio_seconds": 3.5526530612244898,
+ "rtf": 12.300889744597043,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "Thank you for watching!",
+ "raw": {
+ "normalized_reference": "red-eye never committed a more outrageous deed.",
+ "normalized_hypothesis": "thank you for watching!",
+ "errors": 8,
+ "words": 8,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "red eye never committed a more outrageous deed",
+ "normalized_hypothesis": "thank you for watching",
+ "errors": 8,
+ "words": 8,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0449",
+ "category": "phonetic_balance",
+ "text": "He would keep his faith with Gregson for the promised day or two.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0449.wav",
+ "render_seconds": 71.59551129001193,
+ "audio_seconds": 4.597551020408163,
+ "rtf": 15.572532196424824,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "I want to have a try with the correction for the previous character.",
+ "raw": {
+ "normalized_reference": "he would keep his faith with gregson for the promised day or two.",
+ "normalized_hypothesis": "i want to have a try with the correction for the previous character.",
+ "errors": 12,
+ "words": 13,
+ "wer": 0.9230769230769231
+ },
+ "semantic": {
+ "normalized_reference": "he would keep his faith with gregson for the promised day or 2",
+ "normalized_hypothesis": "i want to have a try with the correction for the previous character",
+ "errors": 12,
+ "words": 13,
+ "wer": 0.9230769230769231
+ }
+ },
+ {
+ "id": "release500_0450",
+ "category": "phonetic_balance",
+ "text": "There was one chance, and only one, of saving Jeanne.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0450.wav",
+ "render_seconds": 40.0976740430342,
+ "audio_seconds": 3.8312925170068026,
+ "rtf": 10.465834666772066,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "I wish you all the best. I love you. I love you too.",
+ "raw": {
+ "normalized_reference": "there was one chance, and only one, of saving jeanne.",
+ "normalized_hypothesis": "i wish you all the best. i love you. i love you too.",
+ "errors": 13,
+ "words": 10,
+ "wer": 1.3
+ },
+ "semantic": {
+ "normalized_reference": "there was one chance and only one of saving jeanne",
+ "normalized_hypothesis": "i wish you all the best i love you i love you too",
+ "errors": 13,
+ "words": 10,
+ "wer": 1.3
+ }
+ },
+ {
+ "id": "release500_0451",
+ "category": "phonetic_balance",
+ "text": "I was beneath the water, suffocating and drowning.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0451.wav",
+ "render_seconds": 24.89961629291065,
+ "audio_seconds": 3.4829931972789114,
+ "rtf": 7.148913271597394,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "Thank you for watching. I hope you enjoyed this video.",
+ "raw": {
+ "normalized_reference": "i was beneath the water, suffocating and drowning.",
+ "normalized_hypothesis": "thank you for watching. i hope you enjoyed this video.",
+ "errors": 10,
+ "words": 8,
+ "wer": 1.25
+ },
+ "semantic": {
+ "normalized_reference": "i was beneath the water suffocating and drowning",
+ "normalized_hypothesis": "thank you for watching i hope you enjoyed this video",
+ "errors": 10,
+ "words": 8,
+ "wer": 1.25
+ }
+ },
+ {
+ "id": "release500_0452",
+ "category": "phonetic_balance",
+ "text": "Give them their choice between a fine or an official whipping.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0452.wav",
+ "render_seconds": 52.30043519998435,
+ "audio_seconds": 4.388571428571429,
+ "rtf": 11.917416874996434,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "Thank you for watching. See you in the next video.",
+ "raw": {
+ "normalized_reference": "give them their choice between a fine or an official whipping.",
+ "normalized_hypothesis": "thank you for watching. see you in the next video.",
+ "errors": 11,
+ "words": 11,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "give them their choice between a fine or an official whipping",
+ "normalized_hypothesis": "thank you for watching see you in the next video",
+ "errors": 11,
+ "words": 11,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0453",
+ "category": "phonetic_balance",
+ "text": "They must have been swept away by the chaotic currents.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0453.wav",
+ "render_seconds": 57.59676754404791,
+ "audio_seconds": 3.900952380952381,
+ "rtf": 14.76479636749275,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "See you next time on Shrek 2.0. Bye for now. See you next time on Shrek 2.0. Bye for now. See you next time on Shrek 2.0. Bye for now. See you next time on Shrek 2.0. Bye for now. See you next time on Shrek 2.0. Bye for now. See you next time on Shrek 2.0. Bye for now. See you next time on Shrek 2.0. Bye for now. See you next time on Shrek 2.0. Bye for now. See you next time on Shrek 2.0. Bye for now. See you next time on Shrek 2.0. Bye for now. See you next time on Shrek 2.0. Bye for now. See you next time on Shrek 2.0. Bye for now. See you next time on Shrek 2.0. Bye for now. See you next time on Shrek 2.0. Bye for now. See you next time on Shrek 2.0. Bye for now. See you next time on Shrek 2.0. Bye for now. See you next time on Shrek 2.0. Bye for now. See you next time on Shrek 2.0. Bye for now. See you next time on Shrek 2.0. Bye for now. See you next time on Shrek 2.0. Bye for now. See you next time on Shrek 2.0. Bye for now. See you next time on Shrek 2.0. Bye for now. See you next time on Shrek 2.0. Bye for now. See you next time on Shrek 2.0. Bye for now. See you next time on Shrek 2.0. Bye for now. See you next time on Shrek 2. Bye for now. See you next time on Shrek 2. Bye for now. See you next time on Shrek 2. Bye for now. See you next time on Shrek 2. Bye for now. See you next time on Shrek 2. Bye for now. See you next time",
+ "raw": {
+ "normalized_reference": "they must have been swept away by the chaotic currents.",
+ "normalized_hypothesis": "see you next time on shrek 2.0. bye for now. see you next time on shrek 2.0. bye for now. see you next time on shrek 2.0. bye for now. see you next time on shrek 2.0. bye for now. see you next time on shrek 2.0. bye for now. see you next time on shrek 2.0. bye for now. see you next time on shrek 2.0. bye for now. see you next time on shrek 2.0. bye for now. see you next time on shrek 2.0. bye for now. see you next time on shrek 2.0. bye for now. see you next time on shrek 2.0. bye for now. see you next time on shrek 2.0. bye for now. see you next time on shrek 2.0. bye for now. see you next time on shrek 2.0. bye for now. see you next time on shrek 2.0. bye for now. see you next time on shrek 2.0. bye for now. see you next time on shrek 2.0. bye for now. see you next time on shrek 2.0. bye for now. see you next time on shrek 2.0. bye for now. see you next time on shrek 2.0. bye for now. see you next time on shrek 2.0. bye for now. see you next time on shrek 2.0. bye for now. see you next time on shrek 2.0. bye for now. see you next time on shrek 2.0. bye for now. see you next time on shrek 2.0. bye for now. see you next time on shrek 2. bye for now. see you next time on shrek 2. bye for now. see you next time on shrek 2. bye for now. see you next time on shrek 2. bye for now. see you next time on shrek 2. bye for now. see you next time",
+ "errors": 329,
+ "words": 10,
+ "wer": 32.9
+ },
+ "semantic": {
+ "normalized_reference": "they must have been swept away by the chaotic currents",
+ "normalized_hypothesis": "see you next time on shrek 2 bye for now see you next time on shrek 2 bye for now see you next time on shrek 2 bye for now see you next time on shrek 2 bye for now see you next time on shrek 2 bye for now see you next time on shrek 2 bye for now see you next time on shrek 2 bye for now see you next time on shrek 2 bye for now see you next time on shrek 2 bye for now see you next time on shrek 2 bye for now see you next time on shrek 2 bye for now see you next time on shrek 2 bye for now see you next time on shrek 2 bye for now see you next time on shrek 2 bye for now see you next time on shrek 2 bye for now see you next time on shrek 2 bye for now see you next time on shrek 2 bye for now see you next time on shrek 2 bye for now see you next time on shrek 2 bye for now see you next time on shrek 2 bye for now see you next time on shrek 2 bye for now see you next time on shrek 2 bye for now see you next time on shrek 2 bye for now see you next time on shrek 2 bye for now see you next time on shrek 2 bye for now see you next time on shrek 2 bye for now see you next time on shrek 2 bye for now see you next time on shrek 2 bye for now see you next time on shrek 2 bye for now see you next time on shrek 2 bye for now see you next time",
+ "errors": 304,
+ "words": 10,
+ "wer": 30.4
+ }
+ },
+ {
+ "id": "release500_0454",
+ "category": "phonetic_balance",
+ "text": "You yellow giant thing of the frost.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0454.wav",
+ "render_seconds": 42.49232076504268,
+ "audio_seconds": 2.4380952380952383,
+ "rtf": 17.428490938787036,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "I know you're trying to get possessed.",
+ "raw": {
+ "normalized_reference": "you yellow giant thing of the frost.",
+ "normalized_hypothesis": "i know you're trying to get possessed.",
+ "errors": 7,
+ "words": 7,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "you yellow giant thing of the frost",
+ "normalized_hypothesis": "i know you are trying to get possessed",
+ "errors": 8,
+ "words": 7,
+ "wer": 1.1428571428571428
+ }
+ },
+ {
+ "id": "release500_0455",
+ "category": "phonetic_balance",
+ "text": "He had peeled off his shirt and was wildly waving it.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0455.wav",
+ "render_seconds": 66.29271442000754,
+ "audio_seconds": 3.6919727891156464,
+ "rtf": 17.955905475643256,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "I'm tired of this. I'm tired of this. I'm tired of this. I'm tired of this. I'm tired of this. I'm tired of this. I'm tired of this. I'm tired of this. I'm tired of this. I'm tired of this. I'm tired of this. I'm tired of this. I'm tired of this. I'm tired of this. I'm tired of this. I'm tired of this. I'm tired of this. I'm tired of this. I'm tired of this. I'm tired of this. I'm tired of this. I'm tired of this. I'm tired of this. I'm tired of this. I'm tired of this. I'm tired of this. I'm tired of this. I'm tired of this. I'm tired of this. I'm tired of this. I'm tired of this. I'm tired of this. I'm tired of this. I'm tired of this. I'm tired of this. I'm tired of this. I'm tired of this. I'm tired of this. I'm tired of this. I'm tired of this. I'm tired of this. I'm tired of this. I'm tired of this. I'm tired of this. I'm tired of this. I'm tired of this. I'm tired of this. I'm tired of this. I'm tired of this. I'm tired of this. I'm tired of this. I'm tired of this. I'm tired of this. I'm tired of this. I'm tired of this. I'm tired of this. I'm tired of this. I'm tired of this. I'm tired of this. I'm tired of this. I'm tired of this. I'm tired of this. I'm tired of this. I'm tired of this. I'm tired of this. I'm tired of this. I'm tired of this. I'm tired of this. I'm tired of this. I'm tired of this. I'm tired of this. I'm tired of this. I'm tired of this. I'm tired of this.",
+ "raw": {
+ "normalized_reference": "he had peeled off his shirt and was wildly waving it.",
+ "normalized_hypothesis": "i'm tired of this. i'm tired of this. i'm tired of this. i'm tired of this. i'm tired of this. i'm tired of this. i'm tired of this. i'm tired of this. i'm tired of this. i'm tired of this. i'm tired of this. i'm tired of this. i'm tired of this. i'm tired of this. i'm tired of this. i'm tired of this. i'm tired of this. i'm tired of this. i'm tired of this. i'm tired of this. i'm tired of this. i'm tired of this. i'm tired of this. i'm tired of this. i'm tired of this. i'm tired of this. i'm tired of this. i'm tired of this. i'm tired of this. i'm tired of this. i'm tired of this. i'm tired of this. i'm tired of this. i'm tired of this. i'm tired of this. i'm tired of this. i'm tired of this. i'm tired of this. i'm tired of this. i'm tired of this. i'm tired of this. i'm tired of this. i'm tired of this. i'm tired of this. i'm tired of this. i'm tired of this. i'm tired of this. i'm tired of this. i'm tired of this. i'm tired of this. i'm tired of this. i'm tired of this. i'm tired of this. i'm tired of this. i'm tired of this. i'm tired of this. i'm tired of this. i'm tired of this. i'm tired of this. i'm tired of this. i'm tired of this. i'm tired of this. i'm tired of this. i'm tired of this. i'm tired of this. i'm tired of this. i'm tired of this. i'm tired of this. i'm tired of this. i'm tired of this. i'm tired of this. i'm tired of this. i'm tired of this. i'm tired of this.",
+ "errors": 296,
+ "words": 11,
+ "wer": 26.90909090909091
+ },
+ "semantic": {
+ "normalized_reference": "he had peeled off his shirt and was wildly waving it",
+ "normalized_hypothesis": "i am tired of this i am tired of this i am tired of this i am tired of this i am tired of this i am tired of this i am tired of this i am tired of this i am tired of this i am tired of this i am tired of this i am tired of this i am tired of this i am tired of this i am tired of this i am tired of this i am tired of this i am tired of this i am tired of this i am tired of this i am tired of this i am tired of this i am tired of this i am tired of this i am tired of this i am tired of this i am tired of this i am tired of this i am tired of this i am tired of this i am tired of this i am tired of this i am tired of this i am tired of this i am tired of this i am tired of this i am tired of this i am tired of this i am tired of this i am tired of this i am tired of this i am tired of this i am tired of this i am tired of this i am tired of this i am tired of this i am tired of this i am tired of this i am tired of this i am tired of this i am tired of this i am tired of this i am tired of this i am tired of this i am tired of this i am tired of this i am tired of this i am tired of this i am tired of this i am tired of this i am tired of this i am tired of this i am tired of this i am tired of this i am tired of this i am tired of this i am tired of this i am tired of this i am tired of this i am tired of this i am tired of this i am tired of this i am tired of this i am tired of this",
+ "errors": 370,
+ "words": 11,
+ "wer": 33.63636363636363
+ }
+ },
+ {
+ "id": "release500_0456",
+ "category": "phonetic_balance",
+ "text": "To these he gave castor oil.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0456.wav",
+ "render_seconds": 27.798200483899564,
+ "audio_seconds": 2.2291156462585033,
+ "rtf": 12.47050619852672,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "Come on, son, let's get this going.",
+ "raw": {
+ "normalized_reference": "to these he gave castor oil.",
+ "normalized_hypothesis": "come on, son, let's get this going.",
+ "errors": 7,
+ "words": 6,
+ "wer": 1.1666666666666667
+ },
+ "semantic": {
+ "normalized_reference": "to these he gave castor oil",
+ "normalized_hypothesis": "come on son let us get this going",
+ "errors": 8,
+ "words": 6,
+ "wer": 1.3333333333333333
+ }
+ },
+ {
+ "id": "release500_0457",
+ "category": "phonetic_balance",
+ "text": "Tudor surveyed him with withering disgust.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0457.wav",
+ "render_seconds": 25.503747343900613,
+ "audio_seconds": 2.9257142857142857,
+ "rtf": 8.71710114293478,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "To the soul of the red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red",
+ "raw": {
+ "normalized_reference": "tudor surveyed him with withering disgust.",
+ "normalized_hypothesis": "to the soul of the red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red-red",
+ "errors": 225,
+ "words": 6,
+ "wer": 37.5
+ },
+ "semantic": {
+ "normalized_reference": "tudor surveyed him with withering disgust",
+ "normalized_hypothesis": "to the soul of the red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red red",
+ "errors": 225,
+ "words": 6,
+ "wer": 37.5
+ }
+ },
+ {
+ "id": "release500_0458",
+ "category": "phonetic_balance",
+ "text": "After all, it was simply a mistake in judgment.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0458.wav",
+ "render_seconds": 40.506310326978564,
+ "audio_seconds": 3.4829931972789114,
+ "rtf": 11.629741441534861,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "I'm going to the shelter to start to catch them.",
+ "raw": {
+ "normalized_reference": "after all, it was simply a mistake in judgment.",
+ "normalized_hypothesis": "i'm going to the shelter to start to catch them.",
+ "errors": 10,
+ "words": 9,
+ "wer": 1.1111111111111112
+ },
+ "semantic": {
+ "normalized_reference": "after all it was simply a mistake in judgment",
+ "normalized_hypothesis": "i am going to the shelter to start to catch them",
+ "errors": 11,
+ "words": 9,
+ "wer": 1.2222222222222223
+ }
+ },
+ {
+ "id": "release500_0459",
+ "category": "phonetic_balance",
+ "text": "That is the strange part of it.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0459.wav",
+ "render_seconds": 51.41124410997145,
+ "audio_seconds": 2.2987755102040817,
+ "rtf": 22.364621461191415,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "S-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-",
+ "raw": {
+ "normalized_reference": "that is the strange part of it.",
+ "normalized_hypothesis": "s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-s-",
+ "errors": 222,
+ "words": 7,
+ "wer": 31.714285714285715
+ },
+ "semantic": {
+ "normalized_reference": "that is the strange part of it",
+ "normalized_hypothesis": "s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s s",
+ "errors": 222,
+ "words": 7,
+ "wer": 31.714285714285715
+ }
+ },
+ {
+ "id": "release500_0460",
+ "category": "phonetic_balance",
+ "text": "Not a wheel moved in his empire.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0460.wav",
+ "render_seconds": 39.307403607992455,
+ "audio_seconds": 2.507755102040816,
+ "rtf": 15.674339003838138,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "We are on our way to Europe now.",
+ "raw": {
+ "normalized_reference": "not a wheel moved in his empire.",
+ "normalized_hypothesis": "we are on our way to europe now.",
+ "errors": 8,
+ "words": 7,
+ "wer": 1.1428571428571428
+ },
+ "semantic": {
+ "normalized_reference": "not a wheel moved in his empire",
+ "normalized_hypothesis": "we are on our way to europe now",
+ "errors": 8,
+ "words": 7,
+ "wer": 1.1428571428571428
+ }
+ },
+ {
+ "id": "release500_0461",
+ "category": "phonetic_balance",
+ "text": "Its diameter was not more than two hundred yards.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0461.wav",
+ "render_seconds": 51.40658588602673,
+ "audio_seconds": 3.4133333333333336,
+ "rtf": 15.060523208796893,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "I still don't know what's going on out there.",
+ "raw": {
+ "normalized_reference": "its diameter was not more than two hundred yards.",
+ "normalized_hypothesis": "i still don't know what's going on out there.",
+ "errors": 9,
+ "words": 9,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "its diameter was not more than 200 yards",
+ "normalized_hypothesis": "i still do not know what is going on out there",
+ "errors": 10,
+ "words": 8,
+ "wer": 1.25
+ }
+ },
+ {
+ "id": "release500_0462",
+ "category": "phonetic_balance",
+ "text": "Nimrod replied, with a slight manifestation of sensitiveness.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0462.wav",
+ "render_seconds": 42.29929517302662,
+ "audio_seconds": 4.667210884353741,
+ "rtf": 9.06307776124492,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "Now I have appeared. Let us see your manifestation of self-sacrifice.",
+ "raw": {
+ "normalized_reference": "nimrod replied, with a slight manifestation of sensitiveness.",
+ "normalized_hypothesis": "now i have appeared. let us see your manifestation of self-sacrifice.",
+ "errors": 10,
+ "words": 8,
+ "wer": 1.25
+ },
+ "semantic": {
+ "normalized_reference": "nimrod replied with a slight manifestation of sensitiveness",
+ "normalized_hypothesis": "now i have appeared let us see your manifestation of self sacrifice",
+ "errors": 10,
+ "words": 8,
+ "wer": 1.25
+ }
+ },
+ {
+ "id": "release500_0463",
+ "category": "phonetic_balance",
+ "text": "I tried to read George Moore last night, and was dreadfully bored.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0463.wav",
+ "render_seconds": 52.49574498098809,
+ "audio_seconds": 4.667210884353741,
+ "rtf": 11.247776516157373,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "So talk a lot, talk a lot, speak a lot, talk a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot",
+ "raw": {
+ "normalized_reference": "i tried to read george moore last night, and was dreadfully bored.",
+ "normalized_hypothesis": "so talk a lot, talk a lot, speak a lot, talk a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot, speak a lot",
+ "errors": 334,
+ "words": 12,
+ "wer": 27.833333333333332
+ },
+ "semantic": {
+ "normalized_reference": "i tried to read george moore last night and was dreadfully bored",
+ "normalized_hypothesis": "so talk a lot talk a lot speak a lot talk a lot speak a lot speak a lot speak a lot speak a lot speak a lot speak a lot speak a lot speak a lot speak a lot speak a lot speak a lot speak a lot speak a lot speak a lot speak a lot speak a lot speak a lot speak a lot speak a lot speak a lot speak a lot speak a lot speak a lot speak a lot speak a lot speak a lot speak a lot speak a lot speak a lot speak a lot speak a lot speak a lot speak a lot speak a lot speak a lot speak a lot speak a lot speak a lot speak a lot speak a lot speak a lot speak a lot speak a lot speak a lot speak a lot speak a lot speak a lot speak a lot speak a lot speak a lot speak a lot speak a lot speak a lot speak a lot speak a lot speak a lot speak a lot speak a lot speak a lot speak a lot speak a lot speak a lot speak a lot speak a lot speak a lot speak a lot speak a lot speak a lot speak a lot speak a lot speak a lot speak a lot speak a lot speak a lot speak a lot speak a lot speak a lot speak a lot speak a lot speak a lot speak a lot speak a lot speak a lot speak a lot speak a lot speak a lot speak a lot speak a lot speak a lot speak a lot speak a lot speak a lot speak a lot speak a lot speak a lot speak a lot speak a lot speak a lot speak a lot speak a lot speak a lot speak a lot speak a lot speak a lot speak a lot speak a lot speak a lot",
+ "errors": 334,
+ "words": 12,
+ "wer": 27.833333333333332
+ }
+ },
+ {
+ "id": "release500_0464",
+ "category": "phonetic_balance",
+ "text": "Yes, it was a man who asked, a stranger.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0464.wav",
+ "render_seconds": 30.093770980020054,
+ "audio_seconds": 2.995374149659864,
+ "rtf": 10.046748578449645,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "That's what I love to scratch.",
+ "raw": {
+ "normalized_reference": "yes, it was a man who asked, a stranger.",
+ "normalized_hypothesis": "that's what i love to scratch.",
+ "errors": 9,
+ "words": 9,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "yes it was a man who asked a stranger",
+ "normalized_hypothesis": "that is what i love to scratch",
+ "errors": 9,
+ "words": 9,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0465",
+ "category": "phonetic_balance",
+ "text": "We were now good friends.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0465.wav",
+ "render_seconds": 39.59023493004497,
+ "audio_seconds": 2.020136054421769,
+ "rtf": 19.597806218738587,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "I don't know what I'm talking about. I don't know what I'm talking about. I don't know what I'm talking about. I don't know what I'm talking about. I don't know what I'm talking about. I don't know what I'm talking about. I don't know what I'm talking about. I don't know what I'm talking about. I don't know what I'm talking about. I don't know what I'm talking about. I don't know what I'm talking about. I don't know what I'm talking about. I don't know what I'm talking about. I don't know what I'm talking about. I don't know what I'm talking about. I don't know what I'm talking about. I don't know what I'm talking about. I don't know what I'm talking about. I don't know what I'm talking about. I don't know what I'm talking about. I don't know what I'm talking about. I don't know what I'm talking about. I don't know what I'm talking about. I don't know what I'm talking about. I don't know what I'm talking about. I don't know what I'm talking about. I don't know what I'm talking about. I don't know what I'm talking about. I don't know what I'm talking about. I don't know what I'm talking about. I don't know what I'm talking about. I don't know what I'm talking about. I don't know what I'm talking about. I don't know what I'm talking about. I don't know what I'm talking about. I don't know what I'm talking about. I don't know what I'm talking about. I don't know what I'm talking about. I don't know what I'm talking about. I don't know what I'm talking about. I don't know what I'm talking about. I don't know what I'm talking about. I don't know what I'm talking about. I don't know what I'm talking about. I don't know",
+ "raw": {
+ "normalized_reference": "we were now good friends.",
+ "normalized_hypothesis": "i don't know what i'm talking about. i don't know what i'm talking about. i don't know what i'm talking about. i don't know what i'm talking about. i don't know what i'm talking about. i don't know what i'm talking about. i don't know what i'm talking about. i don't know what i'm talking about. i don't know what i'm talking about. i don't know what i'm talking about. i don't know what i'm talking about. i don't know what i'm talking about. i don't know what i'm talking about. i don't know what i'm talking about. i don't know what i'm talking about. i don't know what i'm talking about. i don't know what i'm talking about. i don't know what i'm talking about. i don't know what i'm talking about. i don't know what i'm talking about. i don't know what i'm talking about. i don't know what i'm talking about. i don't know what i'm talking about. i don't know what i'm talking about. i don't know what i'm talking about. i don't know what i'm talking about. i don't know what i'm talking about. i don't know what i'm talking about. i don't know what i'm talking about. i don't know what i'm talking about. i don't know what i'm talking about. i don't know what i'm talking about. i don't know what i'm talking about. i don't know what i'm talking about. i don't know what i'm talking about. i don't know what i'm talking about. i don't know what i'm talking about. i don't know what i'm talking about. i don't know what i'm talking about. i don't know what i'm talking about. i don't know what i'm talking about. i don't know what i'm talking about. i don't know what i'm talking about. i don't know what i'm talking about. i don't know",
+ "errors": 311,
+ "words": 5,
+ "wer": 62.2
+ },
+ "semantic": {
+ "normalized_reference": "we were now good friends",
+ "normalized_hypothesis": "i do not know what i am talking about i do not know what i am talking about i do not know what i am talking about i do not know what i am talking about i do not know what i am talking about i do not know what i am talking about i do not know what i am talking about i do not know what i am talking about i do not know what i am talking about i do not know what i am talking about i do not know what i am talking about i do not know what i am talking about i do not know what i am talking about i do not know what i am talking about i do not know what i am talking about i do not know what i am talking about i do not know what i am talking about i do not know what i am talking about i do not know what i am talking about i do not know what i am talking about i do not know what i am talking about i do not know what i am talking about i do not know what i am talking about i do not know what i am talking about i do not know what i am talking about i do not know what i am talking about i do not know what i am talking about i do not know what i am talking about i do not know what i am talking about i do not know what i am talking about i do not know what i am talking about i do not know what i am talking about i do not know what i am talking about i do not know what i am talking about i do not know what i am talking about i do not know what i am talking about i do not know what i am talking about i do not know what i am talking about i do not know what i am talking about i do not know what i am talking about i do not know what i am talking about i do not know what i am talking about i do not know what i am talking about i do not know what i am talking about i do not know",
+ "errors": 400,
+ "words": 5,
+ "wer": 80.0
+ }
+ },
+ {
+ "id": "release500_0466",
+ "category": "phonetic_balance",
+ "text": "Yet, in accordance with Ernest's test of truth, it worked.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0466.wav",
+ "render_seconds": 39.81952378200367,
+ "audio_seconds": 3.900952380952381,
+ "rtf": 10.207641594507777,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "I'm going to turn into an honest child of yours, alright?",
+ "raw": {
+ "normalized_reference": "yet, in accordance with ernest's test of truth, it worked.",
+ "normalized_hypothesis": "i'm going to turn into an honest child of yours, alright?",
+ "errors": 11,
+ "words": 10,
+ "wer": 1.1
+ },
+ "semantic": {
+ "normalized_reference": "yet in accordance with ernest is test of truth it worked",
+ "normalized_hypothesis": "i am going to turn into an honest child of yours alright",
+ "errors": 12,
+ "words": 11,
+ "wer": 1.0909090909090908
+ }
+ },
+ {
+ "id": "release500_0467",
+ "category": "phonetic_balance",
+ "text": "It's the strap hangers that'll keep us from going under.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0467.wav",
+ "render_seconds": 66.60312551597599,
+ "audio_seconds": 3.4829931972789114,
+ "rtf": 19.12238173993842,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "Sister, I also want to kiss you.",
+ "raw": {
+ "normalized_reference": "it's the strap hangers that'll keep us from going under.",
+ "normalized_hypothesis": "sister, i also want to kiss you.",
+ "errors": 10,
+ "words": 10,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "it is the strap hangers that will keep us from going under",
+ "normalized_hypothesis": "sister i also want to kiss you",
+ "errors": 12,
+ "words": 12,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0468",
+ "category": "phonetic_balance",
+ "text": "I never allow what can't be changed to annoy me.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0468.wav",
+ "render_seconds": 40.40542334411293,
+ "audio_seconds": 3.0650340136054424,
+ "rtf": 13.18269982152122,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "I love you. I love you. I love you. I love you.",
+ "raw": {
+ "normalized_reference": "i never allow what can't be changed to annoy me.",
+ "normalized_hypothesis": "i love you. i love you. i love you. i love you.",
+ "errors": 11,
+ "words": 10,
+ "wer": 1.1
+ },
+ "semantic": {
+ "normalized_reference": "i never allow what can not be changed to annoy me",
+ "normalized_hypothesis": "i love you i love you i love you i love you",
+ "errors": 11,
+ "words": 11,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0469",
+ "category": "phonetic_balance",
+ "text": "After all superfluous flesh is gone what is left is stringy and resistant.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0469.wav",
+ "render_seconds": 41.10213894897606,
+ "audio_seconds": 5.224489795918367,
+ "rtf": 7.86720628320245,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "I promise you, I'll make sure that you get there as soon as I can. I promise you, I'll make sure that you get there as soon as I can. I promise you, I'll make sure that you get there as soon as I can. I promise you, I'll make sure that you get there as soon as I can. I promise you, I'll make sure that you get there as soon as I can. I promise you, I'll make sure that you get there as soon as I can. I promise you, I'll make sure that you get there as soon as I can. I promise you, I'll make sure that you get there as soon as I can. I promise you, I'll make sure that you get there as soon as I can. I promise you, I'll make sure that you get there as soon as I can. I promise you, I'll make sure that you get there as soon as I can. I promise you, I'll make sure that you get there as soon as I can. I promise you, I'll make sure that you get there as soon as I can. I promise you, I'll make sure that you get there as soon as I can. I promise you, I'll make sure that you get there as soon as I can. I promise you, I'll make sure that you get there as soon as I can. I promise you, I'll make sure that you get there as soon as I can. I promise you, I'll make sure that you get there as soon as I can. I promise you, I'll make sure that you get there as soon as I can. I promise you, I'll make sure that you get there as soon as soon as I can. I promise you, I'll make sure that you get there as soon as soon as I can. I promise you, I promise you, I'll make sure that you get there as soon as soon as I can. I promise you, I promise you, I'll make sure that you get there as soon as soon as I can. I promise you, I promise you, I'll make sure that you",
+ "raw": {
+ "normalized_reference": "after all superfluous flesh is gone what is left is stringy and resistant.",
+ "normalized_hypothesis": "i promise you, i'll make sure that you get there as soon as i can. i promise you, i'll make sure that you get there as soon as i can. i promise you, i'll make sure that you get there as soon as i can. i promise you, i'll make sure that you get there as soon as i can. i promise you, i'll make sure that you get there as soon as i can. i promise you, i'll make sure that you get there as soon as i can. i promise you, i'll make sure that you get there as soon as i can. i promise you, i'll make sure that you get there as soon as i can. i promise you, i'll make sure that you get there as soon as i can. i promise you, i'll make sure that you get there as soon as i can. i promise you, i'll make sure that you get there as soon as i can. i promise you, i'll make sure that you get there as soon as i can. i promise you, i'll make sure that you get there as soon as i can. i promise you, i'll make sure that you get there as soon as i can. i promise you, i'll make sure that you get there as soon as i can. i promise you, i'll make sure that you get there as soon as i can. i promise you, i'll make sure that you get there as soon as i can. i promise you, i'll make sure that you get there as soon as i can. i promise you, i'll make sure that you get there as soon as i can. i promise you, i'll make sure that you get there as soon as soon as i can. i promise you, i'll make sure that you get there as soon as soon as i can. i promise you, i promise you, i'll make sure that you get there as soon as soon as i can. i promise you, i promise you, i'll make sure that you get there as soon as soon as i can. i promise you, i promise you, i'll make sure that you",
+ "errors": 370,
+ "words": 13,
+ "wer": 28.46153846153846
+ },
+ "semantic": {
+ "normalized_reference": "after all superfluous flesh is gone what is left is stringy and resistant",
+ "normalized_hypothesis": "i promise you i will make sure that you get there as soon as i can i promise you i will make sure that you get there as soon as i can i promise you i will make sure that you get there as soon as i can i promise you i will make sure that you get there as soon as i can i promise you i will make sure that you get there as soon as i can i promise you i will make sure that you get there as soon as i can i promise you i will make sure that you get there as soon as i can i promise you i will make sure that you get there as soon as i can i promise you i will make sure that you get there as soon as i can i promise you i will make sure that you get there as soon as i can i promise you i will make sure that you get there as soon as i can i promise you i will make sure that you get there as soon as i can i promise you i will make sure that you get there as soon as i can i promise you i will make sure that you get there as soon as i can i promise you i will make sure that you get there as soon as i can i promise you i will make sure that you get there as soon as i can i promise you i will make sure that you get there as soon as i can i promise you i will make sure that you get there as soon as i can i promise you i will make sure that you get there as soon as i can i promise you i will make sure that you get there as soon as soon as i can i promise you i will make sure that you get there as soon as soon as i can i promise you i promise you i will make sure that you get there as soon as soon as i can i promise you i promise you i will make sure that you get there as soon as soon as i can i promise you i promise you i will make sure that you",
+ "errors": 394,
+ "words": 13,
+ "wer": 30.307692307692307
+ }
+ },
+ {
+ "id": "release500_0470",
+ "category": "phonetic_balance",
+ "text": "Then he shouted, Shut up.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0470.wav",
+ "render_seconds": 33.61183233605698,
+ "audio_seconds": 2.089795918367347,
+ "rtf": 16.083786957683515,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "It was so exciting.",
+ "raw": {
+ "normalized_reference": "then he shouted, shut up.",
+ "normalized_hypothesis": "it was so exciting.",
+ "errors": 5,
+ "words": 5,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "then he shouted shut up",
+ "normalized_hypothesis": "it was so exciting",
+ "errors": 5,
+ "words": 5,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0471",
+ "category": "phonetic_balance",
+ "text": "He had worshiped her, as Dante might have worshiped Beatrice.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0471.wav",
+ "render_seconds": 62.199409024906345,
+ "audio_seconds": 4.458231292517007,
+ "rtf": 13.95158863321111,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "I will have to. I've got to. I've got to. I've got to. I've got to. I've got to.",
+ "raw": {
+ "normalized_reference": "he had worshiped her, as dante might have worshiped beatrice.",
+ "normalized_hypothesis": "i will have to. i've got to. i've got to. i've got to. i've got to. i've got to.",
+ "errors": 19,
+ "words": 10,
+ "wer": 1.9
+ },
+ "semantic": {
+ "normalized_reference": "he had worshiped her as dante might have worshiped beatrice",
+ "normalized_hypothesis": "i will have to i have got to i have got to i have got to i have got to i have got to",
+ "errors": 23,
+ "words": 10,
+ "wer": 2.3
+ }
+ },
+ {
+ "id": "release500_0472",
+ "category": "phonetic_balance",
+ "text": "He looked like one who had passed through an uncomfortable hour or two.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0472.wav",
+ "render_seconds": 30.779865423915908,
+ "audio_seconds": 4.876190476190477,
+ "rtf": 6.312277088889004,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "Yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck,",
+ "raw": {
+ "normalized_reference": "he looked like one who had passed through an uncomfortable hour or two.",
+ "normalized_hypothesis": "yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck, yuck,",
+ "errors": 148,
+ "words": 13,
+ "wer": 11.384615384615385
+ },
+ "semantic": {
+ "normalized_reference": "he looked like one who had passed through an uncomfortable hour or 2",
+ "normalized_hypothesis": "yuck yuck yuck yuck yuck yuck yuck yuck yuck yuck yuck yuck yuck yuck yuck yuck yuck yuck yuck yuck yuck yuck yuck yuck yuck yuck yuck yuck yuck yuck yuck yuck yuck yuck yuck yuck yuck yuck yuck yuck yuck yuck yuck yuck yuck yuck yuck yuck yuck yuck yuck yuck yuck yuck yuck yuck yuck yuck yuck yuck yuck yuck yuck yuck yuck yuck yuck yuck yuck yuck yuck yuck yuck yuck yuck yuck yuck yuck yuck yuck yuck yuck yuck yuck yuck yuck yuck yuck yuck yuck yuck yuck yuck yuck yuck yuck yuck yuck yuck yuck yuck yuck yuck yuck yuck yuck yuck yuck yuck yuck yuck yuck yuck yuck yuck yuck yuck yuck yuck yuck yuck yuck yuck yuck yuck yuck yuck yuck yuck yuck yuck yuck yuck yuck yuck yuck yuck yuck yuck yuck yuck yuck yuck yuck yuck yuck yuck yuck",
+ "errors": 148,
+ "words": 13,
+ "wer": 11.384615384615385
+ }
+ },
+ {
+ "id": "release500_0473",
+ "category": "phonetic_balance",
+ "text": "The history of our westward-faring race is written in it.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0473.wav",
+ "render_seconds": 56.89556074899156,
+ "audio_seconds": 4.179591836734694,
+ "rtf": 13.61270740576458,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "I just want to do this. I just want to do this. I just want to do this. I just want to do this. I just want to do this. I just want to do this. I just want to do this. I just want to do this. I just want to do this. I just want to do this. I just want to do this. I just want to do this. I just want to do this. I just want to do this. I just want to do this. I just want to do this. I just want to do this. I just want to do this. I just want to do this. I just want to do this. I just want to do this. I just want to do this. I just want to do this. I just want to do this. I just want to do this. I just want to do this. I just want to do this. I just want to do this. I just want to do this. I just want to do this. I just want to do this. I just want to do this. I just want to do this. I just want to do this. I just want to do this. I just want to do this. I just want to do this. I just want to do this. I just want to do this. I just want to do this. I just want to do this. I just want to do this. I just want to do this. I just want to do this. I just want to do this. I just want to do this. I just want to do this. I just want to do this. I just want to do this. I just want to do this. I just want to do this. I just want to do this. I just want to do this. I just want to do this. I just want to do this. I just want to do this. I just want to do this. I just want to do this. I just want to do this. I just want to do this. I just want to do this. I just want to do this. I just want to do this. I just want",
+ "raw": {
+ "normalized_reference": "the history of our westward-faring race is written in it.",
+ "normalized_hypothesis": "i just want to do this. i just want to do this. i just want to do this. i just want to do this. i just want to do this. i just want to do this. i just want to do this. i just want to do this. i just want to do this. i just want to do this. i just want to do this. i just want to do this. i just want to do this. i just want to do this. i just want to do this. i just want to do this. i just want to do this. i just want to do this. i just want to do this. i just want to do this. i just want to do this. i just want to do this. i just want to do this. i just want to do this. i just want to do this. i just want to do this. i just want to do this. i just want to do this. i just want to do this. i just want to do this. i just want to do this. i just want to do this. i just want to do this. i just want to do this. i just want to do this. i just want to do this. i just want to do this. i just want to do this. i just want to do this. i just want to do this. i just want to do this. i just want to do this. i just want to do this. i just want to do this. i just want to do this. i just want to do this. i just want to do this. i just want to do this. i just want to do this. i just want to do this. i just want to do this. i just want to do this. i just want to do this. i just want to do this. i just want to do this. i just want to do this. i just want to do this. i just want to do this. i just want to do this. i just want to do this. i just want to do this. i just want to do this. i just want to do this. i just want",
+ "errors": 381,
+ "words": 11,
+ "wer": 34.63636363636363
+ },
+ "semantic": {
+ "normalized_reference": "the history of our westward faring race is written in it",
+ "normalized_hypothesis": "i just want to do this i just want to do this i just want to do this i just want to do this i just want to do this i just want to do this i just want to do this i just want to do this i just want to do this i just want to do this i just want to do this i just want to do this i just want to do this i just want to do this i just want to do this i just want to do this i just want to do this i just want to do this i just want to do this i just want to do this i just want to do this i just want to do this i just want to do this i just want to do this i just want to do this i just want to do this i just want to do this i just want to do this i just want to do this i just want to do this i just want to do this i just want to do this i just want to do this i just want to do this i just want to do this i just want to do this i just want to do this i just want to do this i just want to do this i just want to do this i just want to do this i just want to do this i just want to do this i just want to do this i just want to do this i just want to do this i just want to do this i just want to do this i just want to do this i just want to do this i just want to do this i just want to do this i just want to do this i just want to do this i just want to do this i just want to do this i just want to do this i just want to do this i just want to do this i just want to do this i just want to do this i just want to do this i just want to do this i just want",
+ "errors": 381,
+ "words": 11,
+ "wer": 34.63636363636363
+ }
+ },
+ {
+ "id": "release500_0474",
+ "category": "phonetic_balance",
+ "text": "He may desire to escape pain, or to enjoy pleasure.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0474.wav",
+ "render_seconds": 44.003337262081914,
+ "audio_seconds": 3.8312925170068026,
+ "rtf": 11.485246053845954,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you.",
+ "raw": {
+ "normalized_reference": "he may desire to escape pain, or to enjoy pleasure.",
+ "normalized_hypothesis": "i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you.",
+ "errors": 24,
+ "words": 10,
+ "wer": 2.4
+ },
+ "semantic": {
+ "normalized_reference": "he may desire to escape pain or to enjoy pleasure",
+ "normalized_hypothesis": "i love you i love you i love you i love you i love you i love you i love you i love you",
+ "errors": 24,
+ "words": 10,
+ "wer": 2.4
+ }
+ },
+ {
+ "id": "release500_0475",
+ "category": "phonetic_balance",
+ "text": "And the air was growing chilly.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0475.wav",
+ "render_seconds": 25.10143750894349,
+ "audio_seconds": 2.2987755102040817,
+ "rtf": 10.919481870900489,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "Hello, welcome to my channel.",
+ "raw": {
+ "normalized_reference": "and the air was growing chilly.",
+ "normalized_hypothesis": "hello, welcome to my channel.",
+ "errors": 6,
+ "words": 6,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "and the air was growing chilly",
+ "normalized_hypothesis": "hello welcome to my channel",
+ "errors": 6,
+ "words": 6,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0476",
+ "category": "phonetic_balance",
+ "text": "Dig in; you're sure good, was Daylight's answer.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0476.wav",
+ "render_seconds": 28.491644060006365,
+ "audio_seconds": 2.9257142857142857,
+ "rtf": 9.738354903322488,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "I don't know how I'm going to restore your child to me.",
+ "raw": {
+ "normalized_reference": "dig in; you're sure good, was daylight's answer.",
+ "normalized_hypothesis": "i don't know how i'm going to restore your child to me.",
+ "errors": 12,
+ "words": 8,
+ "wer": 1.5
+ },
+ "semantic": {
+ "normalized_reference": "dig in you are sure good was daylight is answer",
+ "normalized_hypothesis": "i do not know how i am going to restore your child to me",
+ "errors": 14,
+ "words": 10,
+ "wer": 1.4
+ }
+ },
+ {
+ "id": "release500_0477",
+ "category": "phonetic_balance",
+ "text": "My idea was that he would have more influence over the natives.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0477.wav",
+ "render_seconds": 56.70551932498347,
+ "audio_seconds": 4.458231292517007,
+ "rtf": 12.71928610347377,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you.",
+ "raw": {
+ "normalized_reference": "my idea was that he would have more influence over the natives.",
+ "normalized_hypothesis": "i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you.",
+ "errors": 333,
+ "words": 12,
+ "wer": 27.75
+ },
+ "semantic": {
+ "normalized_reference": "my idea was that he would have more influence over the natives",
+ "normalized_hypothesis": "i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you",
+ "errors": 333,
+ "words": 12,
+ "wer": 27.75
+ }
+ },
+ {
+ "id": "release500_0478",
+ "category": "phonetic_balance",
+ "text": "Now run along, and tell them to hurry.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0478.wav",
+ "render_seconds": 28.79980022890959,
+ "audio_seconds": 2.786394557823129,
+ "rtf": 10.335865804808861,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you.",
+ "raw": {
+ "normalized_reference": "now run along, and tell them to hurry.",
+ "normalized_hypothesis": "i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you.",
+ "errors": 333,
+ "words": 8,
+ "wer": 41.625
+ },
+ "semantic": {
+ "normalized_reference": "now run along and tell them to hurry",
+ "normalized_hypothesis": "i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you",
+ "errors": 333,
+ "words": 8,
+ "wer": 41.625
+ }
+ },
+ {
+ "id": "release500_0479",
+ "category": "phonetic_balance",
+ "text": "I have no idea, replied Philip.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0479.wav",
+ "render_seconds": 35.29261709493585,
+ "audio_seconds": 2.36843537414966,
+ "rtf": 14.901237112119627,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you.",
+ "raw": {
+ "normalized_reference": "i have no idea, replied philip.",
+ "normalized_hypothesis": "i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you.",
+ "errors": 332,
+ "words": 6,
+ "wer": 55.333333333333336
+ },
+ "semantic": {
+ "normalized_reference": "i have no idea replied philip",
+ "normalized_hypothesis": "i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you",
+ "errors": 332,
+ "words": 6,
+ "wer": 55.333333333333336
+ }
+ },
+ {
+ "id": "release500_0480",
+ "category": "phonetic_balance",
+ "text": "For the rest, he was a mere automaton.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0480.wav",
+ "render_seconds": 30.688352441997267,
+ "audio_seconds": 2.786394557823129,
+ "rtf": 11.013642111751949,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "For the last time, let's allow a little water.",
+ "raw": {
+ "normalized_reference": "for the rest, he was a mere automaton.",
+ "normalized_hypothesis": "for the last time, let's allow a little water.",
+ "errors": 6,
+ "words": 8,
+ "wer": 0.75
+ },
+ "semantic": {
+ "normalized_reference": "for the rest he was a mere automaton",
+ "normalized_hypothesis": "for the last time let us allow a little water",
+ "errors": 7,
+ "words": 8,
+ "wer": 0.875
+ }
+ },
+ {
+ "id": "release500_0481",
+ "category": "phonetic_balance",
+ "text": "Why, the average review is more nauseating than cod liver oil.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0481.wav",
+ "render_seconds": 69.20198716490995,
+ "audio_seconds": 4.388571428571429,
+ "rtf": 15.768681971170887,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "Yeah, now that I'm in a school, I assume I'm going to have a great life.",
+ "raw": {
+ "normalized_reference": "why, the average review is more nauseating than cod liver oil.",
+ "normalized_hypothesis": "yeah, now that i'm in a school, i assume i'm going to have a great life.",
+ "errors": 16,
+ "words": 11,
+ "wer": 1.4545454545454546
+ },
+ "semantic": {
+ "normalized_reference": "why the average review is more nauseating than cod liver oil",
+ "normalized_hypothesis": "yeah now that i am in a school i assume i am going to have a great life",
+ "errors": 18,
+ "words": 11,
+ "wer": 1.6363636363636365
+ }
+ },
+ {
+ "id": "release500_0482",
+ "category": "phonetic_balance",
+ "text": "I have been robbed, sir, I amended.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0482.wav",
+ "render_seconds": 31.50197906792164,
+ "audio_seconds": 2.716734693877551,
+ "rtf": 11.595530155710337,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "What do you want to see next?",
+ "raw": {
+ "normalized_reference": "i have been robbed, sir, i amended.",
+ "normalized_hypothesis": "what do you want to see next?",
+ "errors": 7,
+ "words": 7,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "i have been robbed sir i amended",
+ "normalized_hypothesis": "what do you want to see next",
+ "errors": 7,
+ "words": 7,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0483",
+ "category": "phonetic_balance",
+ "text": "He moved away as quietly as he had come.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0483.wav",
+ "render_seconds": 43.591895676916465,
+ "audio_seconds": 2.9257142857142857,
+ "rtf": 14.899573717696057,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "I love you guys. I love you guys. I love you guys. I love you guys. I love you guys. I love you guys. I love you guys. I love you guys. I love you guys. I love you guys. I love you guys. I love you guys. I love you guys. I love you guys. I love you guys. I love you guys. I love you guys. I love you guys. I love you guys. I love you guys. I love you guys. I love you guys. I love you guys. I love you guys. I love you guys. I love you guys. I love you guys. I love you guys. I love you guys. I love you guys. I love you guys. I love you guys. I love you guys. I love you guys. I love you guys. I love you guys. I love you guys. I love you guys. I love you guys. I love you guys. I love you guys. I love you guys. I love you guys. I love you guys. I love you guys. I love you guys. I love you guys. I love you guys. I love you guys. I love you guys. I love you guys. I love you guys. I love you guys. I love you guys. I love you guys. I love you guys. I love you guys. I love you guys. I love you guys. I love you guys. I love you guys. I love you guys. I love you guys. I love you guys. I love you guys. I love you guys. I love you guys. I love you guys. I love you guys. I love you guys. I love you guys. I love you guys. I love you guys. I love you guys. I love you guys. I love you guys. I love you guys. I love you guys. I love you guys. I love you guys. I love you guys. I love you guys. I love you guys. I love you guys. I love you guys. I love you guys. I love you guys. I love you guys. I love you guys",
+ "raw": {
+ "normalized_reference": "he moved away as quietly as he had come.",
+ "normalized_hypothesis": "i love you guys. i love you guys. i love you guys. i love you guys. i love you guys. i love you guys. i love you guys. i love you guys. i love you guys. i love you guys. i love you guys. i love you guys. i love you guys. i love you guys. i love you guys. i love you guys. i love you guys. i love you guys. i love you guys. i love you guys. i love you guys. i love you guys. i love you guys. i love you guys. i love you guys. i love you guys. i love you guys. i love you guys. i love you guys. i love you guys. i love you guys. i love you guys. i love you guys. i love you guys. i love you guys. i love you guys. i love you guys. i love you guys. i love you guys. i love you guys. i love you guys. i love you guys. i love you guys. i love you guys. i love you guys. i love you guys. i love you guys. i love you guys. i love you guys. i love you guys. i love you guys. i love you guys. i love you guys. i love you guys. i love you guys. i love you guys. i love you guys. i love you guys. i love you guys. i love you guys. i love you guys. i love you guys. i love you guys. i love you guys. i love you guys. i love you guys. i love you guys. i love you guys. i love you guys. i love you guys. i love you guys. i love you guys. i love you guys. i love you guys. i love you guys. i love you guys. i love you guys. i love you guys. i love you guys. i love you guys. i love you guys. i love you guys. i love you guys. i love you guys. i love you guys. i love you guys. i love you guys. i love you guys. i love you guys",
+ "errors": 356,
+ "words": 9,
+ "wer": 39.55555555555556
+ },
+ "semantic": {
+ "normalized_reference": "he moved away as quietly as he had come",
+ "normalized_hypothesis": "i love you guys i love you guys i love you guys i love you guys i love you guys i love you guys i love you guys i love you guys i love you guys i love you guys i love you guys i love you guys i love you guys i love you guys i love you guys i love you guys i love you guys i love you guys i love you guys i love you guys i love you guys i love you guys i love you guys i love you guys i love you guys i love you guys i love you guys i love you guys i love you guys i love you guys i love you guys i love you guys i love you guys i love you guys i love you guys i love you guys i love you guys i love you guys i love you guys i love you guys i love you guys i love you guys i love you guys i love you guys i love you guys i love you guys i love you guys i love you guys i love you guys i love you guys i love you guys i love you guys i love you guys i love you guys i love you guys i love you guys i love you guys i love you guys i love you guys i love you guys i love you guys i love you guys i love you guys i love you guys i love you guys i love you guys i love you guys i love you guys i love you guys i love you guys i love you guys i love you guys i love you guys i love you guys i love you guys i love you guys i love you guys i love you guys i love you guys i love you guys i love you guys i love you guys i love you guys i love you guys i love you guys i love you guys i love you guys i love you guys i love you guys",
+ "errors": 356,
+ "words": 9,
+ "wer": 39.55555555555556
+ }
+ },
+ {
+ "id": "release500_0484",
+ "category": "phonetic_balance",
+ "text": "A trickle of fresh blood ran over his face.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0484.wav",
+ "render_seconds": 12.693803622038104,
+ "audio_seconds": 3.2043537414965986,
+ "rtf": 3.9614239394522786,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "I'm talking to you, folks. I'm talking to you, folks. I'm talking to you, folks. I'm talking to you, folks. I'm talking to you, folks. I'm talking to you, folks. I'm talking to you, folks. I'm talking to you, folks. I'm talking to you, folks. I'm talking to you, folks. I'm talking to you, folks. I'm talking to you, folks. I'm talking to you, folks. I'm talking to you, folks. I'm talking to you, folks. I'm talking to you, folks. I'm talking to you, folks. I'm talking to you, folks. I'm talking to you, folks. I'm talking to you, folks. I'm talking to you, folks. I'm talking to you, folks. I'm talking to you, folks. I'm talking to you, folks. I'm talking to you, folks. I'm talking to you, folks. I'm talking to you, folks. I'm talking to you, folks. I'm talking to you, folks. I'm talking to you, folks. I'm talking to you, folks. I'm talking to you, folks. I'm talking to you, folks. I'm talking to you, folks. I'm talking to you, folks. I'm talking to you, folks. I'm talking to you, folks. I'm talking to you, folks. I'm talking to you, folks. I'm talking to you, folks. I'm talking to you, folks. I'm talking to you, folks. I'm talking to you, folks. I'm talking to you, folks. I'm talking to you, folks. I'm talking to you, folks. I'm talking to you, folks. I'm talking to you, folks. I'm talking to you, folks. I'm talking to you, folks. I'm talking to you, folks. I'm talking to you, folks. I'm talking to you, folks. I'm talking to you, folks. I'm talking to you, folks. I'm talking to",
+ "raw": {
+ "normalized_reference": "a trickle of fresh blood ran over his face.",
+ "normalized_hypothesis": "i'm talking to you, folks. i'm talking to you, folks. i'm talking to you, folks. i'm talking to you, folks. i'm talking to you, folks. i'm talking to you, folks. i'm talking to you, folks. i'm talking to you, folks. i'm talking to you, folks. i'm talking to you, folks. i'm talking to you, folks. i'm talking to you, folks. i'm talking to you, folks. i'm talking to you, folks. i'm talking to you, folks. i'm talking to you, folks. i'm talking to you, folks. i'm talking to you, folks. i'm talking to you, folks. i'm talking to you, folks. i'm talking to you, folks. i'm talking to you, folks. i'm talking to you, folks. i'm talking to you, folks. i'm talking to you, folks. i'm talking to you, folks. i'm talking to you, folks. i'm talking to you, folks. i'm talking to you, folks. i'm talking to you, folks. i'm talking to you, folks. i'm talking to you, folks. i'm talking to you, folks. i'm talking to you, folks. i'm talking to you, folks. i'm talking to you, folks. i'm talking to you, folks. i'm talking to you, folks. i'm talking to you, folks. i'm talking to you, folks. i'm talking to you, folks. i'm talking to you, folks. i'm talking to you, folks. i'm talking to you, folks. i'm talking to you, folks. i'm talking to you, folks. i'm talking to you, folks. i'm talking to you, folks. i'm talking to you, folks. i'm talking to you, folks. i'm talking to you, folks. i'm talking to you, folks. i'm talking to you, folks. i'm talking to you, folks. i'm talking to you, folks. i'm talking to",
+ "errors": 278,
+ "words": 9,
+ "wer": 30.88888888888889
+ },
+ "semantic": {
+ "normalized_reference": "a trickle of fresh blood ran over his face",
+ "normalized_hypothesis": "i am talking to you folks i am talking to you folks i am talking to you folks i am talking to you folks i am talking to you folks i am talking to you folks i am talking to you folks i am talking to you folks i am talking to you folks i am talking to you folks i am talking to you folks i am talking to you folks i am talking to you folks i am talking to you folks i am talking to you folks i am talking to you folks i am talking to you folks i am talking to you folks i am talking to you folks i am talking to you folks i am talking to you folks i am talking to you folks i am talking to you folks i am talking to you folks i am talking to you folks i am talking to you folks i am talking to you folks i am talking to you folks i am talking to you folks i am talking to you folks i am talking to you folks i am talking to you folks i am talking to you folks i am talking to you folks i am talking to you folks i am talking to you folks i am talking to you folks i am talking to you folks i am talking to you folks i am talking to you folks i am talking to you folks i am talking to you folks i am talking to you folks i am talking to you folks i am talking to you folks i am talking to you folks i am talking to you folks i am talking to you folks i am talking to you folks i am talking to you folks i am talking to you folks i am talking to you folks i am talking to you folks i am talking to you folks i am talking to you folks i am talking to",
+ "errors": 334,
+ "words": 9,
+ "wer": 37.111111111111114
+ }
+ },
+ {
+ "id": "release500_0485",
+ "category": "phonetic_balance",
+ "text": "Suppose you saw me at work through the window.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0485.wav",
+ "render_seconds": 53.307658123085275,
+ "audio_seconds": 3.0650340136054424,
+ "rtf": 17.392191370946232,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "I suppose I'll say that I'm not the same as you are.",
+ "raw": {
+ "normalized_reference": "suppose you saw me at work through the window.",
+ "normalized_hypothesis": "i suppose i'll say that i'm not the same as you are.",
+ "errors": 11,
+ "words": 9,
+ "wer": 1.2222222222222223
+ },
+ "semantic": {
+ "normalized_reference": "suppose you saw me at work through the window",
+ "normalized_hypothesis": "i suppose i will say that i am not the same as you are",
+ "errors": 12,
+ "words": 9,
+ "wer": 1.3333333333333333
+ }
+ },
+ {
+ "id": "release500_0486",
+ "category": "phonetic_balance",
+ "text": "But who was Eileen's double.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0486.wav",
+ "render_seconds": 24.200655221007764,
+ "audio_seconds": 2.089795918367347,
+ "rtf": 11.580391658490042,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "Thank you for watching.",
+ "raw": {
+ "normalized_reference": "but who was eileen's double.",
+ "normalized_hypothesis": "thank you for watching.",
+ "errors": 5,
+ "words": 5,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "but who was eileen is double",
+ "normalized_hypothesis": "thank you for watching",
+ "errors": 6,
+ "words": 6,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0487",
+ "category": "phonetic_balance",
+ "text": "Darkness hid him from Jeanne.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0487.wav",
+ "render_seconds": 17.294839818030596,
+ "audio_seconds": 2.2291156462585033,
+ "rtf": 7.7586103919998095,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "Cut the shit out of Sean's head.",
+ "raw": {
+ "normalized_reference": "darkness hid him from jeanne.",
+ "normalized_hypothesis": "cut the shit out of sean's head.",
+ "errors": 7,
+ "words": 5,
+ "wer": 1.4
+ },
+ "semantic": {
+ "normalized_reference": "darkness hid him from jeanne",
+ "normalized_hypothesis": "cut the shit out of sean is head",
+ "errors": 8,
+ "words": 5,
+ "wer": 1.6
+ }
+ },
+ {
+ "id": "release500_0488",
+ "category": "phonetic_balance",
+ "text": "By this time Charley was as enraged as the Greek.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0488.wav",
+ "render_seconds": 42.10388160403818,
+ "audio_seconds": 3.5526530612244898,
+ "rtf": 11.851391306283725,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "I must go. I'm going to go to work. I'm going to go to work.",
+ "raw": {
+ "normalized_reference": "by this time charley was as enraged as the greek.",
+ "normalized_hypothesis": "i must go. i'm going to go to work. i'm going to go to work.",
+ "errors": 15,
+ "words": 10,
+ "wer": 1.5
+ },
+ "semantic": {
+ "normalized_reference": "by this time charley was as enraged as the greek",
+ "normalized_hypothesis": "i must go i am going to go to work i am going to go to work",
+ "errors": 17,
+ "words": 10,
+ "wer": 1.7
+ }
+ },
+ {
+ "id": "release500_0489",
+ "category": "phonetic_balance",
+ "text": "I am writing these lines in Honolulu, Hawaii.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0489.wav",
+ "render_seconds": 46.409274923964404,
+ "audio_seconds": 3.274013605442177,
+ "rtf": 14.175040337896375,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "I want to start a new life.",
+ "raw": {
+ "normalized_reference": "i am writing these lines in honolulu, hawaii.",
+ "normalized_hypothesis": "i want to start a new life.",
+ "errors": 7,
+ "words": 8,
+ "wer": 0.875
+ },
+ "semantic": {
+ "normalized_reference": "i am writing these lines in honolulu hawaii",
+ "normalized_hypothesis": "i want to start a new life",
+ "errors": 7,
+ "words": 8,
+ "wer": 0.875
+ }
+ },
+ {
+ "id": "release500_0490",
+ "category": "phonetic_balance",
+ "text": "Nobody knows how the natives got them.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0490.wav",
+ "render_seconds": 10.605359158013016,
+ "audio_seconds": 2.716734693877551,
+ "rtf": 3.903715435266209,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "Thank you for watching.",
+ "raw": {
+ "normalized_reference": "nobody knows how the natives got them.",
+ "normalized_hypothesis": "thank you for watching.",
+ "errors": 7,
+ "words": 7,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "nobody knows how the natives got them",
+ "normalized_hypothesis": "thank you for watching",
+ "errors": 7,
+ "words": 7,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0491",
+ "category": "phonetic_balance",
+ "text": "And how would we ever find ourselves.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0491.wav",
+ "render_seconds": 49.29280564200599,
+ "audio_seconds": 2.716734693877551,
+ "rtf": 18.14413669214463,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "And so you are able to make your songs.",
+ "raw": {
+ "normalized_reference": "and how would we ever find ourselves.",
+ "normalized_hypothesis": "and so you are able to make your songs.",
+ "errors": 8,
+ "words": 7,
+ "wer": 1.1428571428571428
+ },
+ "semantic": {
+ "normalized_reference": "and how would we ever find ourselves",
+ "normalized_hypothesis": "and so you are able to make your songs",
+ "errors": 8,
+ "words": 7,
+ "wer": 1.1428571428571428
+ }
+ },
+ {
+ "id": "release500_0492",
+ "category": "phonetic_balance",
+ "text": "They are coming ashore, whoever they are.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0492.wav",
+ "render_seconds": 34.40752541599795,
+ "audio_seconds": 2.8560544217687074,
+ "rtf": 12.047223314004619,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "Thank you for watching.",
+ "raw": {
+ "normalized_reference": "they are coming ashore, whoever they are.",
+ "normalized_hypothesis": "thank you for watching.",
+ "errors": 7,
+ "words": 7,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "they are coming ashore whoever they are",
+ "normalized_hypothesis": "thank you for watching",
+ "errors": 7,
+ "words": 7,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0493",
+ "category": "phonetic_balance",
+ "text": "This is my fifth voyage.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0493.wav",
+ "render_seconds": 19.597948632086627,
+ "audio_seconds": 2.159455782312925,
+ "rtf": 9.075410940387899,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "The fuck we said words.",
+ "raw": {
+ "normalized_reference": "this is my fifth voyage.",
+ "normalized_hypothesis": "the fuck we said words.",
+ "errors": 5,
+ "words": 5,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "this is my 5th voyage",
+ "normalized_hypothesis": "the fuck we said words",
+ "errors": 5,
+ "words": 5,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0494",
+ "category": "phonetic_balance",
+ "text": "And yet if she came he had no words to say.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0494.wav",
+ "render_seconds": 42.38802839803975,
+ "audio_seconds": 2.9257142857142857,
+ "rtf": 14.488095643861243,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "I love you so dearly. I love you so. I love you so dearly. I love you so dearly. I love you so dearly. I love you so dearly. I love you so dearly. I love you so dearly. I love you so dearly. I love you so dearly. I love you so dearly. I love you so dearly. I love you so dearly. I love you so dearly. I love you so dearly. I love you so dearly. I love you so dearly. I love you so dearly. I love you so dearly. I love you so dearly. I love you so dearly. I love you so dearly. I love you so dearly. I love you so dearly. I love you so dearly. I love you so dearly. I love you so dearly. I love you so dearly. I love you so dearly. I love you so dearly. I love you so dearly. I love you so dearly. I love you so dearly. I love you so dearly. I love you so dearly. I love you so dearly. I love you so dearly. I love you so dearly. I love you so dearly. I love you so dearly. I love you so dearly. I love you so dearly. I love you so dearly. I love you so dearly. I love you so dearly. I love you so dearly. I love you so dearly. I love you so dearly. I love you so dearly. I love you so dearly. I love you so dearly. I love you so dearly. I love you so dearly. I love you so dearly. I love you so dearly. I love you so dearly. I love you so dearly. I love you so dearly. I love you so dearly. I love you so dearly. I love you so dearly. I love you so dearly. I love you so dearly. I love you so dear",
+ "raw": {
+ "normalized_reference": "and yet if she came he had no words to say.",
+ "normalized_hypothesis": "i love you so dearly. i love you so. i love you so dearly. i love you so dearly. i love you so dearly. i love you so dearly. i love you so dearly. i love you so dearly. i love you so dearly. i love you so dearly. i love you so dearly. i love you so dearly. i love you so dearly. i love you so dearly. i love you so dearly. i love you so dearly. i love you so dearly. i love you so dearly. i love you so dearly. i love you so dearly. i love you so dearly. i love you so dearly. i love you so dearly. i love you so dearly. i love you so dearly. i love you so dearly. i love you so dearly. i love you so dearly. i love you so dearly. i love you so dearly. i love you so dearly. i love you so dearly. i love you so dearly. i love you so dearly. i love you so dearly. i love you so dearly. i love you so dearly. i love you so dearly. i love you so dearly. i love you so dearly. i love you so dearly. i love you so dearly. i love you so dearly. i love you so dearly. i love you so dearly. i love you so dearly. i love you so dearly. i love you so dearly. i love you so dearly. i love you so dearly. i love you so dearly. i love you so dearly. i love you so dearly. i love you so dearly. i love you so dearly. i love you so dearly. i love you so dearly. i love you so dearly. i love you so dearly. i love you so dearly. i love you so dearly. i love you so dearly. i love you so dearly. i love you so dear",
+ "errors": 319,
+ "words": 11,
+ "wer": 29.0
+ },
+ "semantic": {
+ "normalized_reference": "and yet if she came he had no words to say",
+ "normalized_hypothesis": "i love you so dearly i love you so i love you so dearly i love you so dearly i love you so dearly i love you so dearly i love you so dearly i love you so dearly i love you so dearly i love you so dearly i love you so dearly i love you so dearly i love you so dearly i love you so dearly i love you so dearly i love you so dearly i love you so dearly i love you so dearly i love you so dearly i love you so dearly i love you so dearly i love you so dearly i love you so dearly i love you so dearly i love you so dearly i love you so dearly i love you so dearly i love you so dearly i love you so dearly i love you so dearly i love you so dearly i love you so dearly i love you so dearly i love you so dearly i love you so dearly i love you so dearly i love you so dearly i love you so dearly i love you so dearly i love you so dearly i love you so dearly i love you so dearly i love you so dearly i love you so dearly i love you so dearly i love you so dearly i love you so dearly i love you so dearly i love you so dearly i love you so dearly i love you so dearly i love you so dearly i love you so dearly i love you so dearly i love you so dearly i love you so dearly i love you so dearly i love you so dearly i love you so dearly i love you so dearly i love you so dearly i love you so dearly i love you so dearly i love you so dear",
+ "errors": 319,
+ "words": 11,
+ "wer": 29.0
+ }
+ },
+ {
+ "id": "release500_0495",
+ "category": "phonetic_balance",
+ "text": "You must sleep, he urged.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0495.wav",
+ "render_seconds": 24.098106320016086,
+ "audio_seconds": 2.020136054421769,
+ "rtf": 11.928952145212703,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "I don't know what I want to do. I don't know what I want to do. I don't know what I want to do. I don't know what I want to do. I don't know what I want to do. I don't know what I want to do. I don't know what I want to do. I don't know what I want to do. I don't know what I want to do. I don't know what I want to do. I don't know what I want to do. I don't know what I want to do. I don't know what I want to do. I don't know what I want to do. I don't know what I want to do. I don't know what I want to do. I don't know what I want to do. I don't know what I want to do. I don't know what I want to do. I don't know what I want to do. I don't know what I want to do. I don't know what I want to do. I don't know what I want to do. I don't know what I want to do. I don't know what I want to do. I don't know what I want to do. I don't know what I want to do. I don't know what I want to do. I don't know what I want to do. I don't know what I want to do. I don't know what I want to do. I don't know what I want to do. I don't know what I want to do. I don't know what I want to do. I don't know what I want to do. I don't know what I want to do. I don't know what I want to do. I don't know what I want to do. I don't know what I want to do. I don't know what I want to do. I don't know what I want to do. I don't know what I want to do. I don't know what I want to do. I don't know what I want to do. I don't know",
+ "raw": {
+ "normalized_reference": "you must sleep, he urged.",
+ "normalized_hypothesis": "i don't know what i want to do. i don't know what i want to do. i don't know what i want to do. i don't know what i want to do. i don't know what i want to do. i don't know what i want to do. i don't know what i want to do. i don't know what i want to do. i don't know what i want to do. i don't know what i want to do. i don't know what i want to do. i don't know what i want to do. i don't know what i want to do. i don't know what i want to do. i don't know what i want to do. i don't know what i want to do. i don't know what i want to do. i don't know what i want to do. i don't know what i want to do. i don't know what i want to do. i don't know what i want to do. i don't know what i want to do. i don't know what i want to do. i don't know what i want to do. i don't know what i want to do. i don't know what i want to do. i don't know what i want to do. i don't know what i want to do. i don't know what i want to do. i don't know what i want to do. i don't know what i want to do. i don't know what i want to do. i don't know what i want to do. i don't know what i want to do. i don't know what i want to do. i don't know what i want to do. i don't know what i want to do. i don't know what i want to do. i don't know what i want to do. i don't know what i want to do. i don't know what i want to do. i don't know what i want to do. i don't know what i want to do. i don't know what i want to do. i don't know",
+ "errors": 355,
+ "words": 5,
+ "wer": 71.0
+ },
+ "semantic": {
+ "normalized_reference": "you must sleep he urged",
+ "normalized_hypothesis": "i do not know what i want to do i do not know what i want to do i do not know what i want to do i do not know what i want to do i do not know what i want to do i do not know what i want to do i do not know what i want to do i do not know what i want to do i do not know what i want to do i do not know what i want to do i do not know what i want to do i do not know what i want to do i do not know what i want to do i do not know what i want to do i do not know what i want to do i do not know what i want to do i do not know what i want to do i do not know what i want to do i do not know what i want to do i do not know what i want to do i do not know what i want to do i do not know what i want to do i do not know what i want to do i do not know what i want to do i do not know what i want to do i do not know what i want to do i do not know what i want to do i do not know what i want to do i do not know what i want to do i do not know what i want to do i do not know what i want to do i do not know what i want to do i do not know what i want to do i do not know what i want to do i do not know what i want to do i do not know what i want to do i do not know what i want to do i do not know what i want to do i do not know what i want to do i do not know what i want to do i do not know what i want to do i do not know what i want to do i do not know what i want to do i do not know what i want to do i do not know",
+ "errors": 400,
+ "words": 5,
+ "wer": 80.0
+ }
+ },
+ {
+ "id": "release500_0496",
+ "category": "phonetic_balance",
+ "text": "Could the incident have anything to do with Jeanne and Pierre.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0496.wav",
+ "render_seconds": 14.69217577308882,
+ "audio_seconds": 4.179591836734694,
+ "rtf": 3.515217836334727,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "I'm about to know what I think about this car and car.",
+ "raw": {
+ "normalized_reference": "could the incident have anything to do with jeanne and pierre.",
+ "normalized_hypothesis": "i'm about to know what i think about this car and car.",
+ "errors": 11,
+ "words": 11,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "could the incident have anything to do with jeanne and pierre",
+ "normalized_hypothesis": "i am about to know what i think about this car and car",
+ "errors": 12,
+ "words": 11,
+ "wer": 1.0909090909090908
+ }
+ },
+ {
+ "id": "release500_0497",
+ "category": "phonetic_balance",
+ "text": "Each day she became a more vital part of him.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0497.wav",
+ "render_seconds": 58.90309573803097,
+ "audio_seconds": 3.274013605442177,
+ "rtf": 17.991096811607697,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "It's time for me to go to bed now.",
+ "raw": {
+ "normalized_reference": "each day she became a more vital part of him.",
+ "normalized_hypothesis": "it's time for me to go to bed now.",
+ "errors": 10,
+ "words": 10,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "each day she became a more vital part of him",
+ "normalized_hypothesis": "it is time for me to go to bed now",
+ "errors": 10,
+ "words": 10,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0498",
+ "category": "phonetic_balance",
+ "text": "But why continue the tirade, for tirade it was.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0498.wav",
+ "render_seconds": 31.699932329007424,
+ "audio_seconds": 3.3436734693877552,
+ "rtf": 9.480570581795433,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "I'm not going to tell you what time it is. What time it is.",
+ "raw": {
+ "normalized_reference": "but why continue the tirade, for tirade it was.",
+ "normalized_hypothesis": "i'm not going to tell you what time it is. what time it is.",
+ "errors": 13,
+ "words": 9,
+ "wer": 1.4444444444444444
+ },
+ "semantic": {
+ "normalized_reference": "but why continue the tirade for tirade it was",
+ "normalized_hypothesis": "i am not going to tell you what time it is what time it is",
+ "errors": 14,
+ "words": 9,
+ "wer": 1.5555555555555556
+ }
+ },
+ {
+ "id": "release500_0499",
+ "category": "phonetic_balance",
+ "text": "That is what distinguishes all of us from the lower animals.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0499.wav",
+ "render_seconds": 55.103952111909166,
+ "audio_seconds": 4.31891156462585,
+ "rtf": 12.758759073290461,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "That's what this town has been for so many years.",
+ "raw": {
+ "normalized_reference": "that is what distinguishes all of us from the lower animals.",
+ "normalized_hypothesis": "that's what this town has been for so many years.",
+ "errors": 10,
+ "words": 11,
+ "wer": 0.9090909090909091
+ },
+ "semantic": {
+ "normalized_reference": "that is what distinguishes all of us from the lower animals",
+ "normalized_hypothesis": "that is what this town has been for so many years",
+ "errors": 8,
+ "words": 11,
+ "wer": 0.7272727272727273
+ }
+ },
+ {
+ "id": "release500_0500",
+ "category": "phonetic_balance",
+ "text": "There was something pathetic in the girl's attitude now.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0500.wav",
+ "render_seconds": 39.00000892498065,
+ "audio_seconds": 3.622312925170068,
+ "rtf": 10.766604026390015,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "whisper-large-v3",
+ "hypothesis": "They were so damn pathetic that the boys had to trap them.",
+ "raw": {
+ "normalized_reference": "there was something pathetic in the girl's attitude now.",
+ "normalized_hypothesis": "they were so damn pathetic that the boys had to trap them.",
+ "errors": 10,
+ "words": 9,
+ "wer": 1.1111111111111112
+ },
+ "semantic": {
+ "normalized_reference": "there was something pathetic in the girl is attitude now",
+ "normalized_hypothesis": "they were so damn pathetic that the boys had to trap them",
+ "errors": 10,
+ "words": 10,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0001",
+ "category": "conversational",
+ "text": "It wasn't until later that I realized what had actually happened.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0001.wav",
+ "render_seconds": 144.48957702307962,
+ "audio_seconds": 4.31891156462585,
+ "rtf": 33.455090446057056,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I'm listening to the other way. I'm listening to the other way. I'm listening to the other way.",
+ "raw": {
+ "normalized_reference": "it wasn't until later that i realized what had actually happened.",
+ "normalized_hypothesis": "i'm listening to the other way. i'm listening to the other way. i'm listening to the other way.",
+ "errors": 18,
+ "words": 11,
+ "wer": 1.6363636363636365
+ },
+ "semantic": {
+ "normalized_reference": "it was not until later that i realized what had actually happened",
+ "normalized_hypothesis": "i am listening to the other way i am listening to the other way i am listening to the other way",
+ "errors": 20,
+ "words": 12,
+ "wer": 1.6666666666666667
+ }
+ },
+ {
+ "id": "release500_0002",
+ "category": "descriptive",
+ "text": "The light at that time of afternoon is perfect for about twenty minutes.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0002.wav",
+ "render_seconds": 170.39593176799826,
+ "audio_seconds": 4.806530612244898,
+ "rtf": 35.45091990757437,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The year I got here was after me. It's perfect for both track and language.",
+ "raw": {
+ "normalized_reference": "the light at that time of afternoon is perfect for about twenty minutes.",
+ "normalized_hypothesis": "the year i got here was after me. it's perfect for both track and language.",
+ "errors": 12,
+ "words": 13,
+ "wer": 0.9230769230769231
+ },
+ "semantic": {
+ "normalized_reference": "the light at that time of afternoon is perfect for about 20 minutes",
+ "normalized_hypothesis": "the year i got here was after me it is perfect for both track and language",
+ "errors": 12,
+ "words": 13,
+ "wer": 0.9230769230769231
+ }
+ },
+ {
+ "id": "release500_0003",
+ "category": "emotional",
+ "text": "That was the scariest thing I've ever done, and also maybe the best.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0003.wav",
+ "render_seconds": 179.49796744505875,
+ "audio_seconds": 4.527891156462585,
+ "rtf": 39.64273018993935,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "There is a small house in the middle of the town. I wish you all the best.",
+ "raw": {
+ "normalized_reference": "that was the scariest thing i've ever done, and also maybe the best.",
+ "normalized_hypothesis": "there is a small house in the middle of the town. i wish you all the best.",
+ "errors": 14,
+ "words": 13,
+ "wer": 1.0769230769230769
+ },
+ "semantic": {
+ "normalized_reference": "that was the scariest thing i have ever done and also maybe the best",
+ "normalized_hypothesis": "there is a small house in the middle of the town i wish you all the best",
+ "errors": 15,
+ "words": 14,
+ "wer": 1.0714285714285714
+ }
+ },
+ {
+ "id": "release500_0004",
+ "category": "homographs",
+ "text": "After I record the record, please present the present to the bass player.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0004.wav",
+ "render_seconds": 157.99379248300102,
+ "audio_seconds": 5.015510204081632,
+ "rtf": 31.50104098171814,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I've got a little bit of work to do. Gosh, I'm so depressed. I'm so depressed. I'm so depressed. I'm so depressed. I'm so depressed. I'm so depressed. I'm so depressed. I'm so depressed. I'm so depressed. I'm so depressed. I'm so depressed. I'm so depressed. I'm so depressed. I'm so depressed. I'm so depressed. I'm so depressed. I'm so depressed. I'm so depressed. I'm so depressed. I'm so depressed. I'm so depressed. I'm so depressed. I'm so depressed. I'm so depressed. I'm so depressed. I'm so depressed. I'm so depressed. I'm so depressed. I'm so depressed. I'm so depressed. I'm so depressed. I'm so depressed. I'm so depressed. I'm so depressed. I'm so depressed. I'm so depressed. I'm so depressed. I'm so depressed. I'm so depressed. I'm so depressed. I'm so depressed. I'm so depressed. I'm so depressed. I'm so depressed. I'm so depressed. I'm so depressed. I'm so depressed. I'm so depressed. I'm so depressed. I'm so depressed. I'm so depressed. I'm so depressed. I'm so depressed. I'm so depressed. I'm so depressed. I'm so depressed. I'm so depressed. I'm so depressed. I'm so depressed. I'm so depressed. I'm so depressed. I'm so depressed. I'm so depressed. I'm so depressed. I'm so depressed. I'm so depressed. I'm so depressed. I'm so depressed. I'm so depressed. I'm so depressed. I'm so depressed. I'm so depressed. I'm so depressed. I'm so depressed. I'm so depressed. I'm so depressed. I'm so depressed. I'm so depressed. I'm so depressed. I'm so depressed. I'm so depressed. I'm so depressed. I'm so depressed. I'm so depressed. I'm so depressed. I'm so depressed. I",
+ "raw": {
+ "normalized_reference": "after i record the record, please present the present to the bass player.",
+ "normalized_hypothesis": "i've got a little bit of work to do. gosh, i'm so depressed. i'm so depressed. i'm so depressed. i'm so depressed. i'm so depressed. i'm so depressed. i'm so depressed. i'm so depressed. i'm so depressed. i'm so depressed. i'm so depressed. i'm so depressed. i'm so depressed. i'm so depressed. i'm so depressed. i'm so depressed. i'm so depressed. i'm so depressed. i'm so depressed. i'm so depressed. i'm so depressed. i'm so depressed. i'm so depressed. i'm so depressed. i'm so depressed. i'm so depressed. i'm so depressed. i'm so depressed. i'm so depressed. i'm so depressed. i'm so depressed. i'm so depressed. i'm so depressed. i'm so depressed. i'm so depressed. i'm so depressed. i'm so depressed. i'm so depressed. i'm so depressed. i'm so depressed. i'm so depressed. i'm so depressed. i'm so depressed. i'm so depressed. i'm so depressed. i'm so depressed. i'm so depressed. i'm so depressed. i'm so depressed. i'm so depressed. i'm so depressed. i'm so depressed. i'm so depressed. i'm so depressed. i'm so depressed. i'm so depressed. i'm so depressed. i'm so depressed. i'm so depressed. i'm so depressed. i'm so depressed. i'm so depressed. i'm so depressed. i'm so depressed. i'm so depressed. i'm so depressed. i'm so depressed. i'm so depressed. i'm so depressed. i'm so depressed. i'm so depressed. i'm so depressed. i'm so depressed. i'm so depressed. i'm so depressed. i'm so depressed. i'm so depressed. i'm so depressed. i'm so depressed. i'm so depressed. i'm so depressed. i'm so depressed. i'm so depressed. i'm so depressed. i'm so depressed. i'm so depressed. i",
+ "errors": 269,
+ "words": 13,
+ "wer": 20.692307692307693
+ },
+ "semantic": {
+ "normalized_reference": "after i record the record please present the present to the bass player",
+ "normalized_hypothesis": "i have got a little bit of work to do gosh i am so depressed i am so depressed i am so depressed i am so depressed i am so depressed i am so depressed i am so depressed i am so depressed i am so depressed i am so depressed i am so depressed i am so depressed i am so depressed i am so depressed i am so depressed i am so depressed i am so depressed i am so depressed i am so depressed i am so depressed i am so depressed i am so depressed i am so depressed i am so depressed i am so depressed i am so depressed i am so depressed i am so depressed i am so depressed i am so depressed i am so depressed i am so depressed i am so depressed i am so depressed i am so depressed i am so depressed i am so depressed i am so depressed i am so depressed i am so depressed i am so depressed i am so depressed i am so depressed i am so depressed i am so depressed i am so depressed i am so depressed i am so depressed i am so depressed i am so depressed i am so depressed i am so depressed i am so depressed i am so depressed i am so depressed i am so depressed i am so depressed i am so depressed i am so depressed i am so depressed i am so depressed i am so depressed i am so depressed i am so depressed i am so depressed i am so depressed i am so depressed i am so depressed i am so depressed i am so depressed i am so depressed i am so depressed i am so depressed i am so depressed i am so depressed i am so depressed i am so depressed i am so depressed i am so depressed i am so depressed i am so depressed i am so depressed i am so depressed i am so depressed i am so depressed i am so depressed i",
+ "errors": 355,
+ "words": 13,
+ "wer": 27.307692307692307
+ }
+ },
+ {
+ "id": "release500_0005",
+ "category": "long",
+ "text": "If I'm being completely honest, I had no idea what I was doing for most of it, and I think that might be the most normal thing about the whole experience.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0005.wav",
+ "render_seconds": 302.5043903089827,
+ "audio_seconds": 9.47374149659864,
+ "rtf": 31.930825895723558,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I feel bad to say the worst, but I hope you like how you are listening for the rest of it. I would like to thank you for the rest of your time about the few expires.",
+ "raw": {
+ "normalized_reference": "if i'm being completely honest, i had no idea what i was doing for most of it, and i think that might be the most normal thing about the whole experience.",
+ "normalized_hypothesis": "i feel bad to say the worst, but i hope you like how you are listening for the rest of it. i would like to thank you for the rest of your time about the few expires.",
+ "errors": 30,
+ "words": 31,
+ "wer": 0.967741935483871
+ },
+ "semantic": {
+ "normalized_reference": "if i am being completely honest i had no idea what i was doing for most of it and i think that might be the most normal thing about the whole experience",
+ "normalized_hypothesis": "i feel bad to say the worst but i hope you like how you are listening for the rest of it i would like to thank you for the rest of your time about the few expires",
+ "errors": 30,
+ "words": 32,
+ "wer": 0.9375
+ }
+ },
+ {
+ "id": "release500_0006",
+ "category": "long_range",
+ "text": "The committee postponed its decision after discovering that three pages of the report had been printed in the wrong order.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0006.wav",
+ "render_seconds": 257.10046293900814,
+ "audio_seconds": 7.9412244897959186,
+ "rtf": 32.37541808185495,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The crowd pushed through the discussion. After discovering the plot patches of the report had been pointed in the wrong direction.",
+ "raw": {
+ "normalized_reference": "the committee postponed its decision after discovering that three pages of the report had been printed in the wrong order.",
+ "normalized_hypothesis": "the crowd pushed through the discussion. after discovering the plot patches of the report had been pointed in the wrong direction.",
+ "errors": 10,
+ "words": 20,
+ "wer": 0.5
+ },
+ "semantic": {
+ "normalized_reference": "the committee postponed its decision after discovering that 3 pages of the report had been printed in the wrong order",
+ "normalized_hypothesis": "the crowd pushed through the discussion after discovering the plot patches of the report had been pointed in the wrong direction",
+ "errors": 10,
+ "words": 20,
+ "wer": 0.5
+ }
+ },
+ {
+ "id": "release500_0007",
+ "category": "names_places",
+ "text": "Gwendolyn photographed the eucalyptus trees outside Ljubljana.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0007.wav",
+ "render_seconds": 175.80659322696738,
+ "audio_seconds": 4.527891156462585,
+ "rtf": 38.82747777097987,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Thank you for watching. I hope you enjoyed this video. See you in the next video.",
+ "raw": {
+ "normalized_reference": "gwendolyn photographed the eucalyptus trees outside ljubljana.",
+ "normalized_hypothesis": "thank you for watching. i hope you enjoyed this video. see you in the next video.",
+ "errors": 16,
+ "words": 7,
+ "wer": 2.2857142857142856
+ },
+ "semantic": {
+ "normalized_reference": "gwendolyn photographed the eucalyptus trees outside ljubljana",
+ "normalized_hypothesis": "thank you for watching i hope you enjoyed this video see you in the next video",
+ "errors": 16,
+ "words": 7,
+ "wer": 2.2857142857142856
+ }
+ },
+ {
+ "id": "release500_0008",
+ "category": "narrative",
+ "text": "He had this theory, and the more time passes, the more I think he was right.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0008.wav",
+ "render_seconds": 173.3916494640289,
+ "audio_seconds": 5.294149659863946,
+ "rtf": 32.751557965681855,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Out of style. I do not care about sex. Do you think so as well?",
+ "raw": {
+ "normalized_reference": "he had this theory, and the more time passes, the more i think he was right.",
+ "normalized_hypothesis": "out of style. i do not care about sex. do you think so as well?",
+ "errors": 15,
+ "words": 16,
+ "wer": 0.9375
+ },
+ "semantic": {
+ "normalized_reference": "he had this theory and the more time passes the more i think he was right",
+ "normalized_hypothesis": "out of style i do not care about sex do you think so as well",
+ "errors": 15,
+ "words": 16,
+ "wer": 0.9375
+ }
+ },
+ {
+ "id": "release500_0009",
+ "category": "numbers",
+ "text": "The package weighs twelve point six kilograms and arrived on July twenty-first.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0009.wav",
+ "render_seconds": 205.40336752007715,
+ "audio_seconds": 5.7817687074829935,
+ "rtf": 35.52604365758111,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The podcast lasts close to a six-hour pause, and the rest is a long 20-minute pause.",
+ "raw": {
+ "normalized_reference": "the package weighs twelve point six kilograms and arrived on july twenty-first.",
+ "normalized_hypothesis": "the podcast lasts close to a six-hour pause, and the rest is a long 20-minute pause.",
+ "errors": 15,
+ "words": 13,
+ "wer": 1.1538461538461537
+ },
+ "semantic": {
+ "normalized_reference": "the package weighs 12.6 kilograms and arrived on july 21st",
+ "normalized_hypothesis": "the podcast lasts close to a 6 hour pause and the rest is a long 20 minute pause",
+ "errors": 15,
+ "words": 11,
+ "wer": 1.3636363636363635
+ }
+ },
+ {
+ "id": "release500_0010",
+ "category": "punctuation",
+ "text": "First, close the window; second, turn off the lamp; finally, lock the door.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0010.wav",
+ "render_seconds": 188.7054866560502,
+ "audio_seconds": 5.224489795918367,
+ "rtf": 36.11940955525961,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "First push the button, second close the lid, three and lift the door.",
+ "raw": {
+ "normalized_reference": "first, close the window; second, turn off the lamp; finally, lock the door.",
+ "normalized_hypothesis": "first push the button, second close the lid, three and lift the door.",
+ "errors": 8,
+ "words": 13,
+ "wer": 0.6153846153846154
+ },
+ "semantic": {
+ "normalized_reference": "1st close the window 2nd turn off the lamp finally lock the door",
+ "normalized_hypothesis": "1st push the button 2nd close the lid 3 and lift the door",
+ "errors": 8,
+ "words": 13,
+ "wer": 0.6153846153846154
+ }
+ },
+ {
+ "id": "release500_0011",
+ "category": "question",
+ "text": "When did you first realize something was wrong?",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0011.wav",
+ "render_seconds": 126.30049753002822,
+ "audio_seconds": 3.1346938775510202,
+ "rtf": 40.29117434226161,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Thank you for watching this video.",
+ "raw": {
+ "normalized_reference": "when did you first realize something was wrong?",
+ "normalized_hypothesis": "thank you for watching this video.",
+ "errors": 7,
+ "words": 8,
+ "wer": 0.875
+ },
+ "semantic": {
+ "normalized_reference": "when did you 1st realize something was wrong",
+ "normalized_hypothesis": "thank you for watching this video",
+ "errors": 7,
+ "words": 8,
+ "wer": 0.875
+ }
+ },
+ {
+ "id": "release500_0012",
+ "category": "rare_words",
+ "text": "The archaeologist catalogued a Byzantine reliquary and several anachronistic fragments.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0012.wav",
+ "render_seconds": 207.30320841202047,
+ "audio_seconds": 6.478367346938776,
+ "rtf": 31.99929817347846,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I'll tell you just tell you if there's a time when you're there. And shall we ever question for it?",
+ "raw": {
+ "normalized_reference": "the archaeologist catalogued a byzantine reliquary and several anachronistic fragments.",
+ "normalized_hypothesis": "i'll tell you just tell you if there's a time when you're there. and shall we ever question for it?",
+ "errors": 18,
+ "words": 10,
+ "wer": 1.8
+ },
+ "semantic": {
+ "normalized_reference": "the archaeologist catalogued a byzantine reliquary and several anachronistic fragments",
+ "normalized_hypothesis": "i will tell you just tell you if there is a time when you are there and shall we ever question for it",
+ "errors": 21,
+ "words": 10,
+ "wer": 2.1
+ }
+ },
+ {
+ "id": "release500_0013",
+ "category": "short",
+ "text": "I have no idea.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0013.wav",
+ "render_seconds": 82.4928001590306,
+ "audio_seconds": 1.5325170068027212,
+ "rtf": 53.82830976286176,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Thank you.",
+ "raw": {
+ "normalized_reference": "i have no idea.",
+ "normalized_hypothesis": "thank you.",
+ "errors": 4,
+ "words": 4,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "i have no idea",
+ "normalized_hypothesis": "thank you",
+ "errors": 4,
+ "words": 4,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0014",
+ "category": "technical",
+ "text": "The system runs on three core components that all have to stay in sync.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0014.wav",
+ "render_seconds": 144.5139470390277,
+ "audio_seconds": 4.9458503401360545,
+ "rtf": 29.219231699407285,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The sister wins if what your parents say. You have to stand straight.",
+ "raw": {
+ "normalized_reference": "the system runs on three core components that all have to stay in sync.",
+ "normalized_hypothesis": "the sister wins if what your parents say. you have to stand straight.",
+ "errors": 11,
+ "words": 14,
+ "wer": 0.7857142857142857
+ },
+ "semantic": {
+ "normalized_reference": "the system runs on 3 core components that all have to stay in sync",
+ "normalized_hypothesis": "the sister wins if what your parents say you have to stand straight",
+ "errors": 11,
+ "words": 14,
+ "wer": 0.7857142857142857
+ }
+ },
+ {
+ "id": "release500_0015",
+ "category": "conversational",
+ "text": "We improvised the entire second half and somehow nobody noticed.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0015.wav",
+ "render_seconds": 153.60720988898538,
+ "audio_seconds": 4.597551020408163,
+ "rtf": 33.41065911115183,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Wow, perverse, and tall, short, and half, and so high, low, but unnoticed.",
+ "raw": {
+ "normalized_reference": "we improvised the entire second half and somehow nobody noticed.",
+ "normalized_hypothesis": "wow, perverse, and tall, short, and half, and so high, low, but unnoticed.",
+ "errors": 11,
+ "words": 10,
+ "wer": 1.1
+ },
+ "semantic": {
+ "normalized_reference": "we improvised the entire 2nd half and somehow nobody noticed",
+ "normalized_hypothesis": "wow perverse and tall short and half and so high low but unnoticed",
+ "errors": 11,
+ "words": 10,
+ "wer": 1.1
+ }
+ },
+ {
+ "id": "release500_0016",
+ "category": "descriptive",
+ "text": "There's a smell in old buildings that's impossible to describe but immediately familiar.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0016.wav",
+ "render_seconds": 205.70407292200252,
+ "audio_seconds": 5.851428571428571,
+ "rtf": 35.15450464975629,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "There's a story I read about a special person who took a squall with a rat on the floor.",
+ "raw": {
+ "normalized_reference": "there's a smell in old buildings that's impossible to describe but immediately familiar.",
+ "normalized_hypothesis": "there's a story i read about a special person who took a squall with a rat on the floor.",
+ "errors": 17,
+ "words": 13,
+ "wer": 1.3076923076923077
+ },
+ "semantic": {
+ "normalized_reference": "there is a smell in old buildings that is impossible to describe but immediately familiar",
+ "normalized_hypothesis": "there is a story i read about a special person who took a squall with a rat on the floor",
+ "errors": 17,
+ "words": 15,
+ "wer": 1.1333333333333333
+ }
+ },
+ {
+ "id": "release500_0017",
+ "category": "emotional",
+ "text": "I specifically asked for one thing. One thing. And it still didn't happen.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0017.wav",
+ "render_seconds": 191.3965175850317,
+ "audio_seconds": 5.712108843537415,
+ "rtf": 33.50715520793595,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I specifically ask for Wilson. Wilson. And it's too tangled up in it.",
+ "raw": {
+ "normalized_reference": "i specifically asked for one thing. one thing. and it still didn't happen.",
+ "normalized_hypothesis": "i specifically ask for wilson. wilson. and it's too tangled up in it.",
+ "errors": 10,
+ "words": 13,
+ "wer": 0.7692307692307693
+ },
+ "semantic": {
+ "normalized_reference": "i specifically asked for one thing one thing and it still did not happen",
+ "normalized_hypothesis": "i specifically ask for wilson wilson and it is too tangled up in it",
+ "errors": 11,
+ "words": 14,
+ "wer": 0.7857142857142857
+ }
+ },
+ {
+ "id": "release500_0018",
+ "category": "homographs",
+ "text": "They refuse to place the refuse beside the produce aisle.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0018.wav",
+ "render_seconds": 133.39925820799544,
+ "audio_seconds": 3.9706122448979593,
+ "rtf": 33.59664706101859,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I was forced to push through a house I shouldn't have put us in.",
+ "raw": {
+ "normalized_reference": "they refuse to place the refuse beside the produce aisle.",
+ "normalized_hypothesis": "i was forced to push through a house i shouldn't have put us in.",
+ "errors": 13,
+ "words": 10,
+ "wer": 1.3
+ },
+ "semantic": {
+ "normalized_reference": "they refuse to place the refuse beside the produce aisle",
+ "normalized_hypothesis": "i was forced to push through a house i should not have put us in",
+ "errors": 14,
+ "words": 10,
+ "wer": 1.4
+ }
+ },
+ {
+ "id": "release500_0019",
+ "category": "long",
+ "text": "There were days last year when I genuinely didn't know how I was going to get through it, and somehow I did, and I still don't fully understand how.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0019.wav",
+ "render_seconds": 256.6061690719798,
+ "audio_seconds": 8.916462585034013,
+ "rtf": 28.778920634140803,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I woke up last night. I woke up last night. I woke up last night. I woke up last night. I woke up last night. I woke up last night. I woke up last night. I woke up last night. I woke up last night. I woke up last night. I woke up last night. I woke up last night. I woke up last night. I woke up last night. I woke up last night. I woke up last night. I woke up last night. I woke up last night. I woke up last night. I woke up last night. I woke up last night. I woke up last night. I woke up last night. I woke up last night. I woke up last night. I woke up last night. I woke up last night. I woke up last night. I woke up last night. I woke up last night. I woke up last night. I woke up last night. I woke up last night. I woke up last night. I woke up last night. I woke up last night. I woke up last night. I woke up last night. I woke up last night. I woke up last night. I woke up last night. I woke up last night. I woke up last night. I woke up last night. I woke up last night. I woke up last night. I woke up last night. I woke up last night. I woke up last night. I woke up last night. I woke up last night. I woke up last night. I woke up last night. I woke up last night. I woke up last night. I woke up last night. I woke up last night. I woke up last night. I woke up last night. I woke up last night. I woke up last night. I woke up last night. I woke up last night. I woke up last night. I woke up last night. I woke up last night. I woke up last night. I woke up last night. I woke up last night. I woke up last night. I woke up last night. I woke up last night. I woke up last night. I woke up last night.",
+ "raw": {
+ "normalized_reference": "there were days last year when i genuinely didn't know how i was going to get through it, and somehow i did, and i still don't fully understand how.",
+ "normalized_hypothesis": "i woke up last night. i woke up last night. i woke up last night. i woke up last night. i woke up last night. i woke up last night. i woke up last night. i woke up last night. i woke up last night. i woke up last night. i woke up last night. i woke up last night. i woke up last night. i woke up last night. i woke up last night. i woke up last night. i woke up last night. i woke up last night. i woke up last night. i woke up last night. i woke up last night. i woke up last night. i woke up last night. i woke up last night. i woke up last night. i woke up last night. i woke up last night. i woke up last night. i woke up last night. i woke up last night. i woke up last night. i woke up last night. i woke up last night. i woke up last night. i woke up last night. i woke up last night. i woke up last night. i woke up last night. i woke up last night. i woke up last night. i woke up last night. i woke up last night. i woke up last night. i woke up last night. i woke up last night. i woke up last night. i woke up last night. i woke up last night. i woke up last night. i woke up last night. i woke up last night. i woke up last night. i woke up last night. i woke up last night. i woke up last night. i woke up last night. i woke up last night. i woke up last night. i woke up last night. i woke up last night. i woke up last night. i woke up last night. i woke up last night. i woke up last night. i woke up last night. i woke up last night. i woke up last night. i woke up last night. i woke up last night. i woke up last night. i woke up last night. i woke up last night. i woke up last night. i woke up last night.",
+ "errors": 365,
+ "words": 29,
+ "wer": 12.586206896551724
+ },
+ "semantic": {
+ "normalized_reference": "there were days last year when i genuinely did not know how i was going to get through it and somehow i did and i still do not fully understand how",
+ "normalized_hypothesis": "i woke up last night i woke up last night i woke up last night i woke up last night i woke up last night i woke up last night i woke up last night i woke up last night i woke up last night i woke up last night i woke up last night i woke up last night i woke up last night i woke up last night i woke up last night i woke up last night i woke up last night i woke up last night i woke up last night i woke up last night i woke up last night i woke up last night i woke up last night i woke up last night i woke up last night i woke up last night i woke up last night i woke up last night i woke up last night i woke up last night i woke up last night i woke up last night i woke up last night i woke up last night i woke up last night i woke up last night i woke up last night i woke up last night i woke up last night i woke up last night i woke up last night i woke up last night i woke up last night i woke up last night i woke up last night i woke up last night i woke up last night i woke up last night i woke up last night i woke up last night i woke up last night i woke up last night i woke up last night i woke up last night i woke up last night i woke up last night i woke up last night i woke up last night i woke up last night i woke up last night i woke up last night i woke up last night i woke up last night i woke up last night i woke up last night i woke up last night i woke up last night i woke up last night i woke up last night i woke up last night i woke up last night i woke up last night i woke up last night i woke up last night",
+ "errors": 365,
+ "words": 31,
+ "wer": 11.774193548387096
+ }
+ },
+ {
+ "id": "release500_0020",
+ "category": "long_range",
+ "text": "Before announcing the winner, the judge thanked the volunteers who had worked through the storm to repair the damaged stage.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0020.wav",
+ "render_seconds": 195.6878811479546,
+ "audio_seconds": 7.801904761904762,
+ "rtf": 25.082064844403362,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Before night fell in the middle, the church sat to feed the cows that worked for the school to repair the damaged staff.",
+ "raw": {
+ "normalized_reference": "before announcing the winner, the judge thanked the volunteers who had worked through the storm to repair the damaged stage.",
+ "normalized_hypothesis": "before night fell in the middle, the church sat to feed the cows that worked for the school to repair the damaged staff.",
+ "errors": 14,
+ "words": 20,
+ "wer": 0.7
+ },
+ "semantic": {
+ "normalized_reference": "before announcing the winner the judge thanked the volunteers who had worked through the storm to repair the damaged stage",
+ "normalized_hypothesis": "before night fell in the middle the church sat to feed the cows that worked for the school to repair the damaged staff",
+ "errors": 14,
+ "words": 20,
+ "wer": 0.7
+ }
+ },
+ {
+ "id": "release500_0021",
+ "category": "names_places",
+ "text": "Professor Nguyen mailed the fluorescent poster to Reykjavik on Thursday.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0021.wav",
+ "render_seconds": 169.68825406499673,
+ "audio_seconds": 5.154829931972789,
+ "rtf": 32.9183030874611,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Professor, you are allowed to feel awesome. Who's going to walk over to your doorstep?",
+ "raw": {
+ "normalized_reference": "professor nguyen mailed the fluorescent poster to reykjavik on thursday.",
+ "normalized_hypothesis": "professor, you are allowed to feel awesome. who's going to walk over to your doorstep?",
+ "errors": 13,
+ "words": 10,
+ "wer": 1.3
+ },
+ "semantic": {
+ "normalized_reference": "professor nguyen mailed the fluorescent poster to reykjavik on thursday",
+ "normalized_hypothesis": "professor you are allowed to feel awesome who is going to walk over to your doorstep",
+ "errors": 14,
+ "words": 10,
+ "wer": 1.4
+ }
+ },
+ {
+ "id": "release500_0022",
+ "category": "narrative",
+ "text": "The plan was simple. The execution was anything but.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0022.wav",
+ "render_seconds": 151.58793161890935,
+ "audio_seconds": 3.9706122448979593,
+ "rtf": 38.17747044267351,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The power of the serpent. The power of the serpent. The power of the serpent. The power of the serpent. The power of the serpent. The power of the serpent. The power of the serpent. The power of the serpent. The power of the serpent. The power of the serpent. The power of the serpent. The power of the serpent. The power of the serpent. The power of the serpent. The power of the serpent. The power of the serpent. The power of the serpent. The power of the serpent. The power of the serpent. The power of the serpent. The power of the serpent. The power of the serpent. The power of the serpent. The power of the serpent. The power of the serpent. The power of the serpent. The power of the serpent. The power of the serpent. The power of the serpent. The power of the serpent. The power of the serpent. The power of the serpent. The power of the serpent. The power of the serpent. The power of the serpent. The power of the serpent. The power of the serpent. The power of the serpent. The power of the serpent. The power of the serpent. The power of the serpent. The power of the serpent. The power of the serpent. The power of the serpent. The power of the serpent. The power of the serpent. The power of the serpent. The power of the serpent. The power of the serpent. The power of the serpent. The power of the serpent. The power of the serpent. The power of the serpent. The power of the serpent. The power of the serpent. The power of the serpent. The power of the serpent. The power of the serpent. The power of the serpent. The power of the serpent. The power of the serpent. The power of the serpent. The power of the serpent. The power of the serpent. The power of the serpent. The power of the serpent. The power of the serpent. The power of the serpent. The power of the serpent. The power of the serpent. The power of the serpent. The power of the serpent. The power of the serpent. The power of the serpent.",
+ "raw": {
+ "normalized_reference": "the plan was simple. the execution was anything but.",
+ "normalized_hypothesis": "the power of the serpent. the power of the serpent. the power of the serpent. the power of the serpent. the power of the serpent. the power of the serpent. the power of the serpent. the power of the serpent. the power of the serpent. the power of the serpent. the power of the serpent. the power of the serpent. the power of the serpent. the power of the serpent. the power of the serpent. the power of the serpent. the power of the serpent. the power of the serpent. the power of the serpent. the power of the serpent. the power of the serpent. the power of the serpent. the power of the serpent. the power of the serpent. the power of the serpent. the power of the serpent. the power of the serpent. the power of the serpent. the power of the serpent. the power of the serpent. the power of the serpent. the power of the serpent. the power of the serpent. the power of the serpent. the power of the serpent. the power of the serpent. the power of the serpent. the power of the serpent. the power of the serpent. the power of the serpent. the power of the serpent. the power of the serpent. the power of the serpent. the power of the serpent. the power of the serpent. the power of the serpent. the power of the serpent. the power of the serpent. the power of the serpent. the power of the serpent. the power of the serpent. the power of the serpent. the power of the serpent. the power of the serpent. the power of the serpent. the power of the serpent. the power of the serpent. the power of the serpent. the power of the serpent. the power of the serpent. the power of the serpent. the power of the serpent. the power of the serpent. the power of the serpent. the power of the serpent. the power of the serpent. the power of the serpent. the power of the serpent. the power of the serpent. the power of the serpent. the power of the serpent. the power of the serpent. the power of the serpent. the power of the serpent.",
+ "errors": 368,
+ "words": 9,
+ "wer": 40.888888888888886
+ },
+ "semantic": {
+ "normalized_reference": "the plan was simple the execution was anything but",
+ "normalized_hypothesis": "the power of the serpent the power of the serpent the power of the serpent the power of the serpent the power of the serpent the power of the serpent the power of the serpent the power of the serpent the power of the serpent the power of the serpent the power of the serpent the power of the serpent the power of the serpent the power of the serpent the power of the serpent the power of the serpent the power of the serpent the power of the serpent the power of the serpent the power of the serpent the power of the serpent the power of the serpent the power of the serpent the power of the serpent the power of the serpent the power of the serpent the power of the serpent the power of the serpent the power of the serpent the power of the serpent the power of the serpent the power of the serpent the power of the serpent the power of the serpent the power of the serpent the power of the serpent the power of the serpent the power of the serpent the power of the serpent the power of the serpent the power of the serpent the power of the serpent the power of the serpent the power of the serpent the power of the serpent the power of the serpent the power of the serpent the power of the serpent the power of the serpent the power of the serpent the power of the serpent the power of the serpent the power of the serpent the power of the serpent the power of the serpent the power of the serpent the power of the serpent the power of the serpent the power of the serpent the power of the serpent the power of the serpent the power of the serpent the power of the serpent the power of the serpent the power of the serpent the power of the serpent the power of the serpent the power of the serpent the power of the serpent the power of the serpent the power of the serpent the power of the serpent the power of the serpent the power of the serpent",
+ "errors": 368,
+ "words": 9,
+ "wer": 40.888888888888886
+ }
+ },
+ {
+ "id": "release500_0023",
+ "category": "numbers",
+ "text": "Her score improved from eighty-two percent to ninety-seven percent in two weeks.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0023.wav",
+ "render_seconds": 199.00377235002816,
+ "audio_seconds": 5.50312925170068,
+ "rtf": 36.16192955826511,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "...............................................................................................................................................................................................................................................................................................................................................................................",
+ "raw": {
+ "normalized_reference": "her score improved from eighty-two percent to ninety-seven percent in two weeks.",
+ "normalized_hypothesis": "...............................................................................................................................................................................................................................................................................................................................................................................",
+ "errors": 14,
+ "words": 14,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "her score improved from 82% to 97% in 2 weeks",
+ "normalized_hypothesis": " .",
+ "errors": 10,
+ "words": 10,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0024",
+ "category": "punctuation",
+ "text": "Why would anyone pack spoons, batteries, and a violin for one short trip?",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0024.wav",
+ "render_seconds": 189.39552384999115,
+ "audio_seconds": 5.294149659863946,
+ "rtf": 35.774493737084576,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "There are no books for you scholars. I know very well from this book hunt",
+ "raw": {
+ "normalized_reference": "why would anyone pack spoons, batteries, and a violin for one short trip?",
+ "normalized_hypothesis": "there are no books for you scholars. i know very well from this book hunt",
+ "errors": 15,
+ "words": 13,
+ "wer": 1.1538461538461537
+ },
+ "semantic": {
+ "normalized_reference": "why would anyone pack spoons batteries and a violin for one short trip",
+ "normalized_hypothesis": "there are no books for you scholars i know very well from this book hunt",
+ "errors": 15,
+ "words": 13,
+ "wer": 1.1538461538461537
+ }
+ },
+ {
+ "id": "release500_0025",
+ "category": "question",
+ "text": "How long have you been sitting on this?",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0025.wav",
+ "render_seconds": 113.7015101199504,
+ "audio_seconds": 2.5774149659863945,
+ "rtf": 44.114553387966396,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Oh, you never been shot on the windows.",
+ "raw": {
+ "normalized_reference": "how long have you been sitting on this?",
+ "normalized_hypothesis": "oh, you never been shot on the windows.",
+ "errors": 7,
+ "words": 8,
+ "wer": 0.875
+ },
+ "semantic": {
+ "normalized_reference": "how long have you been sitting on this",
+ "normalized_hypothesis": "0 you never been shot on the windows",
+ "errors": 7,
+ "words": 8,
+ "wer": 0.875
+ }
+ },
+ {
+ "id": "release500_0026",
+ "category": "rare_words",
+ "text": "A mischievous chameleon rested beside the chrysanthemum and the bougainvillea.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0026.wav",
+ "render_seconds": 150.89934963197447,
+ "audio_seconds": 5.572789115646258,
+ "rtf": 27.077886225464173,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Let's have a show now. Let's check the shirt and closet. And the boot, I don't know.",
+ "raw": {
+ "normalized_reference": "a mischievous chameleon rested beside the chrysanthemum and the bougainvillea.",
+ "normalized_hypothesis": "let's have a show now. let's check the shirt and closet. and the boot, i don't know.",
+ "errors": 13,
+ "words": 10,
+ "wer": 1.3
+ },
+ "semantic": {
+ "normalized_reference": "a mischievous chameleon rested beside the chrysanthemum and the bougainvillea",
+ "normalized_hypothesis": "let us have a show now let us check the shirt and closet and the boot i do not know",
+ "errors": 16,
+ "words": 10,
+ "wer": 1.6
+ }
+ },
+ {
+ "id": "release500_0027",
+ "category": "short",
+ "text": "We don't talk about that.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0027.wav",
+ "render_seconds": 87.49977569805924,
+ "audio_seconds": 1.9504761904761905,
+ "rtf": 44.86072484519639,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Thank you for watching.",
+ "raw": {
+ "normalized_reference": "we don't talk about that.",
+ "normalized_hypothesis": "thank you for watching.",
+ "errors": 5,
+ "words": 5,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "we do not talk about that",
+ "normalized_hypothesis": "thank you for watching",
+ "errors": 6,
+ "words": 6,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0028",
+ "category": "technical",
+ "text": "Hash the UTF eight payload with S H A two fifty-six before opening the checkpoint.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0028.wav",
+ "render_seconds": 188.11445014097262,
+ "audio_seconds": 5.7817687074829935,
+ "rtf": 32.535796511110426,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Hello, welcome to my channel. I hope you are enjoying this video.",
+ "raw": {
+ "normalized_reference": "hash the utf eight payload with s h a two fifty-six before opening the checkpoint.",
+ "normalized_hypothesis": "hello, welcome to my channel. i hope you are enjoying this video.",
+ "errors": 16,
+ "words": 16,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "hash the utf 8 payload with s h a 256 before opening the checkpoint",
+ "normalized_hypothesis": "hello welcome to my channel i hope you are enjoying this video",
+ "errors": 14,
+ "words": 14,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0029",
+ "category": "conversational",
+ "text": "I thought I had it all figured out, and then this happened.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0029.wav",
+ "render_seconds": 158.59538096003234,
+ "audio_seconds": 4.179591836734694,
+ "rtf": 37.94518392110149,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "That's all I have to say. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you.",
+ "raw": {
+ "normalized_reference": "i thought i had it all figured out, and then this happened.",
+ "normalized_hypothesis": "that's all i have to say. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you.",
+ "errors": 331,
+ "words": 12,
+ "wer": 27.583333333333332
+ },
+ "semantic": {
+ "normalized_reference": "i thought i had it all figured out and then this happened",
+ "normalized_hypothesis": "that is all i have to say i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you",
+ "errors": 332,
+ "words": 12,
+ "wer": 27.666666666666668
+ }
+ },
+ {
+ "id": "release500_0030",
+ "category": "descriptive",
+ "text": "The kind of quiet you only get at three in the morning in a city.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0030.wav",
+ "render_seconds": 166.09637280600145,
+ "audio_seconds": 4.179591836734694,
+ "rtf": 39.73985482174839,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The corn is here. The corn is here. The corn is here. The corn is here. The corn is here. The corn is here. The corn is here. The corn is here. The corn is here. The corn is here. The corn is here. The corn is here. The corn is here. The corn is here. The corn is here. The corn is here. The corn is here. The corn is here. The corn is here. The corn is here. The corn is here. The corn is here. The corn is here. The corn is here. The corn is here. The corn is here. The corn is here. The corn is here. The corn is here. The corn is here. The corn is here. The corn is here. The corn is here. The corn is here. The corn is here. The corn is here. The corn is here. The corn is here. The corn is here. The corn is here. The corn is here. The corn is here. The corn is here. The corn is here. The corn is here. The corn is here. The corn is here. The corn is here. The corn is here. The corn is here. The corn is here. The corn is here. The corn is here. The corn is here. The corn is here. The corn is here. The corn is here. The corn is here. The corn is here. The corn is here. The corn is here. The corn is here. The corn is here. The corn is here. The corn is here. The corn is here. The corn is here. The corn is here. The corn is here. The corn is here. The corn is here. The corn is here. The corn is here. The corn is here. The corn is here. The corn is here. The corn is here. The corn is here. The corn is here. The corn is here. The corn is here. The corn is here. The corn is here. The corn is here. The corn is here. The corn is here. The corn is here. The corn is here. The corn is here",
+ "raw": {
+ "normalized_reference": "the kind of quiet you only get at three in the morning in a city.",
+ "normalized_hypothesis": "the corn is here. the corn is here. the corn is here. the corn is here. the corn is here. the corn is here. the corn is here. the corn is here. the corn is here. the corn is here. the corn is here. the corn is here. the corn is here. the corn is here. the corn is here. the corn is here. the corn is here. the corn is here. the corn is here. the corn is here. the corn is here. the corn is here. the corn is here. the corn is here. the corn is here. the corn is here. the corn is here. the corn is here. the corn is here. the corn is here. the corn is here. the corn is here. the corn is here. the corn is here. the corn is here. the corn is here. the corn is here. the corn is here. the corn is here. the corn is here. the corn is here. the corn is here. the corn is here. the corn is here. the corn is here. the corn is here. the corn is here. the corn is here. the corn is here. the corn is here. the corn is here. the corn is here. the corn is here. the corn is here. the corn is here. the corn is here. the corn is here. the corn is here. the corn is here. the corn is here. the corn is here. the corn is here. the corn is here. the corn is here. the corn is here. the corn is here. the corn is here. the corn is here. the corn is here. the corn is here. the corn is here. the corn is here. the corn is here. the corn is here. the corn is here. the corn is here. the corn is here. the corn is here. the corn is here. the corn is here. the corn is here. the corn is here. the corn is here. the corn is here. the corn is here. the corn is here. the corn is here. the corn is here. the corn is here",
+ "errors": 354,
+ "words": 15,
+ "wer": 23.6
+ },
+ "semantic": {
+ "normalized_reference": "the kind of quiet you only get at 3 in the morning in a city",
+ "normalized_hypothesis": "the corn is here the corn is here the corn is here the corn is here the corn is here the corn is here the corn is here the corn is here the corn is here the corn is here the corn is here the corn is here the corn is here the corn is here the corn is here the corn is here the corn is here the corn is here the corn is here the corn is here the corn is here the corn is here the corn is here the corn is here the corn is here the corn is here the corn is here the corn is here the corn is here the corn is here the corn is here the corn is here the corn is here the corn is here the corn is here the corn is here the corn is here the corn is here the corn is here the corn is here the corn is here the corn is here the corn is here the corn is here the corn is here the corn is here the corn is here the corn is here the corn is here the corn is here the corn is here the corn is here the corn is here the corn is here the corn is here the corn is here the corn is here the corn is here the corn is here the corn is here the corn is here the corn is here the corn is here the corn is here the corn is here the corn is here the corn is here the corn is here the corn is here the corn is here the corn is here the corn is here the corn is here the corn is here the corn is here the corn is here the corn is here the corn is here the corn is here the corn is here the corn is here the corn is here the corn is here the corn is here the corn is here the corn is here the corn is here the corn is here the corn is here",
+ "errors": 354,
+ "words": 15,
+ "wer": 23.6
+ }
+ },
+ {
+ "id": "release500_0031",
+ "category": "emotional",
+ "text": "The joy of it was almost overwhelming, like too much good all at once.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0031.wav",
+ "render_seconds": 193.0979104069993,
+ "audio_seconds": 4.876190476190477,
+ "rtf": 39.600157407685394,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Actually, I thought this would be a stay for the day. But, let's play it once.",
+ "raw": {
+ "normalized_reference": "the joy of it was almost overwhelming, like too much good all at once.",
+ "normalized_hypothesis": "actually, i thought this would be a stay for the day. but, let's play it once.",
+ "errors": 15,
+ "words": 14,
+ "wer": 1.0714285714285714
+ },
+ "semantic": {
+ "normalized_reference": "the joy of it was almost overwhelming like too much good all at once",
+ "normalized_hypothesis": "actually i thought this would be a stay for the day but let us play it once",
+ "errors": 16,
+ "words": 14,
+ "wer": 1.1428571428571428
+ }
+ },
+ {
+ "id": "release500_0032",
+ "category": "homographs",
+ "text": "Please close the door while I stand close to the microphone.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0032.wav",
+ "render_seconds": 93.71202471398283,
+ "audio_seconds": 4.1099319727891155,
+ "rtf": 22.80135664882726,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "That's physical real estate history.",
+ "raw": {
+ "normalized_reference": "please close the door while i stand close to the microphone.",
+ "normalized_hypothesis": "that's physical real estate history.",
+ "errors": 11,
+ "words": 11,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "please close the door while i stand close to the microphone",
+ "normalized_hypothesis": "that is physical real estate history",
+ "errors": 11,
+ "words": 11,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0033",
+ "category": "long",
+ "text": "I think the most underrated skill in the world is knowing when to stop explaining yourself and just let people draw their own conclusions.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0033.wav",
+ "render_seconds": 270.5096920909127,
+ "audio_seconds": 8.916462585034013,
+ "rtf": 30.338229932681404,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The fact that we still have a lot of space in the world doesn't mean that we're going to still be spreading ourselves and just let that be cool and normal for us.",
+ "raw": {
+ "normalized_reference": "i think the most underrated skill in the world is knowing when to stop explaining yourself and just let people draw their own conclusions.",
+ "normalized_hypothesis": "the fact that we still have a lot of space in the world doesn't mean that we're going to still be spreading ourselves and just let that be cool and normal for us.",
+ "errors": 26,
+ "words": 24,
+ "wer": 1.0833333333333333
+ },
+ "semantic": {
+ "normalized_reference": "i think the most underrated skill in the world is knowing when to stop explaining yourself and just let people draw their own conclusions",
+ "normalized_hypothesis": "the fact that we still have a lot of space in the world does not mean that we are going to still be spreading ourselves and just let that be cool and normal for us",
+ "errors": 28,
+ "words": 24,
+ "wer": 1.1666666666666667
+ }
+ },
+ {
+ "id": "release500_0034",
+ "category": "long_range",
+ "text": "When the final bus disappeared around the corner, Nadia remembered that her umbrella, notebook, and lunch were still inside the quiet office.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0034.wav",
+ "render_seconds": 267.98907835106365,
+ "audio_seconds": 9.404081632653062,
+ "rtf": 28.497102515629596,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It is very best to support the world here. Now I will help them and hold their brain, make them, and let you stay and serve the universe.",
+ "raw": {
+ "normalized_reference": "when the final bus disappeared around the corner, nadia remembered that her umbrella, notebook, and lunch were still inside the quiet office.",
+ "normalized_hypothesis": "it is very best to support the world here. now i will help them and hold their brain, make them, and let you stay and serve the universe.",
+ "errors": 26,
+ "words": 22,
+ "wer": 1.1818181818181819
+ },
+ "semantic": {
+ "normalized_reference": "when the final bus disappeared around the corner nadia remembered that her umbrella notebook and lunch were still inside the quiet office",
+ "normalized_hypothesis": "it is very best to support the world here now i will help them and hold their brain make them and let you stay and serve the universe",
+ "errors": 26,
+ "words": 22,
+ "wer": 1.1818181818181819
+ }
+ },
+ {
+ "id": "release500_0035",
+ "category": "names_places",
+ "text": "Aurelia and Joaquin crossed the Schuylkill River before noon.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0035.wav",
+ "render_seconds": 144.99596785893664,
+ "audio_seconds": 4.31891156462585,
+ "rtf": 33.57234008830596,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Well, I think that's the first thing I'm going to do this morning.",
+ "raw": {
+ "normalized_reference": "aurelia and joaquin crossed the schuylkill river before noon.",
+ "normalized_hypothesis": "well, i think that's the first thing i'm going to do this morning.",
+ "errors": 12,
+ "words": 9,
+ "wer": 1.3333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "aurelia and joaquin crossed the schuylkill river before noon",
+ "normalized_hypothesis": "well i think that is the 1st thing i am going to do this morning",
+ "errors": 14,
+ "words": 9,
+ "wer": 1.5555555555555556
+ }
+ },
+ {
+ "id": "release500_0036",
+ "category": "narrative",
+ "text": "The village had a name, but nobody used it. Everyone just called it home.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0036.wav",
+ "render_seconds": 188.30701990006492,
+ "audio_seconds": 5.224489795918367,
+ "rtf": 36.0431405277468,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "So they are trying now, but nobody asked it. Everyone just who not who.",
+ "raw": {
+ "normalized_reference": "the village had a name, but nobody used it. everyone just called it home.",
+ "normalized_hypothesis": "so they are trying now, but nobody asked it. everyone just who not who.",
+ "errors": 9,
+ "words": 14,
+ "wer": 0.6428571428571429
+ },
+ "semantic": {
+ "normalized_reference": "the village had a name but nobody used it everyone just called it home",
+ "normalized_hypothesis": "so they are trying now but nobody asked it everyone just who not who",
+ "errors": 9,
+ "words": 14,
+ "wer": 0.6428571428571429
+ }
+ },
+ {
+ "id": "release500_0037",
+ "category": "numbers",
+ "text": "Increase the gain by three decibels, wait five seconds, then reduce it by one.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0037.wav",
+ "render_seconds": 191.00252298009582,
+ "audio_seconds": 5.50312925170068,
+ "rtf": 34.70798417483446,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I crushed the iron before I should be. What's your challenge? I want to shut the door.",
+ "raw": {
+ "normalized_reference": "increase the gain by three decibels, wait five seconds, then reduce it by one.",
+ "normalized_hypothesis": "i crushed the iron before i should be. what's your challenge? i want to shut the door.",
+ "errors": 16,
+ "words": 14,
+ "wer": 1.1428571428571428
+ },
+ "semantic": {
+ "normalized_reference": "increase the gain by 3 decibels wait 5 seconds then reduce it by one",
+ "normalized_hypothesis": "i crushed the iron before i should be what is your challenge i want to shut the door",
+ "errors": 17,
+ "words": 14,
+ "wer": 1.2142857142857142
+ }
+ },
+ {
+ "id": "release500_0038",
+ "category": "punctuation",
+ "text": "The result was unexpected: clear, stable, and surprisingly natural.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0038.wav",
+ "render_seconds": 154.6857005050406,
+ "audio_seconds": 5.433469387755102,
+ "rtf": 28.469047944512425,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "For the sake of this ill-expected clown's taboo, I shall pour some of that milk for you.",
+ "raw": {
+ "normalized_reference": "the result was unexpected: clear, stable, and surprisingly natural.",
+ "normalized_hypothesis": "for the sake of this ill-expected clown's taboo, i shall pour some of that milk for you.",
+ "errors": 17,
+ "words": 9,
+ "wer": 1.8888888888888888
+ },
+ "semantic": {
+ "normalized_reference": "the result was unexpected clear stable and surprisingly natural",
+ "normalized_hypothesis": "for the sake of this ill expected clown is taboo i shall pour some of that milk for you",
+ "errors": 18,
+ "words": 9,
+ "wer": 2.0
+ }
+ },
+ {
+ "id": "release500_0039",
+ "category": "question",
+ "text": "Who told you that was the right way to do it?",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0039.wav",
+ "render_seconds": 126.19904938701075,
+ "audio_seconds": 2.9257142857142857,
+ "rtf": 43.13444070845094,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Okay, I know there's a lot, lot of jokes.",
+ "raw": {
+ "normalized_reference": "who told you that was the right way to do it?",
+ "normalized_hypothesis": "okay, i know there's a lot, lot of jokes.",
+ "errors": 11,
+ "words": 11,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "who told you that was the right way to do it",
+ "normalized_hypothesis": "okay i know there is a lot lot of jokes",
+ "errors": 11,
+ "words": 11,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0040",
+ "category": "rare_words",
+ "text": "The meticulous cartographer examined an iridescent archipelago through binoculars.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0040.wav",
+ "render_seconds": 180.29475947702304,
+ "audio_seconds": 5.851428571428571,
+ "rtf": 30.812092684061554,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Electrical is called for at sound and elevation or capability for vehicles.",
+ "raw": {
+ "normalized_reference": "the meticulous cartographer examined an iridescent archipelago through binoculars.",
+ "normalized_hypothesis": "electrical is called for at sound and elevation or capability for vehicles.",
+ "errors": 12,
+ "words": 9,
+ "wer": 1.3333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "the meticulous cartographer examined an iridescent archipelago through binoculars",
+ "normalized_hypothesis": "electrical is called for at sound and elevation or capability for vehicles",
+ "errors": 12,
+ "words": 9,
+ "wer": 1.3333333333333333
+ }
+ },
+ {
+ "id": "release500_0041",
+ "category": "short",
+ "text": "I mean. Yeah. Kind of.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0041.wav",
+ "render_seconds": 114.09826706699096,
+ "audio_seconds": 2.2987755102040817,
+ "rtf": 49.63436688816191,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Good luck. I love you.",
+ "raw": {
+ "normalized_reference": "i mean. yeah. kind of.",
+ "normalized_hypothesis": "good luck. i love you.",
+ "errors": 5,
+ "words": 5,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "i mean yeah kind of",
+ "normalized_hypothesis": "good luck i love you",
+ "errors": 5,
+ "words": 5,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0042",
+ "category": "technical",
+ "text": "The logs are your best friend when something behaves unexpectedly.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0042.wav",
+ "render_seconds": 173.2955003969837,
+ "audio_seconds": 4.527891156462585,
+ "rtf": 38.2728944686848,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "For you, it's our best friend. We'll show you the house you'll expect of you.",
+ "raw": {
+ "normalized_reference": "the logs are your best friend when something behaves unexpectedly.",
+ "normalized_hypothesis": "for you, it's our best friend. we'll show you the house you'll expect of you.",
+ "errors": 13,
+ "words": 10,
+ "wer": 1.3
+ },
+ "semantic": {
+ "normalized_reference": "the logs are your best friend when something behaves unexpectedly",
+ "normalized_hypothesis": "for you it is our best friend we will show you the house you will expect of you",
+ "errors": 16,
+ "words": 10,
+ "wer": 1.6
+ }
+ },
+ {
+ "id": "release500_0043",
+ "category": "conversational",
+ "text": "He kept talking and I kept nodding and neither of us was listening.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0043.wav",
+ "render_seconds": 166.1927850000793,
+ "audio_seconds": 4.527891156462585,
+ "rtf": 36.70423586990934,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I talk to another. I talk to another. I know this is the show.",
+ "raw": {
+ "normalized_reference": "he kept talking and i kept nodding and neither of us was listening.",
+ "normalized_hypothesis": "i talk to another. i talk to another. i know this is the show.",
+ "errors": 13,
+ "words": 13,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "he kept talking and i kept nodding and neither of us was listening",
+ "normalized_hypothesis": "i talk to another i talk to another i know this is the show",
+ "errors": 13,
+ "words": 13,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0044",
+ "category": "descriptive",
+ "text": "It's one of those places where everyone seems to know each other already.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0044.wav",
+ "render_seconds": 114.20044160599355,
+ "audio_seconds": 4.876190476190477,
+ "rtf": 23.42001243872914,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "That's really nice. That's wonderful. Yes. That's true. That's really nice.",
+ "raw": {
+ "normalized_reference": "it's one of those places where everyone seems to know each other already.",
+ "normalized_hypothesis": "that's really nice. that's wonderful. yes. that's true. that's really nice.",
+ "errors": 13,
+ "words": 13,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "it is one of those places where everyone seems to know each other already",
+ "normalized_hypothesis": "that is really nice that is wonderful yes that is true that is really nice",
+ "errors": 14,
+ "words": 14,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0045",
+ "category": "emotional",
+ "text": "He said he was proud of me and nobody had ever said that to me before.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0045.wav",
+ "render_seconds": 166.0905514130136,
+ "audio_seconds": 4.876190476190477,
+ "rtf": 34.06153886399693,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I said I was going to fly, and nobody ever said that to me before.",
+ "raw": {
+ "normalized_reference": "he said he was proud of me and nobody had ever said that to me before.",
+ "normalized_hypothesis": "i said i was going to fly, and nobody ever said that to me before.",
+ "errors": 6,
+ "words": 16,
+ "wer": 0.375
+ },
+ "semantic": {
+ "normalized_reference": "he said he was proud of me and nobody had ever said that to me before",
+ "normalized_hypothesis": "i said i was going to fly and nobody ever said that to me before",
+ "errors": 6,
+ "words": 16,
+ "wer": 0.375
+ }
+ },
+ {
+ "id": "release500_0046",
+ "category": "homographs",
+ "text": "The minute detail took a minute to explain clearly.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0046.wav",
+ "render_seconds": 137.69410212896764,
+ "audio_seconds": 3.622312925170068,
+ "rtf": 38.012757310994296,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Thank you for watching.",
+ "raw": {
+ "normalized_reference": "the minute detail took a minute to explain clearly.",
+ "normalized_hypothesis": "thank you for watching.",
+ "errors": 9,
+ "words": 9,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "the minute detail took a minute to explain clearly",
+ "normalized_hypothesis": "thank you for watching",
+ "errors": 9,
+ "words": 9,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0047",
+ "category": "long",
+ "text": "The strange thing about really wanting something is that getting it never quite feels the way you imagined it would, and not getting it never quite feels as bad.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0047.wav",
+ "render_seconds": 318.20927972893696,
+ "audio_seconds": 9.961360544217687,
+ "rtf": 31.944359238522818,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The stories that I wrote were not written so tenacious, but kind of novel. That's how I used to laugh when watching a movie. I knew it kind of novel. That's how I used to talk.",
+ "raw": {
+ "normalized_reference": "the strange thing about really wanting something is that getting it never quite feels the way you imagined it would, and not getting it never quite feels as bad.",
+ "normalized_hypothesis": "the stories that i wrote were not written so tenacious, but kind of novel. that's how i used to laugh when watching a movie. i knew it kind of novel. that's how i used to talk.",
+ "errors": 34,
+ "words": 29,
+ "wer": 1.1724137931034482
+ },
+ "semantic": {
+ "normalized_reference": "the strange thing about really wanting something is that getting it never quite feels the way you imagined it would and not getting it never quite feels as bad",
+ "normalized_hypothesis": "the stories that i wrote were not written so tenacious but kind of novel that is how i used to laugh when watching a movie i knew it kind of novel that is how i used to talk",
+ "errors": 35,
+ "words": 29,
+ "wer": 1.206896551724138
+ }
+ },
+ {
+ "id": "release500_0048",
+ "category": "long_range",
+ "text": "Even though the recipe looked simple at first, preparing the sauce required patience, careful timing, and a much larger pan.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0048.wav",
+ "render_seconds": 271.20498704595957,
+ "audio_seconds": 8.289523809523809,
+ "rtf": 32.71659425531452,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "As I know the worship I look shall be enforced, preparing the shoes for your passions, coffee turning, and elixir for pain.",
+ "raw": {
+ "normalized_reference": "even though the recipe looked simple at first, preparing the sauce required patience, careful timing, and a much larger pan.",
+ "normalized_hypothesis": "as i know the worship i look shall be enforced, preparing the shoes for your passions, coffee turning, and elixir for pain.",
+ "errors": 19,
+ "words": 20,
+ "wer": 0.95
+ },
+ "semantic": {
+ "normalized_reference": "even though the recipe looked simple at 1st preparing the sauce required patience careful timing and a much larger pan",
+ "normalized_hypothesis": "as i know the worship i look shall be enforced preparing the shoes for your passions coffee turning and elixir for pain",
+ "errors": 19,
+ "words": 20,
+ "wer": 0.95
+ }
+ },
+ {
+ "id": "release500_0049",
+ "category": "names_places",
+ "text": "Xiomara met Siobhan near the old library in Worcestershire.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0049.wav",
+ "render_seconds": 157.49648706498556,
+ "audio_seconds": 4.667210884353741,
+ "rtf": 33.74531191486835,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I don't know what I'm talking about. I don't know what I'm talking about. I don't know what I'm talking about. I don't know what I'm talking about. I don't know what I'm talking about. I don't know what I'm talking about. I don't know what I'm talking about. I don't know what I'm talking about. I don't know what I'm talking about. I don't know what I'm talking about. I don't know what I'm talking about. I don't know what I'm talking about. I don't know what I'm talking about. I don't know what I'm talking about. I don't know what I'm talking about. I don't know what I'm talking about. I don't know what I'm talking about. I don't know what I'm talking about. I don't know what I'm talking about. I don't know what I'm talking about. I don't know what I'm talking about. I don't know what I'm talking about. I don't know what I'm talking about. I don't know what I'm talking about. I don't know what I'm talking about. I don't know what I'm talking about. I don't know what I'm talking about. I don't know what I'm talking about. I don't know what I'm talking about. I don't know what I'm talking about. I don't know what I'm talking about. I don't know what I'm talking about. I don't know what I'm talking about. I don't know what I'm talking about. I don't know what I'm talking about. I don't know what I'm talking about. I don't know what I'm talking about. I don't know what I'm talking about. I don't know what I'm talking about. I don't know what I'm talking about. I don't know what I'm talking about. I don't know what I'm talking about. I don't know what I'm talking about. I don't know what I'm talking about. I don't know",
+ "raw": {
+ "normalized_reference": "xiomara met siobhan near the old library in worcestershire.",
+ "normalized_hypothesis": "i don't know what i'm talking about. i don't know what i'm talking about. i don't know what i'm talking about. i don't know what i'm talking about. i don't know what i'm talking about. i don't know what i'm talking about. i don't know what i'm talking about. i don't know what i'm talking about. i don't know what i'm talking about. i don't know what i'm talking about. i don't know what i'm talking about. i don't know what i'm talking about. i don't know what i'm talking about. i don't know what i'm talking about. i don't know what i'm talking about. i don't know what i'm talking about. i don't know what i'm talking about. i don't know what i'm talking about. i don't know what i'm talking about. i don't know what i'm talking about. i don't know what i'm talking about. i don't know what i'm talking about. i don't know what i'm talking about. i don't know what i'm talking about. i don't know what i'm talking about. i don't know what i'm talking about. i don't know what i'm talking about. i don't know what i'm talking about. i don't know what i'm talking about. i don't know what i'm talking about. i don't know what i'm talking about. i don't know what i'm talking about. i don't know what i'm talking about. i don't know what i'm talking about. i don't know what i'm talking about. i don't know what i'm talking about. i don't know what i'm talking about. i don't know what i'm talking about. i don't know what i'm talking about. i don't know what i'm talking about. i don't know what i'm talking about. i don't know what i'm talking about. i don't know what i'm talking about. i don't know what i'm talking about. i don't know",
+ "errors": 311,
+ "words": 9,
+ "wer": 34.55555555555556
+ },
+ "semantic": {
+ "normalized_reference": "xiomara met siobhan near the old library in worcestershire",
+ "normalized_hypothesis": "i do not know what i am talking about i do not know what i am talking about i do not know what i am talking about i do not know what i am talking about i do not know what i am talking about i do not know what i am talking about i do not know what i am talking about i do not know what i am talking about i do not know what i am talking about i do not know what i am talking about i do not know what i am talking about i do not know what i am talking about i do not know what i am talking about i do not know what i am talking about i do not know what i am talking about i do not know what i am talking about i do not know what i am talking about i do not know what i am talking about i do not know what i am talking about i do not know what i am talking about i do not know what i am talking about i do not know what i am talking about i do not know what i am talking about i do not know what i am talking about i do not know what i am talking about i do not know what i am talking about i do not know what i am talking about i do not know what i am talking about i do not know what i am talking about i do not know what i am talking about i do not know what i am talking about i do not know what i am talking about i do not know what i am talking about i do not know what i am talking about i do not know what i am talking about i do not know what i am talking about i do not know what i am talking about i do not know what i am talking about i do not know what i am talking about i do not know what i am talking about i do not know what i am talking about i do not know what i am talking about i do not know what i am talking about i do not know what i am talking about i do not know",
+ "errors": 400,
+ "words": 9,
+ "wer": 44.44444444444444
+ }
+ },
+ {
+ "id": "release500_0050",
+ "category": "narrative",
+ "text": "She walked into the room and everyone went quiet, not because she was loud, but because she wasn't.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0050.wav",
+ "render_seconds": 195.6048700939864,
+ "audio_seconds": 6.408707482993197,
+ "rtf": 30.521734782413382,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Hello, welcome to the world another world like that. No, because I was young. But because I was young.",
+ "raw": {
+ "normalized_reference": "she walked into the room and everyone went quiet, not because she was loud, but because she wasn't.",
+ "normalized_hypothesis": "hello, welcome to the world another world like that. no, because i was young. but because i was young.",
+ "errors": 14,
+ "words": 18,
+ "wer": 0.7777777777777778
+ },
+ "semantic": {
+ "normalized_reference": "she walked into the room and everyone went quiet not because she was loud but because she was not",
+ "normalized_hypothesis": "hello welcome to the world another world like that no because i was young but because i was young",
+ "errors": 13,
+ "words": 19,
+ "wer": 0.6842105263157895
+ }
+ },
+ {
+ "id": "release500_0051",
+ "category": "numbers",
+ "text": "At exactly seven forty-five, Morgan placed eighteen dollars and fifty cents beside the notebook.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0051.wav",
+ "render_seconds": 221.61156587593723,
+ "audio_seconds": 6.826666666666667,
+ "rtf": 32.46263172010799,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Thank you for watching this video. If you like this video, please leave a like and subscribe to the channel.",
+ "raw": {
+ "normalized_reference": "at exactly seven forty-five, morgan placed eighteen dollars and fifty cents beside the notebook.",
+ "normalized_hypothesis": "thank you for watching this video. if you like this video, please leave a like and subscribe to the channel.",
+ "errors": 19,
+ "words": 15,
+ "wer": 1.2666666666666666
+ },
+ "semantic": {
+ "normalized_reference": "at exactly 745 morgan placed $18.50 beside the notebook",
+ "normalized_hypothesis": "thank you for watching this video if you like this video please leave a like and subscribe to the channel",
+ "errors": 19,
+ "words": 10,
+ "wer": 1.9
+ }
+ },
+ {
+ "id": "release500_0052",
+ "category": "punctuation",
+ "text": "Does punctuation work? Yes: commas, semicolons, and periods should create sensible boundaries.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0052.wav",
+ "render_seconds": 234.1030401220778,
+ "audio_seconds": 7.453605442176871,
+ "rtf": 31.408026885537236,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "There's books that I want. House, books, celebrities, and poets and poets and some people of course.",
+ "raw": {
+ "normalized_reference": "does punctuation work? yes: commas, semicolons, and periods should create sensible boundaries.",
+ "normalized_hypothesis": "there's books that i want. house, books, celebrities, and poets and poets and some people of course.",
+ "errors": 16,
+ "words": 12,
+ "wer": 1.3333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "does punctuation work yes commas semicolons and periods should create sensible boundaries",
+ "normalized_hypothesis": "there is books that i want house books celebrities and poets and poets and some people of course",
+ "errors": 17,
+ "words": 12,
+ "wer": 1.4166666666666667
+ }
+ },
+ {
+ "id": "release500_0053",
+ "category": "question",
+ "text": "Have you ever just stopped and thought about how strange all of this is?",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0053.wav",
+ "render_seconds": 190.98922071699053,
+ "audio_seconds": 4.876190476190477,
+ "rtf": 39.167711279851574,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Have you ever discussed and thought about this kind of relationship?",
+ "raw": {
+ "normalized_reference": "have you ever just stopped and thought about how strange all of this is?",
+ "normalized_hypothesis": "have you ever discussed and thought about this kind of relationship?",
+ "errors": 7,
+ "words": 14,
+ "wer": 0.5
+ },
+ "semantic": {
+ "normalized_reference": "have you ever just stopped and thought about how strange all of this is",
+ "normalized_hypothesis": "have you ever discussed and thought about this kind of relationship",
+ "errors": 7,
+ "words": 14,
+ "wer": 0.5
+ }
+ },
+ {
+ "id": "release500_0054",
+ "category": "rare_words",
+ "text": "Her idiosyncratic explanation sounded plausible but remained fundamentally ambiguous.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0054.wav",
+ "render_seconds": 220.29730763600674,
+ "audio_seconds": 6.339047619047619,
+ "rtf": 34.75242984041572,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "All I saw, all the touch, all the action, showed up close to me. But all I felt was no backness.",
+ "raw": {
+ "normalized_reference": "her idiosyncratic explanation sounded plausible but remained fundamentally ambiguous.",
+ "normalized_hypothesis": "all i saw, all the touch, all the action, showed up close to me. but all i felt was no backness.",
+ "errors": 20,
+ "words": 9,
+ "wer": 2.2222222222222223
+ },
+ "semantic": {
+ "normalized_reference": "her idiosyncratic explanation sounded plausible but remained fundamentally ambiguous",
+ "normalized_hypothesis": "all i saw all the touch all the action showed up close to me but all i felt was no backness",
+ "errors": 20,
+ "words": 9,
+ "wer": 2.2222222222222223
+ }
+ },
+ {
+ "id": "release500_0055",
+ "category": "short",
+ "text": "That tracks.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0055.wav",
+ "render_seconds": 77.29931494104676,
+ "audio_seconds": 1.4628571428571429,
+ "rtf": 52.84132857298118,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Oh, gosh. Oh, gosh. Oh, gosh. Oh, gosh. Oh, gosh. Oh, gosh. Oh, gosh. Oh, gosh. Oh, gosh. Oh, gosh. Oh, gosh. Oh, gosh. Oh, gosh. Oh, gosh. Oh, gosh. Oh, gosh. Oh, gosh. Oh, gosh. Oh, gosh. Oh, gosh. Oh, gosh. Oh, gosh. Oh, gosh. Oh, gosh. Oh, gosh. Oh, gosh. Oh, gosh. Oh, gosh. Oh, gosh. Oh, gosh. Oh, gosh. Oh, gosh. Oh, gosh. Oh, gosh. Oh, gosh. Oh, gosh. Oh, gosh. Oh, gosh. Oh, gosh. Oh, gosh. Oh, gosh. Oh, gosh. Oh, gosh. Oh, gosh. Oh, gosh. Oh, gosh. Oh, gosh. Oh, gosh. Oh, gosh. Oh, gosh. Oh, gosh. Oh, gosh. Oh, gosh. Oh, gosh. Oh, gosh. Oh, gosh. Oh, gosh. Oh, gosh. Oh, gosh. Oh, gosh. Oh, gosh. Oh, gosh. Oh, gosh. Oh, gosh. Oh, gosh. Oh, gosh. Oh, gosh. Oh, gosh. Oh, gosh. Oh, gosh. Oh, gosh. Oh, gosh. Oh, gosh. Oh, gosh. Oh, gosh. Oh, gosh. Oh, gosh. Oh, gosh. Oh, gosh. Oh, gosh. Oh, gosh. Oh, gosh. Oh, gosh. Oh, gosh. Oh, gosh. Oh, gosh. Oh, gosh. Oh, gosh. Oh, gosh. Oh, gosh. Oh, gosh. Oh, gosh. Oh, gosh. Oh, gosh. Oh, gosh. Oh, gosh. Oh, gosh. Oh, gosh. Oh, gosh. Oh, gosh. Oh, gosh. Oh, gosh. Oh, gosh. Oh, gosh. Oh, gosh. Oh, gosh. Oh, gosh. Oh, gosh. Oh, gosh. Oh, gosh. Oh, gosh.",
+ "raw": {
+ "normalized_reference": "that tracks.",
+ "normalized_hypothesis": "oh, gosh. oh, gosh. oh, gosh. oh, gosh. oh, gosh. oh, gosh. oh, gosh. oh, gosh. oh, gosh. oh, gosh. oh, gosh. oh, gosh. oh, gosh. oh, gosh. oh, gosh. oh, gosh. oh, gosh. oh, gosh. oh, gosh. oh, gosh. oh, gosh. oh, gosh. oh, gosh. oh, gosh. oh, gosh. oh, gosh. oh, gosh. oh, gosh. oh, gosh. oh, gosh. oh, gosh. oh, gosh. oh, gosh. oh, gosh. oh, gosh. oh, gosh. oh, gosh. oh, gosh. oh, gosh. oh, gosh. oh, gosh. oh, gosh. oh, gosh. oh, gosh. oh, gosh. oh, gosh. oh, gosh. oh, gosh. oh, gosh. oh, gosh. oh, gosh. oh, gosh. oh, gosh. oh, gosh. oh, gosh. oh, gosh. oh, gosh. oh, gosh. oh, gosh. oh, gosh. oh, gosh. oh, gosh. oh, gosh. oh, gosh. oh, gosh. oh, gosh. oh, gosh. oh, gosh. oh, gosh. oh, gosh. oh, gosh. oh, gosh. oh, gosh. oh, gosh. oh, gosh. oh, gosh. oh, gosh. oh, gosh. oh, gosh. oh, gosh. oh, gosh. oh, gosh. oh, gosh. oh, gosh. oh, gosh. oh, gosh. oh, gosh. oh, gosh. oh, gosh. oh, gosh. oh, gosh. oh, gosh. oh, gosh. oh, gosh. oh, gosh. oh, gosh. oh, gosh. oh, gosh. oh, gosh. oh, gosh. oh, gosh. oh, gosh. oh, gosh. oh, gosh. oh, gosh. oh, gosh. oh, gosh. oh, gosh. oh, gosh. oh, gosh. oh, gosh.",
+ "errors": 222,
+ "words": 2,
+ "wer": 111.0
+ },
+ "semantic": {
+ "normalized_reference": "that tracks",
+ "normalized_hypothesis": "0 gosh 0 gosh 0 gosh 0 gosh 0 gosh 0 gosh 0 gosh 0 gosh 0 gosh 0 gosh 0 gosh 0 gosh 0 gosh 0 gosh 0 gosh 0 gosh 0 gosh 0 gosh 0 gosh 0 gosh 0 gosh 0 gosh 0 gosh 0 gosh 0 gosh 0 gosh 0 gosh 0 gosh 0 gosh 0 gosh 0 gosh 0 gosh 0 gosh 0 gosh 0 gosh 0 gosh 0 gosh 0 gosh 0 gosh 0 gosh 0 gosh 0 gosh 0 gosh 0 gosh 0 gosh 0 gosh 0 gosh 0 gosh 0 gosh 0 gosh 0 gosh 0 gosh 0 gosh 0 gosh 0 gosh 0 gosh 0 gosh 0 gosh 0 gosh 0 gosh 0 gosh 0 gosh 0 gosh 0 gosh 0 gosh 0 gosh 0 gosh 0 gosh 0 gosh 0 gosh 0 gosh 0 gosh 0 gosh 0 gosh 0 gosh 0 gosh 0 gosh 0 gosh 0 gosh 0 gosh 0 gosh 0 gosh 0 gosh 0 gosh 0 gosh 0 gosh 0 gosh 0 gosh 0 gosh 0 gosh 0 gosh 0 gosh 0 gosh 0 gosh 0 gosh 0 gosh 0 gosh 0 gosh 0 gosh 0 gosh 0 gosh 0 gosh 0 gosh 0 gosh 0 gosh 0 gosh 0 gosh 0 gosh 0 gosh 0 gosh 0 gosh",
+ "errors": 222,
+ "words": 2,
+ "wer": 111.0
+ }
+ },
+ {
+ "id": "release500_0056",
+ "category": "technical",
+ "text": "The key thing to remember is that order matters here more than you'd think.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0056.wav",
+ "render_seconds": 135.3948296570452,
+ "audio_seconds": 4.597551020408163,
+ "rtf": 29.449337061413416,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The Crescent River is a beautiful place for living, I think.",
+ "raw": {
+ "normalized_reference": "the key thing to remember is that order matters here more than you'd think.",
+ "normalized_hypothesis": "the crescent river is a beautiful place for living, i think.",
+ "errors": 11,
+ "words": 14,
+ "wer": 0.7857142857142857
+ },
+ "semantic": {
+ "normalized_reference": "the key thing to remember is that order matters here more than you would think",
+ "normalized_hypothesis": "the crescent river is a beautiful place for living i think",
+ "errors": 12,
+ "words": 15,
+ "wer": 0.8
+ }
+ },
+ {
+ "id": "release500_0057",
+ "category": "conversational",
+ "text": "I would not say that again in this room if I were you.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0057.wav",
+ "render_seconds": 150.48785761697218,
+ "audio_seconds": 3.7616326530612243,
+ "rtf": 40.00599513472025,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I love you. I love you too. I love you too. I love you too. I love you too. I love you too. I love you too. I love you too. I love you too. I love you too. I love you too. I love you too. I love you too. I love you too. I love you too. I love you too. I love you too. I love you too. I love you too. I love you too. I love you too. I love you too. I love you too. I love you too. I love you too. I love you too. I love you too. I love you too. I love you too. I love you too. I love you too. I love you too. I love you too. I love you too. I love you too. I love you too. I love you too. I love you too. I love you too. I love you too. I love you too. I love you too. I love you too. I love you too. I love you too. I love you too. I love you too. I love you too. I love you too. I love you too. I love you too. I love you too. I love you too. I love you too. I love you too. I love you too. I love you too. I love you too. I love you too. I love you too. I love you too. I love you too. I love you too. I love you too. I love you too. I love you too. I love you too. I love you too. I love you too. I love you too. I love you too. I love you too. I love you too. I love you too. I love you too. I love you too. I love you too. I love you too. I love you too. I love you too. I love you too. I love you too. I love you too. I love you too. I love you too. I love you too. I love you too. I love you too. I love you too.",
+ "raw": {
+ "normalized_reference": "i would not say that again in this room if i were you.",
+ "normalized_hypothesis": "i love you. i love you too. i love you too. i love you too. i love you too. i love you too. i love you too. i love you too. i love you too. i love you too. i love you too. i love you too. i love you too. i love you too. i love you too. i love you too. i love you too. i love you too. i love you too. i love you too. i love you too. i love you too. i love you too. i love you too. i love you too. i love you too. i love you too. i love you too. i love you too. i love you too. i love you too. i love you too. i love you too. i love you too. i love you too. i love you too. i love you too. i love you too. i love you too. i love you too. i love you too. i love you too. i love you too. i love you too. i love you too. i love you too. i love you too. i love you too. i love you too. i love you too. i love you too. i love you too. i love you too. i love you too. i love you too. i love you too. i love you too. i love you too. i love you too. i love you too. i love you too. i love you too. i love you too. i love you too. i love you too. i love you too. i love you too. i love you too. i love you too. i love you too. i love you too. i love you too. i love you too. i love you too. i love you too. i love you too. i love you too. i love you too. i love you too. i love you too. i love you too. i love you too. i love you too. i love you too. i love you too. i love you too. i love you too. i love you too. i love you too.",
+ "errors": 352,
+ "words": 13,
+ "wer": 27.076923076923077
+ },
+ "semantic": {
+ "normalized_reference": "i would not say that again in this room if i were you",
+ "normalized_hypothesis": "i love you i love you too i love you too i love you too i love you too i love you too i love you too i love you too i love you too i love you too i love you too i love you too i love you too i love you too i love you too i love you too i love you too i love you too i love you too i love you too i love you too i love you too i love you too i love you too i love you too i love you too i love you too i love you too i love you too i love you too i love you too i love you too i love you too i love you too i love you too i love you too i love you too i love you too i love you too i love you too i love you too i love you too i love you too i love you too i love you too i love you too i love you too i love you too i love you too i love you too i love you too i love you too i love you too i love you too i love you too i love you too i love you too i love you too i love you too i love you too i love you too i love you too i love you too i love you too i love you too i love you too i love you too i love you too i love you too i love you too i love you too i love you too i love you too i love you too i love you too i love you too i love you too i love you too i love you too i love you too i love you too i love you too i love you too i love you too i love you too i love you too i love you too i love you too i love you too",
+ "errors": 352,
+ "words": 13,
+ "wer": 27.076923076923077
+ }
+ },
+ {
+ "id": "release500_0058",
+ "category": "descriptive",
+ "text": "The room was filled with the quiet that means something is about to change.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0058.wav",
+ "render_seconds": 172.1077914451016,
+ "audio_seconds": 4.73687074829932,
+ "rtf": 36.333647378259904,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The world is fake, but the plot of my show is about to change.",
+ "raw": {
+ "normalized_reference": "the room was filled with the quiet that means something is about to change.",
+ "normalized_hypothesis": "the world is fake, but the plot of my show is about to change.",
+ "errors": 8,
+ "words": 14,
+ "wer": 0.5714285714285714
+ },
+ "semantic": {
+ "normalized_reference": "the room was filled with the quiet that means something is about to change",
+ "normalized_hypothesis": "the world is fake but the plot of my show is about to change",
+ "errors": 8,
+ "words": 14,
+ "wer": 0.5714285714285714
+ }
+ },
+ {
+ "id": "release500_0059",
+ "category": "emotional",
+ "text": "I didn't expect to feel that way about it, but I did.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0059.wav",
+ "render_seconds": 150.90618067095056,
+ "audio_seconds": 3.4829931972789114,
+ "rtf": 43.326579216073696,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Well, I'm not supposed to say that you're a believer, but you're a dog.",
+ "raw": {
+ "normalized_reference": "i didn't expect to feel that way about it, but i did.",
+ "normalized_hypothesis": "well, i'm not supposed to say that you're a believer, but you're a dog.",
+ "errors": 11,
+ "words": 12,
+ "wer": 0.9166666666666666
+ },
+ "semantic": {
+ "normalized_reference": "i did not expect to feel that way about it but i did",
+ "normalized_hypothesis": "well i am not supposed to say that you are a believer but you are a dog",
+ "errors": 12,
+ "words": 13,
+ "wer": 0.9230769230769231
+ }
+ },
+ {
+ "id": "release500_0060",
+ "category": "homographs",
+ "text": "I object to the strange object that was left near the entrance.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0060.wav",
+ "render_seconds": 172.6041172950063,
+ "audio_seconds": 4.1099319727891155,
+ "rtf": 41.996830711013516,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Welcome to the squad. Welcome to the Slytherin Encourage.",
+ "raw": {
+ "normalized_reference": "i object to the strange object that was left near the entrance.",
+ "normalized_hypothesis": "welcome to the squad. welcome to the slytherin encourage.",
+ "errors": 10,
+ "words": 12,
+ "wer": 0.8333333333333334
+ },
+ "semantic": {
+ "normalized_reference": "i object to the strange object that was left near the entrance",
+ "normalized_hypothesis": "welcome to the squad welcome to the slytherin encourage",
+ "errors": 10,
+ "words": 12,
+ "wer": 0.8333333333333334
+ }
+ },
+ {
+ "id": "release500_0061",
+ "category": "long",
+ "text": "I've learned more from the things that went wrong than from the things that worked, which is a frustrating way to have to learn things but apparently the only way that stuck.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0061.wav",
+ "render_seconds": 321.70934425399173,
+ "audio_seconds": 10.65795918367347,
+ "rtf": 30.184891751772355,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "As long as you're for the things that work, we'll be for the things that work. There's a fish caught on the line to hoof to your face, but the power of our ear will never stop.",
+ "raw": {
+ "normalized_reference": "i've learned more from the things that went wrong than from the things that worked, which is a frustrating way to have to learn things but apparently the only way that stuck.",
+ "normalized_hypothesis": "as long as you're for the things that work, we'll be for the things that work. there's a fish caught on the line to hoof to your face, but the power of our ear will never stop.",
+ "errors": 28,
+ "words": 32,
+ "wer": 0.875
+ },
+ "semantic": {
+ "normalized_reference": "i have learned more from the things that went wrong than from the things that worked which is a frustrating way to have to learn things but apparently the only way that stuck",
+ "normalized_hypothesis": "as long as you are for the things that work we will be for the things that work there is a fish caught on the line to hoof to your face but the power of our ear will never stop",
+ "errors": 29,
+ "words": 33,
+ "wer": 0.8787878787878788
+ }
+ },
+ {
+ "id": "release500_0062",
+ "category": "long_range",
+ "text": "Because every nearby hotel was full, the exhausted hikers followed the narrow road until they found a farmhouse with one empty room.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0062.wav",
+ "render_seconds": 193.3061724229483,
+ "audio_seconds": 8.916462585034013,
+ "rtf": 21.679693104685516,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "But first of all, I will tell you the truth. I choose to help you through the long walk into the third floor house without a wall.",
+ "raw": {
+ "normalized_reference": "because every nearby hotel was full, the exhausted hikers followed the narrow road until they found a farmhouse with one empty room.",
+ "normalized_hypothesis": "but first of all, i will tell you the truth. i choose to help you through the long walk into the third floor house without a wall.",
+ "errors": 26,
+ "words": 22,
+ "wer": 1.1818181818181819
+ },
+ "semantic": {
+ "normalized_reference": "because every nearby hotel was full the exhausted hikers followed the narrow road until they found a farmhouse with one empty room",
+ "normalized_hypothesis": "but 1st of all i will tell you the truth i choose to help you through the long walk into the 3rd floor house without a wall",
+ "errors": 26,
+ "words": 22,
+ "wer": 1.1818181818181819
+ }
+ },
+ {
+ "id": "release500_0063",
+ "category": "names_places",
+ "text": "The itinerary sends Beatrice from Albuquerque to Des Moines and then to Boise.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0063.wav",
+ "render_seconds": 180.29913160100114,
+ "audio_seconds": 5.363809523809524,
+ "rtf": 33.6140071344196,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I can't allow ourselves back for so long because I can't do this. I can't do this.",
+ "raw": {
+ "normalized_reference": "the itinerary sends beatrice from albuquerque to des moines and then to boise.",
+ "normalized_hypothesis": "i can't allow ourselves back for so long because i can't do this. i can't do this.",
+ "errors": 17,
+ "words": 13,
+ "wer": 1.3076923076923077
+ },
+ "semantic": {
+ "normalized_reference": "the itinerary sends beatrice from albuquerque to des moines and then to boise",
+ "normalized_hypothesis": "i can not allow ourselves back for so long because i can not do this i can not do this",
+ "errors": 20,
+ "words": 13,
+ "wer": 1.5384615384615385
+ }
+ },
+ {
+ "id": "release500_0064",
+ "category": "narrative",
+ "text": "We had been driving for six hours when we finally saw the sign for the town.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0064.wav",
+ "render_seconds": 169.9961345680058,
+ "audio_seconds": 5.085170068027211,
+ "rtf": 33.42978352618907,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Thank you for watching. Please like, share, and subscribe.",
+ "raw": {
+ "normalized_reference": "we had been driving for six hours when we finally saw the sign for the town.",
+ "normalized_hypothesis": "thank you for watching. please like, share, and subscribe.",
+ "errors": 15,
+ "words": 16,
+ "wer": 0.9375
+ },
+ "semantic": {
+ "normalized_reference": "we had been driving for 6 hours when we finally saw the sign for the town",
+ "normalized_hypothesis": "thank you for watching please like share and subscribe",
+ "errors": 15,
+ "words": 16,
+ "wer": 0.9375
+ }
+ },
+ {
+ "id": "release500_0065",
+ "category": "numbers",
+ "text": "Call extension zero four nine after eight thirty in the morning.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0065.wav",
+ "render_seconds": 179.89916955307126,
+ "audio_seconds": 4.458231292517007,
+ "rtf": 40.35213916671978,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The extension shows you where else to act, so I do.",
+ "raw": {
+ "normalized_reference": "call extension zero four nine after eight thirty in the morning.",
+ "normalized_hypothesis": "the extension shows you where else to act, so i do.",
+ "errors": 10,
+ "words": 11,
+ "wer": 0.9090909090909091
+ },
+ "semantic": {
+ "normalized_reference": "call extension 049 after 830 in the morning",
+ "normalized_hypothesis": "the extension shows you where else to act so i do",
+ "errors": 10,
+ "words": 8,
+ "wer": 1.25
+ }
+ },
+ {
+ "id": "release500_0066",
+ "category": "punctuation",
+ "text": "Stop - listen carefully - then answer the question without guessing.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0066.wav",
+ "render_seconds": 177.09646569692995,
+ "audio_seconds": 4.597551020408163,
+ "rtf": 38.51973907648068,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Stoop, listen closely. Arise, for the years come the day is come.",
+ "raw": {
+ "normalized_reference": "stop - listen carefully - then answer the question without guessing.",
+ "normalized_hypothesis": "stoop, listen closely. arise, for the years come the day is come.",
+ "errors": 10,
+ "words": 9,
+ "wer": 1.1111111111111112
+ },
+ "semantic": {
+ "normalized_reference": "stop listen carefully then answer the question without guessing",
+ "normalized_hypothesis": "stoop listen closely arise for the years come the day is come",
+ "errors": 10,
+ "words": 9,
+ "wer": 1.1111111111111112
+ }
+ },
+ {
+ "id": "release500_0067",
+ "category": "question",
+ "text": "Does it matter at this point, honestly?",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0067.wav",
+ "render_seconds": 122.09159150102641,
+ "audio_seconds": 2.716734693877551,
+ "rtf": 44.94056478027564,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "................................................................................................................................................................................................................................................................................................................................................................",
+ "raw": {
+ "normalized_reference": "does it matter at this point, honestly?",
+ "normalized_hypothesis": "................................................................................................................................................................................................................................................................................................................................................................",
+ "errors": 7,
+ "words": 7,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "does it matter at this point honestly",
+ "normalized_hypothesis": " .",
+ "errors": 7,
+ "words": 7,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0068",
+ "category": "rare_words",
+ "text": "An entrepreneurial meteorologist described the unprecedented atmospheric phenomenon.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0068.wav",
+ "render_seconds": 179.09063885593787,
+ "audio_seconds": 5.851428571428571,
+ "rtf": 30.60631035135657,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Thank you for watching. I hope you enjoyed this video.",
+ "raw": {
+ "normalized_reference": "an entrepreneurial meteorologist described the unprecedented atmospheric phenomenon.",
+ "normalized_hypothesis": "thank you for watching. i hope you enjoyed this video.",
+ "errors": 10,
+ "words": 8,
+ "wer": 1.25
+ },
+ "semantic": {
+ "normalized_reference": "an entrepreneurial meteorologist described the unprecedented atmospheric phenomenon",
+ "normalized_hypothesis": "thank you for watching i hope you enjoyed this video",
+ "errors": 10,
+ "words": 8,
+ "wer": 1.25
+ }
+ },
+ {
+ "id": "release500_0069",
+ "category": "short",
+ "text": "That's not what I expected.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0069.wav",
+ "render_seconds": 94.20020881399978,
+ "audio_seconds": 2.089795918367347,
+ "rtf": 45.076271795761606,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "That's a relax factor.",
+ "raw": {
+ "normalized_reference": "that's not what i expected.",
+ "normalized_hypothesis": "that's a relax factor.",
+ "errors": 4,
+ "words": 5,
+ "wer": 0.8
+ },
+ "semantic": {
+ "normalized_reference": "that is not what i expected",
+ "normalized_hypothesis": "that is a relax factor",
+ "errors": 4,
+ "words": 6,
+ "wer": 0.6666666666666666
+ }
+ },
+ {
+ "id": "release500_0070",
+ "category": "technical",
+ "text": "There's a known issue with that version that causes failures under load.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0070.wav",
+ "render_seconds": 192.00061726605054,
+ "audio_seconds": 4.806530612244898,
+ "rtf": 39.945780596282596,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I don't know what I'm talking about. I don't know what I'm talking about. I don't know what I'm talking about. I don't know what I'm talking about. I don't know what I'm talking about. I don't know what I'm talking about. I don't know what I'm talking about. I don't know what I'm talking about. I don't know what I'm talking about. I don't know what I'm talking about. I don't know what I'm talking about. I don't know what I'm talking about. I don't know what I'm talking about. I don't know what I'm talking about. I don't know what I'm talking about. I don't know what I'm talking about. I don't know what I'm talking about. I don't know what I'm talking about. I don't know what I'm talking about. I don't know what I'm talking about. I don't know what I'm talking about. I don't know what I'm talking about. I don't know what I'm talking about. I don't know what I'm talking about. I don't know what I'm talking about. I don't know what I'm talking about. I don't know what I'm talking about. I don't know what I'm talking about. I don't know what I'm talking about. I don't know what I'm talking about. I don't know what I'm talking about. I don't know what I'm talking about. I don't know what I'm talking about. I don't know what I'm talking about. I don't know what I'm talking about. I don't know what I'm talking about. I don't know what I'm talking about. I don't know what I'm talking about. I don't know what I'm talking about. I don't know what I'm talking about. I don't know what I'm talking about. I don't know what I'm talking about. I don't know what I'm talking about. I don't know what I'm talking about. I don't know",
+ "raw": {
+ "normalized_reference": "there's a known issue with that version that causes failures under load.",
+ "normalized_hypothesis": "i don't know what i'm talking about. i don't know what i'm talking about. i don't know what i'm talking about. i don't know what i'm talking about. i don't know what i'm talking about. i don't know what i'm talking about. i don't know what i'm talking about. i don't know what i'm talking about. i don't know what i'm talking about. i don't know what i'm talking about. i don't know what i'm talking about. i don't know what i'm talking about. i don't know what i'm talking about. i don't know what i'm talking about. i don't know what i'm talking about. i don't know what i'm talking about. i don't know what i'm talking about. i don't know what i'm talking about. i don't know what i'm talking about. i don't know what i'm talking about. i don't know what i'm talking about. i don't know what i'm talking about. i don't know what i'm talking about. i don't know what i'm talking about. i don't know what i'm talking about. i don't know what i'm talking about. i don't know what i'm talking about. i don't know what i'm talking about. i don't know what i'm talking about. i don't know what i'm talking about. i don't know what i'm talking about. i don't know what i'm talking about. i don't know what i'm talking about. i don't know what i'm talking about. i don't know what i'm talking about. i don't know what i'm talking about. i don't know what i'm talking about. i don't know what i'm talking about. i don't know what i'm talking about. i don't know what i'm talking about. i don't know what i'm talking about. i don't know what i'm talking about. i don't know what i'm talking about. i don't know what i'm talking about. i don't know",
+ "errors": 311,
+ "words": 12,
+ "wer": 25.916666666666668
+ },
+ "semantic": {
+ "normalized_reference": "there is a known issue with that version that causes failures under load",
+ "normalized_hypothesis": "i do not know what i am talking about i do not know what i am talking about i do not know what i am talking about i do not know what i am talking about i do not know what i am talking about i do not know what i am talking about i do not know what i am talking about i do not know what i am talking about i do not know what i am talking about i do not know what i am talking about i do not know what i am talking about i do not know what i am talking about i do not know what i am talking about i do not know what i am talking about i do not know what i am talking about i do not know what i am talking about i do not know what i am talking about i do not know what i am talking about i do not know what i am talking about i do not know what i am talking about i do not know what i am talking about i do not know what i am talking about i do not know what i am talking about i do not know what i am talking about i do not know what i am talking about i do not know what i am talking about i do not know what i am talking about i do not know what i am talking about i do not know what i am talking about i do not know what i am talking about i do not know what i am talking about i do not know what i am talking about i do not know what i am talking about i do not know what i am talking about i do not know what i am talking about i do not know what i am talking about i do not know what i am talking about i do not know what i am talking about i do not know what i am talking about i do not know what i am talking about i do not know what i am talking about i do not know what i am talking about i do not know what i am talking about i do not know what i am talking about i do not know",
+ "errors": 400,
+ "words": 13,
+ "wer": 30.76923076923077
+ }
+ },
+ {
+ "id": "release500_0071",
+ "category": "conversational",
+ "text": "Don't look now but I think that's him over there.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0071.wav",
+ "render_seconds": 125.60043001594022,
+ "audio_seconds": 2.8560544217687074,
+ "rtf": 43.97690361171688,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Oh, yes. Oh, yes. Oh, yes. Oh, yes. Oh, yes. Oh, yes. Oh, yes. Oh, yes. Oh, yes. Oh, yes. Oh, yes. Oh, yes. Oh, yes. Oh, yes. Oh, yes. Oh, yes. Oh, yes. Oh, yes. Oh, yes. Oh, yes. Oh, yes. Oh, yes. Oh, yes. Oh, yes. Oh, yes. Oh, yes. Oh, yes. Oh, yes. Oh, yes. Oh, yes. Oh, yes. Oh, yes. Oh, yes. Oh, yes. Oh, yes. Oh, yes. Oh, yes. Oh, yes. Oh, yes. Oh, yes. Oh, yes. Oh, yes. Oh, yes. Oh, yes. Oh, yes. Oh, yes. Oh, yes. Oh, yes. Oh, yes. Oh, yes. Oh, yes. Oh, yes. Oh, yes. Oh, yes. Oh, yes. Oh, yes. Oh, yes. Oh, yes. Oh, yes. Oh, yes. Oh, yes. Oh, yes. Oh, yes. Oh, yes. Oh, yes. Oh, yes. Oh, yes. Oh, yes. Oh, yes. Oh, yes. Oh, yes. Oh, yes. Oh, yes. Oh, yes. Oh, yes. Oh, yes. Oh, yes. Oh, yes. Oh, yes. Oh, yes. Oh, yes. Oh, yes. Oh, yes. Oh, yes. Oh, yes. Oh, yes. Oh, yes. Oh, yes. Oh, yes. Oh, yes. Oh, yes. Oh, yes. Oh, yes. Oh, yes. Oh, yes. Oh, yes. Oh, yes. Oh, yes. Oh, yes. Oh, yes. Oh, yes. Oh, yes. Oh, yes. Oh, yes. Oh, yes. Oh, yes. Oh, yes. Oh, yes. Oh, yes. Oh, yes. Oh, yes.",
+ "raw": {
+ "normalized_reference": "don't look now but i think that's him over there.",
+ "normalized_hypothesis": "oh, yes. oh, yes. oh, yes. oh, yes. oh, yes. oh, yes. oh, yes. oh, yes. oh, yes. oh, yes. oh, yes. oh, yes. oh, yes. oh, yes. oh, yes. oh, yes. oh, yes. oh, yes. oh, yes. oh, yes. oh, yes. oh, yes. oh, yes. oh, yes. oh, yes. oh, yes. oh, yes. oh, yes. oh, yes. oh, yes. oh, yes. oh, yes. oh, yes. oh, yes. oh, yes. oh, yes. oh, yes. oh, yes. oh, yes. oh, yes. oh, yes. oh, yes. oh, yes. oh, yes. oh, yes. oh, yes. oh, yes. oh, yes. oh, yes. oh, yes. oh, yes. oh, yes. oh, yes. oh, yes. oh, yes. oh, yes. oh, yes. oh, yes. oh, yes. oh, yes. oh, yes. oh, yes. oh, yes. oh, yes. oh, yes. oh, yes. oh, yes. oh, yes. oh, yes. oh, yes. oh, yes. oh, yes. oh, yes. oh, yes. oh, yes. oh, yes. oh, yes. oh, yes. oh, yes. oh, yes. oh, yes. oh, yes. oh, yes. oh, yes. oh, yes. oh, yes. oh, yes. oh, yes. oh, yes. oh, yes. oh, yes. oh, yes. oh, yes. oh, yes. oh, yes. oh, yes. oh, yes. oh, yes. oh, yes. oh, yes. oh, yes. oh, yes. oh, yes. oh, yes. oh, yes. oh, yes. oh, yes. oh, yes. oh, yes. oh, yes. oh, yes.",
+ "errors": 222,
+ "words": 10,
+ "wer": 22.2
+ },
+ "semantic": {
+ "normalized_reference": "do not look now but i think that is him over there",
+ "normalized_hypothesis": "0 yes 0 yes 0 yes 0 yes 0 yes 0 yes 0 yes 0 yes 0 yes 0 yes 0 yes 0 yes 0 yes 0 yes 0 yes 0 yes 0 yes 0 yes 0 yes 0 yes 0 yes 0 yes 0 yes 0 yes 0 yes 0 yes 0 yes 0 yes 0 yes 0 yes 0 yes 0 yes 0 yes 0 yes 0 yes 0 yes 0 yes 0 yes 0 yes 0 yes 0 yes 0 yes 0 yes 0 yes 0 yes 0 yes 0 yes 0 yes 0 yes 0 yes 0 yes 0 yes 0 yes 0 yes 0 yes 0 yes 0 yes 0 yes 0 yes 0 yes 0 yes 0 yes 0 yes 0 yes 0 yes 0 yes 0 yes 0 yes 0 yes 0 yes 0 yes 0 yes 0 yes 0 yes 0 yes 0 yes 0 yes 0 yes 0 yes 0 yes 0 yes 0 yes 0 yes 0 yes 0 yes 0 yes 0 yes 0 yes 0 yes 0 yes 0 yes 0 yes 0 yes 0 yes 0 yes 0 yes 0 yes 0 yes 0 yes 0 yes 0 yes 0 yes 0 yes 0 yes 0 yes 0 yes 0 yes 0 yes 0 yes 0 yes 0 yes",
+ "errors": 222,
+ "words": 12,
+ "wer": 18.5
+ }
+ },
+ {
+ "id": "release500_0072",
+ "category": "descriptive",
+ "text": "The city at night looks completely different from how it feels during the day.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0072.wav",
+ "render_seconds": 205.99246078100987,
+ "audio_seconds": 5.154829931972789,
+ "rtf": 39.96105856050524,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The shadow looks completely different for her face color.",
+ "raw": {
+ "normalized_reference": "the city at night looks completely different from how it feels during the day.",
+ "normalized_hypothesis": "the shadow looks completely different for her face color.",
+ "errors": 10,
+ "words": 14,
+ "wer": 0.7142857142857143
+ },
+ "semantic": {
+ "normalized_reference": "the city at night looks completely different from how it feels during the day",
+ "normalized_hypothesis": "the shadow looks completely different for her face color",
+ "errors": 10,
+ "words": 14,
+ "wer": 0.7142857142857143
+ }
+ },
+ {
+ "id": "release500_0073",
+ "category": "emotional",
+ "text": "It's the kind of happy that makes you nervous because you know it can't last.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0073.wav",
+ "render_seconds": 182.49806611204986,
+ "audio_seconds": 4.876190476190477,
+ "rtf": 37.426361214385224,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "That's the kind of help I would like to know from you. I can't help you. I can't help you. I can't help you. I can't help you. I can't help you. I can't help you. I can't help you. I can't help you. I can't help you. I can't help you. I can't help you. I can't help you. I can't help you. I can't help you. I can't help you. I can't help you. I can't help you. I can't help you. I can't help you. I can't help you. I can't help you. I can't help you. I can't help you. I can't help you. I can't help you. I can't help you. I can't help you. I can't help you. I can't help you. I can't help you. I can't help you. I can't help you. I can't help you. I can't help you. I can't help you. I can't help you. I can't help you. I can't help you. I can't help you. I can't help you. I can't help you. I can't help you. I can't help you. I can't help you. I can't help you. I can't help you. I can't help you. I can't help you. I can't help you. I can't help you. I can't help you. I can't help you. I can't help you. I can't help you. I can't help you. I can't help you. I can't help you. I can't help you. I can't help you. I can't help you. I can't help you. I can't help you. I can't help you. I can't help you. I can't help you. I can't help you. I can't help you. I can't help you. I can't help you. I can't help you. I can't help you. I can't you",
+ "raw": {
+ "normalized_reference": "it's the kind of happy that makes you nervous because you know it can't last.",
+ "normalized_hypothesis": "that's the kind of help i would like to know from you. i can't help you. i can't help you. i can't help you. i can't help you. i can't help you. i can't help you. i can't help you. i can't help you. i can't help you. i can't help you. i can't help you. i can't help you. i can't help you. i can't help you. i can't help you. i can't help you. i can't help you. i can't help you. i can't help you. i can't help you. i can't help you. i can't help you. i can't help you. i can't help you. i can't help you. i can't help you. i can't help you. i can't help you. i can't help you. i can't help you. i can't help you. i can't help you. i can't help you. i can't help you. i can't help you. i can't help you. i can't help you. i can't help you. i can't help you. i can't help you. i can't help you. i can't help you. i can't help you. i can't help you. i can't help you. i can't help you. i can't help you. i can't help you. i can't help you. i can't help you. i can't help you. i can't help you. i can't help you. i can't help you. i can't help you. i can't help you. i can't help you. i can't help you. i can't help you. i can't help you. i can't help you. i can't help you. i can't help you. i can't help you. i can't help you. i can't help you. i can't help you. i can't help you. i can't help you. i can't help you. i can't help you. i can't you",
+ "errors": 293,
+ "words": 15,
+ "wer": 19.533333333333335
+ },
+ "semantic": {
+ "normalized_reference": "it is the kind of happy that makes you nervous because you know it can not last",
+ "normalized_hypothesis": "that is the kind of help i would like to know from you i can not help you i can not help you i can not help you i can not help you i can not help you i can not help you i can not help you i can not help you i can not help you i can not help you i can not help you i can not help you i can not help you i can not help you i can not help you i can not help you i can not help you i can not help you i can not help you i can not help you i can not help you i can not help you i can not help you i can not help you i can not help you i can not help you i can not help you i can not help you i can not help you i can not help you i can not help you i can not help you i can not help you i can not help you i can not help you i can not help you i can not help you i can not help you i can not help you i can not help you i can not help you i can not help you i can not help you i can not help you i can not help you i can not help you i can not help you i can not help you i can not help you i can not help you i can not help you i can not help you i can not help you i can not help you i can not help you i can not help you i can not help you i can not help you i can not help you i can not help you i can not help you i can not help you i can not help you i can not help you i can not help you i can not help you i can not help you i can not help you i can not help you i can not help you i can not help you i can not you",
+ "errors": 364,
+ "words": 17,
+ "wer": 21.41176470588235
+ }
+ },
+ {
+ "id": "release500_0074",
+ "category": "homographs",
+ "text": "The dove dove beneath the branch while the wind began to wind the loose cable.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0074.wav",
+ "render_seconds": 163.8029365990078,
+ "audio_seconds": 5.085170068027211,
+ "rtf": 32.21188955486696,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "But both, both the last of wands were the ones that had to run the most happy.",
+ "raw": {
+ "normalized_reference": "the dove dove beneath the branch while the wind began to wind the loose cable.",
+ "normalized_hypothesis": "but both, both the last of wands were the ones that had to run the most happy.",
+ "errors": 14,
+ "words": 15,
+ "wer": 0.9333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "the dove dove beneath the branch while the wind began to wind the loose cable",
+ "normalized_hypothesis": "but both both the last of wands were the ones that had to run the most happy",
+ "errors": 14,
+ "words": 15,
+ "wer": 0.9333333333333333
+ }
+ },
+ {
+ "id": "release500_0075",
+ "category": "long",
+ "text": "I've thought a lot about what I would do differently if I could go back, and the honest answer is probably not as much as I'd like to think.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0075.wav",
+ "render_seconds": 267.306908751023,
+ "audio_seconds": 8.70748299319728,
+ "rtf": 30.698527801875294,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I saw a year ago, when I went to California, I saw an old dog, and I knew this dog saw this problem, and I know this works, and I do like this dog.",
+ "raw": {
+ "normalized_reference": "i've thought a lot about what i would do differently if i could go back, and the honest answer is probably not as much as i'd like to think.",
+ "normalized_hypothesis": "i saw a year ago, when i went to california, i saw an old dog, and i knew this dog saw this problem, and i know this works, and i do like this dog.",
+ "errors": 30,
+ "words": 29,
+ "wer": 1.0344827586206897
+ },
+ "semantic": {
+ "normalized_reference": "i have thought a lot about what i would do differently if i could go back and the honest answer is probably not as much as i would like to think",
+ "normalized_hypothesis": "i saw a year ago when i went to california i saw an old dog and i knew this dog saw this problem and i know this works and i do like this dog",
+ "errors": 29,
+ "words": 31,
+ "wer": 0.9354838709677419
+ }
+ },
+ {
+ "id": "release500_0076",
+ "category": "long_range",
+ "text": "Although the weather changed without warning, the musicians finished packing before anyone opened the enormous glass door.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0076.wav",
+ "render_seconds": 250.19262322608847,
+ "audio_seconds": 7.801904761904762,
+ "rtf": 32.06814628752355,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Either the weather changed the same way, or the seasons fell apart and the floor on the roof opened and the walls closed.",
+ "raw": {
+ "normalized_reference": "although the weather changed without warning, the musicians finished packing before anyone opened the enormous glass door.",
+ "normalized_hypothesis": "either the weather changed the same way, or the seasons fell apart and the floor on the roof opened and the walls closed.",
+ "errors": 18,
+ "words": 17,
+ "wer": 1.0588235294117647
+ },
+ "semantic": {
+ "normalized_reference": "although the weather changed without warning the musicians finished packing before anyone opened the enormous glass door",
+ "normalized_hypothesis": "either the weather changed the same way or the seasons fell apart and the floor on the roof opened and the walls closed",
+ "errors": 18,
+ "words": 17,
+ "wer": 1.0588235294117647
+ }
+ },
+ {
+ "id": "release500_0077",
+ "category": "names_places",
+ "text": "Saskatchewan looked unusually bright beneath the pale February sunrise.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0077.wav",
+ "render_seconds": 190.80327391496394,
+ "audio_seconds": 5.154829931972789,
+ "rtf": 37.01446535248588,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Just actually looking at who I bought the last both of our shoes.",
+ "raw": {
+ "normalized_reference": "saskatchewan looked unusually bright beneath the pale february sunrise.",
+ "normalized_hypothesis": "just actually looking at who i bought the last both of our shoes.",
+ "errors": 12,
+ "words": 9,
+ "wer": 1.3333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "saskatchewan looked unusually bright beneath the pale february sunrise",
+ "normalized_hypothesis": "just actually looking at who i bought the last both of our shoes",
+ "errors": 12,
+ "words": 9,
+ "wer": 1.3333333333333333
+ }
+ },
+ {
+ "id": "release500_0078",
+ "category": "narrative",
+ "text": "She handled it with a grace I genuinely did not expect.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0078.wav",
+ "render_seconds": 158.5992601349717,
+ "audio_seconds": 3.900952380952381,
+ "rtf": 40.656548618584054,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I'll be done with the class. I'm trying to learn how to do that sport.",
+ "raw": {
+ "normalized_reference": "she handled it with a grace i genuinely did not expect.",
+ "normalized_hypothesis": "i'll be done with the class. i'm trying to learn how to do that sport.",
+ "errors": 14,
+ "words": 11,
+ "wer": 1.2727272727272727
+ },
+ "semantic": {
+ "normalized_reference": "she handled it with a grace i genuinely did not expect",
+ "normalized_hypothesis": "i will be done with the class i am trying to learn how to do that sport",
+ "errors": 15,
+ "words": 11,
+ "wer": 1.3636363636363635
+ }
+ },
+ {
+ "id": "release500_0079",
+ "category": "numbers",
+ "text": "Apartment four hundred seven is across from room twenty-nine on the third floor.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0079.wav",
+ "render_seconds": 186.9999477979727,
+ "audio_seconds": 5.642448979591837,
+ "rtf": 33.141628479820106,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The clock likes to look at something as a push for only 20 minutes or so.",
+ "raw": {
+ "normalized_reference": "apartment four hundred seven is across from room twenty-nine on the third floor.",
+ "normalized_hypothesis": "the clock likes to look at something as a push for only 20 minutes or so.",
+ "errors": 16,
+ "words": 14,
+ "wer": 1.1428571428571428
+ },
+ "semantic": {
+ "normalized_reference": "apartment 407 is across from room 29 on the 3rd floor",
+ "normalized_hypothesis": "the clock likes to look at something as a push for only 20 minutes or so",
+ "errors": 16,
+ "words": 11,
+ "wer": 1.4545454545454546
+ }
+ },
+ {
+ "id": "release500_0080",
+ "category": "punctuation",
+ "text": "\"Not yet,\" Rowan whispered. \"Give me another minute... please.\"",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0080.wav",
+ "render_seconds": 132.82155725499615,
+ "audio_seconds": 5.50312925170068,
+ "rtf": 24.135641955701683,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I'm out. I'm out of here. I'm out of here. I'm out of here. I'm out of here. I'm out of here. I'm out of here. I'm out of here. I'm out of here. I'm out of here. I'm out of here. I'm out of here. I'm out of here. I'm out of here. I'm out of here. I'm out of here. I'm out of here. I'm out of here. I'm out of here. I'm out of here. I'm out of here. I'm out of here. I'm out of here. I'm out of here. I'm out of here. I'm out of here. I'm out of here. I'm out of here. I'm out of here. I'm out of here. I'm out of here. I'm out of here. I'm out of here. I'm out of here. I'm out of here. I'm out of here. I'm out of here. I'm out of here. I'm out of here. I'm out of here. I'm out of here. I'm out of here. I'm out of here. I'm out of here. I'm out of here. I'm out of here. I'm out of here. I'm out of here. I'm out of here. I'm out of here. I'm out of here. I'm out of here. I'm out of here. I'm out of here. I'm out of here. I'm out of here. I'm out of here. I'm out of here. I'm out of here. I'm out of here. I'm out of here. I'm out of here. I'm out of here. I'm out of here. I'm out of here. I'm out of here. I'm out of here. I'm out of here. I'm out of here. I'm out of here. I'm out of here. I'm out of here. I'm out of here. I'm out of here. I'm",
+ "raw": {
+ "normalized_reference": "\"not yet,\" rowan whispered. \"give me another minute... please.\"",
+ "normalized_hypothesis": "i'm out. i'm out of here. i'm out of here. i'm out of here. i'm out of here. i'm out of here. i'm out of here. i'm out of here. i'm out of here. i'm out of here. i'm out of here. i'm out of here. i'm out of here. i'm out of here. i'm out of here. i'm out of here. i'm out of here. i'm out of here. i'm out of here. i'm out of here. i'm out of here. i'm out of here. i'm out of here. i'm out of here. i'm out of here. i'm out of here. i'm out of here. i'm out of here. i'm out of here. i'm out of here. i'm out of here. i'm out of here. i'm out of here. i'm out of here. i'm out of here. i'm out of here. i'm out of here. i'm out of here. i'm out of here. i'm out of here. i'm out of here. i'm out of here. i'm out of here. i'm out of here. i'm out of here. i'm out of here. i'm out of here. i'm out of here. i'm out of here. i'm out of here. i'm out of here. i'm out of here. i'm out of here. i'm out of here. i'm out of here. i'm out of here. i'm out of here. i'm out of here. i'm out of here. i'm out of here. i'm out of here. i'm out of here. i'm out of here. i'm out of here. i'm out of here. i'm out of here. i'm out of here. i'm out of here. i'm out of here. i'm out of here. i'm out of here. i'm out of here. i'm out of here. i'm out of here. i'm",
+ "errors": 295,
+ "words": 9,
+ "wer": 32.77777777777778
+ },
+ "semantic": {
+ "normalized_reference": "not yet rowan whispered give me another minute please",
+ "normalized_hypothesis": "i am out i am out of here i am out of here i am out of here i am out of here i am out of here i am out of here i am out of here i am out of here i am out of here i am out of here i am out of here i am out of here i am out of here i am out of here i am out of here i am out of here i am out of here i am out of here i am out of here i am out of here i am out of here i am out of here i am out of here i am out of here i am out of here i am out of here i am out of here i am out of here i am out of here i am out of here i am out of here i am out of here i am out of here i am out of here i am out of here i am out of here i am out of here i am out of here i am out of here i am out of here i am out of here i am out of here i am out of here i am out of here i am out of here i am out of here i am out of here i am out of here i am out of here i am out of here i am out of here i am out of here i am out of here i am out of here i am out of here i am out of here i am out of here i am out of here i am out of here i am out of here i am out of here i am out of here i am out of here i am out of here i am out of here i am out of here i am out of here i am out of here i am out of here i am out of here i am out of here i am out of here i am out of here i am",
+ "errors": 370,
+ "words": 9,
+ "wer": 41.111111111111114
+ }
+ },
+ {
+ "id": "release500_0081",
+ "category": "question",
+ "text": "What are we actually trying to accomplish here?",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0081.wav",
+ "render_seconds": 103.99980497697834,
+ "audio_seconds": 3.274013605442177,
+ "rtf": 31.76523298623632,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "You're a lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, l",
+ "raw": {
+ "normalized_reference": "what are we actually trying to accomplish here?",
+ "normalized_hypothesis": "you're a lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, lousy, l",
+ "errors": 113,
+ "words": 8,
+ "wer": 14.125
+ },
+ "semantic": {
+ "normalized_reference": "what are we actually trying to accomplish here",
+ "normalized_hypothesis": "you are a lousy lousy lousy lousy lousy lousy lousy lousy lousy lousy lousy lousy lousy lousy lousy lousy lousy lousy lousy lousy lousy lousy lousy lousy lousy lousy lousy lousy lousy lousy lousy lousy lousy lousy lousy lousy lousy lousy lousy lousy lousy lousy lousy lousy lousy lousy lousy lousy lousy lousy lousy lousy lousy lousy lousy lousy lousy lousy lousy lousy lousy lousy lousy lousy lousy lousy lousy lousy lousy lousy lousy lousy lousy lousy lousy lousy lousy lousy lousy lousy lousy lousy lousy lousy lousy lousy lousy lousy lousy lousy lousy lousy lousy lousy lousy lousy lousy lousy lousy lousy lousy lousy lousy lousy lousy lousy lousy lousy lousy lousy l",
+ "errors": 113,
+ "words": 8,
+ "wer": 14.125
+ }
+ },
+ {
+ "id": "release500_0082",
+ "category": "rare_words",
+ "text": "The physician recommended a noninvasive procedure for the recurring neurological symptoms.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0082.wav",
+ "render_seconds": 227.49787733901758,
+ "audio_seconds": 6.408707482993197,
+ "rtf": 35.498246400343476,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Thank you for watching. If you like this video, please subscribe to my channel. Thank you for watching. If you like this video, please subscribe to my channel. Thank you for watching. If you like this video, please subscribe to my channel. Thank you for watching. If you like this video, please subscribe to my channel. Thank you for watching. If you like this video, please subscribe to my channel. Thank you for watching. If you like this video, please subscribe to my channel. Thank you for watching. If you like this video, please subscribe to my channel. Thank you for watching. If you like this video, please subscribe to my channel. Thank you for watching. If you like this video, please subscribe to my channel. Thank you for watching. If you like this video, please subscribe to my channel. Thank you for watching. If you like this video, please subscribe to my channel. Thank you for watching. If you like this video, please subscribe to my channel. Thank you for watching. If you like this video, please subscribe to my channel. Thank you for watching. If you like this video, please subscribe to my channel. Thank you for watching. If you like this video, please subscribe to my channel. Thank you for watching. If you like this video, please subscribe to my channel. Thank you for watching. If you like this video, please subscribe to my channel. Thank you for watching. If you like this video, please subscribe to my channel. Thank you for watching. If you like this video, please subscribe to my channel. Thank you for watching. If you like this video, please subscribe to my channel. Thank you for watching. If you like this video, please subscribe to my channel. Thank you for watching. If you like this video, please subscribe to my channel. Thank you for watching. If you like this video, please subscribe to my channel. Thank you for watching. If you like this video, please subscribe to my channel. Thank you for watching. If you like this video, please subscribe to my channel. Thank you for watching. If you like this video, please subscribe to my channel. Thank you",
+ "raw": {
+ "normalized_reference": "the physician recommended a noninvasive procedure for the recurring neurological symptoms.",
+ "normalized_hypothesis": "thank you for watching. if you like this video, please subscribe to my channel. thank you for watching. if you like this video, please subscribe to my channel. thank you for watching. if you like this video, please subscribe to my channel. thank you for watching. if you like this video, please subscribe to my channel. thank you for watching. if you like this video, please subscribe to my channel. thank you for watching. if you like this video, please subscribe to my channel. thank you for watching. if you like this video, please subscribe to my channel. thank you for watching. if you like this video, please subscribe to my channel. thank you for watching. if you like this video, please subscribe to my channel. thank you for watching. if you like this video, please subscribe to my channel. thank you for watching. if you like this video, please subscribe to my channel. thank you for watching. if you like this video, please subscribe to my channel. thank you for watching. if you like this video, please subscribe to my channel. thank you for watching. if you like this video, please subscribe to my channel. thank you for watching. if you like this video, please subscribe to my channel. thank you for watching. if you like this video, please subscribe to my channel. thank you for watching. if you like this video, please subscribe to my channel. thank you for watching. if you like this video, please subscribe to my channel. thank you for watching. if you like this video, please subscribe to my channel. thank you for watching. if you like this video, please subscribe to my channel. thank you for watching. if you like this video, please subscribe to my channel. thank you for watching. if you like this video, please subscribe to my channel. thank you for watching. if you like this video, please subscribe to my channel. thank you for watching. if you like this video, please subscribe to my channel. thank you for watching. if you like this video, please subscribe to my channel. thank you for watching. if you like this video, please subscribe to my channel. thank you",
+ "errors": 365,
+ "words": 11,
+ "wer": 33.18181818181818
+ },
+ "semantic": {
+ "normalized_reference": "the physician recommended a noninvasive procedure for the recurring neurological symptoms",
+ "normalized_hypothesis": "thank you for watching if you like this video please subscribe to my channel thank you for watching if you like this video please subscribe to my channel thank you for watching if you like this video please subscribe to my channel thank you for watching if you like this video please subscribe to my channel thank you for watching if you like this video please subscribe to my channel thank you for watching if you like this video please subscribe to my channel thank you for watching if you like this video please subscribe to my channel thank you for watching if you like this video please subscribe to my channel thank you for watching if you like this video please subscribe to my channel thank you for watching if you like this video please subscribe to my channel thank you for watching if you like this video please subscribe to my channel thank you for watching if you like this video please subscribe to my channel thank you for watching if you like this video please subscribe to my channel thank you for watching if you like this video please subscribe to my channel thank you for watching if you like this video please subscribe to my channel thank you for watching if you like this video please subscribe to my channel thank you for watching if you like this video please subscribe to my channel thank you for watching if you like this video please subscribe to my channel thank you for watching if you like this video please subscribe to my channel thank you for watching if you like this video please subscribe to my channel thank you for watching if you like this video please subscribe to my channel thank you for watching if you like this video please subscribe to my channel thank you for watching if you like this video please subscribe to my channel thank you for watching if you like this video please subscribe to my channel thank you for watching if you like this video please subscribe to my channel thank you for watching if you like this video please subscribe to my channel thank you",
+ "errors": 365,
+ "words": 11,
+ "wer": 33.18181818181818
+ }
+ },
+ {
+ "id": "release500_0083",
+ "category": "short",
+ "text": "Are you serious right now?",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0083.wav",
+ "render_seconds": 80.69410763599444,
+ "audio_seconds": 2.020136054421769,
+ "rtf": 39.944887602677746,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I'll see you next time.",
+ "raw": {
+ "normalized_reference": "are you serious right now?",
+ "normalized_hypothesis": "i'll see you next time.",
+ "errors": 5,
+ "words": 5,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "are you serious right now",
+ "normalized_hypothesis": "i will see you next time",
+ "errors": 6,
+ "words": 5,
+ "wer": 1.2
+ }
+ },
+ {
+ "id": "release500_0084",
+ "category": "technical",
+ "text": "The performance improvement is real, but it comes with tradeoffs.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0084.wav",
+ "render_seconds": 181.11119199392851,
+ "audio_seconds": 4.597551020408163,
+ "rtf": 39.39297055975895,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The perfumes are perfect as well, but at least they're close.",
+ "raw": {
+ "normalized_reference": "the performance improvement is real, but it comes with tradeoffs.",
+ "normalized_hypothesis": "the perfumes are perfect as well, but at least they're close.",
+ "errors": 9,
+ "words": 10,
+ "wer": 0.9
+ },
+ "semantic": {
+ "normalized_reference": "the performance improvement is real but it comes with tradeoffs",
+ "normalized_hypothesis": "the perfumes are perfect as well but at least they are close",
+ "errors": 10,
+ "words": 10,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0085",
+ "category": "conversational",
+ "text": "I don't know. It's just been one of those weeks, you know?",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0085.wav",
+ "render_seconds": 163.90710327599663,
+ "audio_seconds": 3.900952380952381,
+ "rtf": 42.017201767528434,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "That's the first battle you ever hear this much, I know.",
+ "raw": {
+ "normalized_reference": "i don't know. it's just been one of those weeks, you know?",
+ "normalized_hypothesis": "that's the first battle you ever hear this much, i know.",
+ "errors": 11,
+ "words": 12,
+ "wer": 0.9166666666666666
+ },
+ "semantic": {
+ "normalized_reference": "i do not know it is just been one of those weeks you know",
+ "normalized_hypothesis": "that is the 1st battle you ever hear this much i know",
+ "errors": 13,
+ "words": 14,
+ "wer": 0.9285714285714286
+ }
+ },
+ {
+ "id": "release500_0086",
+ "category": "descriptive",
+ "text": "The moment you stop expecting perfection, things get a lot more interesting.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0086.wav",
+ "render_seconds": 161.07786938804202,
+ "audio_seconds": 5.154829931972789,
+ "rtf": 31.247950274548902,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Believe in a stupid spot in perfection, and it's going to hurt your ancestors.",
+ "raw": {
+ "normalized_reference": "the moment you stop expecting perfection, things get a lot more interesting.",
+ "normalized_hypothesis": "believe in a stupid spot in perfection, and it's going to hurt your ancestors.",
+ "errors": 13,
+ "words": 12,
+ "wer": 1.0833333333333333
+ },
+ "semantic": {
+ "normalized_reference": "the moment you stop expecting perfection things get a lot more interesting",
+ "normalized_hypothesis": "believe in a stupid spot in perfection and it is going to hurt your ancestors",
+ "errors": 14,
+ "words": 12,
+ "wer": 1.1666666666666667
+ }
+ },
+ {
+ "id": "release500_0087",
+ "category": "emotional",
+ "text": "It's the kind of thing that stays with you even years later.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0087.wav",
+ "render_seconds": 131.09238334093243,
+ "audio_seconds": 3.8312925170068026,
+ "rtf": 34.21622931661411,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It's the kind of thing that starts with a laughing house, you know.",
+ "raw": {
+ "normalized_reference": "it's the kind of thing that stays with you even years later.",
+ "normalized_hypothesis": "it's the kind of thing that starts with a laughing house, you know.",
+ "errors": 6,
+ "words": 12,
+ "wer": 0.5
+ },
+ "semantic": {
+ "normalized_reference": "it is the kind of thing that stays with you even years later",
+ "normalized_hypothesis": "it is the kind of thing that starts with a laughing house you know",
+ "errors": 6,
+ "words": 13,
+ "wer": 0.46153846153846156
+ }
+ },
+ {
+ "id": "release500_0088",
+ "category": "long",
+ "text": "She told me once that the most important skill she'd developed was learning to disagree with someone without making them feel they were wrong.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0088.wav",
+ "render_seconds": 274.00661964400206,
+ "audio_seconds": 8.70748299319728,
+ "rtf": 31.46794772474086,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I take my wish that the rest of your space and the world is your entertainment. I show you the way that you will find your way.",
+ "raw": {
+ "normalized_reference": "she told me once that the most important skill she'd developed was learning to disagree with someone without making them feel they were wrong.",
+ "normalized_hypothesis": "i take my wish that the rest of your space and the world is your entertainment. i show you the way that you will find your way.",
+ "errors": 25,
+ "words": 24,
+ "wer": 1.0416666666666667
+ },
+ "semantic": {
+ "normalized_reference": "she told me once that the most important skill she would developed was learning to disagree with someone without making them feel they were wrong",
+ "normalized_hypothesis": "i take my wish that the rest of your space and the world is your entertainment i show you the way that you will find your way",
+ "errors": 25,
+ "words": 25,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0089",
+ "category": "narrative",
+ "text": "He stood at the edge of the cliff and wondered, not for the first time, how he'd gotten here.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0089.wav",
+ "render_seconds": 235.8968853870174,
+ "audio_seconds": 5.7817687074829935,
+ "rtf": 40.80012489633325,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Ashton and the Arch of the Curse of Murdoch wrote for the first time, who had told them how.",
+ "raw": {
+ "normalized_reference": "he stood at the edge of the cliff and wondered, not for the first time, how he'd gotten here.",
+ "normalized_hypothesis": "ashton and the arch of the curse of murdoch wrote for the first time, who had told them how.",
+ "errors": 13,
+ "words": 19,
+ "wer": 0.6842105263157895
+ },
+ "semantic": {
+ "normalized_reference": "he stood at the edge of the cliff and wondered not for the 1st time how he would gotten here",
+ "normalized_hypothesis": "ashton and the arch of the curse of murdoch wrote for the 1st time who had told them how",
+ "errors": 13,
+ "words": 20,
+ "wer": 0.65
+ }
+ },
+ {
+ "id": "release500_0090",
+ "category": "question",
+ "text": "Why does asking for help feel so hard when you actually need it?",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0090.wav",
+ "render_seconds": 170.8886543479748,
+ "audio_seconds": 4.249251700680272,
+ "rtf": 40.21617602003121,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Let us ask you for hope, I say to you, when I walk to the ladder.",
+ "raw": {
+ "normalized_reference": "why does asking for help feel so hard when you actually need it?",
+ "normalized_hypothesis": "let us ask you for hope, i say to you, when i walk to the ladder.",
+ "errors": 14,
+ "words": 13,
+ "wer": 1.0769230769230769
+ },
+ "semantic": {
+ "normalized_reference": "why does asking for help feel so hard when you actually need it",
+ "normalized_hypothesis": "let us ask you for hope i say to you when i walk to the ladder",
+ "errors": 14,
+ "words": 13,
+ "wer": 1.0769230769230769
+ }
+ },
+ {
+ "id": "release500_0091",
+ "category": "short",
+ "text": "That's fair.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0091.wav",
+ "render_seconds": 91.19974928500596,
+ "audio_seconds": 1.4628571428571429,
+ "rtf": 62.343578612797046,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Thank you for watching!",
+ "raw": {
+ "normalized_reference": "that's fair.",
+ "normalized_hypothesis": "thank you for watching!",
+ "errors": 4,
+ "words": 2,
+ "wer": 2.0
+ },
+ "semantic": {
+ "normalized_reference": "that is fair",
+ "normalized_hypothesis": "thank you for watching",
+ "errors": 4,
+ "words": 3,
+ "wer": 1.3333333333333333
+ }
+ },
+ {
+ "id": "release500_0092",
+ "category": "technical",
+ "text": "If you run into that error, the fix is usually just restarting the service.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0092.wav",
+ "render_seconds": 150.22765503602568,
+ "audio_seconds": 5.363809523809524,
+ "rtf": 28.00764165266814,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "As I will on to the hour, the frictions of the lectures will stir in the service.",
+ "raw": {
+ "normalized_reference": "if you run into that error, the fix is usually just restarting the service.",
+ "normalized_hypothesis": "as i will on to the hour, the frictions of the lectures will stir in the service.",
+ "errors": 14,
+ "words": 14,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "if you run into that error the fix is usually just restarting the service",
+ "normalized_hypothesis": "as i will on to the hour the frictions of the lectures will stir in the service",
+ "errors": 14,
+ "words": 14,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0093",
+ "category": "conversational",
+ "text": "It was a disaster in the best possible way.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0093.wav",
+ "render_seconds": 122.3073402219452,
+ "audio_seconds": 3.2043537414965986,
+ "rtf": 38.169113053298965,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I was a disaster in the first place, you know.",
+ "raw": {
+ "normalized_reference": "it was a disaster in the best possible way.",
+ "normalized_hypothesis": "i was a disaster in the first place, you know.",
+ "errors": 5,
+ "words": 9,
+ "wer": 0.5555555555555556
+ },
+ "semantic": {
+ "normalized_reference": "it was a disaster in the best possible way",
+ "normalized_hypothesis": "i was a disaster in the 1st place you know",
+ "errors": 5,
+ "words": 9,
+ "wer": 0.5555555555555556
+ }
+ },
+ {
+ "id": "release500_0094",
+ "category": "descriptive",
+ "text": "It had the look of a place that used to mean something to someone.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0094.wav",
+ "render_seconds": 161.29286159400363,
+ "audio_seconds": 4.458231292517007,
+ "rtf": 36.17866615954366,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I've had the look of the past that has to lie so down to show you.",
+ "raw": {
+ "normalized_reference": "it had the look of a place that used to mean something to someone.",
+ "normalized_hypothesis": "i've had the look of the past that has to lie so down to show you.",
+ "errors": 9,
+ "words": 14,
+ "wer": 0.6428571428571429
+ },
+ "semantic": {
+ "normalized_reference": "it had the look of a place that used to mean something to someone",
+ "normalized_hypothesis": "i have had the look of the past that has to lie so down to show you",
+ "errors": 10,
+ "words": 14,
+ "wer": 0.7142857142857143
+ }
+ },
+ {
+ "id": "release500_0095",
+ "category": "emotional",
+ "text": "There's this feeling you get when something finally works after weeks of failure.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0095.wav",
+ "render_seconds": 204.904479224002,
+ "audio_seconds": 5.015510204081632,
+ "rtf": 40.85416455882201,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "There's a fire in the garden shooting through the works of the wax fire.",
+ "raw": {
+ "normalized_reference": "there's this feeling you get when something finally works after weeks of failure.",
+ "normalized_hypothesis": "there's a fire in the garden shooting through the works of the wax fire.",
+ "errors": 12,
+ "words": 13,
+ "wer": 0.9230769230769231
+ },
+ "semantic": {
+ "normalized_reference": "there is this feeling you get when something finally works after weeks of failure",
+ "normalized_hypothesis": "there is a fire in the garden shooting through the works of the wax fire",
+ "errors": 12,
+ "words": 14,
+ "wer": 0.8571428571428571
+ }
+ },
+ {
+ "id": "release500_0096",
+ "category": "long",
+ "text": "He had a theory that most arguments aren't actually about what they seem to be about, and the longer I live the more I think he was onto something.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0096.wav",
+ "render_seconds": 337.59983496705536,
+ "audio_seconds": 9.125442176870749,
+ "rtf": 36.99544947232611,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I have a flower that lives to our children's share. It's a bitter shout about food. And the little one loves to eat all the things I wish her to eat.",
+ "raw": {
+ "normalized_reference": "he had a theory that most arguments aren't actually about what they seem to be about, and the longer i live the more i think he was onto something.",
+ "normalized_hypothesis": "i have a flower that lives to our children's share. it's a bitter shout about food. and the little one loves to eat all the things i wish her to eat.",
+ "errors": 26,
+ "words": 29,
+ "wer": 0.896551724137931
+ },
+ "semantic": {
+ "normalized_reference": "he had a theory that most arguments are not actually about what they seem to be about and the longer i live the more i think he was onto something",
+ "normalized_hypothesis": "i have a flower that lives to our children is share it is a bitter shout about food and the little one loves to eat all the things i wish her to eat",
+ "errors": 27,
+ "words": 30,
+ "wer": 0.9
+ }
+ },
+ {
+ "id": "release500_0097",
+ "category": "narrative",
+ "text": "The thing nobody tells you about starting over is how ordinary most of it feels.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0097.wav",
+ "render_seconds": 187.19245061906986,
+ "audio_seconds": 5.433469387755102,
+ "rtf": 34.45173557817917,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The thing I hate the most about storming over is how you're blowing loose to the face.",
+ "raw": {
+ "normalized_reference": "the thing nobody tells you about starting over is how ordinary most of it feels.",
+ "normalized_hypothesis": "the thing i hate the most about storming over is how you're blowing loose to the face.",
+ "errors": 11,
+ "words": 15,
+ "wer": 0.7333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "the thing nobody tells you about starting over is how ordinary most of it feels",
+ "normalized_hypothesis": "the thing i hate the most about storming over is how you are blowing loose to the face",
+ "errors": 12,
+ "words": 15,
+ "wer": 0.8
+ }
+ },
+ {
+ "id": "release500_0098",
+ "category": "question",
+ "text": "What exactly are we waiting for at this point?",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0098.wav",
+ "render_seconds": 56.579715312924236,
+ "audio_seconds": 3.2043537414965986,
+ "rtf": 17.65713771866479,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "You're not sure what kind of life you're supposed to live.",
+ "raw": {
+ "normalized_reference": "what exactly are we waiting for at this point?",
+ "normalized_hypothesis": "you're not sure what kind of life you're supposed to live.",
+ "errors": 11,
+ "words": 9,
+ "wer": 1.2222222222222223
+ },
+ "semantic": {
+ "normalized_reference": "what exactly are we waiting for at this point",
+ "normalized_hypothesis": "you are not sure what kind of life you are supposed to live",
+ "errors": 12,
+ "words": 9,
+ "wer": 1.3333333333333333
+ }
+ },
+ {
+ "id": "release500_0099",
+ "category": "short",
+ "text": "I stand by what I said.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0099.wav",
+ "render_seconds": 103.61576511000749,
+ "audio_seconds": 2.089795918367347,
+ "rtf": 49.58176260146842,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Restart the research.",
+ "raw": {
+ "normalized_reference": "i stand by what i said.",
+ "normalized_hypothesis": "restart the research.",
+ "errors": 6,
+ "words": 6,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "i stand by what i said",
+ "normalized_hypothesis": "restart the research",
+ "errors": 6,
+ "words": 6,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0100",
+ "category": "technical",
+ "text": "Double-check that value before you push anything to production.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0100.wav",
+ "render_seconds": 162.6109438110143,
+ "audio_seconds": 4.31891156462585,
+ "rtf": 37.65090842398422,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I've been trapped for so long before I pushed anything to the ocean.",
+ "raw": {
+ "normalized_reference": "double-check that value before you push anything to production.",
+ "normalized_hypothesis": "i've been trapped for so long before i pushed anything to the ocean.",
+ "errors": 10,
+ "words": 10,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "double check that value before you push anything to production",
+ "normalized_hypothesis": "i have been trapped for so long before i pushed anything to the ocean",
+ "errors": 11,
+ "words": 10,
+ "wer": 1.1
+ }
+ },
+ {
+ "id": "release500_0101",
+ "category": "conversational",
+ "text": "I understand your point; I just do not agree with the conclusion.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0101.wav",
+ "render_seconds": 168.29973123699892,
+ "audio_seconds": 4.458231292517007,
+ "rtf": 37.75033644384589,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Well, this time I'll do it. I'll just tell you what I want to tell you.",
+ "raw": {
+ "normalized_reference": "i understand your point; i just do not agree with the conclusion.",
+ "normalized_hypothesis": "well, this time i'll do it. i'll just tell you what i want to tell you.",
+ "errors": 15,
+ "words": 12,
+ "wer": 1.25
+ },
+ "semantic": {
+ "normalized_reference": "i understand your point i just do not agree with the conclusion",
+ "normalized_hypothesis": "well this time i will do it i will just tell you what i want to tell you",
+ "errors": 15,
+ "words": 12,
+ "wer": 1.25
+ }
+ },
+ {
+ "id": "release500_0102",
+ "category": "descriptive",
+ "text": "The sky that day was a strange shade of orange I'd never seen before.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0102.wav",
+ "render_seconds": 171.00677004898898,
+ "audio_seconds": 4.597551020408163,
+ "rtf": 37.1951870223742,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The school of time is as quiet as I feel. I can never shine as I feel.",
+ "raw": {
+ "normalized_reference": "the sky that day was a strange shade of orange i'd never seen before.",
+ "normalized_hypothesis": "the school of time is as quiet as i feel. i can never shine as i feel.",
+ "errors": 15,
+ "words": 14,
+ "wer": 1.0714285714285714
+ },
+ "semantic": {
+ "normalized_reference": "the sky that day was a strange shade of orange i would never seen before",
+ "normalized_hypothesis": "the school of time is as quiet as i feel i can never shine as i feel",
+ "errors": 14,
+ "words": 15,
+ "wer": 0.9333333333333333
+ }
+ },
+ {
+ "id": "release500_0103",
+ "category": "emotional",
+ "text": "It hurt more than I expected, and I thought I was prepared.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0103.wav",
+ "render_seconds": 163.80737377295736,
+ "audio_seconds": 4.388571428571429,
+ "rtf": 37.32589897170252,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I thought we were never expected, but I feel I was prepared.",
+ "raw": {
+ "normalized_reference": "it hurt more than i expected, and i thought i was prepared.",
+ "normalized_hypothesis": "i thought we were never expected, but i feel i was prepared.",
+ "errors": 7,
+ "words": 12,
+ "wer": 0.5833333333333334
+ },
+ "semantic": {
+ "normalized_reference": "it hurt more than i expected and i thought i was prepared",
+ "normalized_hypothesis": "i thought we were never expected but i feel i was prepared",
+ "errors": 7,
+ "words": 12,
+ "wer": 0.5833333333333334
+ }
+ },
+ {
+ "id": "release500_0104",
+ "category": "long",
+ "text": "The hardest part wasn't making the decision, it was sitting with the uncertainty of not knowing whether it was the right one for months afterward.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0104.wav",
+ "render_seconds": 190.2936620110413,
+ "audio_seconds": 8.986122448979591,
+ "rtf": 21.17639316529121,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The first book was a lack of discussion. I was stuck on it for a short time. Without knowing whether it was to work or for itself to work.",
+ "raw": {
+ "normalized_reference": "the hardest part wasn't making the decision, it was sitting with the uncertainty of not knowing whether it was the right one for months afterward.",
+ "normalized_hypothesis": "the first book was a lack of discussion. i was stuck on it for a short time. without knowing whether it was to work or for itself to work.",
+ "errors": 22,
+ "words": 25,
+ "wer": 0.88
+ },
+ "semantic": {
+ "normalized_reference": "the hardest part was not making the decision it was sitting with the uncertainty of not knowing whether it was the right one for months afterward",
+ "normalized_hypothesis": "the 1st book was a lack of discussion i was stuck on it for a short time without knowing whether it was to work or for itself to work",
+ "errors": 21,
+ "words": 26,
+ "wer": 0.8076923076923077
+ }
+ },
+ {
+ "id": "release500_0105",
+ "category": "narrative",
+ "text": "There was this one day last summer when everything just clicked into place.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0105.wav",
+ "render_seconds": 197.19412988598924,
+ "audio_seconds": 5.154829931972789,
+ "rtf": 38.25424553056432,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I wish this will die less soon. When everything is expecting to crash.",
+ "raw": {
+ "normalized_reference": "there was this one day last summer when everything just clicked into place.",
+ "normalized_hypothesis": "i wish this will die less soon. when everything is expecting to crash.",
+ "errors": 10,
+ "words": 13,
+ "wer": 0.7692307692307693
+ },
+ "semantic": {
+ "normalized_reference": "there was this one day last summer when everything just clicked into place",
+ "normalized_hypothesis": "i wish this will die less soon when everything is expecting to crash",
+ "errors": 10,
+ "words": 13,
+ "wer": 0.7692307692307693
+ }
+ },
+ {
+ "id": "release500_0106",
+ "category": "question",
+ "text": "Have you ever thought about why we keep doing things the same way when they stop working?",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0106.wav",
+ "render_seconds": 178.0883134501055,
+ "audio_seconds": 5.433469387755102,
+ "rtf": 32.77616946760505,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Have I ever seen a baby with a cap doing things the same way when they're still walking?",
+ "raw": {
+ "normalized_reference": "have you ever thought about why we keep doing things the same way when they stop working?",
+ "normalized_hypothesis": "have i ever seen a baby with a cap doing things the same way when they're still walking?",
+ "errors": 10,
+ "words": 17,
+ "wer": 0.5882352941176471
+ },
+ "semantic": {
+ "normalized_reference": "have you ever thought about why we keep doing things the same way when they stop working",
+ "normalized_hypothesis": "have i ever seen a baby with a cap doing things the same way when they are still walking",
+ "errors": 10,
+ "words": 17,
+ "wer": 0.5882352941176471
+ }
+ },
+ {
+ "id": "release500_0107",
+ "category": "short",
+ "text": "I can't believe you just said that.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0107.wav",
+ "render_seconds": 117.0924708719831,
+ "audio_seconds": 2.4380952380952383,
+ "rtf": 48.02620875608681,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I can't believe I just said that. I can't believe I just said that. I can't believe I just said that. I can't believe I just said that. I can't believe I just said that. I can't believe I just said that. I can't believe I just said that. I can't believe I just said that. I can't believe I just said that. I can't believe I just said that. I can't believe I just said that. I can't believe I just said that. I can't believe I just said that. I can't believe I just said that. I can't believe I just said that. I can't believe I just said that. I can't believe I just said that. I can't believe I just said that. I can't believe I just said that. I can't believe I just said that. I can't believe I just said that. I can't believe I just said that. I can't believe I just said that. I can't believe I just said that. I can't believe I just said that. I can't believe I just said that. I can't believe I just said that. I can't believe I just said that. I can't believe I just said that. I can't believe I just said that. I can't believe I just said that. I can't believe I just said that. I can't believe I just said that. I can't believe I just said that. I can't believe I just said that. I can't believe I just said that. I can't believe I just said that. I can't believe I just said that. I can't believe I just said that. I can't believe I just said that. I can't believe I just said that. I can't believe I just said that. I can't believe I just said that. I can't believe I just said that. I can't believe I just said that. I can't believe I just said that. I can't believe I just said that. I can't believe I just said that. I can't believe I just said that. I can't",
+ "raw": {
+ "normalized_reference": "i can't believe you just said that.",
+ "normalized_hypothesis": "i can't believe i just said that. i can't believe i just said that. i can't believe i just said that. i can't believe i just said that. i can't believe i just said that. i can't believe i just said that. i can't believe i just said that. i can't believe i just said that. i can't believe i just said that. i can't believe i just said that. i can't believe i just said that. i can't believe i just said that. i can't believe i just said that. i can't believe i just said that. i can't believe i just said that. i can't believe i just said that. i can't believe i just said that. i can't believe i just said that. i can't believe i just said that. i can't believe i just said that. i can't believe i just said that. i can't believe i just said that. i can't believe i just said that. i can't believe i just said that. i can't believe i just said that. i can't believe i just said that. i can't believe i just said that. i can't believe i just said that. i can't believe i just said that. i can't believe i just said that. i can't believe i just said that. i can't believe i just said that. i can't believe i just said that. i can't believe i just said that. i can't believe i just said that. i can't believe i just said that. i can't believe i just said that. i can't believe i just said that. i can't believe i just said that. i can't believe i just said that. i can't believe i just said that. i can't believe i just said that. i can't believe i just said that. i can't believe i just said that. i can't believe i just said that. i can't believe i just said that. i can't believe i just said that. i can't believe i just said that. i can't believe i just said that. i can't",
+ "errors": 339,
+ "words": 7,
+ "wer": 48.42857142857143
+ },
+ "semantic": {
+ "normalized_reference": "i can not believe you just said that",
+ "normalized_hypothesis": "i can not believe i just said that i can not believe i just said that i can not believe i just said that i can not believe i just said that i can not believe i just said that i can not believe i just said that i can not believe i just said that i can not believe i just said that i can not believe i just said that i can not believe i just said that i can not believe i just said that i can not believe i just said that i can not believe i just said that i can not believe i just said that i can not believe i just said that i can not believe i just said that i can not believe i just said that i can not believe i just said that i can not believe i just said that i can not believe i just said that i can not believe i just said that i can not believe i just said that i can not believe i just said that i can not believe i just said that i can not believe i just said that i can not believe i just said that i can not believe i just said that i can not believe i just said that i can not believe i just said that i can not believe i just said that i can not believe i just said that i can not believe i just said that i can not believe i just said that i can not believe i just said that i can not believe i just said that i can not believe i just said that i can not believe i just said that i can not believe i just said that i can not believe i just said that i can not believe i just said that i can not believe i just said that i can not believe i just said that i can not believe i just said that i can not believe i just said that i can not believe i just said that i can not believe i just said that i can not believe i just said that i can not believe i just said that i can not believe i just said that i can not",
+ "errors": 388,
+ "words": 8,
+ "wer": 48.5
+ }
+ },
+ {
+ "id": "release500_0108",
+ "category": "technical",
+ "text": "You'll want to isolate that variable before drawing any conclusions.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0108.wav",
+ "render_seconds": 195.0017470959574,
+ "audio_seconds": 4.597551020408163,
+ "rtf": 42.414264948960906,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you.",
+ "raw": {
+ "normalized_reference": "you'll want to isolate that variable before drawing any conclusions.",
+ "normalized_hypothesis": "i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you.",
+ "errors": 333,
+ "words": 10,
+ "wer": 33.3
+ },
+ "semantic": {
+ "normalized_reference": "you will want to isolate that variable before drawing any conclusions",
+ "normalized_hypothesis": "i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you",
+ "errors": 332,
+ "words": 11,
+ "wer": 30.181818181818183
+ }
+ },
+ {
+ "id": "release500_0109",
+ "category": "conversational",
+ "text": "The meeting that could have been an email lasted two hours.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0109.wav",
+ "render_seconds": 130.09551886399277,
+ "audio_seconds": 3.9706122448979593,
+ "rtf": 32.76459922047515,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "For now, I'm not going to stay in my lost place.",
+ "raw": {
+ "normalized_reference": "the meeting that could have been an email lasted two hours.",
+ "normalized_hypothesis": "for now, i'm not going to stay in my lost place.",
+ "errors": 11,
+ "words": 11,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "the meeting that could have been an email lasted 2 hours",
+ "normalized_hypothesis": "for now i am not going to stay in my lost place",
+ "errors": 12,
+ "words": 11,
+ "wer": 1.0909090909090908
+ }
+ },
+ {
+ "id": "release500_0110",
+ "category": "descriptive",
+ "text": "There's a specific kind of tired that sleep doesn't fix.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0110.wav",
+ "render_seconds": 73.60374982096255,
+ "audio_seconds": 3.622312925170068,
+ "rtf": 20.319544816108575,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I just wish I thought that was true. I just wish I thought that was true. I just wish I thought that was true. I just wish I thought that was true. I just wish I thought that was true. I just wish I thought that was true. I just wish I thought that was true. I just wish I thought that was true. I just wish I thought that was true. I just wish I thought that was true. I just wish I thought that was true. I just wish I thought that was true. I just wish I thought that was true. I just wish I thought that was true. I just wish I thought that was true. I just wish I thought that was true. I just wish I thought that was true. I just wish I thought that was true. I just wish I thought that was true. I just wish I thought that was true. I just wish I thought that was true. I just wish I thought that was true. I just wish I thought that was true. I just wish I thought that was true. I just wish I thought that was true. I just wish I thought that was true. I just wish I thought that was true. I just wish I thought that was true. I just wish I thought that was true. I just wish I thought that was true. I just wish I thought that was true. I just wish I thought that was true. I just wish I thought that was true. I just wish I thought that was true. I just wish I thought that was true. I just wish I thought that was true. I just wish I thought that was true. I just wish I thought that was true. I just wish I thought that was true. I just wish I thought that was true. I just wish I thought that was true. I just wish I thought that was true. I just wish I thought that was true. I just wish I thought that was true. I just wish I thought that was true. I just wish I thought that was true. I just wish I thought that was true. I just wish I thought that was true. I just wish I thought that was true. I just wish",
+ "raw": {
+ "normalized_reference": "there's a specific kind of tired that sleep doesn't fix.",
+ "normalized_hypothesis": "i just wish i thought that was true. i just wish i thought that was true. i just wish i thought that was true. i just wish i thought that was true. i just wish i thought that was true. i just wish i thought that was true. i just wish i thought that was true. i just wish i thought that was true. i just wish i thought that was true. i just wish i thought that was true. i just wish i thought that was true. i just wish i thought that was true. i just wish i thought that was true. i just wish i thought that was true. i just wish i thought that was true. i just wish i thought that was true. i just wish i thought that was true. i just wish i thought that was true. i just wish i thought that was true. i just wish i thought that was true. i just wish i thought that was true. i just wish i thought that was true. i just wish i thought that was true. i just wish i thought that was true. i just wish i thought that was true. i just wish i thought that was true. i just wish i thought that was true. i just wish i thought that was true. i just wish i thought that was true. i just wish i thought that was true. i just wish i thought that was true. i just wish i thought that was true. i just wish i thought that was true. i just wish i thought that was true. i just wish i thought that was true. i just wish i thought that was true. i just wish i thought that was true. i just wish i thought that was true. i just wish i thought that was true. i just wish i thought that was true. i just wish i thought that was true. i just wish i thought that was true. i just wish i thought that was true. i just wish i thought that was true. i just wish i thought that was true. i just wish i thought that was true. i just wish i thought that was true. i just wish i thought that was true. i just wish i thought that was true. i just wish",
+ "errors": 394,
+ "words": 10,
+ "wer": 39.4
+ },
+ "semantic": {
+ "normalized_reference": "there is a specific kind of tired that sleep does not fix",
+ "normalized_hypothesis": "i just wish i thought that was true i just wish i thought that was true i just wish i thought that was true i just wish i thought that was true i just wish i thought that was true i just wish i thought that was true i just wish i thought that was true i just wish i thought that was true i just wish i thought that was true i just wish i thought that was true i just wish i thought that was true i just wish i thought that was true i just wish i thought that was true i just wish i thought that was true i just wish i thought that was true i just wish i thought that was true i just wish i thought that was true i just wish i thought that was true i just wish i thought that was true i just wish i thought that was true i just wish i thought that was true i just wish i thought that was true i just wish i thought that was true i just wish i thought that was true i just wish i thought that was true i just wish i thought that was true i just wish i thought that was true i just wish i thought that was true i just wish i thought that was true i just wish i thought that was true i just wish i thought that was true i just wish i thought that was true i just wish i thought that was true i just wish i thought that was true i just wish i thought that was true i just wish i thought that was true i just wish i thought that was true i just wish i thought that was true i just wish i thought that was true i just wish i thought that was true i just wish i thought that was true i just wish i thought that was true i just wish i thought that was true i just wish i thought that was true i just wish i thought that was true i just wish i thought that was true i just wish i thought that was true i just wish i thought that was true i just wish i thought that was true i just wish",
+ "errors": 394,
+ "words": 12,
+ "wer": 32.833333333333336
+ }
+ },
+ {
+ "id": "release500_0111",
+ "category": "emotional",
+ "text": "Something about that moment broke something open in me that I hadn't realized was closed.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0111.wav",
+ "render_seconds": 191.18641630304046,
+ "audio_seconds": 5.642448979591837,
+ "rtf": 33.88358795879985,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "So, from the very start, we were born so very young. But we haven't realized this yet.",
+ "raw": {
+ "normalized_reference": "something about that moment broke something open in me that i hadn't realized was closed.",
+ "normalized_hypothesis": "so, from the very start, we were born so very young. but we haven't realized this yet.",
+ "errors": 16,
+ "words": 15,
+ "wer": 1.0666666666666667
+ },
+ "semantic": {
+ "normalized_reference": "something about that moment broke something open in me that i had not realized was closed",
+ "normalized_hypothesis": "so from the very start we were born so very young but we have not realized this yet",
+ "errors": 16,
+ "words": 16,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0112",
+ "category": "long",
+ "text": "The thing about making something good is that you don't usually know it's good while you're making it. You just know you're not done yet.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0112.wav",
+ "render_seconds": 227.912969725905,
+ "audio_seconds": 7.662585034013605,
+ "rtf": 29.74361376927205,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The thing about latin children is that they have no experience with latin. They just know how to do it.",
+ "raw": {
+ "normalized_reference": "the thing about making something good is that you don't usually know it's good while you're making it. you just know you're not done yet.",
+ "normalized_hypothesis": "the thing about latin children is that they have no experience with latin. they just know how to do it.",
+ "errors": 18,
+ "words": 25,
+ "wer": 0.72
+ },
+ "semantic": {
+ "normalized_reference": "the thing about making something good is that you do not usually know it is good while you are making it you just know you are not done yet",
+ "normalized_hypothesis": "the thing about latin children is that they have no experience with latin they just know how to do it",
+ "errors": 22,
+ "words": 29,
+ "wer": 0.7586206896551724
+ }
+ },
+ {
+ "id": "release500_0113",
+ "category": "narrative",
+ "text": "She opened the door slowly, as if she expected to find something she wasn't ready for.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0113.wav",
+ "render_seconds": 229.01116321899462,
+ "audio_seconds": 5.921088435374149,
+ "rtf": 38.67720702342855,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I opened the door soon, as if I expected to find something I wouldn't want to see.",
+ "raw": {
+ "normalized_reference": "she opened the door slowly, as if she expected to find something she wasn't ready for.",
+ "normalized_hypothesis": "i opened the door soon, as if i expected to find something i wouldn't want to see.",
+ "errors": 8,
+ "words": 16,
+ "wer": 0.5
+ },
+ "semantic": {
+ "normalized_reference": "she opened the door slowly as if she expected to find something she was not ready for",
+ "normalized_hypothesis": "i opened the door soon as if i expected to find something i would not want to see",
+ "errors": 8,
+ "words": 17,
+ "wer": 0.47058823529411764
+ }
+ },
+ {
+ "id": "release500_0114",
+ "category": "question",
+ "text": "What would you do differently if you could go back?",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0114.wav",
+ "render_seconds": 135.2906975609949,
+ "audio_seconds": 3.3436734693877552,
+ "rtf": 40.4616954375534,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "We are out of here. We are out of here. We are out of here. We are out of here. We are out of here. We are out of here. We are out of here. We are out of here. We are out of here. We are out of here. We are out of here. We are out of here. We are out of here. We are out of here. We are out of here. We are out of here. We are out of here. We are out of here. We are out of here. We are out of here. We are out of here. We are out of here. We are out of here. We are out of here. We are out of here. We are out of here. We are out of here. We are out of here. We are out of here. We are out of here. We are out of here. We are out of here. We are out of here. We are out of here. We are out of here. We are out of here. We are out of here. We are out of here. We are out of here. We are out of here. We are out of here. We are out of here. We are out of here. We are out of here. We are out of here. We are out of here. We are out of here. We are out of here. We are out of here. We are out of here. We are out of here. We are out of here. We are out of here. We are out of here. We are out of here. We are out of here. We are out of here. We are out of here. We are out of here. We are out of here. We are out of here. We are out of here. We are out of here. We are out of here. We are out of here. We are out of here. We are out of here. We are out of here. We are out of here. We are out of here. We are out of here. We are out of here. We are out of here. We are out of here.",
+ "raw": {
+ "normalized_reference": "what would you do differently if you could go back?",
+ "normalized_hypothesis": "we are out of here. we are out of here. we are out of here. we are out of here. we are out of here. we are out of here. we are out of here. we are out of here. we are out of here. we are out of here. we are out of here. we are out of here. we are out of here. we are out of here. we are out of here. we are out of here. we are out of here. we are out of here. we are out of here. we are out of here. we are out of here. we are out of here. we are out of here. we are out of here. we are out of here. we are out of here. we are out of here. we are out of here. we are out of here. we are out of here. we are out of here. we are out of here. we are out of here. we are out of here. we are out of here. we are out of here. we are out of here. we are out of here. we are out of here. we are out of here. we are out of here. we are out of here. we are out of here. we are out of here. we are out of here. we are out of here. we are out of here. we are out of here. we are out of here. we are out of here. we are out of here. we are out of here. we are out of here. we are out of here. we are out of here. we are out of here. we are out of here. we are out of here. we are out of here. we are out of here. we are out of here. we are out of here. we are out of here. we are out of here. we are out of here. we are out of here. we are out of here. we are out of here. we are out of here. we are out of here. we are out of here. we are out of here. we are out of here. we are out of here.",
+ "errors": 370,
+ "words": 10,
+ "wer": 37.0
+ },
+ "semantic": {
+ "normalized_reference": "what would you do differently if you could go back",
+ "normalized_hypothesis": "we are out of here we are out of here we are out of here we are out of here we are out of here we are out of here we are out of here we are out of here we are out of here we are out of here we are out of here we are out of here we are out of here we are out of here we are out of here we are out of here we are out of here we are out of here we are out of here we are out of here we are out of here we are out of here we are out of here we are out of here we are out of here we are out of here we are out of here we are out of here we are out of here we are out of here we are out of here we are out of here we are out of here we are out of here we are out of here we are out of here we are out of here we are out of here we are out of here we are out of here we are out of here we are out of here we are out of here we are out of here we are out of here we are out of here we are out of here we are out of here we are out of here we are out of here we are out of here we are out of here we are out of here we are out of here we are out of here we are out of here we are out of here we are out of here we are out of here we are out of here we are out of here we are out of here we are out of here we are out of here we are out of here we are out of here we are out of here we are out of here we are out of here we are out of here we are out of here we are out of here we are out of here we are out of here",
+ "errors": 370,
+ "words": 10,
+ "wer": 37.0
+ }
+ },
+ {
+ "id": "release500_0115",
+ "category": "short",
+ "text": "Okay. Okay okay okay.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0115.wav",
+ "render_seconds": 126.49804918107111,
+ "audio_seconds": 2.4380952380952383,
+ "rtf": 51.88396548442369,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Uta. Uta, Uta, Uta.",
+ "raw": {
+ "normalized_reference": "okay. okay okay okay.",
+ "normalized_hypothesis": "uta. uta, uta, uta.",
+ "errors": 4,
+ "words": 4,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "okay okay okay okay",
+ "normalized_hypothesis": "uta uta uta uta",
+ "errors": 4,
+ "words": 4,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0116",
+ "category": "technical",
+ "text": "The latency goes up significantly at scale, which is expected.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0116.wav",
+ "render_seconds": 94.58960826997645,
+ "audio_seconds": 4.597551020408163,
+ "rtf": 20.57391159991498,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The Rotten Shackles of Magnifico, let's go. Let's expect it.",
+ "raw": {
+ "normalized_reference": "the latency goes up significantly at scale, which is expected.",
+ "normalized_hypothesis": "the rotten shackles of magnifico, let's go. let's expect it.",
+ "errors": 9,
+ "words": 10,
+ "wer": 0.9
+ },
+ "semantic": {
+ "normalized_reference": "the latency goes up significantly at scale which is expected",
+ "normalized_hypothesis": "the rotten shackles of magnifico let us go let us expect it",
+ "errors": 11,
+ "words": 10,
+ "wer": 1.1
+ }
+ },
+ {
+ "id": "release500_0117",
+ "category": "conversational",
+ "text": "We ended up staying until two in the morning talking about nothing.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0117.wav",
+ "render_seconds": 169.9065376300132,
+ "audio_seconds": 4.458231292517007,
+ "rtf": 38.11074986513052,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "While the oaks stand to chill, I want to have a little fun.",
+ "raw": {
+ "normalized_reference": "we ended up staying until two in the morning talking about nothing.",
+ "normalized_hypothesis": "while the oaks stand to chill, i want to have a little fun.",
+ "errors": 13,
+ "words": 12,
+ "wer": 1.0833333333333333
+ },
+ "semantic": {
+ "normalized_reference": "we ended up staying until 2 in the morning talking about nothing",
+ "normalized_hypothesis": "while the oaks stand to chill i want to have a little fun",
+ "errors": 13,
+ "words": 12,
+ "wer": 1.0833333333333333
+ }
+ },
+ {
+ "id": "release500_0118",
+ "category": "descriptive",
+ "text": "The rain started slowly, and then all at once, the way it usually does.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0118.wav",
+ "render_seconds": 184.69438406801783,
+ "audio_seconds": 5.154829931972789,
+ "rtf": 35.829384578228755,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The world's coldest sea. And how it looks. The rock I see at dusk.",
+ "raw": {
+ "normalized_reference": "the rain started slowly, and then all at once, the way it usually does.",
+ "normalized_hypothesis": "the world's coldest sea. and how it looks. the rock i see at dusk.",
+ "errors": 12,
+ "words": 14,
+ "wer": 0.8571428571428571
+ },
+ "semantic": {
+ "normalized_reference": "the rain started slowly and then all at once the way it usually does",
+ "normalized_hypothesis": "the world is coldest sea and how it looks the rock i see at dusk",
+ "errors": 13,
+ "words": 14,
+ "wer": 0.9285714285714286
+ }
+ },
+ {
+ "id": "release500_0119",
+ "category": "emotional",
+ "text": "I honestly did not think I was going to make it through that year.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0119.wav",
+ "render_seconds": 184.18769364000764,
+ "audio_seconds": 4.31891156462585,
+ "rtf": 42.64678516425328,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I wish you a good night, and I wish you a good night, and I wish you a good night, and I wish you a good night, and I wish you a good night, and I wish you a good night, and I wish you a good night, and I wish you a good night, and I wish you a good night, and I wish you a good night, and I wish you a good night, and I wish you a good night, and I wish you a good night, and I wish you a good night, and I wish you a good night, and I wish you a good night, and I wish you a good night, and I wish you a good night, and I wish you a good night, and I wish you a good night, and I wish you a good night, and I wish you a good night, and I wish you a good night, and I wish you a good night, and I wish you a good night, and I wish you a good night, and I wish you a good night, and I wish you a good night, and I wish you a good night, and I wish you a good night, and I wish you a good night, and I wish you a good night, and I wish you a good night, and I wish you a good night, and I wish you a good night, and I wish you a good night, and I wish you a good night, and I wish you a good night, and I wish you a good night, and I wish you a good night, and I wish you a good night, and I wish you a good night, and I wish you a good night, and I wish you a good night, and I wish you a good night, and I wish you a good night, and I wish you a good night, and I wish you a good night, and I wish you a good night, and I wish you a good night, and I wish you a good night, and I wish you a good night, and I wish you a good night, and I wish you a good night, and I wish you a good night, and I wish you a",
+ "raw": {
+ "normalized_reference": "i honestly did not think i was going to make it through that year.",
+ "normalized_hypothesis": "i wish you a good night, and i wish you a good night, and i wish you a good night, and i wish you a good night, and i wish you a good night, and i wish you a good night, and i wish you a good night, and i wish you a good night, and i wish you a good night, and i wish you a good night, and i wish you a good night, and i wish you a good night, and i wish you a good night, and i wish you a good night, and i wish you a good night, and i wish you a good night, and i wish you a good night, and i wish you a good night, and i wish you a good night, and i wish you a good night, and i wish you a good night, and i wish you a good night, and i wish you a good night, and i wish you a good night, and i wish you a good night, and i wish you a good night, and i wish you a good night, and i wish you a good night, and i wish you a good night, and i wish you a good night, and i wish you a good night, and i wish you a good night, and i wish you a good night, and i wish you a good night, and i wish you a good night, and i wish you a good night, and i wish you a good night, and i wish you a good night, and i wish you a good night, and i wish you a good night, and i wish you a good night, and i wish you a good night, and i wish you a good night, and i wish you a good night, and i wish you a good night, and i wish you a good night, and i wish you a good night, and i wish you a good night, and i wish you a good night, and i wish you a good night, and i wish you a good night, and i wish you a good night, and i wish you a good night, and i wish you a good night, and i wish you a good night, and i wish you a",
+ "errors": 387,
+ "words": 14,
+ "wer": 27.642857142857142
+ },
+ "semantic": {
+ "normalized_reference": "i honestly did not think i was going to make it through that year",
+ "normalized_hypothesis": "i wish you a good night and i wish you a good night and i wish you a good night and i wish you a good night and i wish you a good night and i wish you a good night and i wish you a good night and i wish you a good night and i wish you a good night and i wish you a good night and i wish you a good night and i wish you a good night and i wish you a good night and i wish you a good night and i wish you a good night and i wish you a good night and i wish you a good night and i wish you a good night and i wish you a good night and i wish you a good night and i wish you a good night and i wish you a good night and i wish you a good night and i wish you a good night and i wish you a good night and i wish you a good night and i wish you a good night and i wish you a good night and i wish you a good night and i wish you a good night and i wish you a good night and i wish you a good night and i wish you a good night and i wish you a good night and i wish you a good night and i wish you a good night and i wish you a good night and i wish you a good night and i wish you a good night and i wish you a good night and i wish you a good night and i wish you a good night and i wish you a good night and i wish you a good night and i wish you a good night and i wish you a good night and i wish you a good night and i wish you a good night and i wish you a good night and i wish you a good night and i wish you a good night and i wish you a good night and i wish you a good night and i wish you a good night and i wish you a good night and i wish you a",
+ "errors": 387,
+ "words": 14,
+ "wer": 27.642857142857142
+ }
+ },
+ {
+ "id": "release500_0120",
+ "category": "long",
+ "text": "There's an assumption that people who seem confident all the time actually feel confident, and in my experience that's almost never true.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0120.wav",
+ "render_seconds": 327.906903670053,
+ "audio_seconds": 8.986122448979591,
+ "rtf": 36.490366732904704,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I feel so proud to be so perfect. I feel so perfect. I feel so perfect. I feel so perfect. I feel so perfect. I feel so perfect. I feel so perfect. I feel so perfect. I feel so perfect. I feel so perfect. I feel so perfect. I feel so perfect. I feel so perfect. I feel so perfect. I feel so perfect. I feel so perfect. I feel so perfect. I feel so perfect. I feel so perfect. I feel so perfect. I feel so perfect. I feel so perfect. I feel so perfect. I feel so perfect. I feel so perfect. I feel so perfect. I feel so perfect. I feel so perfect. I feel so perfect. I feel so perfect. I feel so perfect. I feel so perfect. I feel so perfect. I feel so perfect. I feel so perfect. I feel so perfect. I feel so perfect. I feel so perfect. I feel so perfect. I feel so perfect. I feel so perfect. I feel so perfect. I feel so perfect. I feel so perfect. I feel so perfect. I feel so perfect. I feel so perfect. I feel so perfect. I feel so perfect. I feel so perfect. I feel so perfect. I feel so perfect. I feel so perfect. I feel so perfect. I feel so perfect. I feel so perfect. I feel so perfect. I feel so perfect. I feel so perfect. I feel so perfect. I feel so perfect. I feel so perfect. I feel so perfect. I feel so perfect. I feel so perfect. I feel so perfect. I feel so perfect. I feel so perfect. I feel so perfect. I feel so perfect. I feel so perfect. I feel so perfect. I feel so perfect. I feel so perfect. I feel so perfect. I feel so perfect. I feel so perfect. I feel so perfect. I feel so perfect. I feel so perfect. I feel so perfect. I feel so perfect. I feel so perfect. I feel so perfect. I feel so perfect. I feel so perfect. I feel so perfect. I feel so perfect.",
+ "raw": {
+ "normalized_reference": "there's an assumption that people who seem confident all the time actually feel confident, and in my experience that's almost never true.",
+ "normalized_hypothesis": "i feel so proud to be so perfect. i feel so perfect. i feel so perfect. i feel so perfect. i feel so perfect. i feel so perfect. i feel so perfect. i feel so perfect. i feel so perfect. i feel so perfect. i feel so perfect. i feel so perfect. i feel so perfect. i feel so perfect. i feel so perfect. i feel so perfect. i feel so perfect. i feel so perfect. i feel so perfect. i feel so perfect. i feel so perfect. i feel so perfect. i feel so perfect. i feel so perfect. i feel so perfect. i feel so perfect. i feel so perfect. i feel so perfect. i feel so perfect. i feel so perfect. i feel so perfect. i feel so perfect. i feel so perfect. i feel so perfect. i feel so perfect. i feel so perfect. i feel so perfect. i feel so perfect. i feel so perfect. i feel so perfect. i feel so perfect. i feel so perfect. i feel so perfect. i feel so perfect. i feel so perfect. i feel so perfect. i feel so perfect. i feel so perfect. i feel so perfect. i feel so perfect. i feel so perfect. i feel so perfect. i feel so perfect. i feel so perfect. i feel so perfect. i feel so perfect. i feel so perfect. i feel so perfect. i feel so perfect. i feel so perfect. i feel so perfect. i feel so perfect. i feel so perfect. i feel so perfect. i feel so perfect. i feel so perfect. i feel so perfect. i feel so perfect. i feel so perfect. i feel so perfect. i feel so perfect. i feel so perfect. i feel so perfect. i feel so perfect. i feel so perfect. i feel so perfect. i feel so perfect. i feel so perfect. i feel so perfect. i feel so perfect. i feel so perfect. i feel so perfect. i feel so perfect. i feel so perfect. i feel so perfect. i feel so perfect. i feel so perfect. i feel so perfect.",
+ "errors": 355,
+ "words": 22,
+ "wer": 16.136363636363637
+ },
+ "semantic": {
+ "normalized_reference": "there is an assumption that people who seem confident all the time actually feel confident and in my experience that is almost never true",
+ "normalized_hypothesis": "i feel so proud to be so perfect i feel so perfect i feel so perfect i feel so perfect i feel so perfect i feel so perfect i feel so perfect i feel so perfect i feel so perfect i feel so perfect i feel so perfect i feel so perfect i feel so perfect i feel so perfect i feel so perfect i feel so perfect i feel so perfect i feel so perfect i feel so perfect i feel so perfect i feel so perfect i feel so perfect i feel so perfect i feel so perfect i feel so perfect i feel so perfect i feel so perfect i feel so perfect i feel so perfect i feel so perfect i feel so perfect i feel so perfect i feel so perfect i feel so perfect i feel so perfect i feel so perfect i feel so perfect i feel so perfect i feel so perfect i feel so perfect i feel so perfect i feel so perfect i feel so perfect i feel so perfect i feel so perfect i feel so perfect i feel so perfect i feel so perfect i feel so perfect i feel so perfect i feel so perfect i feel so perfect i feel so perfect i feel so perfect i feel so perfect i feel so perfect i feel so perfect i feel so perfect i feel so perfect i feel so perfect i feel so perfect i feel so perfect i feel so perfect i feel so perfect i feel so perfect i feel so perfect i feel so perfect i feel so perfect i feel so perfect i feel so perfect i feel so perfect i feel so perfect i feel so perfect i feel so perfect i feel so perfect i feel so perfect i feel so perfect i feel so perfect i feel so perfect i feel so perfect i feel so perfect i feel so perfect i feel so perfect i feel so perfect i feel so perfect i feel so perfect i feel so perfect i feel so perfect",
+ "errors": 355,
+ "words": 24,
+ "wer": 14.791666666666666
+ }
+ },
+ {
+ "id": "release500_0121",
+ "category": "narrative",
+ "text": "The moment I opened the door I knew something was different, even if I couldn't say what.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0121.wav",
+ "render_seconds": 177.3054225560045,
+ "audio_seconds": 5.572789115646258,
+ "rtf": 31.816280658853348,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Believe it or not, you think that the old one I was shooting was different. I thought I thought I couldn't show you.",
+ "raw": {
+ "normalized_reference": "the moment i opened the door i knew something was different, even if i couldn't say what.",
+ "normalized_hypothesis": "believe it or not, you think that the old one i was shooting was different. i thought i thought i couldn't show you.",
+ "errors": 17,
+ "words": 17,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "the moment i opened the door i knew something was different even if i could not say what",
+ "normalized_hypothesis": "believe it or not you think that the old one i was shooting was different i thought i thought i could not show you",
+ "errors": 17,
+ "words": 18,
+ "wer": 0.9444444444444444
+ }
+ },
+ {
+ "id": "release500_0122",
+ "category": "question",
+ "text": "How do you explain something that doesn't make sense even to you?",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0122.wav",
+ "render_seconds": 138.50993507902604,
+ "audio_seconds": 4.040272108843538,
+ "rtf": 34.28232835502564,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Hope I'll experience something that doesn't last so long.",
+ "raw": {
+ "normalized_reference": "how do you explain something that doesn't make sense even to you?",
+ "normalized_hypothesis": "hope i'll experience something that doesn't last so long.",
+ "errors": 9,
+ "words": 12,
+ "wer": 0.75
+ },
+ "semantic": {
+ "normalized_reference": "how do you explain something that does not make sense even to you",
+ "normalized_hypothesis": "hope i will experience something that does not last so long",
+ "errors": 9,
+ "words": 13,
+ "wer": 0.6923076923076923
+ }
+ },
+ {
+ "id": "release500_0123",
+ "category": "short",
+ "text": "Hold on a second.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0123.wav",
+ "render_seconds": 84.58820638398174,
+ "audio_seconds": 1.6718367346938776,
+ "rtf": 50.59597305682502,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He got a shock out.",
+ "raw": {
+ "normalized_reference": "hold on a second.",
+ "normalized_hypothesis": "he got a shock out.",
+ "errors": 4,
+ "words": 4,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "hold on a 2nd",
+ "normalized_hypothesis": "he got a shock out",
+ "errors": 4,
+ "words": 4,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0124",
+ "category": "technical",
+ "text": "A deterministic split prevents duplicate normalized text from leaking into evaluation.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0124.wav",
+ "render_seconds": 211.10324295493774,
+ "audio_seconds": 6.339047619047619,
+ "rtf": 33.30204403586007,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Thank you for watching.",
+ "raw": {
+ "normalized_reference": "a deterministic split prevents duplicate normalized text from leaking into evaluation.",
+ "normalized_hypothesis": "thank you for watching.",
+ "errors": 11,
+ "words": 11,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "a deterministic split prevents duplicate normalized text from leaking into evaluation",
+ "normalized_hypothesis": "thank you for watching",
+ "errors": 11,
+ "words": 11,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0125",
+ "category": "conversational",
+ "text": "Honestly, that was much better than I expected, but the ending still felt rushed.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0125.wav",
+ "render_seconds": 229.00751723907888,
+ "audio_seconds": 5.642448979591837,
+ "rtf": 40.58654638568745,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Alessia, have as much power as you expected, but how do you stay so quashed?",
+ "raw": {
+ "normalized_reference": "honestly, that was much better than i expected, but the ending still felt rushed.",
+ "normalized_hypothesis": "alessia, have as much power as you expected, but how do you stay so quashed?",
+ "errors": 12,
+ "words": 14,
+ "wer": 0.8571428571428571
+ },
+ "semantic": {
+ "normalized_reference": "honestly that was much better than i expected but the ending still felt rushed",
+ "normalized_hypothesis": "alessia have as much power as you expected but how do you stay so quashed",
+ "errors": 12,
+ "words": 14,
+ "wer": 0.8571428571428571
+ }
+ },
+ {
+ "id": "release500_0126",
+ "category": "descriptive",
+ "text": "She had a habit of tilting her head slightly when thinking hard about something.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0126.wav",
+ "render_seconds": 204.49203690805007,
+ "audio_seconds": 5.224489795918367,
+ "rtf": 39.14105393943146,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "How the hell did this tootin' whole head shut now? What a fuckin' horrible tootin'!",
+ "raw": {
+ "normalized_reference": "she had a habit of tilting her head slightly when thinking hard about something.",
+ "normalized_hypothesis": "how the hell did this tootin' whole head shut now? what a fuckin' horrible tootin'!",
+ "errors": 14,
+ "words": 14,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "she had a habit of tilting her head slightly when thinking hard about something",
+ "normalized_hypothesis": "how the hell did this tootin whole head shut now what a fuckin horrible tootin",
+ "errors": 14,
+ "words": 14,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0127",
+ "category": "emotional",
+ "text": "I cannot stress enough how much I did not want to be right about this.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0127.wav",
+ "render_seconds": 182.69252880394924,
+ "audio_seconds": 4.667210884353741,
+ "rtf": 39.14383415079953,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "A college classroom. Who would try to load you to the wall like this?",
+ "raw": {
+ "normalized_reference": "i cannot stress enough how much i did not want to be right about this.",
+ "normalized_hypothesis": "a college classroom. who would try to load you to the wall like this?",
+ "errors": 13,
+ "words": 15,
+ "wer": 0.8666666666666667
+ },
+ "semantic": {
+ "normalized_reference": "i cannot stress enough how much i did not want to be right about this",
+ "normalized_hypothesis": "a college classroom who would try to load you to the wall like this",
+ "errors": 13,
+ "words": 15,
+ "wer": 0.8666666666666667
+ }
+ },
+ {
+ "id": "release500_0128",
+ "category": "long",
+ "text": "I used to think the goal was to get to a point where nothing was hard anymore, and now I think that point doesn't exist and I'm not sure I'd want it to.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0128.wav",
+ "render_seconds": 181.790965070948,
+ "audio_seconds": 9.264761904761905,
+ "rtf": 19.621763294047636,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I used to think that doing was to get to a point where nothing was happening anymore. I know I think that point doesn't exist. I don't know how I'm gonna tell you.",
+ "raw": {
+ "normalized_reference": "i used to think the goal was to get to a point where nothing was hard anymore, and now i think that point doesn't exist and i'm not sure i'd want it to.",
+ "normalized_hypothesis": "i used to think that doing was to get to a point where nothing was happening anymore. i know i think that point doesn't exist. i don't know how i'm gonna tell you.",
+ "errors": 13,
+ "words": 33,
+ "wer": 0.3939393939393939
+ },
+ "semantic": {
+ "normalized_reference": "i used to think the goal was to get to a point where nothing was hard anymore and now i think that point does not exist and i am not sure i would want it to",
+ "normalized_hypothesis": "i used to think that doing was to get to a point where nothing was happening anymore i know i think that point does not exist i do not know how i am going to tell you",
+ "errors": 14,
+ "words": 36,
+ "wer": 0.3888888888888889
+ }
+ },
+ {
+ "id": "release500_0129",
+ "category": "narrative",
+ "text": "So this is the part of the story where everything goes completely sideways.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0129.wav",
+ "render_seconds": 197.40703524800483,
+ "audio_seconds": 5.224489795918367,
+ "rtf": 37.784940340438425,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "So this is the core of the score. No other sounders could track the sounders.",
+ "raw": {
+ "normalized_reference": "so this is the part of the story where everything goes completely sideways.",
+ "normalized_hypothesis": "so this is the core of the score. no other sounders could track the sounders.",
+ "errors": 9,
+ "words": 13,
+ "wer": 0.6923076923076923
+ },
+ "semantic": {
+ "normalized_reference": "so this is the part of the story where everything goes completely sideways",
+ "normalized_hypothesis": "so this is the core of the score no other sounders could track the sounders",
+ "errors": 9,
+ "words": 13,
+ "wer": 0.6923076923076923
+ }
+ },
+ {
+ "id": "release500_0130",
+ "category": "question",
+ "text": "Is there a version of this where it actually works out okay?",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0130.wav",
+ "render_seconds": 150.68803610792384,
+ "audio_seconds": 4.179591836734694,
+ "rtf": 36.05328988910287,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "A soulful house and small box that looks over time.",
+ "raw": {
+ "normalized_reference": "is there a version of this where it actually works out okay?",
+ "normalized_hypothesis": "a soulful house and small box that looks over time.",
+ "errors": 11,
+ "words": 12,
+ "wer": 0.9166666666666666
+ },
+ "semantic": {
+ "normalized_reference": "is there a version of this where it actually works out okay",
+ "normalized_hypothesis": "a soulful house and small box that looks over time",
+ "errors": 11,
+ "words": 12,
+ "wer": 0.9166666666666666
+ }
+ },
+ {
+ "id": "release500_0131",
+ "category": "short",
+ "text": "Oh no. Oh no no no.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0131.wav",
+ "render_seconds": 95.0924721530173,
+ "audio_seconds": 2.089795918367347,
+ "rtf": 45.50323374509616,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Oh, oh, oh, oh, oh. Oh, oh, oh, oh, oh.",
+ "raw": {
+ "normalized_reference": "oh no. oh no no no.",
+ "normalized_hypothesis": "oh, oh, oh, oh, oh. oh, oh, oh, oh, oh.",
+ "errors": 8,
+ "words": 6,
+ "wer": 1.3333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "0 no 0 no no no",
+ "normalized_hypothesis": "0000000000",
+ "errors": 6,
+ "words": 6,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0132",
+ "category": "technical",
+ "text": "Always test on a copy before modifying the original.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0132.wav",
+ "render_seconds": 155.70008645695634,
+ "audio_seconds": 3.7616326530612243,
+ "rtf": 41.39162454639139,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "He has touched her head and the floor in a third direction.",
+ "raw": {
+ "normalized_reference": "always test on a copy before modifying the original.",
+ "normalized_hypothesis": "he has touched her head and the floor in a third direction.",
+ "errors": 12,
+ "words": 9,
+ "wer": 1.3333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "always test on a copy before modifying the original",
+ "normalized_hypothesis": "he has touched her head and the floor in a 3rd direction",
+ "errors": 12,
+ "words": 9,
+ "wer": 1.3333333333333333
+ }
+ },
+ {
+ "id": "release500_0133",
+ "category": "conversational",
+ "text": "I wasn't even supposed to be there. That's the wild part.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0133.wav",
+ "render_seconds": 162.20817584195174,
+ "audio_seconds": 3.9706122448979593,
+ "rtf": 40.85218244374813,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Are you sure I'm supposed to bother? That's the right book.",
+ "raw": {
+ "normalized_reference": "i wasn't even supposed to be there. that's the wild part.",
+ "normalized_hypothesis": "are you sure i'm supposed to bother? that's the right book.",
+ "errors": 8,
+ "words": 11,
+ "wer": 0.7272727272727273
+ },
+ "semantic": {
+ "normalized_reference": "i was not even supposed to be there that is the wild part",
+ "normalized_hypothesis": "are you sure i am supposed to bother that is the right book",
+ "errors": 9,
+ "words": 13,
+ "wer": 0.6923076923076923
+ }
+ },
+ {
+ "id": "release500_0134",
+ "category": "descriptive",
+ "text": "It's a small town. Not the kind you leave. The kind you leave and come back to.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0134.wav",
+ "render_seconds": 139.6144138079835,
+ "audio_seconds": 5.7817687074829935,
+ "rtf": 24.147353668314164,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It's a sweet time. With the corncob laugh. The corncob laugh and the corncob time.",
+ "raw": {
+ "normalized_reference": "it's a small town. not the kind you leave. the kind you leave and come back to.",
+ "normalized_hypothesis": "it's a sweet time. with the corncob laugh. the corncob laugh and the corncob time.",
+ "errors": 12,
+ "words": 17,
+ "wer": 0.7058823529411765
+ },
+ "semantic": {
+ "normalized_reference": "it is a small town not the kind you leave the kind you leave and come back to",
+ "normalized_hypothesis": "it is a sweet time with the corncob laugh the corncob laugh and the corncob time",
+ "errors": 12,
+ "words": 18,
+ "wer": 0.6666666666666666
+ }
+ },
+ {
+ "id": "release500_0135",
+ "category": "emotional",
+ "text": "She cried. I cried. We all just sort of stood there crying.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0135.wav",
+ "render_seconds": 179.39809406595305,
+ "audio_seconds": 4.806530612244898,
+ "rtf": 37.32382221990361,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I heard, I heard, I just sort of stood there and heard.",
+ "raw": {
+ "normalized_reference": "she cried. i cried. we all just sort of stood there crying.",
+ "normalized_hypothesis": "i heard, i heard, i just sort of stood there and heard.",
+ "errors": 7,
+ "words": 12,
+ "wer": 0.5833333333333334
+ },
+ "semantic": {
+ "normalized_reference": "she cried i cried we all just sort of stood there crying",
+ "normalized_hypothesis": "i heard i heard i just sort of stood there and heard",
+ "errors": 7,
+ "words": 12,
+ "wer": 0.5833333333333334
+ }
+ },
+ {
+ "id": "release500_0136",
+ "category": "long",
+ "text": "There's a version of being busy that's actually just a very elaborate way of avoiding the one thing you know you should be doing.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0136.wav",
+ "render_seconds": 213.9000673040282,
+ "audio_seconds": 8.150204081632653,
+ "rtf": 26.244749844494645,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "As you thought, it was very special. But actually, I just thought I loved what you had to say and listen. I really have had a good time.",
+ "raw": {
+ "normalized_reference": "there's a version of being busy that's actually just a very elaborate way of avoiding the one thing you know you should be doing.",
+ "normalized_hypothesis": "as you thought, it was very special. but actually, i just thought i loved what you had to say and listen. i really have had a good time.",
+ "errors": 26,
+ "words": 24,
+ "wer": 1.0833333333333333
+ },
+ "semantic": {
+ "normalized_reference": "there is a version of being busy that is actually just a very elaborate way of avoiding the one thing you know you should be doing",
+ "normalized_hypothesis": "as you thought it was very special but actually i just thought i loved what you had to say and listen i really have had a good time",
+ "errors": 27,
+ "words": 26,
+ "wer": 1.0384615384615385
+ }
+ },
+ {
+ "id": "release500_0137",
+ "category": "narrative",
+ "text": "The letter arrived on a Tuesday, which felt appropriate somehow.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0137.wav",
+ "render_seconds": 173.59912472194992,
+ "audio_seconds": 4.667210884353741,
+ "rtf": 37.19547477572097,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "So now we're off to a test. Let's say to put that show on.",
+ "raw": {
+ "normalized_reference": "the letter arrived on a tuesday, which felt appropriate somehow.",
+ "normalized_hypothesis": "so now we're off to a test. let's say to put that show on.",
+ "errors": 13,
+ "words": 10,
+ "wer": 1.3
+ },
+ "semantic": {
+ "normalized_reference": "the letter arrived on a tuesday which felt appropriate somehow",
+ "normalized_hypothesis": "so now we are off to a test let us say to put that show on",
+ "errors": 15,
+ "words": 10,
+ "wer": 1.5
+ }
+ },
+ {
+ "id": "release500_0138",
+ "category": "question",
+ "text": "What exactly did she say when you told her?",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0138.wav",
+ "render_seconds": 119.71911585098132,
+ "audio_seconds": 2.9257142857142857,
+ "rtf": 40.91961967562838,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "That's it. That's it. That's it. That's it. That's it. That's it. That's it. That's it. That's it. That's it. That's it. That's it. That's it. That's it. That's it. That's it. That's it. That's it. That's it. That's it. That's it. That's it. That's it. That's it. That's it. That's it. That's it. That's it. That's it. That's it. That's it. That's it. That's it. That's it. That's it. That's it. That's it. That's it. That's it. That's it. That's it. That's it. That's it. That's it. That's it. That's it. That's it. That's it. That's it. That's it. That's it. That's it. That's it. That's it. That's it. That's it. That's it. That's it. That's it. That's it. That's it. That's it. That's it. That's it. That's it. That's it. That's it. That's it. That's it. That's it. That's it. That's it. That's it. That's it. That's it. That's it. That's it. That's it. That's it. That's it. That's it. That's it. That's it. That's it. That's it. That's it. That's it. That's it. That's it. That's it. That's it. That's it. That's it. That's it. That's it. That's it. That's it. That's it. That's it. That's it. That's it. That's it. That's it. That's it. That's it. That's it. That's it. That's it. That's it. That's it. That's it.",
+ "raw": {
+ "normalized_reference": "what exactly did she say when you told her?",
+ "normalized_hypothesis": "that's it. that's it. that's it. that's it. that's it. that's it. that's it. that's it. that's it. that's it. that's it. that's it. that's it. that's it. that's it. that's it. that's it. that's it. that's it. that's it. that's it. that's it. that's it. that's it. that's it. that's it. that's it. that's it. that's it. that's it. that's it. that's it. that's it. that's it. that's it. that's it. that's it. that's it. that's it. that's it. that's it. that's it. that's it. that's it. that's it. that's it. that's it. that's it. that's it. that's it. that's it. that's it. that's it. that's it. that's it. that's it. that's it. that's it. that's it. that's it. that's it. that's it. that's it. that's it. that's it. that's it. that's it. that's it. that's it. that's it. that's it. that's it. that's it. that's it. that's it. that's it. that's it. that's it. that's it. that's it. that's it. that's it. that's it. that's it. that's it. that's it. that's it. that's it. that's it. that's it. that's it. that's it. that's it. that's it. that's it. that's it. that's it. that's it. that's it. that's it. that's it. that's it. that's it. that's it. that's it. that's it. that's it. that's it. that's it. that's it. that's it.",
+ "errors": 222,
+ "words": 9,
+ "wer": 24.666666666666668
+ },
+ "semantic": {
+ "normalized_reference": "what exactly did she say when you told her",
+ "normalized_hypothesis": "that is it that is it that is it that is it that is it that is it that is it that is it that is it that is it that is it that is it that is it that is it that is it that is it that is it that is it that is it that is it that is it that is it that is it that is it that is it that is it that is it that is it that is it that is it that is it that is it that is it that is it that is it that is it that is it that is it that is it that is it that is it that is it that is it that is it that is it that is it that is it that is it that is it that is it that is it that is it that is it that is it that is it that is it that is it that is it that is it that is it that is it that is it that is it that is it that is it that is it that is it that is it that is it that is it that is it that is it that is it that is it that is it that is it that is it that is it that is it that is it that is it that is it that is it that is it that is it that is it that is it that is it that is it that is it that is it that is it that is it that is it that is it that is it that is it that is it that is it that is it that is it that is it that is it that is it that is it that is it that is it that is it that is it that is it that is it",
+ "errors": 333,
+ "words": 9,
+ "wer": 37.0
+ }
+ },
+ {
+ "id": "release500_0139",
+ "category": "short",
+ "text": "No way. Absolutely not.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0139.wav",
+ "render_seconds": 92.69116011692677,
+ "audio_seconds": 2.5774149659863945,
+ "rtf": 35.96283925566996,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Hey there, hope she doesn't know.",
+ "raw": {
+ "normalized_reference": "no way. absolutely not.",
+ "normalized_hypothesis": "hey there, hope she doesn't know.",
+ "errors": 6,
+ "words": 4,
+ "wer": 1.5
+ },
+ "semantic": {
+ "normalized_reference": "no way absolutely not",
+ "normalized_hypothesis": "hey there hope she does not know",
+ "errors": 6,
+ "words": 4,
+ "wer": 1.5
+ }
+ },
+ {
+ "id": "release500_0140",
+ "category": "technical",
+ "text": "The first thing you want to do is make sure everything is backed up.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0140.wav",
+ "render_seconds": 121.99403522501234,
+ "audio_seconds": 4.597551020408163,
+ "rtf": 26.53456909634945,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The first thing I looked at him was like sure everything is good dude.",
+ "raw": {
+ "normalized_reference": "the first thing you want to do is make sure everything is backed up.",
+ "normalized_hypothesis": "the first thing i looked at him was like sure everything is good dude.",
+ "errors": 8,
+ "words": 14,
+ "wer": 0.5714285714285714
+ },
+ "semantic": {
+ "normalized_reference": "the 1st thing you want to do is make sure everything is backed up",
+ "normalized_hypothesis": "the 1st thing i looked at him was like sure everything is good dude",
+ "errors": 8,
+ "words": 14,
+ "wer": 0.5714285714285714
+ }
+ },
+ {
+ "id": "release500_0141",
+ "category": "conversational",
+ "text": "I have the most insane idea and you need to just hear me out.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0141.wav",
+ "render_seconds": 144.2943132860819,
+ "audio_seconds": 4.179591836734694,
+ "rtf": 34.52354175301764,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I hope the rooster's sad all the time. I know not to curse her out.",
+ "raw": {
+ "normalized_reference": "i have the most insane idea and you need to just hear me out.",
+ "normalized_hypothesis": "i hope the rooster's sad all the time. i know not to curse her out.",
+ "errors": 12,
+ "words": 14,
+ "wer": 0.8571428571428571
+ },
+ "semantic": {
+ "normalized_reference": "i have the most insane idea and you need to just hear me out",
+ "normalized_hypothesis": "i hope the rooster is sad all the time i know not to curse her out",
+ "errors": 13,
+ "words": 14,
+ "wer": 0.9285714285714286
+ }
+ },
+ {
+ "id": "release500_0142",
+ "category": "descriptive",
+ "text": "There's a coffee shop on the corner that's been there longer than anyone can remember.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0142.wav",
+ "render_seconds": 179.10101063700859,
+ "audio_seconds": 5.085170068027211,
+ "rtf": 35.220259743739646,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I was supposed to help you in the pool, but I know you're not going to let me help you.",
+ "raw": {
+ "normalized_reference": "there's a coffee shop on the corner that's been there longer than anyone can remember.",
+ "normalized_hypothesis": "i was supposed to help you in the pool, but i know you're not going to let me help you.",
+ "errors": 19,
+ "words": 15,
+ "wer": 1.2666666666666666
+ },
+ "semantic": {
+ "normalized_reference": "there is a coffee shop on the corner that has been there longer than anyone can remember",
+ "normalized_hypothesis": "i was supposed to help you in the pool but i know you are not going to let me help you",
+ "errors": 20,
+ "words": 17,
+ "wer": 1.1764705882352942
+ }
+ },
+ {
+ "id": "release500_0143",
+ "category": "emotional",
+ "text": "I kept it together until I got to the car, and then I completely fell apart.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0143.wav",
+ "render_seconds": 206.52414280793164,
+ "audio_seconds": 5.294149659863946,
+ "rtf": 39.00987997631316,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I can't get you out of here until I get to the pool. I never want to be at your same pool.",
+ "raw": {
+ "normalized_reference": "i kept it together until i got to the car, and then i completely fell apart.",
+ "normalized_hypothesis": "i can't get you out of here until i get to the pool. i never want to be at your same pool.",
+ "errors": 17,
+ "words": 16,
+ "wer": 1.0625
+ },
+ "semantic": {
+ "normalized_reference": "i kept it together until i got to the car and then i completely fell apart",
+ "normalized_hypothesis": "i can not get you out of here until i get to the pool i never want to be at your same pool",
+ "errors": 18,
+ "words": 16,
+ "wer": 1.125
+ }
+ },
+ {
+ "id": "release500_0144",
+ "category": "long",
+ "text": "The older I get, the more I think that most of what we call wisdom is just pattern recognition built from having been wrong enough times to start noticing.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0144.wav",
+ "render_seconds": 313.5820713500725,
+ "audio_seconds": 9.682721088435374,
+ "rtf": 32.38573831529666,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Hello and welcome. The reason I'm here today is to tell you that I'm going to show you how to make the most out of this video.",
+ "raw": {
+ "normalized_reference": "the older i get, the more i think that most of what we call wisdom is just pattern recognition built from having been wrong enough times to start noticing.",
+ "normalized_hypothesis": "hello and welcome. the reason i'm here today is to tell you that i'm going to show you how to make the most out of this video.",
+ "errors": 28,
+ "words": 29,
+ "wer": 0.9655172413793104
+ },
+ "semantic": {
+ "normalized_reference": "the older i get the more i think that most of what we call wisdom is just pattern recognition built from having been wrong enough times to start noticing",
+ "normalized_hypothesis": "hello and welcome the reason i am here today is to tell you that i am going to show you how to make the most out of this video",
+ "errors": 28,
+ "words": 29,
+ "wer": 0.9655172413793104
+ }
+ },
+ {
+ "id": "release500_0145",
+ "category": "narrative",
+ "text": "I remember the exact moment I realized I had made a huge mistake.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0145.wav",
+ "render_seconds": 187.80746719695162,
+ "audio_seconds": 4.527891156462585,
+ "rtf": 41.477911174807524,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I love that song. I love that song. I love that song. I love that song. I love that song. I love that song. I love that song. I love that song. I love that song. I love that song. I love that song. I love that song. I love that song. I love that song. I love that song. I love that song. I love that song. I love that song. I love that song. I love that song. I love that song. I love that song. I love that song. I love that song. I love that song. I love that song. I love that song. I love that song. I love that song. I love that song. I love that song. I love that song. I love that song. I love that song. I love that song. I love that song. I love that song. I love that song. I love that song. I love that song. I love that song. I love that song. I love that song. I love that song. I love that song. I love that song. I love that song. I love that song. I love that song. I love that song. I love that song. I love that song. I love that song. I love that song. I love that song. I love that song. I love that song. I love that song. I love that song. I love that song. I love that song. I love that song. I love that song. I love that song. I love that song. I love that song. I love that song. I love that song. I love that song. I love that song. I love that song. I love that song. I love that song. I love that song. I love that song. I love that song. I love that song. I love that song. I love that song. I love that song. I love that song. I love that song. I love that song. I love that song. I love that song. I love that song. I love that song. I love that song. I love that song",
+ "raw": {
+ "normalized_reference": "i remember the exact moment i realized i had made a huge mistake.",
+ "normalized_hypothesis": "i love that song. i love that song. i love that song. i love that song. i love that song. i love that song. i love that song. i love that song. i love that song. i love that song. i love that song. i love that song. i love that song. i love that song. i love that song. i love that song. i love that song. i love that song. i love that song. i love that song. i love that song. i love that song. i love that song. i love that song. i love that song. i love that song. i love that song. i love that song. i love that song. i love that song. i love that song. i love that song. i love that song. i love that song. i love that song. i love that song. i love that song. i love that song. i love that song. i love that song. i love that song. i love that song. i love that song. i love that song. i love that song. i love that song. i love that song. i love that song. i love that song. i love that song. i love that song. i love that song. i love that song. i love that song. i love that song. i love that song. i love that song. i love that song. i love that song. i love that song. i love that song. i love that song. i love that song. i love that song. i love that song. i love that song. i love that song. i love that song. i love that song. i love that song. i love that song. i love that song. i love that song. i love that song. i love that song. i love that song. i love that song. i love that song. i love that song. i love that song. i love that song. i love that song. i love that song. i love that song. i love that song. i love that song. i love that song. i love that song. i love that song",
+ "errors": 353,
+ "words": 13,
+ "wer": 27.153846153846153
+ },
+ "semantic": {
+ "normalized_reference": "i remember the exact moment i realized i had made a huge mistake",
+ "normalized_hypothesis": "i love that song i love that song i love that song i love that song i love that song i love that song i love that song i love that song i love that song i love that song i love that song i love that song i love that song i love that song i love that song i love that song i love that song i love that song i love that song i love that song i love that song i love that song i love that song i love that song i love that song i love that song i love that song i love that song i love that song i love that song i love that song i love that song i love that song i love that song i love that song i love that song i love that song i love that song i love that song i love that song i love that song i love that song i love that song i love that song i love that song i love that song i love that song i love that song i love that song i love that song i love that song i love that song i love that song i love that song i love that song i love that song i love that song i love that song i love that song i love that song i love that song i love that song i love that song i love that song i love that song i love that song i love that song i love that song i love that song i love that song i love that song i love that song i love that song i love that song i love that song i love that song i love that song i love that song i love that song i love that song i love that song i love that song i love that song i love that song i love that song i love that song i love that song i love that song i love that song",
+ "errors": 353,
+ "words": 13,
+ "wer": 27.153846153846153
+ }
+ },
+ {
+ "id": "release500_0146",
+ "category": "question",
+ "text": "Do you think it would have gone differently if we'd left earlier?",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0146.wav",
+ "render_seconds": 103.29142461402807,
+ "audio_seconds": 3.9706122448979593,
+ "rtf": 26.013979266485276,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I think I'm going to have to go to school now. I think I'm going to have to go to school now.",
+ "raw": {
+ "normalized_reference": "do you think it would have gone differently if we'd left earlier?",
+ "normalized_hypothesis": "i think i'm going to have to go to school now. i think i'm going to have to go to school now.",
+ "errors": 21,
+ "words": 12,
+ "wer": 1.75
+ },
+ "semantic": {
+ "normalized_reference": "do you think it would have gone differently if we would left earlier",
+ "normalized_hypothesis": "i think i am going to have to go to school now i think i am going to have to go to school now",
+ "errors": 23,
+ "words": 13,
+ "wer": 1.7692307692307692
+ }
+ },
+ {
+ "id": "release500_0147",
+ "category": "short",
+ "text": "You have to be kidding me.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0147.wav",
+ "render_seconds": 100.99945445195772,
+ "audio_seconds": 1.9504761904761905,
+ "rtf": 51.78194686257598,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I have to back her over.",
+ "raw": {
+ "normalized_reference": "you have to be kidding me.",
+ "normalized_hypothesis": "i have to back her over.",
+ "errors": 4,
+ "words": 6,
+ "wer": 0.6666666666666666
+ },
+ "semantic": {
+ "normalized_reference": "you have to be kidding me",
+ "normalized_hypothesis": "i have to back her over",
+ "errors": 4,
+ "words": 6,
+ "wer": 0.6666666666666666
+ }
+ },
+ {
+ "id": "release500_0148",
+ "category": "technical",
+ "text": "If the checksum fails, restart the worker process without deleting the cached tensors.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0148.wav",
+ "render_seconds": 184.21122604899574,
+ "audio_seconds": 5.851428571428571,
+ "rtf": 31.48141070173267,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "As the treacherous eyes, watch out for your oppressors without delaying the treacherous",
+ "raw": {
+ "normalized_reference": "if the checksum fails, restart the worker process without deleting the cached tensors.",
+ "normalized_hypothesis": "as the treacherous eyes, watch out for your oppressors without delaying the treacherous",
+ "errors": 11,
+ "words": 13,
+ "wer": 0.8461538461538461
+ },
+ "semantic": {
+ "normalized_reference": "if the checksum fails restart the worker process without deleting the cached tensors",
+ "normalized_hypothesis": "as the treacherous eyes watch out for your oppressors without delaying the treacherous",
+ "errors": 11,
+ "words": 13,
+ "wer": 0.8461538461538461
+ }
+ },
+ {
+ "id": "release500_0149",
+ "category": "conversational",
+ "text": "I need you to understand how close that was to not working.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0149.wav",
+ "render_seconds": 128.47528654895723,
+ "audio_seconds": 3.900952380952381,
+ "rtf": 32.93433859287234,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "When I talk to you, I start hoping for something to help me.",
+ "raw": {
+ "normalized_reference": "i need you to understand how close that was to not working.",
+ "normalized_hypothesis": "when i talk to you, i start hoping for something to help me.",
+ "errors": 11,
+ "words": 12,
+ "wer": 0.9166666666666666
+ },
+ "semantic": {
+ "normalized_reference": "i need you to understand how close that was to not working",
+ "normalized_hypothesis": "when i talk to you i start hoping for something to help me",
+ "errors": 11,
+ "words": 12,
+ "wer": 0.9166666666666666
+ }
+ },
+ {
+ "id": "release500_0150",
+ "category": "descriptive",
+ "text": "It was the sort of morning where the whole world looks rinsed clean.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0150.wav",
+ "render_seconds": 174.10902754589915,
+ "audio_seconds": 4.527891156462585,
+ "rtf": 38.45256467735453,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I wish the show was a lie, but it seemed like it was true.",
+ "raw": {
+ "normalized_reference": "it was the sort of morning where the whole world looks rinsed clean.",
+ "normalized_hypothesis": "i wish the show was a lie, but it seemed like it was true.",
+ "errors": 13,
+ "words": 13,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "it was the sort of morning where the whole world looks rinsed clean",
+ "normalized_hypothesis": "i wish the show was a lie but it seemed like it was true",
+ "errors": 13,
+ "words": 13,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0151",
+ "category": "emotional",
+ "text": "She started laughing and I started laughing and we couldn't stop for a full minute.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0151.wav",
+ "render_seconds": 179.99561694508884,
+ "audio_seconds": 5.433469387755102,
+ "rtf": 33.12719813066914,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I started laughing and I started laughing and I couldn't stop for a few minutes.",
+ "raw": {
+ "normalized_reference": "she started laughing and i started laughing and we couldn't stop for a full minute.",
+ "normalized_hypothesis": "i started laughing and i started laughing and i couldn't stop for a few minutes.",
+ "errors": 4,
+ "words": 15,
+ "wer": 0.26666666666666666
+ },
+ "semantic": {
+ "normalized_reference": "she started laughing and i started laughing and we could not stop for a full minute",
+ "normalized_hypothesis": "i started laughing and i started laughing and i could not stop for a few minutes",
+ "errors": 4,
+ "words": 16,
+ "wer": 0.25
+ }
+ },
+ {
+ "id": "release500_0152",
+ "category": "long",
+ "text": "We spent three years building something we were genuinely proud of, and then the market shifted and none of it mattered the way we thought it would.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0152.wav",
+ "render_seconds": 231.64165761403274,
+ "audio_seconds": 9.543401360544218,
+ "rtf": 24.272442168441216,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "They spent three hours going and shooting. They were trying to win the polls. And then the law existed. And they were the ones who were the ones who were the ones who were the ones who were the ones who were the ones who were the ones who were the ones who were the ones who were the ones who were the ones who were the ones who were the ones who were the ones who were the ones who were the ones who were the ones who were the ones who were the ones who were the ones who were the ones who were the ones who were the ones who were the ones who were the ones who were the ones who were the ones who were the ones who were the ones who were the ones who were the ones who were the ones who were the ones who were the ones who were the ones who were the ones who were the ones who were the ones who were the ones who were the ones who were the ones who were the ones who were the ones who were the ones who were the ones who were the ones who were the ones who were the ones who were the ones who were the ones who were the ones who were the ones who were the ones who were the ones who were the ones who were the ones who were the ones who were the ones who were the ones who were the ones who were the ones who were the ones who were the men who were the men who were the men who were the men who were the men who were the men who were the men who were the men who were the men who were the men who were the men who were the men who were the men who were the men who were the men who were the men who were the men who were the men who were the men who were the men who were the men who were the men who were the men who were the men who were the men who were the men who were the men who were the men who were the men who were the men who were the men who were the men who were the men who were the men who were the men who were the men who were the men who were the men who were the men who were the men who were the men who were the men who were the men who",
+ "raw": {
+ "normalized_reference": "we spent three years building something we were genuinely proud of, and then the market shifted and none of it mattered the way we thought it would.",
+ "normalized_hypothesis": "they spent three hours going and shooting. they were trying to win the polls. and then the law existed. and they were the ones who were the ones who were the ones who were the ones who were the ones who were the ones who were the ones who were the ones who were the ones who were the ones who were the ones who were the ones who were the ones who were the ones who were the ones who were the ones who were the ones who were the ones who were the ones who were the ones who were the ones who were the ones who were the ones who were the ones who were the ones who were the ones who were the ones who were the ones who were the ones who were the ones who were the ones who were the ones who were the ones who were the ones who were the ones who were the ones who were the ones who were the ones who were the ones who were the ones who were the ones who were the ones who were the ones who were the ones who were the ones who were the ones who were the ones who were the ones who were the ones who were the ones who were the ones who were the ones who were the ones who were the ones who were the ones who were the ones who were the ones who were the ones who were the ones who were the ones who were the ones who were the ones who were the men who were the men who were the men who were the men who were the men who were the men who were the men who were the men who were the men who were the men who were the men who were the men who were the men who were the men who were the men who were the men who were the men who were the men who were the men who were the men who were the men who were the men who were the men who were the men who were the men who were the men who were the men who were the men who were the men who were the men who were the men who were the men who were the men who were the men who were the men who were the men who were the men who were the men who were the men who were the men who were the men who were the men who were the men who",
+ "errors": 433,
+ "words": 27,
+ "wer": 16.037037037037038
+ },
+ "semantic": {
+ "normalized_reference": "we spent 3 years building something we were genuinely proud of and then the market shifted and none of it mattered the way we thought it would",
+ "normalized_hypothesis": "they spent 3 hours going and shooting they were trying to win the polls and then the law existed and they were the ones who were the ones who were the ones who were the ones who were the ones who were the ones who were the ones who were the ones who were the ones who were the ones who were the ones who were the ones who were the ones who were the ones who were the ones who were the ones who were the ones who were the ones who were the ones who were the ones who were the ones who were the ones who were the ones who were the ones who were the ones who were the ones who were the ones who were the ones who were the ones who were the ones who were the ones who were the ones who were the ones who were the ones who were the ones who were the ones who were the ones who were the ones who were the ones who were the ones who were the ones who were the ones who were the ones who were the ones who were the ones who were the ones who were the ones who were the ones who were the ones who were the ones who were the ones who were the ones who were the ones who were the ones who were the ones who were the ones who were the ones who were the ones who were the ones who were the ones who were the ones who were the ones who were the men who were the men who were the men who were the men who were the men who were the men who were the men who were the men who were the men who were the men who were the men who were the men who were the men who were the men who were the men who were the men who were the men who were the men who were the men who were the men who were the men who were the men who were the men who were the men who were the men who were the men who were the men who were the men who were the men who were the men who were the men who were the men who were the men who were the men who were the men who were the men who were the men who were the men who were the men who were the men who were the men who were the men who were the men who",
+ "errors": 433,
+ "words": 27,
+ "wer": 16.037037037037038
+ }
+ },
+ {
+ "id": "release500_0153",
+ "category": "narrative",
+ "text": "And somehow, despite all of that, it actually turned out okay.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0153.wav",
+ "render_seconds": 156.49953109992202,
+ "audio_seconds": 4.527891156462585,
+ "rtf": 34.56344812453205,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I so hope to spark a little fun and actually talk all the time.",
+ "raw": {
+ "normalized_reference": "and somehow, despite all of that, it actually turned out okay.",
+ "normalized_hypothesis": "i so hope to spark a little fun and actually talk all the time.",
+ "errors": 13,
+ "words": 11,
+ "wer": 1.1818181818181819
+ },
+ "semantic": {
+ "normalized_reference": "and somehow despite all of that it actually turned out okay",
+ "normalized_hypothesis": "i so hope to spark a little fun and actually talk all the time",
+ "errors": 13,
+ "words": 11,
+ "wer": 1.1818181818181819
+ }
+ },
+ {
+ "id": "release500_0154",
+ "category": "question",
+ "text": "Is that something you actually want, or something you think you should want?",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0154.wav",
+ "render_seconds": 173.49401189107448,
+ "audio_seconds": 4.876190476190477,
+ "rtf": 35.57982665734926,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I thought so. I thought so. I thought so. I thought so. I thought so. I thought so. I thought so. I thought so. I thought so. I thought so. I thought so. I thought so. I thought so. I thought so. I thought so. I thought so. I thought so. I thought so. I thought so. I thought so. I thought so. I thought so. I thought so. I thought so. I thought so. I thought so. I thought so. I thought so. I thought so. I thought so. I thought so. I thought so. I thought so. I thought so. I thought so. I thought so. I thought so. I thought so. I thought so. I thought so. I thought so. I thought so. I thought so. I thought so. I thought so. I thought so. I thought so. I thought so. I thought so. I thought so. I thought so. I thought so. I thought so. I thought so. I thought so. I thought so. I thought so. I thought so. I thought so. I thought so. I thought so. I thought so. I thought so. I thought so. I thought so. I thought so. I thought so. I thought so. I thought so. I thought so. I thought so. I thought so. I thought so. I thought so. I thought so. I thought so. I thought so. I thought so. I thought so. I thought so. I thought so. I thought so. I thought so. I thought so. I thought so. I thought so. I thought so. I thought so. I thought so. I thought so. I thought so. I thought so. I thought so. I thought so. I thought so. I thought so. I thought so. I thought so. I thought so. I thought so. I thought so. I thought so. I thought so. I thought so. I thought so. I thought so. I thought so. I thought so. I thought so. I thought so. I thought so.",
+ "raw": {
+ "normalized_reference": "is that something you actually want, or something you think you should want?",
+ "normalized_hypothesis": "i thought so. i thought so. i thought so. i thought so. i thought so. i thought so. i thought so. i thought so. i thought so. i thought so. i thought so. i thought so. i thought so. i thought so. i thought so. i thought so. i thought so. i thought so. i thought so. i thought so. i thought so. i thought so. i thought so. i thought so. i thought so. i thought so. i thought so. i thought so. i thought so. i thought so. i thought so. i thought so. i thought so. i thought so. i thought so. i thought so. i thought so. i thought so. i thought so. i thought so. i thought so. i thought so. i thought so. i thought so. i thought so. i thought so. i thought so. i thought so. i thought so. i thought so. i thought so. i thought so. i thought so. i thought so. i thought so. i thought so. i thought so. i thought so. i thought so. i thought so. i thought so. i thought so. i thought so. i thought so. i thought so. i thought so. i thought so. i thought so. i thought so. i thought so. i thought so. i thought so. i thought so. i thought so. i thought so. i thought so. i thought so. i thought so. i thought so. i thought so. i thought so. i thought so. i thought so. i thought so. i thought so. i thought so. i thought so. i thought so. i thought so. i thought so. i thought so. i thought so. i thought so. i thought so. i thought so. i thought so. i thought so. i thought so. i thought so. i thought so. i thought so. i thought so. i thought so. i thought so. i thought so. i thought so. i thought so. i thought so. i thought so. i thought so. i thought so.",
+ "errors": 333,
+ "words": 13,
+ "wer": 25.615384615384617
+ },
+ "semantic": {
+ "normalized_reference": "is that something you actually want or something you think you should want",
+ "normalized_hypothesis": "i thought so i thought so i thought so i thought so i thought so i thought so i thought so i thought so i thought so i thought so i thought so i thought so i thought so i thought so i thought so i thought so i thought so i thought so i thought so i thought so i thought so i thought so i thought so i thought so i thought so i thought so i thought so i thought so i thought so i thought so i thought so i thought so i thought so i thought so i thought so i thought so i thought so i thought so i thought so i thought so i thought so i thought so i thought so i thought so i thought so i thought so i thought so i thought so i thought so i thought so i thought so i thought so i thought so i thought so i thought so i thought so i thought so i thought so i thought so i thought so i thought so i thought so i thought so i thought so i thought so i thought so i thought so i thought so i thought so i thought so i thought so i thought so i thought so i thought so i thought so i thought so i thought so i thought so i thought so i thought so i thought so i thought so i thought so i thought so i thought so i thought so i thought so i thought so i thought so i thought so i thought so i thought so i thought so i thought so i thought so i thought so i thought so i thought so i thought so i thought so i thought so i thought so i thought so i thought so i thought so i thought so i thought so i thought so i thought so i thought so i thought so",
+ "errors": 333,
+ "words": 13,
+ "wer": 25.615384615384617
+ }
+ },
+ {
+ "id": "release500_0155",
+ "category": "short",
+ "text": "This is fine.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0155.wav",
+ "render_seconds": 98.60779886704404,
+ "audio_seconds": 1.5325170068027212,
+ "rtf": 64.343689779188,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Shushu.",
+ "raw": {
+ "normalized_reference": "this is fine.",
+ "normalized_hypothesis": "shushu.",
+ "errors": 3,
+ "words": 3,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "this is fine",
+ "normalized_hypothesis": "shushu",
+ "errors": 3,
+ "words": 3,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0156",
+ "category": "technical",
+ "text": "The tricky part isn't the setup, it's making sure all dependencies line up.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0156.wav",
+ "render_seconds": 184.59595783997793,
+ "audio_seconds": 4.9458503401360545,
+ "rtf": 37.32340146687494,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The cock-a-poo isn't the same. It's like you're fully dependent on your head.",
+ "raw": {
+ "normalized_reference": "the tricky part isn't the setup, it's making sure all dependencies line up.",
+ "normalized_hypothesis": "the cock-a-poo isn't the same. it's like you're fully dependent on your head.",
+ "errors": 11,
+ "words": 13,
+ "wer": 0.8461538461538461
+ },
+ "semantic": {
+ "normalized_reference": "the tricky part is not the setup it is making sure all dependencies line up",
+ "normalized_hypothesis": "the cock a poo is not the same it is like you are fully dependent on your head",
+ "errors": 12,
+ "words": 15,
+ "wer": 0.8
+ }
+ },
+ {
+ "id": "release500_0157",
+ "category": "conversational",
+ "text": "He called it a minor setback. I called it a disaster.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0157.wav",
+ "render_seconds": 158.30804827006068,
+ "audio_seconds": 4.1099319727891155,
+ "rtf": 38.51841084431098,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you.",
+ "raw": {
+ "normalized_reference": "he called it a minor setback. i called it a disaster.",
+ "normalized_hypothesis": "thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you.",
+ "errors": 296,
+ "words": 11,
+ "wer": 26.90909090909091
+ },
+ "semantic": {
+ "normalized_reference": "he called it a minor setback i called it a disaster",
+ "normalized_hypothesis": "thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you",
+ "errors": 296,
+ "words": 11,
+ "wer": 26.90909090909091
+ }
+ },
+ {
+ "id": "release500_0158",
+ "category": "descriptive",
+ "text": "The whole place smelled like pine and old wood and cold air.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0158.wav",
+ "render_seconds": 107.55815625994,
+ "audio_seconds": 4.31891156462585,
+ "rtf": 24.903995983825578,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Thank you for watching. If you enjoyed this video, please like, share, and subscribe. Thank you for watching. If you enjoyed this video, please like, share, and subscribe. If you enjoyed this video, please like, share, and subscribe. If you enjoyed this video, please like, share, and subscribe. If you enjoyed this video, please like, share, and subscribe. If you enjoyed this video, please like, share, and subscribe. If you enjoyed this video, please like, share, and subscribe. If you enjoyed this video, please like, share, and subscribe. If you enjoyed this video, please like, share, and subscribe. If you enjoyed this video, please like, share, and subscribe. If you enjoyed this video, please like, share, and subscribe. If you enjoyed this video, please like, share, and subscribe. If you enjoyed this video, please like, share, and subscribe. If you enjoyed this video, please like, share, and subscribe. If you enjoyed this video, please like, share, and subscribe. If you enjoyed this video, please like, share, and subscribe. If you enjoyed this video, please like, share, and subscribe. If you enjoyed this video, please like, share, and subscribe. If you enjoyed this video, please like, share, and subscribe. If you enjoyed this video, please like, share, and subscribe. If you enjoyed this video, please like, share, and subscribe. If you enjoyed this video, please like, share, and subscribe. If you enjoyed this video, please like, share, and subscribe. If you enjoyed this video, please like, share, and subscribe. If you enjoyed this video, please like, share, and subscribe. If you enjoyed this video, please like, share, and subscribe. If you enjoyed this video, please like, share, and subscribe. If you enjoyed this video, please like, share, and subscribe. If you enjoyed this video, please like, share, and subscribe. If you enjoyed this video, please like, share, and subscribe. If you enjoyed this video, please like, share, and subscribe.",
+ "raw": {
+ "normalized_reference": "the whole place smelled like pine and old wood and cold air.",
+ "normalized_hypothesis": "thank you for watching. if you enjoyed this video, please like, share, and subscribe. thank you for watching. if you enjoyed this video, please like, share, and subscribe. if you enjoyed this video, please like, share, and subscribe. if you enjoyed this video, please like, share, and subscribe. if you enjoyed this video, please like, share, and subscribe. if you enjoyed this video, please like, share, and subscribe. if you enjoyed this video, please like, share, and subscribe. if you enjoyed this video, please like, share, and subscribe. if you enjoyed this video, please like, share, and subscribe. if you enjoyed this video, please like, share, and subscribe. if you enjoyed this video, please like, share, and subscribe. if you enjoyed this video, please like, share, and subscribe. if you enjoyed this video, please like, share, and subscribe. if you enjoyed this video, please like, share, and subscribe. if you enjoyed this video, please like, share, and subscribe. if you enjoyed this video, please like, share, and subscribe. if you enjoyed this video, please like, share, and subscribe. if you enjoyed this video, please like, share, and subscribe. if you enjoyed this video, please like, share, and subscribe. if you enjoyed this video, please like, share, and subscribe. if you enjoyed this video, please like, share, and subscribe. if you enjoyed this video, please like, share, and subscribe. if you enjoyed this video, please like, share, and subscribe. if you enjoyed this video, please like, share, and subscribe. if you enjoyed this video, please like, share, and subscribe. if you enjoyed this video, please like, share, and subscribe. if you enjoyed this video, please like, share, and subscribe. if you enjoyed this video, please like, share, and subscribe. if you enjoyed this video, please like, share, and subscribe. if you enjoyed this video, please like, share, and subscribe. if you enjoyed this video, please like, share, and subscribe.",
+ "errors": 315,
+ "words": 12,
+ "wer": 26.25
+ },
+ "semantic": {
+ "normalized_reference": "the whole place smelled like pine and old wood and cold air",
+ "normalized_hypothesis": "thank you for watching if you enjoyed this video please like share and subscribe thank you for watching if you enjoyed this video please like share and subscribe if you enjoyed this video please like share and subscribe if you enjoyed this video please like share and subscribe if you enjoyed this video please like share and subscribe if you enjoyed this video please like share and subscribe if you enjoyed this video please like share and subscribe if you enjoyed this video please like share and subscribe if you enjoyed this video please like share and subscribe if you enjoyed this video please like share and subscribe if you enjoyed this video please like share and subscribe if you enjoyed this video please like share and subscribe if you enjoyed this video please like share and subscribe if you enjoyed this video please like share and subscribe if you enjoyed this video please like share and subscribe if you enjoyed this video please like share and subscribe if you enjoyed this video please like share and subscribe if you enjoyed this video please like share and subscribe if you enjoyed this video please like share and subscribe if you enjoyed this video please like share and subscribe if you enjoyed this video please like share and subscribe if you enjoyed this video please like share and subscribe if you enjoyed this video please like share and subscribe if you enjoyed this video please like share and subscribe if you enjoyed this video please like share and subscribe if you enjoyed this video please like share and subscribe if you enjoyed this video please like share and subscribe if you enjoyed this video please like share and subscribe if you enjoyed this video please like share and subscribe if you enjoyed this video please like share and subscribe if you enjoyed this video please like share and subscribe",
+ "errors": 315,
+ "words": 12,
+ "wer": 26.25
+ }
+ },
+ {
+ "id": "release500_0159",
+ "category": "emotional",
+ "text": "I've never been so relieved in my entire life.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0159.wav",
+ "render_seconds": 103.70968468405772,
+ "audio_seconds": 3.2043537414965986,
+ "rtf": 32.365242120746615,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I've never been so lost on you until now.",
+ "raw": {
+ "normalized_reference": "i've never been so relieved in my entire life.",
+ "normalized_hypothesis": "i've never been so lost on you until now.",
+ "errors": 5,
+ "words": 9,
+ "wer": 0.5555555555555556
+ },
+ "semantic": {
+ "normalized_reference": "i have never been so relieved in my entire life",
+ "normalized_hypothesis": "i have never been so lost on you until now",
+ "errors": 5,
+ "words": 10,
+ "wer": 0.5
+ }
+ },
+ {
+ "id": "release500_0160",
+ "category": "narrative",
+ "text": "I wrote it all down that night so I wouldn't forget, and I still haven't looked at it.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0160.wav",
+ "render_seconds": 182.30416422104463,
+ "audio_seconds": 5.363809523809524,
+ "rtf": 33.987814707403565,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "There will be power at work so that you will forget. I will stay here for your time.",
+ "raw": {
+ "normalized_reference": "i wrote it all down that night so i wouldn't forget, and i still haven't looked at it.",
+ "normalized_hypothesis": "there will be power at work so that you will forget. i will stay here for your time.",
+ "errors": 17,
+ "words": 18,
+ "wer": 0.9444444444444444
+ },
+ "semantic": {
+ "normalized_reference": "i wrote it all down that night so i would not forget and i still have not looked at it",
+ "normalized_hypothesis": "there will be power at work so that you will forget i will stay here for your time",
+ "errors": 17,
+ "words": 20,
+ "wer": 0.85
+ }
+ },
+ {
+ "id": "release500_0161",
+ "category": "short",
+ "text": "That's kind of wild actually.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0161.wav",
+ "render_seconds": 86.89950804307591,
+ "audio_seconds": 2.2291156462585033,
+ "rtf": 38.983849128210935,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "That's kind of weird actually.",
+ "raw": {
+ "normalized_reference": "that's kind of wild actually.",
+ "normalized_hypothesis": "that's kind of weird actually.",
+ "errors": 1,
+ "words": 5,
+ "wer": 0.2
+ },
+ "semantic": {
+ "normalized_reference": "that is kind of wild actually",
+ "normalized_hypothesis": "that is kind of weird actually",
+ "errors": 1,
+ "words": 6,
+ "wer": 0.16666666666666666
+ }
+ },
+ {
+ "id": "release500_0162",
+ "category": "technical",
+ "text": "Version pinning will save you a lot of pain later down the line.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0162.wav",
+ "render_seconds": 177.2036938010715,
+ "audio_seconds": 4.31891156462585,
+ "rtf": 41.029711108804044,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I'm going to show you how to do this. I'm going to show you how to do this. I'm going to show you how to do this. I'm going to show you how to do this. I'm going to show you how to do this. I'm going to show you how to do this. I'm going to show you how to do this. I'm going to show you how to do this. I'm going to show you how to do this. I'm going to show you how to do this. I'm going to show you how to do this. I'm going to show you how to do this. I'm going to show you how to do this. I'm going to show you how to do this. I'm going to show you how to do this. I'm going to show you how to do this. I'm going to show you how to do this. I'm going to show you how to do this. I'm going to show you how to do this. I'm going to show you how to do this. I'm going to show you how to do this. I'm going to show you how to do this. I'm going to show you how to do this. I'm going to show you how to do this. I'm going to show you how to do this. I'm going to show you how to do this. I'm going to show you how to do this. I'm going to show you how to do this. I'm going to show you how to do this. I'm going to show you how to do this. I'm going to show you how to do this. I'm going to show you how to do this. I'm going to show you how to do this. I'm going to show you how to do this. I'm going to show you how to do this. I'm going to show you how to do this. I'm going to show you how to do this. I'm going to show you how to do this. I'm going to show you how to do this. I'm going to show you how to do this. I'm going to",
+ "raw": {
+ "normalized_reference": "version pinning will save you a lot of pain later down the line.",
+ "normalized_hypothesis": "i'm going to show you how to do this. i'm going to show you how to do this. i'm going to show you how to do this. i'm going to show you how to do this. i'm going to show you how to do this. i'm going to show you how to do this. i'm going to show you how to do this. i'm going to show you how to do this. i'm going to show you how to do this. i'm going to show you how to do this. i'm going to show you how to do this. i'm going to show you how to do this. i'm going to show you how to do this. i'm going to show you how to do this. i'm going to show you how to do this. i'm going to show you how to do this. i'm going to show you how to do this. i'm going to show you how to do this. i'm going to show you how to do this. i'm going to show you how to do this. i'm going to show you how to do this. i'm going to show you how to do this. i'm going to show you how to do this. i'm going to show you how to do this. i'm going to show you how to do this. i'm going to show you how to do this. i'm going to show you how to do this. i'm going to show you how to do this. i'm going to show you how to do this. i'm going to show you how to do this. i'm going to show you how to do this. i'm going to show you how to do this. i'm going to show you how to do this. i'm going to show you how to do this. i'm going to show you how to do this. i'm going to show you how to do this. i'm going to show you how to do this. i'm going to show you how to do this. i'm going to show you how to do this. i'm going to show you how to do this. i'm going to",
+ "errors": 362,
+ "words": 13,
+ "wer": 27.846153846153847
+ },
+ "semantic": {
+ "normalized_reference": "version pinning will save you a lot of pain later down the line",
+ "normalized_hypothesis": "i am going to show you how to do this i am going to show you how to do this i am going to show you how to do this i am going to show you how to do this i am going to show you how to do this i am going to show you how to do this i am going to show you how to do this i am going to show you how to do this i am going to show you how to do this i am going to show you how to do this i am going to show you how to do this i am going to show you how to do this i am going to show you how to do this i am going to show you how to do this i am going to show you how to do this i am going to show you how to do this i am going to show you how to do this i am going to show you how to do this i am going to show you how to do this i am going to show you how to do this i am going to show you how to do this i am going to show you how to do this i am going to show you how to do this i am going to show you how to do this i am going to show you how to do this i am going to show you how to do this i am going to show you how to do this i am going to show you how to do this i am going to show you how to do this i am going to show you how to do this i am going to show you how to do this i am going to show you how to do this i am going to show you how to do this i am going to show you how to do this i am going to show you how to do this i am going to show you how to do this i am going to show you how to do this i am going to show you how to do this i am going to show you how to do this i am going to show you how to do this i am going to",
+ "errors": 403,
+ "words": 13,
+ "wer": 31.0
+ }
+ },
+ {
+ "id": "release500_0163",
+ "category": "conversational",
+ "text": "At some point you just have to commit and see what happens.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0163.wav",
+ "render_seconds": 142.88773106306326,
+ "audio_seconds": 4.040272108843538,
+ "rtf": 35.36586824196912,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "That's it. You have to just have to go out and sell your hobbies.",
+ "raw": {
+ "normalized_reference": "at some point you just have to commit and see what happens.",
+ "normalized_hypothesis": "that's it. you have to just have to go out and sell your hobbies.",
+ "errors": 10,
+ "words": 12,
+ "wer": 0.8333333333333334
+ },
+ "semantic": {
+ "normalized_reference": "at some point you just have to commit and see what happens",
+ "normalized_hypothesis": "that is it you have to just have to go out and sell your hobbies",
+ "errors": 10,
+ "words": 12,
+ "wer": 0.8333333333333334
+ }
+ },
+ {
+ "id": "release500_0164",
+ "category": "emotional",
+ "text": "Nobody told me it was going to be this hard, and honestly, I'm grateful.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0164.wav",
+ "render_seconds": 89.99885719595477,
+ "audio_seconds": 5.085170068027211,
+ "rtf": 17.698298383729334,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I'd like to know what's going on in this world. I wish that I could see.",
+ "raw": {
+ "normalized_reference": "nobody told me it was going to be this hard, and honestly, i'm grateful.",
+ "normalized_hypothesis": "i'd like to know what's going on in this world. i wish that i could see.",
+ "errors": 14,
+ "words": 14,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "nobody told me it was going to be this hard and honestly i am grateful",
+ "normalized_hypothesis": "i would like to know what is going on in this world i wish that i could see",
+ "errors": 15,
+ "words": 15,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0165",
+ "category": "narrative",
+ "text": "Chapter seven. In which everything that could go wrong did, and a few things that couldn't.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0165.wav",
+ "render_seconds": 196.1990482499823,
+ "audio_seconds": 6.060408163265306,
+ "rtf": 32.373900068188085,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Help the servants, and let everything that they wound up in the house answer to it.",
+ "raw": {
+ "normalized_reference": "chapter seven. in which everything that could go wrong did, and a few things that couldn't.",
+ "normalized_hypothesis": "help the servants, and let everything that they wound up in the house answer to it.",
+ "errors": 15,
+ "words": 16,
+ "wer": 0.9375
+ },
+ "semantic": {
+ "normalized_reference": "chapter 7 in which everything that could go wrong did and a few things that could not",
+ "normalized_hypothesis": "help the servants and let everything that they wound up in the house answer to it",
+ "errors": 16,
+ "words": 17,
+ "wer": 0.9411764705882353
+ }
+ },
+ {
+ "id": "release500_0166",
+ "category": "short",
+ "text": "Wait, what?",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0166.wav",
+ "render_seconds": 85.48826819704846,
+ "audio_seconds": 1.3931972789115645,
+ "rtf": 61.361208129717404,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Blah. Blah.",
+ "raw": {
+ "normalized_reference": "wait, what?",
+ "normalized_hypothesis": "blah. blah.",
+ "errors": 2,
+ "words": 2,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "wait what",
+ "normalized_hypothesis": "blah blah",
+ "errors": 2,
+ "words": 2,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0167",
+ "category": "technical",
+ "text": "The graphics processor remained idle because the data loader exhausted every central processing core.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0167.wav",
+ "render_seconds": 252.69250211201143,
+ "audio_seconds": 7.174965986394557,
+ "rtf": 35.21863414978922,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The Earth exposes all I am, because the God of the Earth sustains all who are sent from possession to me.",
+ "raw": {
+ "normalized_reference": "the graphics processor remained idle because the data loader exhausted every central processing core.",
+ "normalized_hypothesis": "the earth exposes all i am, because the god of the earth sustains all who are sent from possession to me.",
+ "errors": 18,
+ "words": 14,
+ "wer": 1.2857142857142858
+ },
+ "semantic": {
+ "normalized_reference": "the graphics processor remained idle because the data loader exhausted every central processing core",
+ "normalized_hypothesis": "the earth exposes all i am because the god of the earth sustains all who are sent from possession to me",
+ "errors": 18,
+ "words": 14,
+ "wer": 1.2857142857142858
+ }
+ },
+ {
+ "id": "release500_0168",
+ "category": "conversational",
+ "text": "Could you speak a little more slowly? I am trying to write every detail down.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0168.wav",
+ "render_seconds": 211.395826210035,
+ "audio_seconds": 5.712108843537415,
+ "rtf": 37.00836801266571,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Thank you for watching. I hope you enjoyed the video.",
+ "raw": {
+ "normalized_reference": "could you speak a little more slowly? i am trying to write every detail down.",
+ "normalized_hypothesis": "thank you for watching. i hope you enjoyed the video.",
+ "errors": 13,
+ "words": 15,
+ "wer": 0.8666666666666667
+ },
+ "semantic": {
+ "normalized_reference": "could you speak a little more slowly i am trying to write every detail down",
+ "normalized_hypothesis": "thank you for watching i hope you enjoyed the video",
+ "errors": 13,
+ "words": 15,
+ "wer": 0.8666666666666667
+ }
+ },
+ {
+ "id": "release500_0169",
+ "category": "emotional",
+ "text": "I was so excited I could barely form a complete sentence.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0169.wav",
+ "render_seconds": 156.512929058983,
+ "audio_seconds": 4.31891156462585,
+ "rtf": 36.238975194793504,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Thank you for watching.",
+ "raw": {
+ "normalized_reference": "i was so excited i could barely form a complete sentence.",
+ "normalized_hypothesis": "thank you for watching.",
+ "errors": 11,
+ "words": 11,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "i was so excited i could barely form a complete sentence",
+ "normalized_hypothesis": "thank you for watching",
+ "errors": 11,
+ "words": 11,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0170",
+ "category": "narrative",
+ "text": "He looked at me like I had said something in a language he didn't speak.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0170.wav",
+ "render_seconds": 131.19891132391058,
+ "audio_seconds": 4.667210884353741,
+ "rtf": 28.110774202155515,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Thank you for watching.",
+ "raw": {
+ "normalized_reference": "he looked at me like i had said something in a language he didn't speak.",
+ "normalized_hypothesis": "thank you for watching.",
+ "errors": 15,
+ "words": 15,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "he looked at me like i had said something in a language he did not speak",
+ "normalized_hypothesis": "thank you for watching",
+ "errors": 16,
+ "words": 16,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0171",
+ "category": "short",
+ "text": "That's actually hilarious.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0171.wav",
+ "render_seconds": 85.59090774389915,
+ "audio_seconds": 2.159455782312925,
+ "rtf": 39.6354064968283,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Attraction Lounge.",
+ "raw": {
+ "normalized_reference": "that's actually hilarious.",
+ "normalized_hypothesis": "attraction lounge.",
+ "errors": 3,
+ "words": 3,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "that is actually hilarious",
+ "normalized_hypothesis": "attraction lounge",
+ "errors": 4,
+ "words": 4,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0172",
+ "category": "technical",
+ "text": "The encoder maps phonemes, stress, and phrase boundaries into a shared hidden representation.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0172.wav",
+ "render_seconds": 210.00737861997914,
+ "audio_seconds": 6.617687074829932,
+ "rtf": 31.734256432089776,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "On Google Maps, you'll notice scores and flashbacks flashing to how I happened to look for some time.",
+ "raw": {
+ "normalized_reference": "the encoder maps phonemes, stress, and phrase boundaries into a shared hidden representation.",
+ "normalized_hypothesis": "on google maps, you'll notice scores and flashbacks flashing to how i happened to look for some time.",
+ "errors": 16,
+ "words": 13,
+ "wer": 1.2307692307692308
+ },
+ "semantic": {
+ "normalized_reference": "the encoder maps phonemes stress and phrase boundaries into a shared hidden representation",
+ "normalized_hypothesis": "on google maps you will notice scores and flashbacks flashing to how i happened to look for some time",
+ "errors": 17,
+ "words": 13,
+ "wer": 1.3076923076923077
+ }
+ },
+ {
+ "id": "release500_0173",
+ "category": "conversational",
+ "text": "She showed up three hours late and acted like nothing had happened.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0173.wav",
+ "render_seconds": 195.00905901391525,
+ "audio_seconds": 4.597551020408163,
+ "rtf": 42.41585534304797,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I hope to have you here. I hope to do everything I can.",
+ "raw": {
+ "normalized_reference": "she showed up three hours late and acted like nothing had happened.",
+ "normalized_hypothesis": "i hope to have you here. i hope to do everything i can.",
+ "errors": 13,
+ "words": 12,
+ "wer": 1.0833333333333333
+ },
+ "semantic": {
+ "normalized_reference": "she showed up 3 hours late and acted like nothing had happened",
+ "normalized_hypothesis": "i hope to have you here i hope to do everything i can",
+ "errors": 13,
+ "words": 12,
+ "wer": 1.0833333333333333
+ }
+ },
+ {
+ "id": "release500_0174",
+ "category": "emotional",
+ "text": "Oh my god. This is the best thing I've ever tasted.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0174.wav",
+ "render_seconds": 158.50374704203568,
+ "audio_seconds": 3.9706122448979593,
+ "rtf": 39.91922083192716,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "This is the best thing I've ever tasted.",
+ "raw": {
+ "normalized_reference": "oh my god. this is the best thing i've ever tasted.",
+ "normalized_hypothesis": "this is the best thing i've ever tasted.",
+ "errors": 3,
+ "words": 11,
+ "wer": 0.2727272727272727
+ },
+ "semantic": {
+ "normalized_reference": "0 my god this is the best thing i have ever tasted",
+ "normalized_hypothesis": "this is the best thing i have ever tasted",
+ "errors": 3,
+ "words": 12,
+ "wer": 0.25
+ }
+ },
+ {
+ "id": "release500_0175",
+ "category": "narrative",
+ "text": "The note was short, which was somehow more frightening than a long one would have been.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0175.wav",
+ "render_seconds": 153.29321441799402,
+ "audio_seconds": 5.572789115646258,
+ "rtf": 27.507449364557036,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I hate the short. Watch the show. I hate the short. I hate the short. I hate the short. I hate the short. I hate the short. I hate the short. I hate the short. I hate the short. I hate the short. I hate the short. I hate the short. I hate the short. I hate the short. I hate the short. I hate the short. I hate the short. I hate the short. I hate the short. I hate the short. I hate the short. I hate the short. I hate the short. I hate the short. I hate the short. I hate the short. I hate the short. I hate the short. I hate the short. I hate the short. I hate the short. I hate the short. I hate the short. I hate the short. I hate the short. I hate the short. I hate the short. I hate the short. I hate the short. I hate the short. I hate the short. I hate the short. I hate the short. I hate the short. I hate the short. I hate the short. I hate the short. I hate the short. I hate the short. I hate the short. I hate the short. I hate the short. I hate the short. I hate the short. I hate the short. I hate the short. I hate the short. I hate the short. I hate the short. I hate the short. I hate the short. I hate the short. I hate the short. I hate the short. I hate the short. I hate the short. I hate the short. I hate the short. I hate the short. I hate the short. I hate the short. I hate the short. I hate the short. I hate the short. I hate the short. I hate the short. I hate the short. I hate the short. I hate the short. I hate the short. I hate the short. I hate the short. I hate the short. I hate the short. I hate the short. I hate the short. I hate the short. I hate the short.",
+ "raw": {
+ "normalized_reference": "the note was short, which was somehow more frightening than a long one would have been.",
+ "normalized_hypothesis": "i hate the short. watch the show. i hate the short. i hate the short. i hate the short. i hate the short. i hate the short. i hate the short. i hate the short. i hate the short. i hate the short. i hate the short. i hate the short. i hate the short. i hate the short. i hate the short. i hate the short. i hate the short. i hate the short. i hate the short. i hate the short. i hate the short. i hate the short. i hate the short. i hate the short. i hate the short. i hate the short. i hate the short. i hate the short. i hate the short. i hate the short. i hate the short. i hate the short. i hate the short. i hate the short. i hate the short. i hate the short. i hate the short. i hate the short. i hate the short. i hate the short. i hate the short. i hate the short. i hate the short. i hate the short. i hate the short. i hate the short. i hate the short. i hate the short. i hate the short. i hate the short. i hate the short. i hate the short. i hate the short. i hate the short. i hate the short. i hate the short. i hate the short. i hate the short. i hate the short. i hate the short. i hate the short. i hate the short. i hate the short. i hate the short. i hate the short. i hate the short. i hate the short. i hate the short. i hate the short. i hate the short. i hate the short. i hate the short. i hate the short. i hate the short. i hate the short. i hate the short. i hate the short. i hate the short. i hate the short. i hate the short. i hate the short. i hate the short. i hate the short. i hate the short. i hate the short. i hate the short. i hate the short. i hate the short.",
+ "errors": 353,
+ "words": 16,
+ "wer": 22.0625
+ },
+ "semantic": {
+ "normalized_reference": "the note was short which was somehow more frightening than a long one would have been",
+ "normalized_hypothesis": "i hate the short watch the show i hate the short i hate the short i hate the short i hate the short i hate the short i hate the short i hate the short i hate the short i hate the short i hate the short i hate the short i hate the short i hate the short i hate the short i hate the short i hate the short i hate the short i hate the short i hate the short i hate the short i hate the short i hate the short i hate the short i hate the short i hate the short i hate the short i hate the short i hate the short i hate the short i hate the short i hate the short i hate the short i hate the short i hate the short i hate the short i hate the short i hate the short i hate the short i hate the short i hate the short i hate the short i hate the short i hate the short i hate the short i hate the short i hate the short i hate the short i hate the short i hate the short i hate the short i hate the short i hate the short i hate the short i hate the short i hate the short i hate the short i hate the short i hate the short i hate the short i hate the short i hate the short i hate the short i hate the short i hate the short i hate the short i hate the short i hate the short i hate the short i hate the short i hate the short i hate the short i hate the short i hate the short i hate the short i hate the short i hate the short i hate the short i hate the short i hate the short i hate the short i hate the short i hate the short i hate the short i hate the short i hate the short i hate the short i hate the short",
+ "errors": 353,
+ "words": 16,
+ "wer": 22.0625
+ }
+ },
+ {
+ "id": "release500_0176",
+ "category": "short",
+ "text": "Noted.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0176.wav",
+ "render_seconds": 84.70032263104804,
+ "audio_seconds": 1.3235374149659864,
+ "rtf": 63.995412349733044,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Thank you.",
+ "raw": {
+ "normalized_reference": "noted.",
+ "normalized_hypothesis": "thank you.",
+ "errors": 2,
+ "words": 1,
+ "wer": 2.0
+ },
+ "semantic": {
+ "normalized_reference": "noted",
+ "normalized_hypothesis": "thank you",
+ "errors": 2,
+ "words": 1,
+ "wer": 2.0
+ }
+ },
+ {
+ "id": "release500_0177",
+ "category": "technical",
+ "text": "It's a simple process once you understand the underlying structure.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0177.wav",
+ "render_seconds": 157.00311071495526,
+ "audio_seconds": 4.597551020408163,
+ "rtf": 34.1492916594141,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I just showed the process which I will show you in the next clip.",
+ "raw": {
+ "normalized_reference": "it's a simple process once you understand the underlying structure.",
+ "normalized_hypothesis": "i just showed the process which i will show you in the next clip.",
+ "errors": 11,
+ "words": 10,
+ "wer": 1.1
+ },
+ "semantic": {
+ "normalized_reference": "it is a simple process once you understand the underlying structure",
+ "normalized_hypothesis": "i just showed the process which i will show you in the next clip",
+ "errors": 11,
+ "words": 11,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0178",
+ "category": "conversational",
+ "text": "You have to hear this story because it's almost too good to be true.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0178.wav",
+ "render_seconds": 166.39960729691666,
+ "audio_seconds": 4.458231292517007,
+ "rtf": 37.32413066504936,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I have to go to school because I'm too used to being at home.",
+ "raw": {
+ "normalized_reference": "you have to hear this story because it's almost too good to be true.",
+ "normalized_hypothesis": "i have to go to school because i'm too used to being at home.",
+ "errors": 10,
+ "words": 14,
+ "wer": 0.7142857142857143
+ },
+ "semantic": {
+ "normalized_reference": "you have to hear this story because it is almost too good to be true",
+ "normalized_hypothesis": "i have to go to school because i am too used to being at home",
+ "errors": 11,
+ "words": 15,
+ "wer": 0.7333333333333333
+ }
+ },
+ {
+ "id": "release500_0179",
+ "category": "emotional",
+ "text": "I have never felt more seen in my entire life than in that moment.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0179.wav",
+ "render_seconds": 129.49014355696272,
+ "audio_seconds": 4.458231292517007,
+ "rtf": 29.045182957265503,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I have never seen you shine like that. How you ask and I don't do it.",
+ "raw": {
+ "normalized_reference": "i have never felt more seen in my entire life than in that moment.",
+ "normalized_hypothesis": "i have never seen you shine like that. how you ask and i don't do it.",
+ "errors": 13,
+ "words": 14,
+ "wer": 0.9285714285714286
+ },
+ "semantic": {
+ "normalized_reference": "i have never felt more seen in my entire life than in that moment",
+ "normalized_hypothesis": "i have never seen you shine like that how you ask and i do not do it",
+ "errors": 14,
+ "words": 14,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0180",
+ "category": "narrative",
+ "text": "She knew. She had always known. She just waited for the right moment.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0180.wav",
+ "render_seconds": 184.3956411619438,
+ "audio_seconds": 5.224489795918367,
+ "rtf": 35.2944781911533,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Hello. How are you this morning? I just woke up from the walk we went.",
+ "raw": {
+ "normalized_reference": "she knew. she had always known. she just waited for the right moment.",
+ "normalized_hypothesis": "hello. how are you this morning? i just woke up from the walk we went.",
+ "errors": 13,
+ "words": 13,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "she knew she had always known she just waited for the right moment",
+ "normalized_hypothesis": "hello how are you this morning i just woke up from the walk we went",
+ "errors": 13,
+ "words": 13,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0181",
+ "category": "short",
+ "text": "Yeah, no. Hard pass.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0181.wav",
+ "render_seconds": 95.00375633803196,
+ "audio_seconds": 1.9504761904761905,
+ "rtf": 48.70798054440115,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I love hot dogs. I love hot dogs. I love hot dogs. I love hot dogs. I love hot dogs. I love hot dogs. I love hot dogs. I love hot dogs. I love hot dogs. I love hot dogs. I love hot dogs. I love hot dogs. I love hot dogs. I love hot dogs. I love hot dogs. I love hot dogs. I love hot dogs. I love hot dogs. I love hot dogs. I love hot dogs. I love hot dogs. I love hot dogs. I love hot dogs. I love hot dogs. I love hot dogs. I love hot dogs. I love hot dogs. I love hot dogs. I love hot dogs. I love hot dogs. I love hot dogs. I love hot dogs. I love hot dogs. I love hot dogs. I love hot dogs. I love hot dogs. I love hot dogs. I love hot dogs. I love hot dogs. I love hot dogs. I love hot dogs. I love hot dogs. I love hot dogs. I love hot dogs. I love hot dogs. I love hot dogs. I love hot dogs. I love hot dogs. I love hot dogs. I love hot dogs. I love hot dogs. I love hot dogs. I love hot dogs. I love hot dogs. I love hot dogs. I love hot dogs. I love hot dogs. I love hot dogs. I love hot dogs. I love hot dogs. I love hot dogs. I love hot dogs. I love hot dogs. I love hot dogs. I love hot dogs. I love hot dogs. I love hot dogs. I love hot dogs. I love hot dogs. I love hot dogs. I love hot dogs. I love hot dogs. I love hot dogs. I love hot dogs. I love hot dogs. I love hot dogs. I love hot dogs. I love hot dogs. I love hot dogs. I love hot dogs. I love hot dogs. I love hot dogs. I love hot dogs. I love hot dogs. I love hot dogs. I love hot dogs. I love hot dogs. I love hot dogs. I love hot dogs",
+ "raw": {
+ "normalized_reference": "yeah, no. hard pass.",
+ "normalized_hypothesis": "i love hot dogs. i love hot dogs. i love hot dogs. i love hot dogs. i love hot dogs. i love hot dogs. i love hot dogs. i love hot dogs. i love hot dogs. i love hot dogs. i love hot dogs. i love hot dogs. i love hot dogs. i love hot dogs. i love hot dogs. i love hot dogs. i love hot dogs. i love hot dogs. i love hot dogs. i love hot dogs. i love hot dogs. i love hot dogs. i love hot dogs. i love hot dogs. i love hot dogs. i love hot dogs. i love hot dogs. i love hot dogs. i love hot dogs. i love hot dogs. i love hot dogs. i love hot dogs. i love hot dogs. i love hot dogs. i love hot dogs. i love hot dogs. i love hot dogs. i love hot dogs. i love hot dogs. i love hot dogs. i love hot dogs. i love hot dogs. i love hot dogs. i love hot dogs. i love hot dogs. i love hot dogs. i love hot dogs. i love hot dogs. i love hot dogs. i love hot dogs. i love hot dogs. i love hot dogs. i love hot dogs. i love hot dogs. i love hot dogs. i love hot dogs. i love hot dogs. i love hot dogs. i love hot dogs. i love hot dogs. i love hot dogs. i love hot dogs. i love hot dogs. i love hot dogs. i love hot dogs. i love hot dogs. i love hot dogs. i love hot dogs. i love hot dogs. i love hot dogs. i love hot dogs. i love hot dogs. i love hot dogs. i love hot dogs. i love hot dogs. i love hot dogs. i love hot dogs. i love hot dogs. i love hot dogs. i love hot dogs. i love hot dogs. i love hot dogs. i love hot dogs. i love hot dogs. i love hot dogs. i love hot dogs. i love hot dogs. i love hot dogs. i love hot dogs",
+ "errors": 356,
+ "words": 4,
+ "wer": 89.0
+ },
+ "semantic": {
+ "normalized_reference": "yeah no hard pass",
+ "normalized_hypothesis": "i love hot dogs i love hot dogs i love hot dogs i love hot dogs i love hot dogs i love hot dogs i love hot dogs i love hot dogs i love hot dogs i love hot dogs i love hot dogs i love hot dogs i love hot dogs i love hot dogs i love hot dogs i love hot dogs i love hot dogs i love hot dogs i love hot dogs i love hot dogs i love hot dogs i love hot dogs i love hot dogs i love hot dogs i love hot dogs i love hot dogs i love hot dogs i love hot dogs i love hot dogs i love hot dogs i love hot dogs i love hot dogs i love hot dogs i love hot dogs i love hot dogs i love hot dogs i love hot dogs i love hot dogs i love hot dogs i love hot dogs i love hot dogs i love hot dogs i love hot dogs i love hot dogs i love hot dogs i love hot dogs i love hot dogs i love hot dogs i love hot dogs i love hot dogs i love hot dogs i love hot dogs i love hot dogs i love hot dogs i love hot dogs i love hot dogs i love hot dogs i love hot dogs i love hot dogs i love hot dogs i love hot dogs i love hot dogs i love hot dogs i love hot dogs i love hot dogs i love hot dogs i love hot dogs i love hot dogs i love hot dogs i love hot dogs i love hot dogs i love hot dogs i love hot dogs i love hot dogs i love hot dogs i love hot dogs i love hot dogs i love hot dogs i love hot dogs i love hot dogs i love hot dogs i love hot dogs i love hot dogs i love hot dogs i love hot dogs i love hot dogs i love hot dogs i love hot dogs i love hot dogs",
+ "errors": 356,
+ "words": 4,
+ "wer": 89.0
+ }
+ },
+ {
+ "id": "release500_0182",
+ "category": "technical",
+ "text": "Please verify the sample rate, hop length, and mel filter configuration before inference.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0182.wav",
+ "render_seconds": 227.49867438699584,
+ "audio_seconds": 6.199727891156463,
+ "rtf": 36.694945065493755,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The Father and the Son be one, and the Son be one, and the Holy Spirit be one, and the Holy Spirit be one, and the Holy Spirit be one, and the Holy Spirit be one, and the Holy Spirit be one, and the Holy Spirit be one, and the Holy Spirit be one, and the Holy Spirit be one, and the Holy Spirit be one, and the Holy Spirit be one, and the Holy Spirit be one, and the Holy Spirit be one, and the Holy Spirit be one, and the Holy Spirit be one, and the Holy Spirit be one, and the Holy Spirit be one, and the Holy Spirit be one, and the Holy Spirit be one, and the Holy Spirit be one, and the Holy Spirit be one, and the Holy Spirit be one, and the Holy Spirit be one, and the Holy Spirit be one, and the Holy Spirit be one, and the Holy Spirit be one, and the Holy Spirit be one, and the Holy Spirit be one, and the Holy Spirit be one, and the Holy Spirit be one, and the Holy Spirit be one, and the Holy Spirit be one, and the Holy Spirit be one, and the Holy Spirit be one, and the Holy Spirit be one, and the Holy Spirit be one, and the Holy Spirit be one, and the Holy Spirit be one, and the Holy Spirit be one, and the Holy Spirit be one, and the Holy Spirit be one, and the Holy Spirit be one, and the Holy Spirit be one, and the Holy Spirit be one, and the Holy Spirit be one, and the Holy Spirit be one, and the Holy Spirit be one, and the Holy Spirit be one, and the Holy Spirit be one, and the Holy Spirit be one, and the Holy Spirit be one, and the Holy Spirit be one, and the Holy Spirit be one, and the Holy Spirit be one, and the Holy Spirit be one, and the Holy Spirit be one, and the Holy Spirit be one, and the Holy Spirit be one, and the Holy Spirit be one, and the Holy Spirit be one, and the Holy Spirit be one, and the Holy Spirit be one, and the Holy",
+ "raw": {
+ "normalized_reference": "please verify the sample rate, hop length, and mel filter configuration before inference.",
+ "normalized_hypothesis": "the father and the son be one, and the son be one, and the holy spirit be one, and the holy spirit be one, and the holy spirit be one, and the holy spirit be one, and the holy spirit be one, and the holy spirit be one, and the holy spirit be one, and the holy spirit be one, and the holy spirit be one, and the holy spirit be one, and the holy spirit be one, and the holy spirit be one, and the holy spirit be one, and the holy spirit be one, and the holy spirit be one, and the holy spirit be one, and the holy spirit be one, and the holy spirit be one, and the holy spirit be one, and the holy spirit be one, and the holy spirit be one, and the holy spirit be one, and the holy spirit be one, and the holy spirit be one, and the holy spirit be one, and the holy spirit be one, and the holy spirit be one, and the holy spirit be one, and the holy spirit be one, and the holy spirit be one, and the holy spirit be one, and the holy spirit be one, and the holy spirit be one, and the holy spirit be one, and the holy spirit be one, and the holy spirit be one, and the holy spirit be one, and the holy spirit be one, and the holy spirit be one, and the holy spirit be one, and the holy spirit be one, and the holy spirit be one, and the holy spirit be one, and the holy spirit be one, and the holy spirit be one, and the holy spirit be one, and the holy spirit be one, and the holy spirit be one, and the holy spirit be one, and the holy spirit be one, and the holy spirit be one, and the holy spirit be one, and the holy spirit be one, and the holy spirit be one, and the holy spirit be one, and the holy spirit be one, and the holy spirit be one, and the holy spirit be one, and the holy spirit be one, and the holy spirit be one, and the holy spirit be one, and the holy",
+ "errors": 379,
+ "words": 13,
+ "wer": 29.153846153846153
+ },
+ "semantic": {
+ "normalized_reference": "please verify the sample rate hop length and mel filter configuration before inference",
+ "normalized_hypothesis": "the father and the son be one and the son be one and the holy spirit be one and the holy spirit be one and the holy spirit be one and the holy spirit be one and the holy spirit be one and the holy spirit be one and the holy spirit be one and the holy spirit be one and the holy spirit be one and the holy spirit be one and the holy spirit be one and the holy spirit be one and the holy spirit be one and the holy spirit be one and the holy spirit be one and the holy spirit be one and the holy spirit be one and the holy spirit be one and the holy spirit be one and the holy spirit be one and the holy spirit be one and the holy spirit be one and the holy spirit be one and the holy spirit be one and the holy spirit be one and the holy spirit be one and the holy spirit be one and the holy spirit be one and the holy spirit be one and the holy spirit be one and the holy spirit be one and the holy spirit be one and the holy spirit be one and the holy spirit be one and the holy spirit be one and the holy spirit be one and the holy spirit be one and the holy spirit be one and the holy spirit be one and the holy spirit be one and the holy spirit be one and the holy spirit be one and the holy spirit be one and the holy spirit be one and the holy spirit be one and the holy spirit be one and the holy spirit be one and the holy spirit be one and the holy spirit be one and the holy spirit be one and the holy spirit be one and the holy spirit be one and the holy spirit be one and the holy spirit be one and the holy spirit be one and the holy spirit be one and the holy spirit be one and the holy spirit be one and the holy spirit be one and the holy spirit be one and the holy spirit be one and the holy",
+ "errors": 379,
+ "words": 13,
+ "wer": 29.153846153846153
+ }
+ },
+ {
+ "id": "release500_0183",
+ "category": "conversational",
+ "text": "That's the worst idea I've ever heard and I'm completely in.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0183.wav",
+ "render_seconds": 140.69971582002472,
+ "audio_seconds": 3.8312925170068026,
+ "rtf": 36.723824974331734,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "That's the worst thing I've ever heard on YouTube, fat man.",
+ "raw": {
+ "normalized_reference": "that's the worst idea i've ever heard and i'm completely in.",
+ "normalized_hypothesis": "that's the worst thing i've ever heard on youtube, fat man.",
+ "errors": 5,
+ "words": 11,
+ "wer": 0.45454545454545453
+ },
+ "semantic": {
+ "normalized_reference": "that is the worst idea i have ever heard and i am completely in",
+ "normalized_hypothesis": "that is the worst thing i have ever heard on youtube fat man",
+ "errors": 6,
+ "words": 14,
+ "wer": 0.42857142857142855
+ }
+ },
+ {
+ "id": "release500_0184",
+ "category": "emotional",
+ "text": "I remember thinking: this is it. This is the thing I've been waiting for.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0184.wav",
+ "render_seconds": 167.0922068769578,
+ "audio_seconds": 5.154829931972789,
+ "rtf": 32.414688570144634,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The world is sinking. The ship is sinking. The ship is sinking. The ship is sinking. The ship is sinking. The ship is sinking. The ship is sinking. The ship is sinking. The ship is sinking. The ship is sinking. The ship is sinking. The ship is sinking. The ship is sinking. The ship is sinking. The ship is sinking. The ship is sinking. The ship is sinking. The ship is sinking. The ship is sinking. The ship is sinking. The ship is sinking. The ship is sinking. The ship is sinking. The ship is sinking. The ship is sinking. The ship is sinking. The ship is sinking. The ship is sinking. The ship is sinking. The ship is sinking. The ship is sinking. The ship is sinking. The ship is sinking. The ship is sinking. The ship is sinking. The ship is sinking. The ship is sinking. The ship is sinking. The ship is sinking. The ship is sinking. The ship is sinking. The ship is sinking. The ship is sinking. The ship is sinking. The ship is sinking. The ship is sinking. The ship is sinking. The ship is sinking. The ship is sinking. The ship is sinking. The ship is sinking. The ship is sinking. The ship is sinking. The ship is sinking. The ship is sinking. The ship is sinking. The ship is sinking. The ship is sinking. The ship is sinking. The ship is sinking. The ship is sinking. The ship is sinking. The ship is sinking. The ship is sinking. The ship is sinking. The ship is sinking. The ship is sinking. The ship is sinking. The ship is sinking. The ship is sinking. The ship is sinking. The ship is sinking. The ship is sinking. The ship is sinking. The ship is sinking. The ship is sinking. The ship is sinking. The ship is sinking. The ship is sinking. The ship is sinking. The ship is sinking. The ship is sinking. The ship is sinking. The ship is sinking. The ship is sinking. The ship is sinking. The ship is sinking. The ship is sinking. The ship is sinking",
+ "raw": {
+ "normalized_reference": "i remember thinking: this is it. this is the thing i've been waiting for.",
+ "normalized_hypothesis": "the world is sinking. the ship is sinking. the ship is sinking. the ship is sinking. the ship is sinking. the ship is sinking. the ship is sinking. the ship is sinking. the ship is sinking. the ship is sinking. the ship is sinking. the ship is sinking. the ship is sinking. the ship is sinking. the ship is sinking. the ship is sinking. the ship is sinking. the ship is sinking. the ship is sinking. the ship is sinking. the ship is sinking. the ship is sinking. the ship is sinking. the ship is sinking. the ship is sinking. the ship is sinking. the ship is sinking. the ship is sinking. the ship is sinking. the ship is sinking. the ship is sinking. the ship is sinking. the ship is sinking. the ship is sinking. the ship is sinking. the ship is sinking. the ship is sinking. the ship is sinking. the ship is sinking. the ship is sinking. the ship is sinking. the ship is sinking. the ship is sinking. the ship is sinking. the ship is sinking. the ship is sinking. the ship is sinking. the ship is sinking. the ship is sinking. the ship is sinking. the ship is sinking. the ship is sinking. the ship is sinking. the ship is sinking. the ship is sinking. the ship is sinking. the ship is sinking. the ship is sinking. the ship is sinking. the ship is sinking. the ship is sinking. the ship is sinking. the ship is sinking. the ship is sinking. the ship is sinking. the ship is sinking. the ship is sinking. the ship is sinking. the ship is sinking. the ship is sinking. the ship is sinking. the ship is sinking. the ship is sinking. the ship is sinking. the ship is sinking. the ship is sinking. the ship is sinking. the ship is sinking. the ship is sinking. the ship is sinking. the ship is sinking. the ship is sinking. the ship is sinking. the ship is sinking. the ship is sinking. the ship is sinking. the ship is sinking. the ship is sinking. the ship is sinking",
+ "errors": 353,
+ "words": 14,
+ "wer": 25.214285714285715
+ },
+ "semantic": {
+ "normalized_reference": "i remember thinking this is it this is the thing i have been waiting for",
+ "normalized_hypothesis": "the world is sinking the ship is sinking the ship is sinking the ship is sinking the ship is sinking the ship is sinking the ship is sinking the ship is sinking the ship is sinking the ship is sinking the ship is sinking the ship is sinking the ship is sinking the ship is sinking the ship is sinking the ship is sinking the ship is sinking the ship is sinking the ship is sinking the ship is sinking the ship is sinking the ship is sinking the ship is sinking the ship is sinking the ship is sinking the ship is sinking the ship is sinking the ship is sinking the ship is sinking the ship is sinking the ship is sinking the ship is sinking the ship is sinking the ship is sinking the ship is sinking the ship is sinking the ship is sinking the ship is sinking the ship is sinking the ship is sinking the ship is sinking the ship is sinking the ship is sinking the ship is sinking the ship is sinking the ship is sinking the ship is sinking the ship is sinking the ship is sinking the ship is sinking the ship is sinking the ship is sinking the ship is sinking the ship is sinking the ship is sinking the ship is sinking the ship is sinking the ship is sinking the ship is sinking the ship is sinking the ship is sinking the ship is sinking the ship is sinking the ship is sinking the ship is sinking the ship is sinking the ship is sinking the ship is sinking the ship is sinking the ship is sinking the ship is sinking the ship is sinking the ship is sinking the ship is sinking the ship is sinking the ship is sinking the ship is sinking the ship is sinking the ship is sinking the ship is sinking the ship is sinking the ship is sinking the ship is sinking the ship is sinking the ship is sinking the ship is sinking the ship is sinking the ship is sinking the ship is sinking",
+ "errors": 353,
+ "words": 15,
+ "wer": 23.533333333333335
+ }
+ },
+ {
+ "id": "release500_0185",
+ "category": "narrative",
+ "text": "We never talked about it again, which I think was the right call.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0185.wav",
+ "render_seconds": 169.90739757893607,
+ "audio_seconds": 4.458231292517007,
+ "rtf": 38.11094275528504,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I know what to do right now. What should I say? What should I do?",
+ "raw": {
+ "normalized_reference": "we never talked about it again, which i think was the right call.",
+ "normalized_hypothesis": "i know what to do right now. what should i say? what should i do?",
+ "errors": 14,
+ "words": 13,
+ "wer": 1.0769230769230769
+ },
+ "semantic": {
+ "normalized_reference": "we never talked about it again which i think was the right call",
+ "normalized_hypothesis": "i know what to do right now what should i say what should i do",
+ "errors": 14,
+ "words": 13,
+ "wer": 1.0769230769230769
+ }
+ },
+ {
+ "id": "release500_0186",
+ "category": "short",
+ "text": "In hindsight, obvious.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0186.wav",
+ "render_seconds": 55.21192568494007,
+ "audio_seconds": 1.9504761904761905,
+ "rtf": 28.306895492767126,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I'm here for a good time.",
+ "raw": {
+ "normalized_reference": "in hindsight, obvious.",
+ "normalized_hypothesis": "i'm here for a good time.",
+ "errors": 6,
+ "words": 3,
+ "wer": 2.0
+ },
+ "semantic": {
+ "normalized_reference": "in hindsight obvious",
+ "normalized_hypothesis": "i am here for a good time",
+ "errors": 7,
+ "words": 3,
+ "wer": 2.3333333333333335
+ }
+ },
+ {
+ "id": "release500_0187",
+ "category": "technical",
+ "text": "Once it's configured correctly, it basically runs itself.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0187.wav",
+ "render_seconds": 167.79994737298694,
+ "audio_seconds": 4.179591836734694,
+ "rtf": 40.14744834607597,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Winsor & Fowler Platinum, a passion to our Winsor & Fowler.",
+ "raw": {
+ "normalized_reference": "once it's configured correctly, it basically runs itself.",
+ "normalized_hypothesis": "winsor & fowler platinum, a passion to our winsor & fowler.",
+ "errors": 9,
+ "words": 8,
+ "wer": 1.125
+ },
+ "semantic": {
+ "normalized_reference": "once it is configured correctly it basically runs itself",
+ "normalized_hypothesis": "winsor fowler platinum a passion to our winsor fowler",
+ "errors": 9,
+ "words": 9,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0188",
+ "category": "conversational",
+ "text": "You got engaged? When? How? Tell me everything right now.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0188.wav",
+ "render_seconds": 167.60357385501266,
+ "audio_seconds": 4.597551020408163,
+ "rtf": 36.454967679756834,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Hello, folks. You're home to the other thing you're home to.",
+ "raw": {
+ "normalized_reference": "you got engaged? when? how? tell me everything right now.",
+ "normalized_hypothesis": "hello, folks. you're home to the other thing you're home to.",
+ "errors": 11,
+ "words": 10,
+ "wer": 1.1
+ },
+ "semantic": {
+ "normalized_reference": "you got engaged when how tell me everything right now",
+ "normalized_hypothesis": "hello folks you are home to the other thing you are home to",
+ "errors": 12,
+ "words": 10,
+ "wer": 1.2
+ }
+ },
+ {
+ "id": "release500_0189",
+ "category": "emotional",
+ "text": "It's hard to explain how much that moment meant to me.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0189.wav",
+ "render_seconds": 140.10762640601024,
+ "audio_seconds": 3.4829931972789114,
+ "rtf": 40.226213050163096,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "That's hard to explain how it looks that way at one time.",
+ "raw": {
+ "normalized_reference": "it's hard to explain how much that moment meant to me.",
+ "normalized_hypothesis": "that's hard to explain how it looks that way at one time.",
+ "errors": 7,
+ "words": 11,
+ "wer": 0.6363636363636364
+ },
+ "semantic": {
+ "normalized_reference": "it is hard to explain how much that moment meant to me",
+ "normalized_hypothesis": "that is hard to explain how it looks that way at one time",
+ "errors": 7,
+ "words": 12,
+ "wer": 0.5833333333333334
+ }
+ },
+ {
+ "id": "release500_0190",
+ "category": "narrative",
+ "text": "He told the story three different times that night and somehow it got better each time.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0190.wav",
+ "render_seconds": 201.40748736192472,
+ "audio_seconds": 5.642448979591837,
+ "rtf": 35.69504803506334,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I toured the school for different tours that night. I saw how the people act here.",
+ "raw": {
+ "normalized_reference": "he told the story three different times that night and somehow it got better each time.",
+ "normalized_hypothesis": "i toured the school for different tours that night. i saw how the people act here.",
+ "errors": 12,
+ "words": 16,
+ "wer": 0.75
+ },
+ "semantic": {
+ "normalized_reference": "he told the story 3 different times that night and somehow it got better each time",
+ "normalized_hypothesis": "i toured the school for different tours that night i saw how the people act here",
+ "errors": 12,
+ "words": 16,
+ "wer": 0.75
+ }
+ },
+ {
+ "id": "release500_0191",
+ "category": "short",
+ "text": "Right. Moving on.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0191.wav",
+ "render_seconds": 88.69991307798773,
+ "audio_seconds": 1.811156462585034,
+ "rtf": 48.97418578149112,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Yeah, who's on who?",
+ "raw": {
+ "normalized_reference": "right. moving on.",
+ "normalized_hypothesis": "yeah, who's on who?",
+ "errors": 3,
+ "words": 3,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "right moving on",
+ "normalized_hypothesis": "yeah who is on who",
+ "errors": 4,
+ "words": 3,
+ "wer": 1.3333333333333333
+ }
+ },
+ {
+ "id": "release500_0192",
+ "category": "conversational",
+ "text": "No, I did not move your keys; check beside the small ceramic bowl.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0192.wav",
+ "render_seconds": 177.68764486000873,
+ "audio_seconds": 4.667210884353741,
+ "rtf": 38.07148407535752,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "No, I don't know. I don't know how it goes. I don't know how it goes. I don't know how it goes. I don't know how it goes. I don't know how it goes. I don't know how it goes. I don't know how it goes. I don't know how it goes. I don't know how it goes. I don't know how it goes. I don't know how it goes. I don't know how it goes. I don't know how it goes. I don't know how it goes. I don't know how it goes. I don't know how it goes. I don't know how it goes. I don't know how it goes. I don't know how it goes. I don't know how it goes. I don't know how it goes. I don't know how it goes. I don't know how it goes. I don't know how it goes. I don't know how it goes. I don't know how it goes. I don't know how it goes. I don't know how it goes. I don't know how it goes. I don't know how it goes. I don't know how it goes. I don't know how it goes. I don't know how it goes. I don't know how it goes. I don't know how it goes. I don't know how it goes. I don't know how it goes. I don't know how it goes. I don't know how it goes. I don't know how it goes. I don't know how it goes. I don't know how it goes. I don't know how it goes. I don't know how it goes. I don't know how it goes. I don't know how it goes. I don't know how it goes. I don't know how it goes. I don't know how it goes. I don't know how it goes. I don't know how it goes. I don't know how it goes. I don't know how it goes. I don't know how it goes. I don't know how",
+ "raw": {
+ "normalized_reference": "no, i did not move your keys; check beside the small ceramic bowl.",
+ "normalized_hypothesis": "no, i don't know. i don't know how it goes. i don't know how it goes. i don't know how it goes. i don't know how it goes. i don't know how it goes. i don't know how it goes. i don't know how it goes. i don't know how it goes. i don't know how it goes. i don't know how it goes. i don't know how it goes. i don't know how it goes. i don't know how it goes. i don't know how it goes. i don't know how it goes. i don't know how it goes. i don't know how it goes. i don't know how it goes. i don't know how it goes. i don't know how it goes. i don't know how it goes. i don't know how it goes. i don't know how it goes. i don't know how it goes. i don't know how it goes. i don't know how it goes. i don't know how it goes. i don't know how it goes. i don't know how it goes. i don't know how it goes. i don't know how it goes. i don't know how it goes. i don't know how it goes. i don't know how it goes. i don't know how it goes. i don't know how it goes. i don't know how it goes. i don't know how it goes. i don't know how it goes. i don't know how it goes. i don't know how it goes. i don't know how it goes. i don't know how it goes. i don't know how it goes. i don't know how it goes. i don't know how it goes. i don't know how it goes. i don't know how it goes. i don't know how it goes. i don't know how it goes. i don't know how it goes. i don't know how it goes. i don't know how it goes. i don't know how it goes. i don't know how",
+ "errors": 330,
+ "words": 13,
+ "wer": 25.384615384615383
+ },
+ "semantic": {
+ "normalized_reference": "no i did not move your keys check beside the small ceramic bowl",
+ "normalized_hypothesis": "no i do not know i do not know how it goes i do not know how it goes i do not know how it goes i do not know how it goes i do not know how it goes i do not know how it goes i do not know how it goes i do not know how it goes i do not know how it goes i do not know how it goes i do not know how it goes i do not know how it goes i do not know how it goes i do not know how it goes i do not know how it goes i do not know how it goes i do not know how it goes i do not know how it goes i do not know how it goes i do not know how it goes i do not know how it goes i do not know how it goes i do not know how it goes i do not know how it goes i do not know how it goes i do not know how it goes i do not know how it goes i do not know how it goes i do not know how it goes i do not know how it goes i do not know how it goes i do not know how it goes i do not know how it goes i do not know how it goes i do not know how it goes i do not know how it goes i do not know how it goes i do not know how it goes i do not know how it goes i do not know how it goes i do not know how it goes i do not know how it goes i do not know how it goes i do not know how it goes i do not know how it goes i do not know how it goes i do not know how it goes i do not know how it goes i do not know how it goes i do not know how it goes i do not know how it goes i do not know how it goes i do not know how it goes i do not know how it goes i do not know how",
+ "errors": 385,
+ "words": 13,
+ "wer": 29.615384615384617
+ }
+ },
+ {
+ "id": "release500_0193",
+ "category": "emotional",
+ "text": "I was devastated. I'm not going to pretend otherwise.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0193.wav",
+ "render_seconds": 159.89793053700123,
+ "audio_seconds": 4.1099319727891155,
+ "rtf": 38.90524991548461,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The rest of the standard. The whole point of the challenge is a yes.",
+ "raw": {
+ "normalized_reference": "i was devastated. i'm not going to pretend otherwise.",
+ "normalized_hypothesis": "the rest of the standard. the whole point of the challenge is a yes.",
+ "errors": 14,
+ "words": 9,
+ "wer": 1.5555555555555556
+ },
+ "semantic": {
+ "normalized_reference": "i was devastated i am not going to pretend otherwise",
+ "normalized_hypothesis": "the rest of the standard the whole point of the challenge is a yes",
+ "errors": 14,
+ "words": 10,
+ "wer": 1.4
+ }
+ },
+ {
+ "id": "release500_0194",
+ "category": "narrative",
+ "text": "That was the last summer before everything changed, though we didn't know it yet.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0194.wav",
+ "render_seconds": 151.09806479304098,
+ "audio_seconds": 5.015510204081632,
+ "rtf": 30.126160379471877,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I wish the last show before everything changed. Do you have a new account?",
+ "raw": {
+ "normalized_reference": "that was the last summer before everything changed, though we didn't know it yet.",
+ "normalized_hypothesis": "i wish the last show before everything changed. do you have a new account?",
+ "errors": 9,
+ "words": 14,
+ "wer": 0.6428571428571429
+ },
+ "semantic": {
+ "normalized_reference": "that was the last summer before everything changed though we did not know it yet",
+ "normalized_hypothesis": "i wish the last show before everything changed do you have a new account",
+ "errors": 10,
+ "words": 15,
+ "wer": 0.6666666666666666
+ }
+ },
+ {
+ "id": "release500_0195",
+ "category": "short",
+ "text": "Bold of you to assume.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0195.wav",
+ "render_seconds": 100.08805331797339,
+ "audio_seconds": 1.8808163265306121,
+ "rtf": 53.21521932053707,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Be the fire to the show.",
+ "raw": {
+ "normalized_reference": "bold of you to assume.",
+ "normalized_hypothesis": "be the fire to the show.",
+ "errors": 5,
+ "words": 5,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "bold of you to assume",
+ "normalized_hypothesis": "be the fire to the show",
+ "errors": 5,
+ "words": 5,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0196",
+ "category": "conversational",
+ "text": "It's one of those decisions that makes total sense in the moment.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0196.wav",
+ "render_seconds": 154.00022177398205,
+ "audio_seconds": 4.388571428571429,
+ "rtf": 35.09119636776934,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "That's one of those distractions that makes totally sense to me.",
+ "raw": {
+ "normalized_reference": "it's one of those decisions that makes total sense in the moment.",
+ "normalized_hypothesis": "that's one of those distractions that makes totally sense to me.",
+ "errors": 6,
+ "words": 12,
+ "wer": 0.5
+ },
+ "semantic": {
+ "normalized_reference": "it is one of those decisions that makes total sense in the moment",
+ "normalized_hypothesis": "that is one of those distractions that makes totally sense to me",
+ "errors": 6,
+ "words": 13,
+ "wer": 0.46153846153846156
+ }
+ },
+ {
+ "id": "release500_0197",
+ "category": "emotional",
+ "text": "I was furious. Not the loud kind. The quiet kind that lasts for days.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0197.wav",
+ "render_seconds": 203.4922595010139,
+ "audio_seconds": 5.363809523809524,
+ "rtf": 37.938010027710334,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I wish you all the best. I love you all. I love you all. I love you all. I love you all. I love you all. I love you all. I love you all. I love you all. I love you all. I love you all. I love you all. I love you all. I love you all. I love you all. I love you all. I love you all. I love you all. I love you all. I love you all. I love you all. I love you all. I love you all. I love you all. I love you all. I love you all. I love you all. I love you all. I love you all. I love you all. I love you all. I love you all. I love you all. I love you all. I love you all. I love you all. I love you all. I love you all. I love you all. I love you all. I love you all. I love you all. I love you all. I love you all. I love you all. I love you all. I love you all. I love you all. I love you all. I love you all. I love you all. I love you all. I love you all. I love you all. I love you all. I love you all. I love you all. I love you all. I love you all. I love you all. I love you all. I love you all. I love you all. I love you all. I love you all. I love you all. I love you all. I love you all. I love you all. I love you all. I love you all. I love you all. I love you all. I love you all. I love you all. I love you all. I love you all. I love you all. I love you all. I love you all. I love you all. I love you all. I love you all. I love you all. I love you all. I love you all. I love you all. I love you all. I love",
+ "raw": {
+ "normalized_reference": "i was furious. not the loud kind. the quiet kind that lasts for days.",
+ "normalized_hypothesis": "i wish you all the best. i love you all. i love you all. i love you all. i love you all. i love you all. i love you all. i love you all. i love you all. i love you all. i love you all. i love you all. i love you all. i love you all. i love you all. i love you all. i love you all. i love you all. i love you all. i love you all. i love you all. i love you all. i love you all. i love you all. i love you all. i love you all. i love you all. i love you all. i love you all. i love you all. i love you all. i love you all. i love you all. i love you all. i love you all. i love you all. i love you all. i love you all. i love you all. i love you all. i love you all. i love you all. i love you all. i love you all. i love you all. i love you all. i love you all. i love you all. i love you all. i love you all. i love you all. i love you all. i love you all. i love you all. i love you all. i love you all. i love you all. i love you all. i love you all. i love you all. i love you all. i love you all. i love you all. i love you all. i love you all. i love you all. i love you all. i love you all. i love you all. i love you all. i love you all. i love you all. i love you all. i love you all. i love you all. i love you all. i love you all. i love you all. i love you all. i love you all. i love you all. i love you all. i love you all. i love you all. i love you all. i love you all. i love you all. i love you all. i love",
+ "errors": 354,
+ "words": 14,
+ "wer": 25.285714285714285
+ },
+ "semantic": {
+ "normalized_reference": "i was furious not the loud kind the quiet kind that lasts for days",
+ "normalized_hypothesis": "i wish you all the best i love you all i love you all i love you all i love you all i love you all i love you all i love you all i love you all i love you all i love you all i love you all i love you all i love you all i love you all i love you all i love you all i love you all i love you all i love you all i love you all i love you all i love you all i love you all i love you all i love you all i love you all i love you all i love you all i love you all i love you all i love you all i love you all i love you all i love you all i love you all i love you all i love you all i love you all i love you all i love you all i love you all i love you all i love you all i love you all i love you all i love you all i love you all i love you all i love you all i love you all i love you all i love you all i love you all i love you all i love you all i love you all i love you all i love you all i love you all i love you all i love you all i love you all i love you all i love you all i love you all i love you all i love you all i love you all i love you all i love you all i love you all i love you all i love you all i love you all i love you all i love you all i love you all i love you all i love you all i love you all i love you all i love you all i love you all i love you all i love you all i love you all i love you all i love",
+ "errors": 354,
+ "words": 14,
+ "wer": 25.285714285714285
+ }
+ },
+ {
+ "id": "release500_0198",
+ "category": "narrative",
+ "text": "That's the part of the story I usually leave out.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0198.wav",
+ "render_seconds": 126.10402616497595,
+ "audio_seconds": 3.3436734693877552,
+ "rtf": 37.71421680959364,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "That's the door of the school. I'll see you on the other.",
+ "raw": {
+ "normalized_reference": "that's the part of the story i usually leave out.",
+ "normalized_hypothesis": "that's the door of the school. i'll see you on the other.",
+ "errors": 8,
+ "words": 10,
+ "wer": 0.8
+ },
+ "semantic": {
+ "normalized_reference": "that is the part of the story i usually leave out",
+ "normalized_hypothesis": "that is the door of the school i will see you on the other",
+ "errors": 8,
+ "words": 11,
+ "wer": 0.7272727272727273
+ }
+ },
+ {
+ "id": "release500_0199",
+ "category": "short",
+ "text": "I've said too much.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0199.wav",
+ "render_seconds": 79.7018426869763,
+ "audio_seconds": 1.7414965986394557,
+ "rtf": 45.766292480412176,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I'll shut the door.",
+ "raw": {
+ "normalized_reference": "i've said too much.",
+ "normalized_hypothesis": "i'll shut the door.",
+ "errors": 4,
+ "words": 4,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "i have said too much",
+ "normalized_hypothesis": "i will shut the door",
+ "errors": 4,
+ "words": 5,
+ "wer": 0.8
+ }
+ },
+ {
+ "id": "release500_0200",
+ "category": "conversational",
+ "text": "I've been faking that I understood this the entire time.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0200.wav",
+ "render_seconds": 132.69919514399953,
+ "audio_seconds": 3.622312925170068,
+ "rtf": 36.63382978922763,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I've been thinking that I lost a person too long ago.",
+ "raw": {
+ "normalized_reference": "i've been faking that i understood this the entire time.",
+ "normalized_hypothesis": "i've been thinking that i lost a person too long ago.",
+ "errors": 7,
+ "words": 10,
+ "wer": 0.7
+ },
+ "semantic": {
+ "normalized_reference": "i have been faking that i understood this the entire time",
+ "normalized_hypothesis": "i have been thinking that i lost a person too long ago",
+ "errors": 7,
+ "words": 11,
+ "wer": 0.6363636363636364
+ }
+ },
+ {
+ "id": "release500_0201",
+ "category": "phonetic_balance",
+ "text": "Peace be unto you and grace before the Lord.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0201.wav",
+ "render_seconds": 141.20551825605799,
+ "audio_seconds": 3.0650340136054424,
+ "rtf": 46.06980465119079,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "That's about all I want to ask of you.",
+ "raw": {
+ "normalized_reference": "peace be unto you and grace before the lord.",
+ "normalized_hypothesis": "that's about all i want to ask of you.",
+ "errors": 9,
+ "words": 9,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "peace be unto you and grace before the lord",
+ "normalized_hypothesis": "that is about all i want to ask of you",
+ "errors": 10,
+ "words": 9,
+ "wer": 1.1111111111111112
+ }
+ },
+ {
+ "id": "release500_0202",
+ "category": "phonetic_balance",
+ "text": "From now on we're pals.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0202.wav",
+ "render_seconds": 103.29508127400186,
+ "audio_seconds": 1.8808163265306121,
+ "rtf": 54.92034486139422,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Peace.",
+ "raw": {
+ "normalized_reference": "from now on we're pals.",
+ "normalized_hypothesis": "peace.",
+ "errors": 5,
+ "words": 5,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "from now on we are pals",
+ "normalized_hypothesis": "peace",
+ "errors": 6,
+ "words": 6,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0203",
+ "category": "phonetic_balance",
+ "text": "Here we allow our solicitors to look after our legal work.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0203.wav",
+ "render_seconds": 151.49941342906095,
+ "audio_seconds": 4.179591836734694,
+ "rtf": 36.24741825207025,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I love you. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love you so much. I love",
+ "raw": {
+ "normalized_reference": "here we allow our solicitors to look after our legal work.",
+ "normalized_hypothesis": "i love you. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love you so much. i love",
+ "errors": 370,
+ "words": 11,
+ "wer": 33.63636363636363
+ },
+ "semantic": {
+ "normalized_reference": "here we allow our solicitors to look after our legal work",
+ "normalized_hypothesis": "i love you i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love you so much i love",
+ "errors": 370,
+ "words": 11,
+ "wer": 33.63636363636363
+ }
+ },
+ {
+ "id": "release500_0204",
+ "category": "phonetic_balance",
+ "text": "Her words sent a strange chill through Philip.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0204.wav",
+ "render_seconds": 122.99952331290115,
+ "audio_seconds": 3.2043537414965986,
+ "rtf": 38.38512637354889,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "A lot so to score a chase for failure.",
+ "raw": {
+ "normalized_reference": "her words sent a strange chill through philip.",
+ "normalized_hypothesis": "a lot so to score a chase for failure.",
+ "errors": 9,
+ "words": 8,
+ "wer": 1.125
+ },
+ "semantic": {
+ "normalized_reference": "her words sent a strange chill through philip",
+ "normalized_hypothesis": "a lot so to score a chase for failure",
+ "errors": 9,
+ "words": 8,
+ "wer": 1.125
+ }
+ },
+ {
+ "id": "release500_0205",
+ "category": "phonetic_balance",
+ "text": "It was introduced by Representative Dick of Ohio.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0205.wav",
+ "render_seconds": 156.49971805210225,
+ "audio_seconds": 3.7616326530612243,
+ "rtf": 41.604200220014164,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I wish I could ask all of the senators to be here.",
+ "raw": {
+ "normalized_reference": "it was introduced by representative dick of ohio.",
+ "normalized_hypothesis": "i wish i could ask all of the senators to be here.",
+ "errors": 11,
+ "words": 8,
+ "wer": 1.375
+ },
+ "semantic": {
+ "normalized_reference": "it was introduced by representative dick of ohio",
+ "normalized_hypothesis": "i wish i could ask all of the senators to be here",
+ "errors": 11,
+ "words": 8,
+ "wer": 1.375
+ }
+ },
+ {
+ "id": "release500_0206",
+ "category": "phonetic_balance",
+ "text": "It's the aurora borealis.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0206.wav",
+ "render_seconds": 44.29847009794321,
+ "audio_seconds": 2.020136054421769,
+ "rtf": 21.92845872978735,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry.",
+ "raw": {
+ "normalized_reference": "it's the aurora borealis.",
+ "normalized_hypothesis": "i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry.",
+ "errors": 222,
+ "words": 4,
+ "wer": 55.5
+ },
+ "semantic": {
+ "normalized_reference": "it is the aurora borealis",
+ "normalized_hypothesis": "i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry",
+ "errors": 333,
+ "words": 5,
+ "wer": 66.6
+ }
+ },
+ {
+ "id": "release500_0207",
+ "category": "phonetic_balance",
+ "text": "The boy grew and prospered.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0207.wav",
+ "render_seconds": 104.49678320402745,
+ "audio_seconds": 2.159455782312925,
+ "rtf": 48.39033244390132,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "A vampire passport.",
+ "raw": {
+ "normalized_reference": "the boy grew and prospered.",
+ "normalized_hypothesis": "a vampire passport.",
+ "errors": 5,
+ "words": 5,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "the boy grew and prospered",
+ "normalized_hypothesis": "a vampire passport",
+ "errors": 5,
+ "words": 5,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0208",
+ "category": "phonetic_balance",
+ "text": "His voice was passionately rebellious.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0208.wav",
+ "render_seconds": 124.50408441293985,
+ "audio_seconds": 2.995374149659864,
+ "rtf": 41.5654533264493,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I feel as if I'm in a palace.",
+ "raw": {
+ "normalized_reference": "his voice was passionately rebellious.",
+ "normalized_hypothesis": "i feel as if i'm in a palace.",
+ "errors": 8,
+ "words": 5,
+ "wer": 1.6
+ },
+ "semantic": {
+ "normalized_reference": "his voice was passionately rebellious",
+ "normalized_hypothesis": "i feel as if i am in a palace",
+ "errors": 9,
+ "words": 5,
+ "wer": 1.8
+ }
+ },
+ {
+ "id": "release500_0209",
+ "category": "phonetic_balance",
+ "text": "That's where they cut off the Scottish Chiefs and killed all hands.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0209.wav",
+ "render_seconds": 160.80715630296618,
+ "audio_seconds": 4.388571428571429,
+ "rtf": 36.642255667993595,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "That's when I closed the store. I'm so glad to be alive.",
+ "raw": {
+ "normalized_reference": "that's where they cut off the scottish chiefs and killed all hands.",
+ "normalized_hypothesis": "that's when i closed the store. i'm so glad to be alive.",
+ "errors": 11,
+ "words": 12,
+ "wer": 0.9166666666666666
+ },
+ "semantic": {
+ "normalized_reference": "that is where they cut off the scottish chiefs and killed all hands",
+ "normalized_hypothesis": "that is when i closed the store i am so glad to be alive",
+ "errors": 12,
+ "words": 13,
+ "wer": 0.9230769230769231
+ }
+ },
+ {
+ "id": "release500_0210",
+ "category": "phonetic_balance",
+ "text": "And Tom King patiently endured.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0210.wav",
+ "render_seconds": 99.49595871998463,
+ "audio_seconds": 2.4380952380952383,
+ "rtf": 40.80888931874369,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "A true compassion angle.",
+ "raw": {
+ "normalized_reference": "and tom king patiently endured.",
+ "normalized_hypothesis": "a true compassion angle.",
+ "errors": 5,
+ "words": 5,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "and tom king patiently endured",
+ "normalized_hypothesis": "a true compassion angle",
+ "errors": 5,
+ "words": 5,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0211",
+ "category": "phonetic_balance",
+ "text": "From that moment his friendship for Belize turns to hatred and jealousy.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0211.wav",
+ "render_seconds": 190.4977680540178,
+ "audio_seconds": 5.154829931972789,
+ "rtf": 36.95519940870542,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "For that reason, his friendship for the years turns to hatred and chastisement.",
+ "raw": {
+ "normalized_reference": "from that moment his friendship for belize turns to hatred and jealousy.",
+ "normalized_hypothesis": "for that reason, his friendship for the years turns to hatred and chastisement.",
+ "errors": 5,
+ "words": 12,
+ "wer": 0.4166666666666667
+ },
+ "semantic": {
+ "normalized_reference": "from that moment his friendship for belize turns to hatred and jealousy",
+ "normalized_hypothesis": "for that reason his friendship for the years turns to hatred and chastisement",
+ "errors": 5,
+ "words": 12,
+ "wer": 0.4166666666666667
+ }
+ },
+ {
+ "id": "release500_0212",
+ "category": "phonetic_balance",
+ "text": "Then you don't believe in altruism.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0212.wav",
+ "render_seconds": 73.53096668096259,
+ "audio_seconds": 2.4380952380952383,
+ "rtf": 30.15918555273856,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Thank you for watching.",
+ "raw": {
+ "normalized_reference": "then you don't believe in altruism.",
+ "normalized_hypothesis": "thank you for watching.",
+ "errors": 5,
+ "words": 6,
+ "wer": 0.8333333333333334
+ },
+ "semantic": {
+ "normalized_reference": "then you do not believe in altruism",
+ "normalized_hypothesis": "thank you for watching",
+ "errors": 6,
+ "words": 7,
+ "wer": 0.8571428571428571
+ }
+ },
+ {
+ "id": "release500_0213",
+ "category": "phonetic_balance",
+ "text": "What Pascal glimpsed with the vision of a seer, I have lived.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0213.wav",
+ "render_seconds": 162.4048843840137,
+ "audio_seconds": 4.31891156462585,
+ "rtf": 37.603197461646324,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The past is lost but the future is a show. The past is lost.",
+ "raw": {
+ "normalized_reference": "what pascal glimpsed with the vision of a seer, i have lived.",
+ "normalized_hypothesis": "the past is lost but the future is a show. the past is lost.",
+ "errors": 12,
+ "words": 12,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "what pascal glimpsed with the vision of a seer i have lived",
+ "normalized_hypothesis": "the past is lost but the future is a show the past is lost",
+ "errors": 12,
+ "words": 12,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0214",
+ "category": "phonetic_balance",
+ "text": "You were making them talk shop, Ruth charged him.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0214.wav",
+ "render_seconds": 111.89531640789937,
+ "audio_seconds": 3.274013605442177,
+ "rtf": 34.1768025098928,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you.",
+ "raw": {
+ "normalized_reference": "you were making them talk shop, ruth charged him.",
+ "normalized_hypothesis": "thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you.",
+ "errors": 295,
+ "words": 9,
+ "wer": 32.77777777777778
+ },
+ "semantic": {
+ "normalized_reference": "you were making them talk shop ruth charged him",
+ "normalized_hypothesis": "thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you",
+ "errors": 295,
+ "words": 9,
+ "wer": 32.77777777777778
+ }
+ },
+ {
+ "id": "release500_0215",
+ "category": "phonetic_balance",
+ "text": "Philip snatched at the letter which Gregson held out to him.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0215.wav",
+ "render_seconds": 151.79221775499173,
+ "audio_seconds": 3.9706122448979593,
+ "rtf": 38.22891997324524,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "They expect that the little witch quacks her head out to hell.",
+ "raw": {
+ "normalized_reference": "philip snatched at the letter which gregson held out to him.",
+ "normalized_hypothesis": "they expect that the little witch quacks her head out to hell.",
+ "errors": 9,
+ "words": 11,
+ "wer": 0.8181818181818182
+ },
+ "semantic": {
+ "normalized_reference": "philip snatched at the letter which gregson held out to him",
+ "normalized_hypothesis": "they expect that the little witch quacks her head out to hell",
+ "errors": 9,
+ "words": 11,
+ "wer": 0.8181818181818182
+ }
+ },
+ {
+ "id": "release500_0216",
+ "category": "phonetic_balance",
+ "text": "For a time the exciting thrill of his adventure was gone.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0216.wav",
+ "render_seconds": 145.6996357330354,
+ "audio_seconds": 4.040272108843538,
+ "rtf": 36.06183737330988,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "For to not shatter free of his invention is good.",
+ "raw": {
+ "normalized_reference": "for a time the exciting thrill of his adventure was gone.",
+ "normalized_hypothesis": "for to not shatter free of his invention is good.",
+ "errors": 8,
+ "words": 11,
+ "wer": 0.7272727272727273
+ },
+ "semantic": {
+ "normalized_reference": "for a time the exciting thrill of his adventure was gone",
+ "normalized_hypothesis": "for to not shatter free of his invention is good",
+ "errors": 8,
+ "words": 11,
+ "wer": 0.7272727272727273
+ }
+ },
+ {
+ "id": "release500_0217",
+ "category": "phonetic_balance",
+ "text": "Obviously, it was a disease that could be contracted by contact.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0217.wav",
+ "render_seconds": 180.42634402099065,
+ "audio_seconds": 5.085170068027211,
+ "rtf": 35.48088689411069,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Oh, my God. Oh, my God. Oh, my God. Oh, my God. Oh, my God. Oh, my God. Oh, my God. Oh, my God. Oh, my God. Oh, my God. Oh, my God. Oh, my God. Oh, my God. Oh, my God. Oh, my God. Oh, my God. Oh, my God. Oh, my God. Oh, my God. Oh, my God. Oh, my God. Oh, my God. Oh, my God. Oh, my God. Oh, my God. Oh, my God. Oh, my God. Oh, my God. Oh, my God. Oh, my God. Oh, my God. Oh, my God. Oh, my God. Oh, my God. Oh, my God. Oh, my God. Oh, my God. Oh, my God. Oh, my God. Oh, my God. Oh, my God. Oh, my God. Oh, my God. Oh, my God. Oh, my God. Oh, my God. Oh, my God. Oh, my God. Oh, my God. Oh, my God. Oh, my God. Oh, my God. Oh, my God. Oh, my God. Oh, my God. Oh, my God. Oh, my God. Oh, my God. Oh, my God. Oh, my God. Oh, my God. Oh, my God. Oh, my God. Oh, my God. Oh, my God. Oh, my God. Oh, my God. Oh, my God. Oh, my God. Oh, my God. Oh, my God. Oh, my God. Oh, my God. Oh, my God. Oh, my God. Oh, my God. Oh, my God. Oh, my God. Oh, my God. Oh, my God. Oh, my God. Oh, my God. Oh, my God. Oh, my God. Oh, my God. Oh, my God. Oh, my God. Oh, my God. Oh, my God",
+ "raw": {
+ "normalized_reference": "obviously, it was a disease that could be contracted by contact.",
+ "normalized_hypothesis": "oh, my god. oh, my god. oh, my god. oh, my god. oh, my god. oh, my god. oh, my god. oh, my god. oh, my god. oh, my god. oh, my god. oh, my god. oh, my god. oh, my god. oh, my god. oh, my god. oh, my god. oh, my god. oh, my god. oh, my god. oh, my god. oh, my god. oh, my god. oh, my god. oh, my god. oh, my god. oh, my god. oh, my god. oh, my god. oh, my god. oh, my god. oh, my god. oh, my god. oh, my god. oh, my god. oh, my god. oh, my god. oh, my god. oh, my god. oh, my god. oh, my god. oh, my god. oh, my god. oh, my god. oh, my god. oh, my god. oh, my god. oh, my god. oh, my god. oh, my god. oh, my god. oh, my god. oh, my god. oh, my god. oh, my god. oh, my god. oh, my god. oh, my god. oh, my god. oh, my god. oh, my god. oh, my god. oh, my god. oh, my god. oh, my god. oh, my god. oh, my god. oh, my god. oh, my god. oh, my god. oh, my god. oh, my god. oh, my god. oh, my god. oh, my god. oh, my god. oh, my god. oh, my god. oh, my god. oh, my god. oh, my god. oh, my god. oh, my god. oh, my god. oh, my god. oh, my god. oh, my god. oh, my god. oh, my god",
+ "errors": 267,
+ "words": 11,
+ "wer": 24.272727272727273
+ },
+ "semantic": {
+ "normalized_reference": "obviously it was a disease that could be contracted by contact",
+ "normalized_hypothesis": "0 my god 0 my god 0 my god 0 my god 0 my god 0 my god 0 my god 0 my god 0 my god 0 my god 0 my god 0 my god 0 my god 0 my god 0 my god 0 my god 0 my god 0 my god 0 my god 0 my god 0 my god 0 my god 0 my god 0 my god 0 my god 0 my god 0 my god 0 my god 0 my god 0 my god 0 my god 0 my god 0 my god 0 my god 0 my god 0 my god 0 my god 0 my god 0 my god 0 my god 0 my god 0 my god 0 my god 0 my god 0 my god 0 my god 0 my god 0 my god 0 my god 0 my god 0 my god 0 my god 0 my god 0 my god 0 my god 0 my god 0 my god 0 my god 0 my god 0 my god 0 my god 0 my god 0 my god 0 my god 0 my god 0 my god 0 my god 0 my god 0 my god 0 my god 0 my god 0 my god 0 my god 0 my god 0 my god 0 my god 0 my god 0 my god 0 my god 0 my god 0 my god 0 my god 0 my god 0 my god 0 my god 0 my god 0 my god 0 my god 0 my god",
+ "errors": 267,
+ "words": 11,
+ "wer": 24.272727272727273
+ }
+ },
+ {
+ "id": "release500_0218",
+ "category": "phonetic_balance",
+ "text": "Blind with rage, he darted in.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0218.wav",
+ "render_seconds": 72.27195741294418,
+ "audio_seconds": 2.2291156462585033,
+ "rtf": 32.42180706696409,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Thank you for watching.",
+ "raw": {
+ "normalized_reference": "blind with rage, he darted in.",
+ "normalized_hypothesis": "thank you for watching.",
+ "errors": 6,
+ "words": 6,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "blind with rage he darted in",
+ "normalized_hypothesis": "thank you for watching",
+ "errors": 6,
+ "words": 6,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0219",
+ "category": "phonetic_balance",
+ "text": "You live on an income which your father earned.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0219.wav",
+ "render_seconds": 118.79254577099346,
+ "audio_seconds": 3.1346938775510202,
+ "rtf": 37.896059523298696,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I love you and I love you too. I love you too.",
+ "raw": {
+ "normalized_reference": "you live on an income which your father earned.",
+ "normalized_hypothesis": "i love you and i love you too. i love you too.",
+ "errors": 11,
+ "words": 9,
+ "wer": 1.2222222222222223
+ },
+ "semantic": {
+ "normalized_reference": "you live on an income which your father earned",
+ "normalized_hypothesis": "i love you and i love you too i love you too",
+ "errors": 11,
+ "words": 9,
+ "wer": 1.2222222222222223
+ }
+ },
+ {
+ "id": "release500_0220",
+ "category": "phonetic_balance",
+ "text": "Something about it seemed to fascinate him, to challenge his presence.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0220.wav",
+ "render_seconds": 174.38811884389725,
+ "audio_seconds": 4.806530612244898,
+ "rtf": 36.2814955135486,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "So tell a bird that's out of fashion at home to try and catch questions.",
+ "raw": {
+ "normalized_reference": "something about it seemed to fascinate him, to challenge his presence.",
+ "normalized_hypothesis": "so tell a bird that's out of fashion at home to try and catch questions.",
+ "errors": 14,
+ "words": 11,
+ "wer": 1.2727272727272727
+ },
+ "semantic": {
+ "normalized_reference": "something about it seemed to fascinate him to challenge his presence",
+ "normalized_hypothesis": "so tell a bird that is out of fashion at home to try and catch questions",
+ "errors": 15,
+ "words": 11,
+ "wer": 1.3636363636363635
+ }
+ },
+ {
+ "id": "release500_0221",
+ "category": "phonetic_balance",
+ "text": "Her beautiful hair was done up in shining coils.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0221.wav",
+ "render_seconds": 152.71167974406853,
+ "audio_seconds": 3.4829931972789114,
+ "rtf": 43.84495492651968,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I'll pass you how the stirrup and sawdust.",
+ "raw": {
+ "normalized_reference": "her beautiful hair was done up in shining coils.",
+ "normalized_hypothesis": "i'll pass you how the stirrup and sawdust.",
+ "errors": 9,
+ "words": 9,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "her beautiful hair was done up in shining coils",
+ "normalized_hypothesis": "i will pass you how the stirrup and sawdust",
+ "errors": 9,
+ "words": 9,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0222",
+ "category": "phonetic_balance",
+ "text": "It does, was her audacious answer.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0222.wav",
+ "render_seconds": 138.70300833694637,
+ "audio_seconds": 2.716734693877551,
+ "rtf": 51.05504363364162,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "At first, this whole talk was so-so.",
+ "raw": {
+ "normalized_reference": "it does, was her audacious answer.",
+ "normalized_hypothesis": "at first, this whole talk was so-so.",
+ "errors": 8,
+ "words": 6,
+ "wer": 1.3333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "it does was her audacious answer",
+ "normalized_hypothesis": "at 1st this whole talk was so so",
+ "errors": 8,
+ "words": 6,
+ "wer": 1.3333333333333333
+ }
+ },
+ {
+ "id": "release500_0223",
+ "category": "phonetic_balance",
+ "text": "He caught himself with a jerk.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0223.wav",
+ "render_seconds": 105.6640717900591,
+ "audio_seconds": 2.2291156462585033,
+ "rtf": 47.401790018123435,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I could have a show for the call.",
+ "raw": {
+ "normalized_reference": "he caught himself with a jerk.",
+ "normalized_hypothesis": "i could have a show for the call.",
+ "errors": 8,
+ "words": 6,
+ "wer": 1.3333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "he caught himself with a jerk",
+ "normalized_hypothesis": "i could have a show for the call",
+ "errors": 8,
+ "words": 6,
+ "wer": 1.3333333333333333
+ }
+ },
+ {
+ "id": "release500_0224",
+ "category": "phonetic_balance",
+ "text": "He did not believe in the burning of daylight for such a luxury.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0224.wav",
+ "render_seconds": 94.21396563295275,
+ "audio_seconds": 4.388571428571429,
+ "rtf": 21.468026023133763,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I can't believe I'm the boy who's trying to look for shit to look for.",
+ "raw": {
+ "normalized_reference": "he did not believe in the burning of daylight for such a luxury.",
+ "normalized_hypothesis": "i can't believe i'm the boy who's trying to look for shit to look for.",
+ "errors": 13,
+ "words": 13,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "he did not believe in the burning of daylight for such a luxury",
+ "normalized_hypothesis": "i can not believe i am the boy who is trying to look for shit to look for",
+ "errors": 14,
+ "words": 13,
+ "wer": 1.0769230769230769
+ }
+ },
+ {
+ "id": "release500_0225",
+ "category": "phonetic_balance",
+ "text": "He had comparatively no advantages at first.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0225.wav",
+ "render_seconds": 150.89984683995135,
+ "audio_seconds": 3.4133333333333336,
+ "rtf": 44.208939503892,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I took all of you who thought that I should first.",
+ "raw": {
+ "normalized_reference": "he had comparatively no advantages at first.",
+ "normalized_hypothesis": "i took all of you who thought that i should first.",
+ "errors": 10,
+ "words": 7,
+ "wer": 1.4285714285714286
+ },
+ "semantic": {
+ "normalized_reference": "he had comparatively no advantages at 1st",
+ "normalized_hypothesis": "i took all of you who thought that i should 1st",
+ "errors": 10,
+ "words": 7,
+ "wer": 1.4285714285714286
+ }
+ },
+ {
+ "id": "release500_0226",
+ "category": "phonetic_balance",
+ "text": "Such men believe, when they come together.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0226.wav",
+ "render_seconds": 130.6986567300046,
+ "audio_seconds": 2.8560544217687074,
+ "rtf": 45.761962984257515,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Such lovely eyes will not hurt at all.",
+ "raw": {
+ "normalized_reference": "such men believe, when they come together.",
+ "normalized_hypothesis": "such lovely eyes will not hurt at all.",
+ "errors": 7,
+ "words": 7,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "such men believe when they come together",
+ "normalized_hypothesis": "such lovely eyes will not hurt at all",
+ "errors": 7,
+ "words": 7,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0227",
+ "category": "phonetic_balance",
+ "text": "The ship should be in within a week or ten days.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0227.wav",
+ "render_seconds": 143.38940703007393,
+ "audio_seconds": 3.274013605442177,
+ "rtf": 43.796215993643756,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I hope somebody will tell me what to do next.",
+ "raw": {
+ "normalized_reference": "the ship should be in within a week or ten days.",
+ "normalized_hypothesis": "i hope somebody will tell me what to do next.",
+ "errors": 11,
+ "words": 11,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "the ship should be in within a week or 10 days",
+ "normalized_hypothesis": "i hope somebody will tell me what to do next",
+ "errors": 11,
+ "words": 11,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0228",
+ "category": "phonetic_balance",
+ "text": "Such a dog the wise driver kills, or turns loose.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0228.wav",
+ "render_seconds": 155.70062283601146,
+ "audio_seconds": 3.622312925170068,
+ "rtf": 42.98375818226729,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Subscribe to watch more videos like this.",
+ "raw": {
+ "normalized_reference": "such a dog the wise driver kills, or turns loose.",
+ "normalized_hypothesis": "subscribe to watch more videos like this.",
+ "errors": 10,
+ "words": 10,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "such a dog the wise driver kills or turns loose",
+ "normalized_hypothesis": "subscribe to watch more videos like this",
+ "errors": 10,
+ "words": 10,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0229",
+ "category": "phonetic_balance",
+ "text": "You have heard always how he was the lover of the Princess Naomi.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0229.wav",
+ "render_seconds": 154.11101181805134,
+ "audio_seconds": 4.458231292517007,
+ "rtf": 34.56774709663932,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I love her as hell. I wish to live with the princess now.",
+ "raw": {
+ "normalized_reference": "you have heard always how he was the lover of the princess naomi.",
+ "normalized_hypothesis": "i love her as hell. i wish to live with the princess now.",
+ "errors": 11,
+ "words": 13,
+ "wer": 0.8461538461538461
+ },
+ "semantic": {
+ "normalized_reference": "you have heard always how he was the lover of the princess naomi",
+ "normalized_hypothesis": "i love her as hell i wish to live with the princess now",
+ "errors": 11,
+ "words": 13,
+ "wer": 0.8461538461538461
+ }
+ },
+ {
+ "id": "release500_0230",
+ "category": "phonetic_balance",
+ "text": "I want to know how all this is possible.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0230.wav",
+ "render_seconds": 81.68238760298118,
+ "audio_seconds": 2.9257142857142857,
+ "rtf": 27.918784825237708,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I'll be up tonight. Have a good night. This is Percival.",
+ "raw": {
+ "normalized_reference": "i want to know how all this is possible.",
+ "normalized_hypothesis": "i'll be up tonight. have a good night. this is percival.",
+ "errors": 9,
+ "words": 9,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "i want to know how all this is possible",
+ "normalized_hypothesis": "i will be up tonight have a good night this is percival",
+ "errors": 9,
+ "words": 9,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0231",
+ "category": "phonetic_balance",
+ "text": "It was my idea to a tee.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0231.wav",
+ "render_seconds": 89.59984059398994,
+ "audio_seconds": 2.159455782312925,
+ "rtf": 41.49186166619368,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I wish you all the best.",
+ "raw": {
+ "normalized_reference": "it was my idea to a tee.",
+ "normalized_hypothesis": "i wish you all the best.",
+ "errors": 7,
+ "words": 7,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "it was my idea to a tee",
+ "normalized_hypothesis": "i wish you all the best",
+ "errors": 7,
+ "words": 7,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0232",
+ "category": "phonetic_balance",
+ "text": "They likewise are disinclined to being eaten.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0232.wav",
+ "render_seconds": 139.69522951799445,
+ "audio_seconds": 3.2043537414965986,
+ "rtf": 43.59544569281841,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Thank you for watching.",
+ "raw": {
+ "normalized_reference": "they likewise are disinclined to being eaten.",
+ "normalized_hypothesis": "thank you for watching.",
+ "errors": 7,
+ "words": 7,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "they likewise are disinclined to being eaten",
+ "normalized_hypothesis": "thank you for watching",
+ "errors": 7,
+ "words": 7,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0233",
+ "category": "phonetic_balance",
+ "text": "This state of mind comes of an undue prominence of the ego.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0233.wav",
+ "render_seconds": 165.50575402600225,
+ "audio_seconds": 4.1099319727891155,
+ "rtf": 40.26970643839766,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The start of the long course of the mental purge of the heart.",
+ "raw": {
+ "normalized_reference": "this state of mind comes of an undue prominence of the ego.",
+ "normalized_hypothesis": "the start of the long course of the mental purge of the heart.",
+ "errors": 9,
+ "words": 12,
+ "wer": 0.75
+ },
+ "semantic": {
+ "normalized_reference": "this state of mind comes of an undue prominence of the ego",
+ "normalized_hypothesis": "the start of the long course of the mental purge of the heart",
+ "errors": 9,
+ "words": 12,
+ "wer": 0.75
+ }
+ },
+ {
+ "id": "release500_0234",
+ "category": "phonetic_balance",
+ "text": "Then it is as I said, Womble announced with finality.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0234.wav",
+ "render_seconds": 143.99591594492085,
+ "audio_seconds": 3.8312925170068026,
+ "rtf": 37.58416129954433,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Follow us on social media. Follow us on social media. Follow us on social media. Follow us on social media. Follow us on social media. Follow us on social media. Follow us on social media. Follow us on social media. Follow us on social media. Follow us on social media. Follow us on social media. Follow us on social media. Follow us on social media. Follow us on social media. Follow us on social media. Follow us on social media. Follow us on social media. Follow us on social media. Follow us on social media. Follow us on social media. Follow us on social media. Follow us on social media. Follow us on social media. Follow us on social media. Follow us on social media. Follow us on social media. Follow us on social media. Follow us on social media. Follow us on social media. Follow us on social media. Follow us on social media. Follow us on social media. Follow us on social media. Follow us on social media. Follow us on social media. Follow us on social media. Follow us on social media. Follow us on social media. Follow us on social media. Follow us on social media. Follow us on social media. Follow us on social media. Follow us on social media. Follow us on social media. Follow us on social media. Follow us on social media. Follow us on social media. Follow us on social media. Follow us on social media. Follow us on social media. Follow us on social media. Follow us on social media. Follow us on social media. Follow us on social media. Follow us on social media. Follow us on social media. Follow us on social media. Follow us on social media. Follow us on social media. Follow us on social media. Follow us on social media. Follow us on social media. Follow us on social media. Follow us on social media. Follow us on social media. Follow us on social media. Follow us on social media. Follow us on social media. Follow us on social media. Follow us on social media. Follow us on social media. Follow us on social media. Follow us on social media. Follow us on social media.",
+ "raw": {
+ "normalized_reference": "then it is as i said, womble announced with finality.",
+ "normalized_hypothesis": "follow us on social media. follow us on social media. follow us on social media. follow us on social media. follow us on social media. follow us on social media. follow us on social media. follow us on social media. follow us on social media. follow us on social media. follow us on social media. follow us on social media. follow us on social media. follow us on social media. follow us on social media. follow us on social media. follow us on social media. follow us on social media. follow us on social media. follow us on social media. follow us on social media. follow us on social media. follow us on social media. follow us on social media. follow us on social media. follow us on social media. follow us on social media. follow us on social media. follow us on social media. follow us on social media. follow us on social media. follow us on social media. follow us on social media. follow us on social media. follow us on social media. follow us on social media. follow us on social media. follow us on social media. follow us on social media. follow us on social media. follow us on social media. follow us on social media. follow us on social media. follow us on social media. follow us on social media. follow us on social media. follow us on social media. follow us on social media. follow us on social media. follow us on social media. follow us on social media. follow us on social media. follow us on social media. follow us on social media. follow us on social media. follow us on social media. follow us on social media. follow us on social media. follow us on social media. follow us on social media. follow us on social media. follow us on social media. follow us on social media. follow us on social media. follow us on social media. follow us on social media. follow us on social media. follow us on social media. follow us on social media. follow us on social media. follow us on social media. follow us on social media. follow us on social media. follow us on social media.",
+ "errors": 370,
+ "words": 10,
+ "wer": 37.0
+ },
+ "semantic": {
+ "normalized_reference": "then it is as i said womble announced with finality",
+ "normalized_hypothesis": "follow us on social media follow us on social media follow us on social media follow us on social media follow us on social media follow us on social media follow us on social media follow us on social media follow us on social media follow us on social media follow us on social media follow us on social media follow us on social media follow us on social media follow us on social media follow us on social media follow us on social media follow us on social media follow us on social media follow us on social media follow us on social media follow us on social media follow us on social media follow us on social media follow us on social media follow us on social media follow us on social media follow us on social media follow us on social media follow us on social media follow us on social media follow us on social media follow us on social media follow us on social media follow us on social media follow us on social media follow us on social media follow us on social media follow us on social media follow us on social media follow us on social media follow us on social media follow us on social media follow us on social media follow us on social media follow us on social media follow us on social media follow us on social media follow us on social media follow us on social media follow us on social media follow us on social media follow us on social media follow us on social media follow us on social media follow us on social media follow us on social media follow us on social media follow us on social media follow us on social media follow us on social media follow us on social media follow us on social media follow us on social media follow us on social media follow us on social media follow us on social media follow us on social media follow us on social media follow us on social media follow us on social media follow us on social media follow us on social media follow us on social media",
+ "errors": 370,
+ "words": 10,
+ "wer": 37.0
+ }
+ },
+ {
+ "id": "release500_0235",
+ "category": "phonetic_balance",
+ "text": "I can see that knife now.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0235.wav",
+ "render_seconds": 94.4952214009827,
+ "audio_seconds": 2.020136054421769,
+ "rtf": 46.77666199469442,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Let us know if you have any questions.",
+ "raw": {
+ "normalized_reference": "i can see that knife now.",
+ "normalized_hypothesis": "let us know if you have any questions.",
+ "errors": 8,
+ "words": 6,
+ "wer": 1.3333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "i can see that knife now",
+ "normalized_hypothesis": "let us know if you have any questions",
+ "errors": 8,
+ "words": 6,
+ "wer": 1.3333333333333333
+ }
+ },
+ {
+ "id": "release500_0236",
+ "category": "phonetic_balance",
+ "text": "If I meet her again I shall apologize, said Eileen.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0236.wav",
+ "render_seconds": 153.71506068203598,
+ "audio_seconds": 3.7616326530612243,
+ "rtf": 40.86392129676521,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I feel like we're out of here, baby. I feel like we're out of here, baby. I feel like we're out of here, baby. I feel like we're out of here, baby. I feel like we're out of here, baby. I feel like we're out of here, baby. I feel like we're out of here, baby. I feel like we're out of here, baby. I feel like we're out of here, baby. I feel like we're out of here, baby. I feel like we're out of here, baby. I feel like we're out of here, baby. I feel like we're out of here, baby. I feel like we're out of here, baby. I feel like we're out of here, baby. I feel like we're out of here, baby. I feel like we're out of here, baby. I feel like we're out of here, baby. I feel like we're out of here, baby. I feel like we're out of here, baby. I feel like we're out of here, baby. I feel like we're out of here, baby. I feel like we're out of here, baby. I feel like we're out of here, baby. I feel like we're out of here, baby. I feel like we're out of here, baby. I feel like we're out of here, baby. I feel like we're out of here, baby. I feel like we're out of here, baby. I feel like we're out of here, baby. I feel like we're out of here, baby. I feel like we're out of here, baby. I feel like we're out of here, baby. I feel like we're out of here, baby. I feel like we're out of here, baby. I feel like we're out of here, baby. I feel like we're out of here, baby. I feel like we're out of here, baby. I feel like we're out of here, baby. I feel like we're out of here, baby. I feel like we",
+ "raw": {
+ "normalized_reference": "if i meet her again i shall apologize, said eileen.",
+ "normalized_hypothesis": "i feel like we're out of here, baby. i feel like we're out of here, baby. i feel like we're out of here, baby. i feel like we're out of here, baby. i feel like we're out of here, baby. i feel like we're out of here, baby. i feel like we're out of here, baby. i feel like we're out of here, baby. i feel like we're out of here, baby. i feel like we're out of here, baby. i feel like we're out of here, baby. i feel like we're out of here, baby. i feel like we're out of here, baby. i feel like we're out of here, baby. i feel like we're out of here, baby. i feel like we're out of here, baby. i feel like we're out of here, baby. i feel like we're out of here, baby. i feel like we're out of here, baby. i feel like we're out of here, baby. i feel like we're out of here, baby. i feel like we're out of here, baby. i feel like we're out of here, baby. i feel like we're out of here, baby. i feel like we're out of here, baby. i feel like we're out of here, baby. i feel like we're out of here, baby. i feel like we're out of here, baby. i feel like we're out of here, baby. i feel like we're out of here, baby. i feel like we're out of here, baby. i feel like we're out of here, baby. i feel like we're out of here, baby. i feel like we're out of here, baby. i feel like we're out of here, baby. i feel like we're out of here, baby. i feel like we're out of here, baby. i feel like we're out of here, baby. i feel like we're out of here, baby. i feel like we're out of here, baby. i feel like we",
+ "errors": 322,
+ "words": 10,
+ "wer": 32.2
+ },
+ "semantic": {
+ "normalized_reference": "if i meet her again i shall apologize said eileen",
+ "normalized_hypothesis": "i feel like we are out of here baby i feel like we are out of here baby i feel like we are out of here baby i feel like we are out of here baby i feel like we are out of here baby i feel like we are out of here baby i feel like we are out of here baby i feel like we are out of here baby i feel like we are out of here baby i feel like we are out of here baby i feel like we are out of here baby i feel like we are out of here baby i feel like we are out of here baby i feel like we are out of here baby i feel like we are out of here baby i feel like we are out of here baby i feel like we are out of here baby i feel like we are out of here baby i feel like we are out of here baby i feel like we are out of here baby i feel like we are out of here baby i feel like we are out of here baby i feel like we are out of here baby i feel like we are out of here baby i feel like we are out of here baby i feel like we are out of here baby i feel like we are out of here baby i feel like we are out of here baby i feel like we are out of here baby i feel like we are out of here baby i feel like we are out of here baby i feel like we are out of here baby i feel like we are out of here baby i feel like we are out of here baby i feel like we are out of here baby i feel like we are out of here baby i feel like we are out of here baby i feel like we are out of here baby i feel like we are out of here baby i feel like we are out of here baby i feel like we",
+ "errors": 362,
+ "words": 10,
+ "wer": 36.2
+ }
+ },
+ {
+ "id": "release500_0237",
+ "category": "phonetic_balance",
+ "text": "A wildly exciting time was his during the week preceding Thursday the eighteenth.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0237.wav",
+ "render_seconds": 187.4109411360696,
+ "audio_seconds": 5.572789115646258,
+ "rtf": 33.62964886108671,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Thank you for watching. If you like this video, please like, share, and subscribe.",
+ "raw": {
+ "normalized_reference": "a wildly exciting time was his during the week preceding thursday the eighteenth.",
+ "normalized_hypothesis": "thank you for watching. if you like this video, please like, share, and subscribe.",
+ "errors": 14,
+ "words": 13,
+ "wer": 1.0769230769230769
+ },
+ "semantic": {
+ "normalized_reference": "a wildly exciting time was his during the week preceding thursday the 18th",
+ "normalized_hypothesis": "thank you for watching if you like this video please like share and subscribe",
+ "errors": 14,
+ "words": 13,
+ "wer": 1.0769230769230769
+ }
+ },
+ {
+ "id": "release500_0238",
+ "category": "phonetic_balance",
+ "text": "In what bucolic school of fence he had been taught was beyond imagining.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0238.wav",
+ "render_seconds": 170.8976426590234,
+ "audio_seconds": 4.876190476190477,
+ "rtf": 35.047368123432534,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I hope that the next school of thought should happen to this part of the world.",
+ "raw": {
+ "normalized_reference": "in what bucolic school of fence he had been taught was beyond imagining.",
+ "normalized_hypothesis": "i hope that the next school of thought should happen to this part of the world.",
+ "errors": 14,
+ "words": 13,
+ "wer": 1.0769230769230769
+ },
+ "semantic": {
+ "normalized_reference": "in what bucolic school of fence he had been taught was beyond imagining",
+ "normalized_hypothesis": "i hope that the next school of thought should happen to this part of the world",
+ "errors": 14,
+ "words": 13,
+ "wer": 1.0769230769230769
+ }
+ },
+ {
+ "id": "release500_0239",
+ "category": "phonetic_balance",
+ "text": "So she said, the irate skipper dashed on.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0239.wav",
+ "render_seconds": 124.01602367509622,
+ "audio_seconds": 3.0650340136054424,
+ "rtf": 40.46154891743418,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Slash. Slash. Slash. Slash. Slash. Slash. Slash. Slash. Slash. Slash. Slash. Slash. Slash. Slash. Slash. Slash. Slash. Slash. Slash. Slash. Slash. Slash. Slash. Slash. Slash. Slash. Slash. Slash. Slash. Slash. Slash. Slash. Slash. Slash. Slash. Slash. Slash. Slash. Slash. Slash. Slash. Slash. Slash. Slash. Slash. Slash. Slash. Slash. Slash. Slash. Slash. Slash. Slash. Slash. Slash. Slash. Slash. Slash. Slash. Slash. Slash. Slash. Slash. Slash. Slash. Slash. Slash. Slash. Slash. Slash. Slash. Slash. Slash. Slash. Slash. Slash. Slash. Slash. Slash. Slash. Slash. Slash. Slash. Slash. Slash. Slash. Slash. Slash. Slash. Slash. Slash. Slash. Slash. Slash. Slash. Slash. Slash. Slash. Slash. Slash. Slash. Slash. Slash. Slash. Slash. Slash. Slash. Slash. Slash. Slash. Slash. Slash. Slash. Slash. Slash. Slash. Slash. Slash. Slash. Slash. Slash. Slash. Slash. Slash. Slash. Slash. Slash. Slash. Slash. Slash. Slash. Slash. Slash. Slash. Slash. Slash. Slash. Slash. Slash. Slash. Slash. Slash. Slash. Slash. Slash. Slash. Slash. Slash.",
+ "raw": {
+ "normalized_reference": "so she said, the irate skipper dashed on.",
+ "normalized_hypothesis": "slash. slash. slash. slash. slash. slash. slash. slash. slash. slash. slash. slash. slash. slash. slash. slash. slash. slash. slash. slash. slash. slash. slash. slash. slash. slash. slash. slash. slash. slash. slash. slash. slash. slash. slash. slash. slash. slash. slash. slash. slash. slash. slash. slash. slash. slash. slash. slash. slash. slash. slash. slash. slash. slash. slash. slash. slash. slash. slash. slash. slash. slash. slash. slash. slash. slash. slash. slash. slash. slash. slash. slash. slash. slash. slash. slash. slash. slash. slash. slash. slash. slash. slash. slash. slash. slash. slash. slash. slash. slash. slash. slash. slash. slash. slash. slash. slash. slash. slash. slash. slash. slash. slash. slash. slash. slash. slash. slash. slash. slash. slash. slash. slash. slash. slash. slash. slash. slash. slash. slash. slash. slash. slash. slash. slash. slash. slash. slash. slash. slash. slash. slash. slash. slash. slash. slash. slash. slash. slash. slash. slash. slash. slash. slash. slash. slash. slash. slash.",
+ "errors": 148,
+ "words": 8,
+ "wer": 18.5
+ },
+ "semantic": {
+ "normalized_reference": "so she said the irate skipper dashed on",
+ "normalized_hypothesis": "slash slash slash slash slash slash slash slash slash slash slash slash slash slash slash slash slash slash slash slash slash slash slash slash slash slash slash slash slash slash slash slash slash slash slash slash slash slash slash slash slash slash slash slash slash slash slash slash slash slash slash slash slash slash slash slash slash slash slash slash slash slash slash slash slash slash slash slash slash slash slash slash slash slash slash slash slash slash slash slash slash slash slash slash slash slash slash slash slash slash slash slash slash slash slash slash slash slash slash slash slash slash slash slash slash slash slash slash slash slash slash slash slash slash slash slash slash slash slash slash slash slash slash slash slash slash slash slash slash slash slash slash slash slash slash slash slash slash slash slash slash slash slash slash slash slash slash slash",
+ "errors": 148,
+ "words": 8,
+ "wer": 18.5
+ }
+ },
+ {
+ "id": "release500_0240",
+ "category": "phonetic_balance",
+ "text": "It is not an attempt to smash the market.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0240.wav",
+ "render_seconds": 111.5925358888926,
+ "audio_seconds": 3.0650340136054424,
+ "rtf": 36.40825367468752,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I've never been told to swallow food.",
+ "raw": {
+ "normalized_reference": "it is not an attempt to smash the market.",
+ "normalized_hypothesis": "i've never been told to swallow food.",
+ "errors": 8,
+ "words": 9,
+ "wer": 0.8888888888888888
+ },
+ "semantic": {
+ "normalized_reference": "it is not an attempt to smash the market",
+ "normalized_hypothesis": "i have never been told to swallow food",
+ "errors": 8,
+ "words": 9,
+ "wer": 0.8888888888888888
+ }
+ },
+ {
+ "id": "release500_0241",
+ "category": "phonetic_balance",
+ "text": "He destroyed everything that had belonged to the woman.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0241.wav",
+ "render_seconds": 137.30032210005447,
+ "audio_seconds": 3.622312925170068,
+ "rtf": 37.904047755235894,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I can screw up everything. I can't be here today.",
+ "raw": {
+ "normalized_reference": "he destroyed everything that had belonged to the woman.",
+ "normalized_hypothesis": "i can screw up everything. i can't be here today.",
+ "errors": 10,
+ "words": 9,
+ "wer": 1.1111111111111112
+ },
+ "semantic": {
+ "normalized_reference": "he destroyed everything that had belonged to the woman",
+ "normalized_hypothesis": "i can screw up everything i can not be here today",
+ "errors": 10,
+ "words": 9,
+ "wer": 1.1111111111111112
+ }
+ },
+ {
+ "id": "release500_0242",
+ "category": "phonetic_balance",
+ "text": "Did Thorpe go to see any one in Churchill.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0242.wav",
+ "render_seconds": 65.49456997099333,
+ "audio_seconds": 2.995374149659864,
+ "rtf": 21.865238430541467,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I'm too old for this. I'm not willing to talk to you.",
+ "raw": {
+ "normalized_reference": "did thorpe go to see any one in churchill.",
+ "normalized_hypothesis": "i'm too old for this. i'm not willing to talk to you.",
+ "errors": 12,
+ "words": 9,
+ "wer": 1.3333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "did thorpe go to see any one in churchill",
+ "normalized_hypothesis": "i am too old for this i am not willing to talk to you",
+ "errors": 14,
+ "words": 9,
+ "wer": 1.5555555555555556
+ }
+ },
+ {
+ "id": "release500_0243",
+ "category": "phonetic_balance",
+ "text": "Many other similar disconcerting omissions will be noticed in the Manuscript.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0243.wav",
+ "render_seconds": 208.19875388999935,
+ "audio_seconds": 5.712108843537415,
+ "rtf": 36.44866713727838,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello,",
+ "raw": {
+ "normalized_reference": "many other similar disconcerting omissions will be noticed in the manuscript.",
+ "normalized_hypothesis": "hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello, hello,",
+ "errors": 222,
+ "words": 11,
+ "wer": 20.181818181818183
+ },
+ "semantic": {
+ "normalized_reference": "many other similar disconcerting omissions will be noticed in the manuscript",
+ "normalized_hypothesis": "hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello hello",
+ "errors": 222,
+ "words": 11,
+ "wer": 20.181818181818183
+ }
+ },
+ {
+ "id": "release500_0244",
+ "category": "phonetic_balance",
+ "text": "The ringing of the big bell aroused him.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0244.wav",
+ "render_seconds": 126.7997991410084,
+ "audio_seconds": 2.716734693877551,
+ "rtf": 46.67360395064161,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The one I left at that day was true.",
+ "raw": {
+ "normalized_reference": "the ringing of the big bell aroused him.",
+ "normalized_hypothesis": "the one i left at that day was true.",
+ "errors": 8,
+ "words": 8,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "the ringing of the big bell aroused him",
+ "normalized_hypothesis": "the one i left at that day was true",
+ "errors": 8,
+ "words": 8,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0245",
+ "category": "phonetic_balance",
+ "text": "A sinewy hand, dripping with water, was clutching the rail.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0245.wav",
+ "render_seconds": 123.57655678398442,
+ "audio_seconds": 4.249251700680272,
+ "rtf": 29.081957363034242,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "You're sure alive. Hope I'm not little. You're still trying to lie.",
+ "raw": {
+ "normalized_reference": "a sinewy hand, dripping with water, was clutching the rail.",
+ "normalized_hypothesis": "you're sure alive. hope i'm not little. you're still trying to lie.",
+ "errors": 12,
+ "words": 10,
+ "wer": 1.2
+ },
+ "semantic": {
+ "normalized_reference": "a sinewy hand dripping with water was clutching the rail",
+ "normalized_hypothesis": "you are sure alive hope i am not little you are still trying to lie",
+ "errors": 15,
+ "words": 10,
+ "wer": 1.5
+ }
+ },
+ {
+ "id": "release500_0246",
+ "category": "phonetic_balance",
+ "text": "Enters now the psychology of the situation.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0246.wav",
+ "render_seconds": 132.20554474997334,
+ "audio_seconds": 3.1346938775510202,
+ "rtf": 42.174945916332646,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "And cause no to suck who you have to suck what.",
+ "raw": {
+ "normalized_reference": "enters now the psychology of the situation.",
+ "normalized_hypothesis": "and cause no to suck who you have to suck what.",
+ "errors": 11,
+ "words": 7,
+ "wer": 1.5714285714285714
+ },
+ "semantic": {
+ "normalized_reference": "enters now the psychology of the situation",
+ "normalized_hypothesis": "and cause no to suck who you have to suck what",
+ "errors": 11,
+ "words": 7,
+ "wer": 1.5714285714285714
+ }
+ },
+ {
+ "id": "release500_0247",
+ "category": "phonetic_balance",
+ "text": "Wash your hands of me.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0247.wav",
+ "render_seconds": 89.69241090700962,
+ "audio_seconds": 1.8808163265306121,
+ "rtf": 47.68802229213836,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "You're so hot, Jessica.",
+ "raw": {
+ "normalized_reference": "wash your hands of me.",
+ "normalized_hypothesis": "you're so hot, jessica.",
+ "errors": 5,
+ "words": 5,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "wash your hands of me",
+ "normalized_hypothesis": "you are so hot jessica",
+ "errors": 5,
+ "words": 5,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0248",
+ "category": "phonetic_balance",
+ "text": "He confessed that the sketch had startled him.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0248.wav",
+ "render_seconds": 66.59408195200376,
+ "audio_seconds": 3.1346938775510202,
+ "rtf": 21.24420583104287,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I confess that the scotch has scared her.",
+ "raw": {
+ "normalized_reference": "he confessed that the sketch had startled him.",
+ "normalized_hypothesis": "i confess that the scotch has scared her.",
+ "errors": 6,
+ "words": 8,
+ "wer": 0.75
+ },
+ "semantic": {
+ "normalized_reference": "he confessed that the sketch had startled him",
+ "normalized_hypothesis": "i confess that the scotch has scared her",
+ "errors": 6,
+ "words": 8,
+ "wer": 0.75
+ }
+ },
+ {
+ "id": "release500_0249",
+ "category": "phonetic_balance",
+ "text": "Her face was against his breast.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0249.wav",
+ "render_seconds": 110.28980051109102,
+ "audio_seconds": 2.507755102040816,
+ "rtf": 43.97949401890837,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Oh, fast, where the ice has burst.",
+ "raw": {
+ "normalized_reference": "her face was against his breast.",
+ "normalized_hypothesis": "oh, fast, where the ice has burst.",
+ "errors": 7,
+ "words": 6,
+ "wer": 1.1666666666666667
+ },
+ "semantic": {
+ "normalized_reference": "her face was against his breast",
+ "normalized_hypothesis": "0 fast where the ice has burst",
+ "errors": 7,
+ "words": 6,
+ "wer": 1.1666666666666667
+ }
+ },
+ {
+ "id": "release500_0250",
+ "category": "phonetic_balance",
+ "text": "He was a pariah; a wanderer without a friend or a home.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0250.wav",
+ "render_seconds": 156.49989560095128,
+ "audio_seconds": 3.900952380952381,
+ "rtf": 40.118381440673545,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I wish you power. I wish you power. I wish you power. I wish you power. I wish you power. I wish you power. I wish you power. I wish you power. I wish you power. I wish you power. I wish you power. I wish you power. I wish you power. I wish you power. I wish you power. I wish you power. I wish you power. I wish you power. I wish you power. I wish you power. I wish you power. I wish you power. I wish you power. I wish you power. I wish you power. I wish you power. I wish you power. I wish you power. I wish you power. I wish you power. I wish you power. I wish you power. I wish you power. I wish you power. I wish you power. I wish you power. I wish you power. I wish you power. I wish you power. I wish you power. I wish you power. I wish you power. I wish you power. I wish you power. I wish you power. I wish you power. I wish you power. I wish you power. I wish you power. I wish you power. I wish you power. I wish you power. I wish you power. I wish you power. I wish you power. I wish you power. I wish you power. I wish you power. I wish you power. I wish you power. I wish you power. I wish you power. I wish you power. I wish you power. I wish you power. I wish you power. I wish you power. I wish you power. I wish you power. I wish you power. I wish you power. I wish you power. I wish you power. I wish you power. I wish you power. I wish you power. I wish you power. I wish you power. I wish you power. I wish you power. I wish you power. I wish you power. I wish you power. I wish you power. I wish you power. I wish you power. I wish you power. I wish you power. I wish you power",
+ "raw": {
+ "normalized_reference": "he was a pariah; a wanderer without a friend or a home.",
+ "normalized_hypothesis": "i wish you power. i wish you power. i wish you power. i wish you power. i wish you power. i wish you power. i wish you power. i wish you power. i wish you power. i wish you power. i wish you power. i wish you power. i wish you power. i wish you power. i wish you power. i wish you power. i wish you power. i wish you power. i wish you power. i wish you power. i wish you power. i wish you power. i wish you power. i wish you power. i wish you power. i wish you power. i wish you power. i wish you power. i wish you power. i wish you power. i wish you power. i wish you power. i wish you power. i wish you power. i wish you power. i wish you power. i wish you power. i wish you power. i wish you power. i wish you power. i wish you power. i wish you power. i wish you power. i wish you power. i wish you power. i wish you power. i wish you power. i wish you power. i wish you power. i wish you power. i wish you power. i wish you power. i wish you power. i wish you power. i wish you power. i wish you power. i wish you power. i wish you power. i wish you power. i wish you power. i wish you power. i wish you power. i wish you power. i wish you power. i wish you power. i wish you power. i wish you power. i wish you power. i wish you power. i wish you power. i wish you power. i wish you power. i wish you power. i wish you power. i wish you power. i wish you power. i wish you power. i wish you power. i wish you power. i wish you power. i wish you power. i wish you power. i wish you power. i wish you power. i wish you power. i wish you power. i wish you power. i wish you power. i wish you power",
+ "errors": 356,
+ "words": 12,
+ "wer": 29.666666666666668
+ },
+ "semantic": {
+ "normalized_reference": "he was a pariah a wanderer without a friend or a home",
+ "normalized_hypothesis": "i wish you power i wish you power i wish you power i wish you power i wish you power i wish you power i wish you power i wish you power i wish you power i wish you power i wish you power i wish you power i wish you power i wish you power i wish you power i wish you power i wish you power i wish you power i wish you power i wish you power i wish you power i wish you power i wish you power i wish you power i wish you power i wish you power i wish you power i wish you power i wish you power i wish you power i wish you power i wish you power i wish you power i wish you power i wish you power i wish you power i wish you power i wish you power i wish you power i wish you power i wish you power i wish you power i wish you power i wish you power i wish you power i wish you power i wish you power i wish you power i wish you power i wish you power i wish you power i wish you power i wish you power i wish you power i wish you power i wish you power i wish you power i wish you power i wish you power i wish you power i wish you power i wish you power i wish you power i wish you power i wish you power i wish you power i wish you power i wish you power i wish you power i wish you power i wish you power i wish you power i wish you power i wish you power i wish you power i wish you power i wish you power i wish you power i wish you power i wish you power i wish you power i wish you power i wish you power i wish you power i wish you power i wish you power i wish you power i wish you power i wish you power",
+ "errors": 356,
+ "words": 12,
+ "wer": 29.666666666666668
+ }
+ },
+ {
+ "id": "release500_0251",
+ "category": "phonetic_balance",
+ "text": "A cry of joy burst from Philip's lips.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0251.wav",
+ "render_seconds": 116.60638577397913,
+ "audio_seconds": 2.786394557823129,
+ "rtf": 41.84848317572005,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Thank you for watching.",
+ "raw": {
+ "normalized_reference": "a cry of joy burst from philip's lips.",
+ "normalized_hypothesis": "thank you for watching.",
+ "errors": 8,
+ "words": 8,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "a cry of joy burst from philip is lips",
+ "normalized_hypothesis": "thank you for watching",
+ "errors": 9,
+ "words": 9,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0252",
+ "category": "phonetic_balance",
+ "text": "And the big Persian knew of his existence before he did of hers.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0252.wav",
+ "render_seconds": 156.19583395903464,
+ "audio_seconds": 4.31891156462585,
+ "rtf": 36.165555052888884,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "And the bad portion of his existence was who had had his horse.",
+ "raw": {
+ "normalized_reference": "and the big persian knew of his existence before he did of hers.",
+ "normalized_hypothesis": "and the bad portion of his existence was who had had his horse.",
+ "errors": 9,
+ "words": 13,
+ "wer": 0.6923076923076923
+ },
+ "semantic": {
+ "normalized_reference": "and the big persian knew of his existence before he did of hers",
+ "normalized_hypothesis": "and the bad portion of his existence was who had had his horse",
+ "errors": 9,
+ "words": 13,
+ "wer": 0.6923076923076923
+ }
+ },
+ {
+ "id": "release500_0253",
+ "category": "phonetic_balance",
+ "text": "The men stared into each other's face.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0253.wav",
+ "render_seconds": 99.39987088704947,
+ "audio_seconds": 2.5774149659863945,
+ "rtf": 38.565722710083065,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Blast all and block all your farts.",
+ "raw": {
+ "normalized_reference": "the men stared into each other's face.",
+ "normalized_hypothesis": "blast all and block all your farts.",
+ "errors": 7,
+ "words": 7,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "the men stared into each other is face",
+ "normalized_hypothesis": "blast all and block all your farts",
+ "errors": 8,
+ "words": 8,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0254",
+ "category": "phonetic_balance",
+ "text": "It is the fire, partly, she said.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0254.wav",
+ "render_seconds": 76.80247374996543,
+ "audio_seconds": 2.716734693877551,
+ "rtf": 28.270141329239078,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I asked the fool to put my hat on.",
+ "raw": {
+ "normalized_reference": "it is the fire, partly, she said.",
+ "normalized_hypothesis": "i asked the fool to put my hat on.",
+ "errors": 8,
+ "words": 7,
+ "wer": 1.1428571428571428
+ },
+ "semantic": {
+ "normalized_reference": "it is the fire partly she said",
+ "normalized_hypothesis": "i asked the fool to put my hat on",
+ "errors": 8,
+ "words": 7,
+ "wer": 1.1428571428571428
+ }
+ },
+ {
+ "id": "release500_0255",
+ "category": "phonetic_balance",
+ "text": "It seemed nearer to him since he had seen and talked with Gregson.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0255.wav",
+ "render_seconds": 173.5110007530311,
+ "audio_seconds": 4.597551020408163,
+ "rtf": 37.7398749862328,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It's so nice to have some sunshine on top of the ocean.",
+ "raw": {
+ "normalized_reference": "it seemed nearer to him since he had seen and talked with gregson.",
+ "normalized_hypothesis": "it's so nice to have some sunshine on top of the ocean.",
+ "errors": 12,
+ "words": 13,
+ "wer": 0.9230769230769231
+ },
+ "semantic": {
+ "normalized_reference": "it seemed nearer to him since he had seen and talked with gregson",
+ "normalized_hypothesis": "it is so nice to have some sunshine on top of the ocean",
+ "errors": 12,
+ "words": 13,
+ "wer": 0.9230769230769231
+ }
+ },
+ {
+ "id": "release500_0256",
+ "category": "phonetic_balance",
+ "text": "Among my minor afflictions, I may mention a new and mysterious one.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0256.wav",
+ "render_seconds": 181.49999442393892,
+ "audio_seconds": 4.9458503401360545,
+ "rtf": 36.69742955039478,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Thank you for watching. Please subscribe to my channel. Thank you for watching. Please subscribe to my channel. Thank you for watching. Please subscribe to my channel. Thank you for watching. Please subscribe to my channel. Thank you for watching. Please subscribe to my channel. Thank you for watching. Please subscribe to my channel. Thank you for watching. Please subscribe to my channel. Thank you for watching. Please subscribe to my channel. Thank you for watching. Please subscribe to my channel. Thank you for watching. Please subscribe to my channel. Thank you for watching. Please subscribe to my channel. Thank you for watching. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel. Please subscribe to my channel.",
+ "raw": {
+ "normalized_reference": "among my minor afflictions, i may mention a new and mysterious one.",
+ "normalized_hypothesis": "thank you for watching. please subscribe to my channel. thank you for watching. please subscribe to my channel. thank you for watching. please subscribe to my channel. thank you for watching. please subscribe to my channel. thank you for watching. please subscribe to my channel. thank you for watching. please subscribe to my channel. thank you for watching. please subscribe to my channel. thank you for watching. please subscribe to my channel. thank you for watching. please subscribe to my channel. thank you for watching. please subscribe to my channel. thank you for watching. please subscribe to my channel. thank you for watching. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel. please subscribe to my channel.",
+ "errors": 367,
+ "words": 12,
+ "wer": 30.583333333333332
+ },
+ "semantic": {
+ "normalized_reference": "among my minor afflictions i may mention a new and mysterious one",
+ "normalized_hypothesis": "thank you for watching please subscribe to my channel thank you for watching please subscribe to my channel thank you for watching please subscribe to my channel thank you for watching please subscribe to my channel thank you for watching please subscribe to my channel thank you for watching please subscribe to my channel thank you for watching please subscribe to my channel thank you for watching please subscribe to my channel thank you for watching please subscribe to my channel thank you for watching please subscribe to my channel thank you for watching please subscribe to my channel thank you for watching please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel please subscribe to my channel",
+ "errors": 367,
+ "words": 12,
+ "wer": 30.583333333333332
+ }
+ },
+ {
+ "id": "release500_0257",
+ "category": "phonetic_balance",
+ "text": "Between him and all domestic animals there must be no hostilities.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0257.wav",
+ "render_seconds": 166.90159599797335,
+ "audio_seconds": 4.667210884353741,
+ "rtf": 35.760457398119875,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "That's one hell of a load to rush to college, though, isn't it? That's one hell of a load to rush to college, though, isn't it? That's one hell of a load to rush to college, though, isn't it? That's one hell of a load to rush to college, though, isn't it? That's one hell of a load to rush to college, though, isn't it? That's one hell of a load to rush to college, though, isn't it? That's one hell of a load to rush to college, though, isn't it? That's one hell of a load to rush to college, though, isn't it? That's one hell of a load to rush to college, though, isn't it? That's one hell of a load to rush to college, though, isn't it? That's one hell of a load to rush to college, though, isn't it? That's one hell of a load to rush to college, though, isn't it? That's one hell of a load to rush to college, though, isn't it? That's one hell of a load to rush to college, though, isn't it? That's one hell of a load to rush to college, though, isn't it? That's one hell of a load to rush to college, though, isn't it? That's one hell of a load to rush to college, though, isn't it? That's one hell of a load to rush to college, though, isn't it? That's one hell of a load to rush to college, though, isn't it? That's one hell of a load to rush to college, though, isn't it? That's one hell of a load to college, though, isn't it? That's one hell of a load to college, though, isn't it? That's one hell of a load to college, though, isn't it? That's one hell of a load to college, though, isn't it? That's one hell of a load to college, though, isn't it? That's one hell",
+ "raw": {
+ "normalized_reference": "between him and all domestic animals there must be no hostilities.",
+ "normalized_hypothesis": "that's one hell of a load to rush to college, though, isn't it? that's one hell of a load to rush to college, though, isn't it? that's one hell of a load to rush to college, though, isn't it? that's one hell of a load to rush to college, though, isn't it? that's one hell of a load to rush to college, though, isn't it? that's one hell of a load to rush to college, though, isn't it? that's one hell of a load to rush to college, though, isn't it? that's one hell of a load to rush to college, though, isn't it? that's one hell of a load to rush to college, though, isn't it? that's one hell of a load to rush to college, though, isn't it? that's one hell of a load to rush to college, though, isn't it? that's one hell of a load to rush to college, though, isn't it? that's one hell of a load to rush to college, though, isn't it? that's one hell of a load to rush to college, though, isn't it? that's one hell of a load to rush to college, though, isn't it? that's one hell of a load to rush to college, though, isn't it? that's one hell of a load to rush to college, though, isn't it? that's one hell of a load to rush to college, though, isn't it? that's one hell of a load to rush to college, though, isn't it? that's one hell of a load to rush to college, though, isn't it? that's one hell of a load to college, though, isn't it? that's one hell of a load to college, though, isn't it? that's one hell of a load to college, though, isn't it? that's one hell of a load to college, though, isn't it? that's one hell of a load to college, though, isn't it? that's one hell",
+ "errors": 318,
+ "words": 11,
+ "wer": 28.90909090909091
+ },
+ "semantic": {
+ "normalized_reference": "between him and all domestic animals there must be no hostilities",
+ "normalized_hypothesis": "that is one hell of a load to rush to college though is not it that is one hell of a load to rush to college though is not it that is one hell of a load to rush to college though is not it that is one hell of a load to rush to college though is not it that is one hell of a load to rush to college though is not it that is one hell of a load to rush to college though is not it that is one hell of a load to rush to college though is not it that is one hell of a load to rush to college though is not it that is one hell of a load to rush to college though is not it that is one hell of a load to rush to college though is not it that is one hell of a load to rush to college though is not it that is one hell of a load to rush to college though is not it that is one hell of a load to rush to college though is not it that is one hell of a load to rush to college though is not it that is one hell of a load to rush to college though is not it that is one hell of a load to rush to college though is not it that is one hell of a load to rush to college though is not it that is one hell of a load to rush to college though is not it that is one hell of a load to rush to college though is not it that is one hell of a load to rush to college though is not it that is one hell of a load to college though is not it that is one hell of a load to college though is not it that is one hell of a load to college though is not it that is one hell of a load to college though is not it that is one hell of a load to college though is not it that is one hell",
+ "errors": 369,
+ "words": 11,
+ "wer": 33.54545454545455
+ }
+ },
+ {
+ "id": "release500_0258",
+ "category": "phonetic_balance",
+ "text": "With these arguments he convinced himself that he should go on alone.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0258.wav",
+ "render_seconds": 171.09754963498563,
+ "audio_seconds": 4.597551020408163,
+ "rtf": 37.21493222706985,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "But I sure wish I convinced himself that I should kill you.",
+ "raw": {
+ "normalized_reference": "with these arguments he convinced himself that he should go on alone.",
+ "normalized_hypothesis": "but i sure wish i convinced himself that i should kill you.",
+ "errors": 9,
+ "words": 12,
+ "wer": 0.75
+ },
+ "semantic": {
+ "normalized_reference": "with these arguments he convinced himself that he should go on alone",
+ "normalized_hypothesis": "but i sure wish i convinced himself that i should kill you",
+ "errors": 9,
+ "words": 12,
+ "wer": 0.75
+ }
+ },
+ {
+ "id": "release500_0259",
+ "category": "phonetic_balance",
+ "text": "We are consumed in our own flesh-pots.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0259.wav",
+ "render_seconds": 122.80917294404935,
+ "audio_seconds": 2.8560544217687074,
+ "rtf": 42.999591327113315,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I can shake all your fish, bitch.",
+ "raw": {
+ "normalized_reference": "we are consumed in our own flesh-pots.",
+ "normalized_hypothesis": "i can shake all your fish, bitch.",
+ "errors": 8,
+ "words": 8,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "we are consumed in our own flesh pots",
+ "normalized_hypothesis": "i can shake all your fish bitch",
+ "errors": 8,
+ "words": 8,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0260",
+ "category": "phonetic_balance",
+ "text": "Mercedes screamed, cried, laughed, and manifested the chaotic abandonment of hysteria.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0260.wav",
+ "render_seconds": 159.50137514597736,
+ "audio_seconds": 6.199727891156463,
+ "rtf": 25.727157376293313,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I saw a squawk, horn, lisp, and now the first step to cover the boundary of his tower.",
+ "raw": {
+ "normalized_reference": "mercedes screamed, cried, laughed, and manifested the chaotic abandonment of hysteria.",
+ "normalized_hypothesis": "i saw a squawk, horn, lisp, and now the first step to cover the boundary of his tower.",
+ "errors": 15,
+ "words": 11,
+ "wer": 1.3636363636363635
+ },
+ "semantic": {
+ "normalized_reference": "mercedes screamed cried laughed and manifested the chaotic abandonment of hysteria",
+ "normalized_hypothesis": "i saw a squawk horn lisp and now the 1st step to cover the boundary of his tower",
+ "errors": 15,
+ "words": 11,
+ "wer": 1.3636363636363635
+ }
+ },
+ {
+ "id": "release500_0261",
+ "category": "phonetic_balance",
+ "text": "Philip made no effort to follow.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0261.wav",
+ "render_seconds": 117.49560932500754,
+ "audio_seconds": 2.2987755102040817,
+ "rtf": 51.11225902810165,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Say it like I want to see you.",
+ "raw": {
+ "normalized_reference": "philip made no effort to follow.",
+ "normalized_hypothesis": "say it like i want to see you.",
+ "errors": 7,
+ "words": 6,
+ "wer": 1.1666666666666667
+ },
+ "semantic": {
+ "normalized_reference": "philip made no effort to follow",
+ "normalized_hypothesis": "say it like i want to see you",
+ "errors": 7,
+ "words": 6,
+ "wer": 1.1666666666666667
+ }
+ },
+ {
+ "id": "release500_0262",
+ "category": "phonetic_balance",
+ "text": "She said with chattering teeth.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0262.wav",
+ "render_seconds": 112.71082032192498,
+ "audio_seconds": 2.2987755102040817,
+ "rtf": 49.030807846007846,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I'm so excited. I'm so excited. I'm so excited. I'm so excited. I'm so excited. I'm so excited. I'm so excited. I'm so excited. I'm so excited. I'm so excited. I'm so excited. I'm so excited. I'm so excited. I'm so excited. I'm so excited. I'm so excited. I'm so excited. I'm so excited. I'm so excited. I'm so excited. I'm so excited. I'm so excited. I'm so excited. I'm so excited. I'm so excited. I'm so excited. I'm so excited. I'm so excited. I'm so excited. I'm so excited. I'm so excited. I'm so excited. I'm so excited. I'm so excited. I'm so excited. I'm so excited. I'm so excited. I'm so excited. I'm so excited. I'm so excited. I'm so excited. I'm so excited. I'm so excited. I'm so excited. I'm so excited. I'm so excited. I'm so excited. I'm so excited. I'm so excited. I'm so excited. I'm so excited. I'm so excited. I'm so excited. I'm so excited. I'm so excited. I'm so excited. I'm so excited. I'm so excited. I'm so excited. I'm so excited. I'm so excited. I'm so excited. I'm so excited. I'm so excited. I'm so excited. I'm so excited. I'm so excited. I'm so excited. I'm so excited. I'm so excited. I'm so excited. I'm so excited. I'm so excited. I'm so excited. I'm so excited. I'm so excited. I'm so excited. I'm so excited. I'm so excited. I'm so excited. I'm so excited. I'm so excited. I'm so excited. I'm so excited. I'm so excited. I'm so excited. I'm so excited. I'm so excited. I'm so excited",
+ "raw": {
+ "normalized_reference": "she said with chattering teeth.",
+ "normalized_hypothesis": "i'm so excited. i'm so excited. i'm so excited. i'm so excited. i'm so excited. i'm so excited. i'm so excited. i'm so excited. i'm so excited. i'm so excited. i'm so excited. i'm so excited. i'm so excited. i'm so excited. i'm so excited. i'm so excited. i'm so excited. i'm so excited. i'm so excited. i'm so excited. i'm so excited. i'm so excited. i'm so excited. i'm so excited. i'm so excited. i'm so excited. i'm so excited. i'm so excited. i'm so excited. i'm so excited. i'm so excited. i'm so excited. i'm so excited. i'm so excited. i'm so excited. i'm so excited. i'm so excited. i'm so excited. i'm so excited. i'm so excited. i'm so excited. i'm so excited. i'm so excited. i'm so excited. i'm so excited. i'm so excited. i'm so excited. i'm so excited. i'm so excited. i'm so excited. i'm so excited. i'm so excited. i'm so excited. i'm so excited. i'm so excited. i'm so excited. i'm so excited. i'm so excited. i'm so excited. i'm so excited. i'm so excited. i'm so excited. i'm so excited. i'm so excited. i'm so excited. i'm so excited. i'm so excited. i'm so excited. i'm so excited. i'm so excited. i'm so excited. i'm so excited. i'm so excited. i'm so excited. i'm so excited. i'm so excited. i'm so excited. i'm so excited. i'm so excited. i'm so excited. i'm so excited. i'm so excited. i'm so excited. i'm so excited. i'm so excited. i'm so excited. i'm so excited. i'm so excited. i'm so excited",
+ "errors": 267,
+ "words": 5,
+ "wer": 53.4
+ },
+ "semantic": {
+ "normalized_reference": "she said with chattering teeth",
+ "normalized_hypothesis": "i am so excited i am so excited i am so excited i am so excited i am so excited i am so excited i am so excited i am so excited i am so excited i am so excited i am so excited i am so excited i am so excited i am so excited i am so excited i am so excited i am so excited i am so excited i am so excited i am so excited i am so excited i am so excited i am so excited i am so excited i am so excited i am so excited i am so excited i am so excited i am so excited i am so excited i am so excited i am so excited i am so excited i am so excited i am so excited i am so excited i am so excited i am so excited i am so excited i am so excited i am so excited i am so excited i am so excited i am so excited i am so excited i am so excited i am so excited i am so excited i am so excited i am so excited i am so excited i am so excited i am so excited i am so excited i am so excited i am so excited i am so excited i am so excited i am so excited i am so excited i am so excited i am so excited i am so excited i am so excited i am so excited i am so excited i am so excited i am so excited i am so excited i am so excited i am so excited i am so excited i am so excited i am so excited i am so excited i am so excited i am so excited i am so excited i am so excited i am so excited i am so excited i am so excited i am so excited i am so excited i am so excited i am so excited i am so excited i am so excited i am so excited",
+ "errors": 356,
+ "words": 5,
+ "wer": 71.2
+ }
+ },
+ {
+ "id": "release500_0263",
+ "category": "phonetic_balance",
+ "text": "Halfway around the track one donkey got into an argument with its rider.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0263.wav",
+ "render_seconds": 165.69103769597132,
+ "audio_seconds": 5.015510204081632,
+ "rtf": 33.035729358327615,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Have no way to cut your dicks out until you're up to it with a twirl.",
+ "raw": {
+ "normalized_reference": "halfway around the track one donkey got into an argument with its rider.",
+ "normalized_hypothesis": "have no way to cut your dicks out until you're up to it with a twirl.",
+ "errors": 15,
+ "words": 13,
+ "wer": 1.1538461538461537
+ },
+ "semantic": {
+ "normalized_reference": "halfway around the track one donkey got into an argument with its rider",
+ "normalized_hypothesis": "have no way to cut your dicks out until you are up to it with a twirl",
+ "errors": 16,
+ "words": 13,
+ "wer": 1.2307692307692308
+ }
+ },
+ {
+ "id": "release500_0264",
+ "category": "phonetic_balance",
+ "text": "Earth and gravel seemed to fill the pan.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0264.wav",
+ "render_seconds": 128.8076142220525,
+ "audio_seconds": 2.8560544217687074,
+ "rtf": 45.09984587138367,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I think I'll be sure to say goodbye.",
+ "raw": {
+ "normalized_reference": "earth and gravel seemed to fill the pan.",
+ "normalized_hypothesis": "i think i'll be sure to say goodbye.",
+ "errors": 8,
+ "words": 8,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "earth and gravel seemed to fill the pan",
+ "normalized_hypothesis": "i think i will be sure to say goodbye",
+ "errors": 9,
+ "words": 8,
+ "wer": 1.125
+ }
+ },
+ {
+ "id": "release500_0265",
+ "category": "phonetic_balance",
+ "text": "Her achievements with cocoanuts were a revelation.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0265.wav",
+ "render_seconds": 135.29178818303626,
+ "audio_seconds": 3.5526530612244898,
+ "rtf": 38.08190269398424,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I have lunch with a bitch. What was that?",
+ "raw": {
+ "normalized_reference": "her achievements with cocoanuts were a revelation.",
+ "normalized_hypothesis": "i have lunch with a bitch. what was that?",
+ "errors": 8,
+ "words": 7,
+ "wer": 1.1428571428571428
+ },
+ "semantic": {
+ "normalized_reference": "her achievements with cocoanuts were a revelation",
+ "normalized_hypothesis": "i have lunch with a bitch what was that",
+ "errors": 8,
+ "words": 7,
+ "wer": 1.1428571428571428
+ }
+ },
+ {
+ "id": "release500_0266",
+ "category": "phonetic_balance",
+ "text": "For a full minute he crouched and listened.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0266.wav",
+ "render_seconds": 94.99260276998393,
+ "audio_seconds": 3.1346938775510202,
+ "rtf": 30.30362978990373,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "4-3-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-",
+ "raw": {
+ "normalized_reference": "for a full minute he crouched and listened.",
+ "normalized_hypothesis": "4-3-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-1-2-",
+ "errors": 222,
+ "words": 8,
+ "wer": 27.75
+ },
+ "semantic": {
+ "normalized_reference": "for a full minute he crouched and listened",
+ "normalized_hypothesis": "4 3 one 2 one 2 one 2 one 2 one 2 one 2 one 2 one 2 one 2 one 2 one 2 one 2 one 2 one 2 one 2 one 2 one 2 one 2 one 2 one 2 one 2 one 2 one 2 one 2 one 2 one 2 one 2 one 2 one 2 one 2 one 2 one 2 one 2 one 2 one 2 one 2 one 2 one 2 one 2 one 2 one 2 one 2 one 2 one 2 one 2 one 2 one 2 one 2 one 2 one 2 one 2 one 2 one 2 one 2 one 2 one 2 one 2 one 2 one 2 one 2 one 2 one 2 one 2 one 2 one 2 one 2 one 2 one 2 one 2 one 2 one 2 one 2 one 2 one 2 one 2 one 2 one 2 one 2 one 2 one 2 one 2 one 2 one 2 one 2 one 2 one 2 one 2 one 2 one 2 one 2 one 2 one 2 one 2 one 2 one 2 one 2 one 2 one 2 one 2 one 2 one 2 one 2 one 2 one 2 one 2 one 2 one 2 one 2 one 2 one 2",
+ "errors": 222,
+ "words": 8,
+ "wer": 27.75
+ }
+ },
+ {
+ "id": "release500_0267",
+ "category": "phonetic_balance",
+ "text": "It was edged with ice.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0267.wav",
+ "render_seconds": 93.4998834610451,
+ "audio_seconds": 1.9504761904761905,
+ "rtf": 47.9369519697741,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I wish I could do this.",
+ "raw": {
+ "normalized_reference": "it was edged with ice.",
+ "normalized_hypothesis": "i wish i could do this.",
+ "errors": 6,
+ "words": 5,
+ "wer": 1.2
+ },
+ "semantic": {
+ "normalized_reference": "it was edged with ice",
+ "normalized_hypothesis": "i wish i could do this",
+ "errors": 6,
+ "words": 5,
+ "wer": 1.2
+ }
+ },
+ {
+ "id": "release500_0268",
+ "category": "phonetic_balance",
+ "text": "The Portuguese boy passed the Hawaiian.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0268.wav",
+ "render_seconds": 123.9883021019632,
+ "audio_seconds": 2.8560544217687074,
+ "rtf": 43.41244380951932,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The box has been passed away.",
+ "raw": {
+ "normalized_reference": "the portuguese boy passed the hawaiian.",
+ "normalized_hypothesis": "the box has been passed away.",
+ "errors": 5,
+ "words": 6,
+ "wer": 0.8333333333333334
+ },
+ "semantic": {
+ "normalized_reference": "the portuguese boy passed the hawaiian",
+ "normalized_hypothesis": "the box has been passed away",
+ "errors": 5,
+ "words": 6,
+ "wer": 0.8333333333333334
+ }
+ },
+ {
+ "id": "release500_0269",
+ "category": "phonetic_balance",
+ "text": "But he did not broach it, preferring to mature it carefully.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0269.wav",
+ "render_seconds": 162.41152951307595,
+ "audio_seconds": 4.1099319727891155,
+ "rtf": 39.51684129770618,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "But I got no bullshit. Before I took your coffee.",
+ "raw": {
+ "normalized_reference": "but he did not broach it, preferring to mature it carefully.",
+ "normalized_hypothesis": "but i got no bullshit. before i took your coffee.",
+ "errors": 10,
+ "words": 11,
+ "wer": 0.9090909090909091
+ },
+ "semantic": {
+ "normalized_reference": "but he did not broach it preferring to mature it carefully",
+ "normalized_hypothesis": "but i got no bullshit before i took your coffee",
+ "errors": 10,
+ "words": 11,
+ "wer": 0.9090909090909091
+ }
+ },
+ {
+ "id": "release500_0270",
+ "category": "phonetic_balance",
+ "text": "Come on, Del Mar challenged.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0270.wav",
+ "render_seconds": 92.99759810091928,
+ "audio_seconds": 2.159455782312925,
+ "rtf": 43.06529397944536,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Care for your child.",
+ "raw": {
+ "normalized_reference": "come on, del mar challenged.",
+ "normalized_hypothesis": "care for your child.",
+ "errors": 5,
+ "words": 5,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "come on del mar challenged",
+ "normalized_hypothesis": "care for your child",
+ "errors": 5,
+ "words": 5,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0271",
+ "category": "phonetic_balance",
+ "text": "For a full half minute Jeanne looked at him without speaking.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0271.wav",
+ "render_seconds": 162.0087238959968,
+ "audio_seconds": 4.179591836734694,
+ "rtf": 38.761852885272674,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Thank you for watching! Please like, subscribe, and share! Thank you for watching! Please like, subscribe, and share! Thank you for watching! Please like, subscribe, and share! Thank you for watching! Please like, subscribe, and share! Thank you for watching! Please like, subscribe, and share! Thank you for watching! Please like, subscribe, and share! Thank you for watching! Please like, subscribe, and share! Thank you for watching! Please like, subscribe, and share! Thank you for watching! Please like, subscribe, and share! Thank you for watching! Please like, subscribe, and share! Thank you for watching! Please like, subscribe, and share! Thank you for watching! Please like, subscribe, and share! Thank you for watching! Please like, subscribe, and share! Please like, subscribe, and share! Please like, subscribe, and share! Please like, subscribe, and share! Please like, subscribe, and share! Please like, subscribe, and share! Please like, subscribe, and share! Please like, subscribe, and share! Please like, subscribe, and share! Please like, subscribe, and share! Please like, subscribe, and share! Please like, subscribe, and share! Please like, subscribe, and share! Please like, subscribe, and share! Please like, subscribe, and share! Please like, subscribe, and share! Please like, subscribe, and share! Please like, subscribe, and share! Please like, subscribe, and share! Please like, subscribe, subscribe, subscribe, subscribe, subscribe, subscribe, subscribe! Please like, subscribe, subscribe, subscribe, subscribe, subscribe, subscribe, subscribe, subscribe, subscribe, subscribe, subscribe, subscribe, subscribe, subscribe, subscribe, subscribe, subscribe, subscribe, subscribe, subscribe, subscribe, subscribe, subscribe, subscribe, subscribe, subscribe, subscribe, subscribe, subscribe, subscribe, subscribe, subscribe, subscribe, subscribe, subscribe, subscribe, subscribe, subscribe, subscribe, subscribe, subscribe, subscribe, subscribe, subscribe, subscribe, subscribe, subscribe, subscribe, subscribe, subscribe, subscribe, subscribe, subscribe, subscribe, subscribe, subscribe",
+ "raw": {
+ "normalized_reference": "for a full half minute jeanne looked at him without speaking.",
+ "normalized_hypothesis": "thank you for watching! please like, subscribe, and share! thank you for watching! please like, subscribe, and share! thank you for watching! please like, subscribe, and share! thank you for watching! please like, subscribe, and share! thank you for watching! please like, subscribe, and share! thank you for watching! please like, subscribe, and share! thank you for watching! please like, subscribe, and share! thank you for watching! please like, subscribe, and share! thank you for watching! please like, subscribe, and share! thank you for watching! please like, subscribe, and share! thank you for watching! please like, subscribe, and share! thank you for watching! please like, subscribe, and share! thank you for watching! please like, subscribe, and share! please like, subscribe, and share! please like, subscribe, and share! please like, subscribe, and share! please like, subscribe, and share! please like, subscribe, and share! please like, subscribe, and share! please like, subscribe, and share! please like, subscribe, and share! please like, subscribe, and share! please like, subscribe, and share! please like, subscribe, and share! please like, subscribe, and share! please like, subscribe, and share! please like, subscribe, and share! please like, subscribe, and share! please like, subscribe, and share! please like, subscribe, and share! please like, subscribe, and share! please like, subscribe, subscribe, subscribe, subscribe, subscribe, subscribe, subscribe! please like, subscribe, subscribe, subscribe, subscribe, subscribe, subscribe, subscribe, subscribe, subscribe, subscribe, subscribe, subscribe, subscribe, subscribe, subscribe, subscribe, subscribe, subscribe, subscribe, subscribe, subscribe, subscribe, subscribe, subscribe, subscribe, subscribe, subscribe, subscribe, subscribe, subscribe, subscribe, subscribe, subscribe, subscribe, subscribe, subscribe, subscribe, subscribe, subscribe, subscribe, subscribe, subscribe, subscribe, subscribe, subscribe, subscribe, subscribe, subscribe, subscribe, subscribe, subscribe, subscribe, subscribe, subscribe, subscribe, subscribe",
+ "errors": 273,
+ "words": 11,
+ "wer": 24.818181818181817
+ },
+ "semantic": {
+ "normalized_reference": "for a full half minute jeanne looked at him without speaking",
+ "normalized_hypothesis": "thank you for watching please like subscribe and share thank you for watching please like subscribe and share thank you for watching please like subscribe and share thank you for watching please like subscribe and share thank you for watching please like subscribe and share thank you for watching please like subscribe and share thank you for watching please like subscribe and share thank you for watching please like subscribe and share thank you for watching please like subscribe and share thank you for watching please like subscribe and share thank you for watching please like subscribe and share thank you for watching please like subscribe and share thank you for watching please like subscribe and share please like subscribe and share please like subscribe and share please like subscribe and share please like subscribe and share please like subscribe and share please like subscribe and share please like subscribe and share please like subscribe and share please like subscribe and share please like subscribe and share please like subscribe and share please like subscribe and share please like subscribe and share please like subscribe and share please like subscribe and share please like subscribe and share please like subscribe and share please like subscribe and share please like subscribe subscribe subscribe subscribe subscribe subscribe subscribe please like subscribe subscribe subscribe subscribe subscribe subscribe subscribe subscribe subscribe subscribe subscribe subscribe subscribe subscribe subscribe subscribe subscribe subscribe subscribe subscribe subscribe subscribe subscribe subscribe subscribe subscribe subscribe subscribe subscribe subscribe subscribe subscribe subscribe subscribe subscribe subscribe subscribe subscribe subscribe subscribe subscribe subscribe subscribe subscribe subscribe subscribe subscribe subscribe subscribe subscribe subscribe subscribe subscribe subscribe subscribe subscribe",
+ "errors": 273,
+ "words": 11,
+ "wer": 24.818181818181817
+ }
+ },
+ {
+ "id": "release500_0272",
+ "category": "phonetic_balance",
+ "text": "He bore no grudges and had few enemies.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0272.wav",
+ "render_seconds": 87.70124634995591,
+ "audio_seconds": 2.786394557823129,
+ "rtf": 31.474812532821094,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Happy birthday to you.",
+ "raw": {
+ "normalized_reference": "he bore no grudges and had few enemies.",
+ "normalized_hypothesis": "happy birthday to you.",
+ "errors": 8,
+ "words": 8,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "he bore no grudges and had few enemies",
+ "normalized_hypothesis": "happy birthday to you",
+ "errors": 8,
+ "words": 8,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0273",
+ "category": "phonetic_balance",
+ "text": "I'll go over tomorrow afternoon.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0273.wav",
+ "render_seconds": 127.19252713792957,
+ "audio_seconds": 2.2987755102040817,
+ "rtf": 55.33055601703257,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It is a two-hour afternoon.",
+ "raw": {
+ "normalized_reference": "i'll go over tomorrow afternoon.",
+ "normalized_hypothesis": "it is a two-hour afternoon.",
+ "errors": 5,
+ "words": 5,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "i will go over tomorrow afternoon",
+ "normalized_hypothesis": "it is a 2 hour afternoon",
+ "errors": 5,
+ "words": 6,
+ "wer": 0.8333333333333334
+ }
+ },
+ {
+ "id": "release500_0274",
+ "category": "phonetic_balance",
+ "text": "Well, there are better men in Hawaii, that's all.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0274.wav",
+ "render_seconds": 125.60755661095027,
+ "audio_seconds": 3.274013605442177,
+ "rtf": 38.3650075253692,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Thank you for watching. Please subscribe for more videos. Thank you for watching. Please subscribe for more videos. Thank you for watching. Please subscribe for more videos. Thank you for watching. Please subscribe for more videos. Thank you for watching. Please subscribe for more videos. Thank you for watching. Please subscribe for more videos. Thank you for watching. Please subscribe for more videos. Thank you for watching. Please subscribe for more videos. Thank you for watching. Please subscribe for more videos. Thank you for watching. Please subscribe for more videos. Thank you for watching. Please subscribe for more videos. Thank you for watching. Please subscribe for more videos. Thank you for watching. Please subscribe for more videos. Please subscribe for more videos. Please subscribe for more videos. Please subscribe for more videos. Please subscribe for more videos. Please subscribe for more videos. Please subscribe for more videos. Please subscribe for more videos. Please subscribe for more videos. Please subscribe for more videos. Please subscribe for more videos. Please subscribe for more videos. Please subscribe for more videos. Please subscribe for more videos. Please subscribe for more videos. Please subscribe for more videos. Please subscribe for more videos. Please subscribe for more videos. Please subscribe for more videos. Please subscribe for more videos. Please subscribe for more videos. Please subscribe for more videos. Please subscribe for more videos. Please subscribe for more videos. Please subscribe for more videos. Please subscribe for more videos. Please subscribe for more videos. Please subscribe for more videos. Please subscribe for more videos. Please subscribe for more videos. Please subscribe for more videos. Please subscribe for more videos. Please subscribe for more videos. Please subscribe for more videos. Please subscribe for more videos. Please subscribe for more videos. Please subscribe for more videos. Please subscribe for more videos. Please subscribe for more videos. Please subscribe for more videos. Please subscribe for more videos. Please subscribe for more videos. Please subscribe for more videos. Please subscribe for more videos. Please subscribe for more videos. Please subscribe for more videos. Please subscribe for more videos. Please subscribe for more videos. Please subscribe for more videos. Please subscribe for more videos. Please subscribe for more videos. Please",
+ "raw": {
+ "normalized_reference": "well, there are better men in hawaii, that's all.",
+ "normalized_hypothesis": "thank you for watching. please subscribe for more videos. thank you for watching. please subscribe for more videos. thank you for watching. please subscribe for more videos. thank you for watching. please subscribe for more videos. thank you for watching. please subscribe for more videos. thank you for watching. please subscribe for more videos. thank you for watching. please subscribe for more videos. thank you for watching. please subscribe for more videos. thank you for watching. please subscribe for more videos. thank you for watching. please subscribe for more videos. thank you for watching. please subscribe for more videos. thank you for watching. please subscribe for more videos. thank you for watching. please subscribe for more videos. please subscribe for more videos. please subscribe for more videos. please subscribe for more videos. please subscribe for more videos. please subscribe for more videos. please subscribe for more videos. please subscribe for more videos. please subscribe for more videos. please subscribe for more videos. please subscribe for more videos. please subscribe for more videos. please subscribe for more videos. please subscribe for more videos. please subscribe for more videos. please subscribe for more videos. please subscribe for more videos. please subscribe for more videos. please subscribe for more videos. please subscribe for more videos. please subscribe for more videos. please subscribe for more videos. please subscribe for more videos. please subscribe for more videos. please subscribe for more videos. please subscribe for more videos. please subscribe for more videos. please subscribe for more videos. please subscribe for more videos. please subscribe for more videos. please subscribe for more videos. please subscribe for more videos. please subscribe for more videos. please subscribe for more videos. please subscribe for more videos. please subscribe for more videos. please subscribe for more videos. please subscribe for more videos. please subscribe for more videos. please subscribe for more videos. please subscribe for more videos. please subscribe for more videos. please subscribe for more videos. please subscribe for more videos. please subscribe for more videos. please subscribe for more videos. please subscribe for more videos. please subscribe for more videos. please subscribe for more videos. please subscribe for more videos. please subscribe for more videos. please",
+ "errors": 368,
+ "words": 9,
+ "wer": 40.888888888888886
+ },
+ "semantic": {
+ "normalized_reference": "well there are better men in hawaii that is all",
+ "normalized_hypothesis": "thank you for watching please subscribe for more videos thank you for watching please subscribe for more videos thank you for watching please subscribe for more videos thank you for watching please subscribe for more videos thank you for watching please subscribe for more videos thank you for watching please subscribe for more videos thank you for watching please subscribe for more videos thank you for watching please subscribe for more videos thank you for watching please subscribe for more videos thank you for watching please subscribe for more videos thank you for watching please subscribe for more videos thank you for watching please subscribe for more videos thank you for watching please subscribe for more videos please subscribe for more videos please subscribe for more videos please subscribe for more videos please subscribe for more videos please subscribe for more videos please subscribe for more videos please subscribe for more videos please subscribe for more videos please subscribe for more videos please subscribe for more videos please subscribe for more videos please subscribe for more videos please subscribe for more videos please subscribe for more videos please subscribe for more videos please subscribe for more videos please subscribe for more videos please subscribe for more videos please subscribe for more videos please subscribe for more videos please subscribe for more videos please subscribe for more videos please subscribe for more videos please subscribe for more videos please subscribe for more videos please subscribe for more videos please subscribe for more videos please subscribe for more videos please subscribe for more videos please subscribe for more videos please subscribe for more videos please subscribe for more videos please subscribe for more videos please subscribe for more videos please subscribe for more videos please subscribe for more videos please subscribe for more videos please subscribe for more videos please subscribe for more videos please subscribe for more videos please subscribe for more videos please subscribe for more videos please subscribe for more videos please subscribe for more videos please subscribe for more videos please subscribe for more videos please subscribe for more videos please subscribe for more videos please subscribe for more videos please subscribe for more videos please",
+ "errors": 368,
+ "words": 10,
+ "wer": 36.8
+ }
+ },
+ {
+ "id": "release500_0275",
+ "category": "phonetic_balance",
+ "text": "He was an athlete and a giant.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0275.wav",
+ "render_seconds": 107.79519376193639,
+ "audio_seconds": 2.36843537414966,
+ "rtf": 45.513251042637435,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I wish you all the best.",
+ "raw": {
+ "normalized_reference": "he was an athlete and a giant.",
+ "normalized_hypothesis": "i wish you all the best.",
+ "errors": 7,
+ "words": 7,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "he was an athlete and a giant",
+ "normalized_hypothesis": "i wish you all the best",
+ "errors": 7,
+ "words": 7,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0276",
+ "category": "phonetic_balance",
+ "text": "McVeigh when he returns from a trip to Honolulu.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0276.wav",
+ "render_seconds": 135.294383415021,
+ "audio_seconds": 3.3436734693877552,
+ "rtf": 40.462797774267756,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "What's that? What's that? What's that? What's that? What's that? What's that? What's that? What's that? What's that? What's that? What's that? What's that? What's that? What's that? What's that? What's that? What's that? What's that? What's that? What's that? What's that? What's that? What's that? What's that? What's that? What's that? What's that? What's that? What's that? What's that? What's that? What's that? What's that? What's that? What's that? What's that? What's that? What's that? What's that? What's that? What's that? What's that? What's that? What's that? What's that? What's that? What's that? What's that? What's that? What's that? What's that? What's that? What's that? What's that? What's that? What's that? What's that? What's that? What's that? What's that? What's that? What's that? What's that? What's that? What's that? What's that? What's that? What's that? What's that? What's that? What's that? What's that? What's that? What's that? What's that? What's that? What's that? What's that? What's that? What's that? What's that? What's that? What's that? What's that? What's that? What's that? What's that? What's that? What's that? What's that? What's that? What's that? What's that? What's that? What's that? What's that? What's that? What's that? What's that? What's that? What's that? What's that? What's that? What's that? What's that? What's that? What's that? What's that? What's that? What's that? What's that?",
+ "raw": {
+ "normalized_reference": "mcveigh when he returns from a trip to honolulu.",
+ "normalized_hypothesis": "what's that? what's that? what's that? what's that? what's that? what's that? what's that? what's that? what's that? what's that? what's that? what's that? what's that? what's that? what's that? what's that? what's that? what's that? what's that? what's that? what's that? what's that? what's that? what's that? what's that? what's that? what's that? what's that? what's that? what's that? what's that? what's that? what's that? what's that? what's that? what's that? what's that? what's that? what's that? what's that? what's that? what's that? what's that? what's that? what's that? what's that? what's that? what's that? what's that? what's that? what's that? what's that? what's that? what's that? what's that? what's that? what's that? what's that? what's that? what's that? what's that? what's that? what's that? what's that? what's that? what's that? what's that? what's that? what's that? what's that? what's that? what's that? what's that? what's that? what's that? what's that? what's that? what's that? what's that? what's that? what's that? what's that? what's that? what's that? what's that? what's that? what's that? what's that? what's that? what's that? what's that? what's that? what's that? what's that? what's that? what's that? what's that? what's that? what's that? what's that? what's that? what's that? what's that? what's that? what's that? what's that? what's that? what's that? what's that? what's that? what's that?",
+ "errors": 222,
+ "words": 9,
+ "wer": 24.666666666666668
+ },
+ "semantic": {
+ "normalized_reference": "mcveigh when he returns from a trip to honolulu",
+ "normalized_hypothesis": "what is that what is that what is that what is that what is that what is that what is that what is that what is that what is that what is that what is that what is that what is that what is that what is that what is that what is that what is that what is that what is that what is that what is that what is that what is that what is that what is that what is that what is that what is that what is that what is that what is that what is that what is that what is that what is that what is that what is that what is that what is that what is that what is that what is that what is that what is that what is that what is that what is that what is that what is that what is that what is that what is that what is that what is that what is that what is that what is that what is that what is that what is that what is that what is that what is that what is that what is that what is that what is that what is that what is that what is that what is that what is that what is that what is that what is that what is that what is that what is that what is that what is that what is that what is that what is that what is that what is that what is that what is that what is that what is that what is that what is that what is that what is that what is that what is that what is that what is that what is that what is that what is that what is that what is that what is that what is that what is that what is that what is that what is that what is that",
+ "errors": 333,
+ "words": 9,
+ "wer": 37.0
+ }
+ },
+ {
+ "id": "release500_0277",
+ "category": "phonetic_balance",
+ "text": "You're a devil for fighting, and will surely win.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0277.wav",
+ "render_seconds": 123.59655887796544,
+ "audio_seconds": 3.274013605442177,
+ "rtf": 37.75077741659931,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I'll tell you something I wish I had known.",
+ "raw": {
+ "normalized_reference": "you're a devil for fighting, and will surely win.",
+ "normalized_hypothesis": "i'll tell you something i wish i had known.",
+ "errors": 9,
+ "words": 9,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "you are a devil for fighting and will surely win",
+ "normalized_hypothesis": "i will tell you something i wish i had known",
+ "errors": 10,
+ "words": 10,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0278",
+ "category": "phonetic_balance",
+ "text": "All an appearance can know is mirage.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0278.wav",
+ "render_seconds": 71.3037364219781,
+ "audio_seconds": 2.8560544217687074,
+ "rtf": 24.96581853570594,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Be a part of us. Be a part of us. Be a part of us. Be a part of us. Be a part of us. Be a part of us. Be a part of us. Be a part of us. Be a part of us. Be a part of us. Be a part of us. Be a part of us. Be a part of us. Be a part of us. Be a part of us. Be a part of us. Be a part of us. Be a part of us. Be a part of us. Be a part of us. Be a part of us. Be a part of us. Be a part of us. Be a part of us. Be a part of us. Be a part of us. Be a part of us. Be a part of us. Be a part of us. Be a part of us. Be a part of us. Be a part of us. Be a part of us. Be a part of us. Be a part of us. Be a part of us. Be a part of us. Be a part of us. Be a part of us. Be a part of us. Be a part of us. Be a part of us. Be a part of us. Be a part of us. Be a part of us. Be a part of us. Be a part of us. Be a part of us. Be a part of us. Be a part of us. Be a part of us. Be a part of us. Be a part of us. Be a part of us. Be a part of us. Be a part of us. Be a part of us. Be a part of us. Be a part of us. Be a part of us. Be a part of us. Be a part of us. Be a part of us. Be a part of us. Be a part of us. Be a part of us. Be a part of us. Be a part of us. Be a part of us. Be a part of us. Be a part of us. Be a part of us. Be a part of us. Be a part of us.",
+ "raw": {
+ "normalized_reference": "all an appearance can know is mirage.",
+ "normalized_hypothesis": "be a part of us. be a part of us. be a part of us. be a part of us. be a part of us. be a part of us. be a part of us. be a part of us. be a part of us. be a part of us. be a part of us. be a part of us. be a part of us. be a part of us. be a part of us. be a part of us. be a part of us. be a part of us. be a part of us. be a part of us. be a part of us. be a part of us. be a part of us. be a part of us. be a part of us. be a part of us. be a part of us. be a part of us. be a part of us. be a part of us. be a part of us. be a part of us. be a part of us. be a part of us. be a part of us. be a part of us. be a part of us. be a part of us. be a part of us. be a part of us. be a part of us. be a part of us. be a part of us. be a part of us. be a part of us. be a part of us. be a part of us. be a part of us. be a part of us. be a part of us. be a part of us. be a part of us. be a part of us. be a part of us. be a part of us. be a part of us. be a part of us. be a part of us. be a part of us. be a part of us. be a part of us. be a part of us. be a part of us. be a part of us. be a part of us. be a part of us. be a part of us. be a part of us. be a part of us. be a part of us. be a part of us. be a part of us. be a part of us. be a part of us.",
+ "errors": 370,
+ "words": 7,
+ "wer": 52.857142857142854
+ },
+ "semantic": {
+ "normalized_reference": "all an appearance can know is mirage",
+ "normalized_hypothesis": "be a part of us be a part of us be a part of us be a part of us be a part of us be a part of us be a part of us be a part of us be a part of us be a part of us be a part of us be a part of us be a part of us be a part of us be a part of us be a part of us be a part of us be a part of us be a part of us be a part of us be a part of us be a part of us be a part of us be a part of us be a part of us be a part of us be a part of us be a part of us be a part of us be a part of us be a part of us be a part of us be a part of us be a part of us be a part of us be a part of us be a part of us be a part of us be a part of us be a part of us be a part of us be a part of us be a part of us be a part of us be a part of us be a part of us be a part of us be a part of us be a part of us be a part of us be a part of us be a part of us be a part of us be a part of us be a part of us be a part of us be a part of us be a part of us be a part of us be a part of us be a part of us be a part of us be a part of us be a part of us be a part of us be a part of us be a part of us be a part of us be a part of us be a part of us be a part of us be a part of us be a part of us be a part of us",
+ "errors": 370,
+ "words": 7,
+ "wer": 52.857142857142854
+ }
+ },
+ {
+ "id": "release500_0279",
+ "category": "phonetic_balance",
+ "text": "You fired me out of your house, in short.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0279.wav",
+ "render_seconds": 110.00324166705832,
+ "audio_seconds": 2.9257142857142857,
+ "rtf": 37.598764241670324,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I feel a powerful ocean flow.",
+ "raw": {
+ "normalized_reference": "you fired me out of your house, in short.",
+ "normalized_hypothesis": "i feel a powerful ocean flow.",
+ "errors": 9,
+ "words": 9,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "you fired me out of your house in short",
+ "normalized_hypothesis": "i feel a powerful ocean flow",
+ "errors": 9,
+ "words": 9,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0280",
+ "category": "phonetic_balance",
+ "text": "He made sure that the magazine was loaded, and resumed his paddling.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0280.wav",
+ "render_seconds": 162.699455367052,
+ "audio_seconds": 4.876190476190477,
+ "rtf": 33.36609924519621,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I'm not sure that there is a solution. I wish I could help you.",
+ "raw": {
+ "normalized_reference": "he made sure that the magazine was loaded, and resumed his paddling.",
+ "normalized_hypothesis": "i'm not sure that there is a solution. i wish i could help you.",
+ "errors": 12,
+ "words": 12,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "he made sure that the magazine was loaded and resumed his paddling",
+ "normalized_hypothesis": "i am not sure that there is a solution i wish i could help you",
+ "errors": 13,
+ "words": 12,
+ "wer": 1.0833333333333333
+ }
+ },
+ {
+ "id": "release500_0281",
+ "category": "phonetic_balance",
+ "text": "Change chairs, Daylight commanded.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0281.wav",
+ "render_seconds": 110.99231745500583,
+ "audio_seconds": 2.507755102040816,
+ "rtf": 44.25963179765044,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha",
+ "raw": {
+ "normalized_reference": "change chairs, daylight commanded.",
+ "normalized_hypothesis": "ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha",
+ "errors": 444,
+ "words": 4,
+ "wer": 111.0
+ },
+ "semantic": {
+ "normalized_reference": "change chairs daylight commanded",
+ "normalized_hypothesis": "ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha ha",
+ "errors": 444,
+ "words": 4,
+ "wer": 111.0
+ }
+ },
+ {
+ "id": "release500_0282",
+ "category": "phonetic_balance",
+ "text": "By that answer my professional medical prestige stood or fell.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0282.wav",
+ "render_seconds": 176.0012555270223,
+ "audio_seconds": 4.667210884353741,
+ "rtf": 37.71015706983483,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The actual, professional, organic, plastic, sterile food.",
+ "raw": {
+ "normalized_reference": "by that answer my professional medical prestige stood or fell.",
+ "normalized_hypothesis": "the actual, professional, organic, plastic, sterile food.",
+ "errors": 9,
+ "words": 10,
+ "wer": 0.9
+ },
+ "semantic": {
+ "normalized_reference": "by that answer my professional medical prestige stood or fell",
+ "normalized_hypothesis": "the actual professional organic plastic sterile food",
+ "errors": 9,
+ "words": 10,
+ "wer": 0.9
+ }
+ },
+ {
+ "id": "release500_0283",
+ "category": "phonetic_balance",
+ "text": "The moon had already begun its westward decline.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0283.wav",
+ "render_seconds": 140.7004451659741,
+ "audio_seconds": 3.4133333333333336,
+ "rtf": 41.22083354471897,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Hello everyone and welcome back to my channel.",
+ "raw": {
+ "normalized_reference": "the moon had already begun its westward decline.",
+ "normalized_hypothesis": "hello everyone and welcome back to my channel.",
+ "errors": 8,
+ "words": 8,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "the moon had already begun its westward decline",
+ "normalized_hypothesis": "hello everyone and welcome back to my channel",
+ "errors": 8,
+ "words": 8,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0284",
+ "category": "phonetic_balance",
+ "text": "At sea, Monday, March 16, 1908.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0284.wav",
+ "render_seconds": 110.49829378596041,
+ "audio_seconds": 3.8312925170068026,
+ "rtf": 28.84099642495771,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "A child looks at a child.",
+ "raw": {
+ "normalized_reference": "at sea, monday, march 16, 1908.",
+ "normalized_hypothesis": "a child looks at a child.",
+ "errors": 6,
+ "words": 6,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "at sea monday march 16 1908",
+ "normalized_hypothesis": "a child looks at a child",
+ "errors": 6,
+ "words": 6,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0285",
+ "category": "phonetic_balance",
+ "text": "And wherever I ranged, the way lay along alcohol-drenched roads.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0285.wav",
+ "render_seconds": 181.09901139803696,
+ "audio_seconds": 4.527891156462585,
+ "rtf": 39.99632613508328,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Thank you for watching. Don't forget to like, share, and subscribe.",
+ "raw": {
+ "normalized_reference": "and wherever i ranged, the way lay along alcohol-drenched roads.",
+ "normalized_hypothesis": "thank you for watching. don't forget to like, share, and subscribe.",
+ "errors": 11,
+ "words": 11,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "and wherever i ranged the way lay along alcohol drenched roads",
+ "normalized_hypothesis": "thank you for watching do not forget to like share and subscribe",
+ "errors": 12,
+ "words": 11,
+ "wer": 1.0909090909090908
+ }
+ },
+ {
+ "id": "release500_0286",
+ "category": "phonetic_balance",
+ "text": "The date was nearly eighteen years old.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0286.wav",
+ "render_seconds": 128.10178916691802,
+ "audio_seconds": 2.8560544217687074,
+ "rtf": 44.852712956214155,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The dark was now the action lawsuit.",
+ "raw": {
+ "normalized_reference": "the date was nearly eighteen years old.",
+ "normalized_hypothesis": "the dark was now the action lawsuit.",
+ "errors": 5,
+ "words": 7,
+ "wer": 0.7142857142857143
+ },
+ "semantic": {
+ "normalized_reference": "the date was nearly 18 years old",
+ "normalized_hypothesis": "the dark was now the action lawsuit",
+ "errors": 5,
+ "words": 7,
+ "wer": 0.7142857142857143
+ }
+ },
+ {
+ "id": "release500_0287",
+ "category": "phonetic_balance",
+ "text": "Very early in my life, I separated from my mother.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0287.wav",
+ "render_seconds": 151.10867715091445,
+ "audio_seconds": 3.900952380952381,
+ "rtf": 38.736355226674846,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you.",
+ "raw": {
+ "normalized_reference": "very early in my life, i separated from my mother.",
+ "normalized_hypothesis": "i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you.",
+ "errors": 23,
+ "words": 10,
+ "wer": 2.3
+ },
+ "semantic": {
+ "normalized_reference": "very early in my life i separated from my mother",
+ "normalized_hypothesis": "i love you i love you i love you i love you i love you i love you i love you i love you",
+ "errors": 23,
+ "words": 10,
+ "wer": 2.3
+ }
+ },
+ {
+ "id": "release500_0288",
+ "category": "phonetic_balance",
+ "text": "There was pride and strength, the ring of triumph in his voice.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0288.wav",
+ "render_seconds": 165.89893930708058,
+ "audio_seconds": 4.458231292517007,
+ "rtf": 37.21182873251472,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I was poor and strong, the one with clothes and a fish.",
+ "raw": {
+ "normalized_reference": "there was pride and strength, the ring of triumph in his voice.",
+ "normalized_hypothesis": "i was poor and strong, the one with clothes and a fish.",
+ "errors": 9,
+ "words": 12,
+ "wer": 0.75
+ },
+ "semantic": {
+ "normalized_reference": "there was pride and strength the ring of triumph in his voice",
+ "normalized_hypothesis": "i was poor and strong the one with clothes and a fish",
+ "errors": 9,
+ "words": 12,
+ "wer": 0.75
+ }
+ },
+ {
+ "id": "release500_0289",
+ "category": "phonetic_balance",
+ "text": "From my earliest recollection my sleep was a period of terror.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0289.wav",
+ "render_seconds": 180.30714368098415,
+ "audio_seconds": 4.73687074829932,
+ "rtf": 38.06461127226659,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "For the last couple of hours, I slapped Mr. Power out of town.",
+ "raw": {
+ "normalized_reference": "from my earliest recollection my sleep was a period of terror.",
+ "normalized_hypothesis": "for the last couple of hours, i slapped mr. power out of town.",
+ "errors": 12,
+ "words": 11,
+ "wer": 1.0909090909090908
+ },
+ "semantic": {
+ "normalized_reference": "from my earliest recollection my sleep was a period of terror",
+ "normalized_hypothesis": "for the last couple of hours i slapped mister power out of town",
+ "errors": 12,
+ "words": 11,
+ "wer": 1.0909090909090908
+ }
+ },
+ {
+ "id": "release500_0290",
+ "category": "phonetic_balance",
+ "text": "We'll have to watch our chances.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0290.wav",
+ "render_seconds": 78.39858104998711,
+ "audio_seconds": 2.2987755102040817,
+ "rtf": 34.10449637295249,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "You have to look for chances.",
+ "raw": {
+ "normalized_reference": "we'll have to watch our chances.",
+ "normalized_hypothesis": "you have to look for chances.",
+ "errors": 3,
+ "words": 6,
+ "wer": 0.5
+ },
+ "semantic": {
+ "normalized_reference": "we will have to watch our chances",
+ "normalized_hypothesis": "you have to look for chances",
+ "errors": 4,
+ "words": 7,
+ "wer": 0.5714285714285714
+ }
+ },
+ {
+ "id": "release500_0291",
+ "category": "phonetic_balance",
+ "text": "Close beside him gleamed the white fangs of the wolf-dog.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0291.wav",
+ "render_seconds": 144.80133695702534,
+ "audio_seconds": 3.7616326530612243,
+ "rtf": 38.49427903045921,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Push the circle down the left side of the list.",
+ "raw": {
+ "normalized_reference": "close beside him gleamed the white fangs of the wolf-dog.",
+ "normalized_hypothesis": "push the circle down the left side of the list.",
+ "errors": 8,
+ "words": 11,
+ "wer": 0.7272727272727273
+ },
+ "semantic": {
+ "normalized_reference": "close beside him gleamed the white fangs of the wolf dog",
+ "normalized_hypothesis": "push the circle down the left side of the list",
+ "errors": 8,
+ "words": 11,
+ "wer": 0.7272727272727273
+ }
+ },
+ {
+ "id": "release500_0292",
+ "category": "phonetic_balance",
+ "text": "Would you be satisfied with that one hundredth part of me.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0292.wav",
+ "render_seconds": 139.69537051802035,
+ "audio_seconds": 3.900952380952381,
+ "rtf": 35.810580821269866,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you.",
+ "raw": {
+ "normalized_reference": "would you be satisfied with that one hundredth part of me.",
+ "normalized_hypothesis": "thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you.",
+ "errors": 295,
+ "words": 11,
+ "wer": 26.818181818181817
+ },
+ "semantic": {
+ "normalized_reference": "would you be satisfied with that 100th part of me",
+ "normalized_hypothesis": "thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you",
+ "errors": 295,
+ "words": 10,
+ "wer": 29.5
+ }
+ },
+ {
+ "id": "release500_0293",
+ "category": "phonetic_balance",
+ "text": "Two gigantic owls were tearing at the carcass.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0293.wav",
+ "render_seconds": 111.19883950799704,
+ "audio_seconds": 3.3436734693877552,
+ "rtf": 33.256488866527434,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-",
+ "raw": {
+ "normalized_reference": "two gigantic owls were tearing at the carcass.",
+ "normalized_hypothesis": "ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-ch-",
+ "errors": 222,
+ "words": 8,
+ "wer": 27.75
+ },
+ "semantic": {
+ "normalized_reference": "2 gigantic owls were tearing at the carcass",
+ "normalized_hypothesis": "ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch ch",
+ "errors": 222,
+ "words": 8,
+ "wer": 27.75
+ }
+ },
+ {
+ "id": "release500_0294",
+ "category": "phonetic_balance",
+ "text": "The lush vegetation of that sheltered spot make a natural shield.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0294.wav",
+ "render_seconds": 181.1006852670107,
+ "audio_seconds": 4.597551020408163,
+ "rtf": 39.39068527203269,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Thank you for watching.",
+ "raw": {
+ "normalized_reference": "the lush vegetation of that sheltered spot make a natural shield.",
+ "normalized_hypothesis": "thank you for watching.",
+ "errors": 11,
+ "words": 11,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "the lush vegetation of that sheltered spot make a natural shield",
+ "normalized_hypothesis": "thank you for watching",
+ "errors": 11,
+ "words": 11,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0295",
+ "category": "phonetic_balance",
+ "text": "Below him the shadow was broken into a pool of rippling starlight.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0295.wav",
+ "render_seconds": 175.69179761398118,
+ "audio_seconds": 4.527891156462585,
+ "rtf": 38.80212477352048,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Thank you for watching. Please subscribe to our channel.",
+ "raw": {
+ "normalized_reference": "below him the shadow was broken into a pool of rippling starlight.",
+ "normalized_hypothesis": "thank you for watching. please subscribe to our channel.",
+ "errors": 12,
+ "words": 12,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "below him the shadow was broken into a pool of rippling starlight",
+ "normalized_hypothesis": "thank you for watching please subscribe to our channel",
+ "errors": 12,
+ "words": 12,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0296",
+ "category": "phonetic_balance",
+ "text": "Lots of men take women buggy riding.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0296.wav",
+ "render_seconds": 86.09542912605684,
+ "audio_seconds": 2.507755102040816,
+ "rtf": 34.33167339824858,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It's a long time going back home.",
+ "raw": {
+ "normalized_reference": "lots of men take women buggy riding.",
+ "normalized_hypothesis": "it's a long time going back home.",
+ "errors": 7,
+ "words": 7,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "lots of men take women buggy riding",
+ "normalized_hypothesis": "it is a long time going back home",
+ "errors": 8,
+ "words": 7,
+ "wer": 1.1428571428571428
+ }
+ },
+ {
+ "id": "release500_0297",
+ "category": "phonetic_balance",
+ "text": "But it contributed to the smash.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0297.wav",
+ "render_seconds": 122.70294041000307,
+ "audio_seconds": 2.4380952380952383,
+ "rtf": 50.32737790254031,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I can't get back to this world.",
+ "raw": {
+ "normalized_reference": "but it contributed to the smash.",
+ "normalized_hypothesis": "i can't get back to this world.",
+ "errors": 6,
+ "words": 6,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "but it contributed to the smash",
+ "normalized_hypothesis": "i can not get back to this world",
+ "errors": 7,
+ "words": 6,
+ "wer": 1.1666666666666667
+ }
+ },
+ {
+ "id": "release500_0298",
+ "category": "phonetic_balance",
+ "text": "No man ate of the seal meat or the oil.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0298.wav",
+ "render_seconds": 76.19988947198726,
+ "audio_seconds": 2.786394557823129,
+ "rtf": 27.347128301714182,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Allah is the Greatest, Allah is the Greatest, Allah is the Greatest, Allah is the Greatest, Allah is the Greatest, Allah is the Greatest, Allah is the Greatest, Allah is the Greatest, Allah is the Greatest, Allah is the Greatest, Allah is the Greatest, Allah is the Greatest, Allah is the Greatest, Allah is the Greatest, Allah is the Greatest, Allah is the Greatest, Allah is the Greatest, Allah is the Greatest, Allah is the Greatest, Allah is the Greatest, Allah is the Greatest, Allah is the Greatest, Allah is the Greatest, Allah is the Greatest, Allah is the Greatest, Allah is the Greatest, Allah is the Greatest, Allah is the Greatest, Allah is the Greatest, Allah is the Greatest, Allah is the Greatest, Allah is the Greatest, Allah is the Greatest, Allah is the Greatest, Allah is the Greatest, Allah is the Greatest, Allah is the Greatest, Allah is the Greatest, Allah is the Greatest, Allah is the Greatest, Allah is the Greatest, Allah is the Greatest, Allah is the Greatest, Allah is the Greatest, Allah is the Greatest, Allah is the Greatest, Allah is the Greatest, Allah is the Greatest, Allah is the Greatest, Allah is the Greatest, Allah is the Greatest, Allah is the Greatest, Allah is the Greatest, Allah is the Greatest, Allah is the Greatest, Allah is the Greatest, Allah is the Greatest, Allah is the Greatest, Allah is the Greatest, Allah is the Greatest, Allah is the Greatest, Allah is the Greatest, Allah is the Greatest, Allah is the Greatest, Allah is the Greatest, Allah is the Greatest, Allah is the Greatest, Allah is the Greatest, Allah is the Greatest, Allah is the Greatest, Allah is the Greatest, Allah is the Greatest, Allah is the Greatest, Allah is the Greatest,",
+ "raw": {
+ "normalized_reference": "no man ate of the seal meat or the oil.",
+ "normalized_hypothesis": "allah is the greatest, allah is the greatest, allah is the greatest, allah is the greatest, allah is the greatest, allah is the greatest, allah is the greatest, allah is the greatest, allah is the greatest, allah is the greatest, allah is the greatest, allah is the greatest, allah is the greatest, allah is the greatest, allah is the greatest, allah is the greatest, allah is the greatest, allah is the greatest, allah is the greatest, allah is the greatest, allah is the greatest, allah is the greatest, allah is the greatest, allah is the greatest, allah is the greatest, allah is the greatest, allah is the greatest, allah is the greatest, allah is the greatest, allah is the greatest, allah is the greatest, allah is the greatest, allah is the greatest, allah is the greatest, allah is the greatest, allah is the greatest, allah is the greatest, allah is the greatest, allah is the greatest, allah is the greatest, allah is the greatest, allah is the greatest, allah is the greatest, allah is the greatest, allah is the greatest, allah is the greatest, allah is the greatest, allah is the greatest, allah is the greatest, allah is the greatest, allah is the greatest, allah is the greatest, allah is the greatest, allah is the greatest, allah is the greatest, allah is the greatest, allah is the greatest, allah is the greatest, allah is the greatest, allah is the greatest, allah is the greatest, allah is the greatest, allah is the greatest, allah is the greatest, allah is the greatest, allah is the greatest, allah is the greatest, allah is the greatest, allah is the greatest, allah is the greatest, allah is the greatest, allah is the greatest, allah is the greatest, allah is the greatest,",
+ "errors": 294,
+ "words": 10,
+ "wer": 29.4
+ },
+ "semantic": {
+ "normalized_reference": "no man ate of the seal meat or the oil",
+ "normalized_hypothesis": "allah is the greatest allah is the greatest allah is the greatest allah is the greatest allah is the greatest allah is the greatest allah is the greatest allah is the greatest allah is the greatest allah is the greatest allah is the greatest allah is the greatest allah is the greatest allah is the greatest allah is the greatest allah is the greatest allah is the greatest allah is the greatest allah is the greatest allah is the greatest allah is the greatest allah is the greatest allah is the greatest allah is the greatest allah is the greatest allah is the greatest allah is the greatest allah is the greatest allah is the greatest allah is the greatest allah is the greatest allah is the greatest allah is the greatest allah is the greatest allah is the greatest allah is the greatest allah is the greatest allah is the greatest allah is the greatest allah is the greatest allah is the greatest allah is the greatest allah is the greatest allah is the greatest allah is the greatest allah is the greatest allah is the greatest allah is the greatest allah is the greatest allah is the greatest allah is the greatest allah is the greatest allah is the greatest allah is the greatest allah is the greatest allah is the greatest allah is the greatest allah is the greatest allah is the greatest allah is the greatest allah is the greatest allah is the greatest allah is the greatest allah is the greatest allah is the greatest allah is the greatest allah is the greatest allah is the greatest allah is the greatest allah is the greatest allah is the greatest allah is the greatest allah is the greatest allah is the greatest",
+ "errors": 294,
+ "words": 10,
+ "wer": 29.4
+ }
+ },
+ {
+ "id": "release500_0299",
+ "category": "phonetic_balance",
+ "text": "You're going in for grab sharing.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0299.wav",
+ "render_seconds": 98.39253861107863,
+ "audio_seconds": 2.2291156462585033,
+ "rtf": 44.13971916451587,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I think I should go home.",
+ "raw": {
+ "normalized_reference": "you're going in for grab sharing.",
+ "normalized_hypothesis": "i think i should go home.",
+ "errors": 6,
+ "words": 6,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "you are going in for grab sharing",
+ "normalized_hypothesis": "i think i should go home",
+ "errors": 7,
+ "words": 7,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0300",
+ "category": "phonetic_balance",
+ "text": "He it was that lived to found the family of the Patino.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0300.wav",
+ "render_seconds": 135.30550636490807,
+ "audio_seconds": 3.6919727891156464,
+ "rtf": 36.648565440082336,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I wish that love to fall to fall after the channel.",
+ "raw": {
+ "normalized_reference": "he it was that lived to found the family of the patino.",
+ "normalized_hypothesis": "i wish that love to fall to fall after the channel.",
+ "errors": 9,
+ "words": 12,
+ "wer": 0.75
+ },
+ "semantic": {
+ "normalized_reference": "he it was that lived to found the family of the patino",
+ "normalized_hypothesis": "i wish that love to fall to fall after the channel",
+ "errors": 9,
+ "words": 12,
+ "wer": 0.75
+ }
+ },
+ {
+ "id": "release500_0301",
+ "category": "phonetic_balance",
+ "text": "For such countries nothing remained but reorganization.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0301.wav",
+ "render_seconds": 149.49207354802638,
+ "audio_seconds": 3.9706122448979593,
+ "rtf": 37.649627898094636,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "For such good cause, nothing will land below the sun.",
+ "raw": {
+ "normalized_reference": "for such countries nothing remained but reorganization.",
+ "normalized_hypothesis": "for such good cause, nothing will land below the sun.",
+ "errors": 7,
+ "words": 7,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "for such countries nothing remained but reorganization",
+ "normalized_hypothesis": "for such good cause nothing will land below the sun",
+ "errors": 7,
+ "words": 7,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0302",
+ "category": "phonetic_balance",
+ "text": "Stand off butcher and baker and all the rest.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0302.wav",
+ "render_seconds": 92.71865101600997,
+ "audio_seconds": 3.3436734693877552,
+ "rtf": 27.72957702505181,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Stardust become the core of the worst.",
+ "raw": {
+ "normalized_reference": "stand off butcher and baker and all the rest.",
+ "normalized_hypothesis": "stardust become the core of the worst.",
+ "errors": 8,
+ "words": 9,
+ "wer": 0.8888888888888888
+ },
+ "semantic": {
+ "normalized_reference": "stand off butcher and baker and all the rest",
+ "normalized_hypothesis": "stardust become the core of the worst",
+ "errors": 8,
+ "words": 9,
+ "wer": 0.8888888888888888
+ }
+ },
+ {
+ "id": "release500_0303",
+ "category": "phonetic_balance",
+ "text": "All eyes, however, were staring at him in certitude of expectancy.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0303.wav",
+ "render_seconds": 176.4957405719906,
+ "audio_seconds": 4.806530612244898,
+ "rtf": 36.71998678680171,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Alas, however, the starling had her own sort of expectancy.",
+ "raw": {
+ "normalized_reference": "all eyes, however, were staring at him in certitude of expectancy.",
+ "normalized_hypothesis": "alas, however, the starling had her own sort of expectancy.",
+ "errors": 8,
+ "words": 11,
+ "wer": 0.7272727272727273
+ },
+ "semantic": {
+ "normalized_reference": "all eyes however were staring at him in certitude of expectancy",
+ "normalized_hypothesis": "alas however the starling had her own sort of expectancy",
+ "errors": 8,
+ "words": 11,
+ "wer": 0.7272727272727273
+ }
+ },
+ {
+ "id": "release500_0304",
+ "category": "phonetic_balance",
+ "text": "Two days ago Jeanne learned where her father's men were hiding.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0304.wav",
+ "render_seconds": 166.2022139140172,
+ "audio_seconds": 3.9706122448979593,
+ "rtf": 41.85808224602612,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "................................................................................................................................................................................................................................................................................................................................................................",
+ "raw": {
+ "normalized_reference": "two days ago jeanne learned where her father's men were hiding.",
+ "normalized_hypothesis": "................................................................................................................................................................................................................................................................................................................................................................",
+ "errors": 11,
+ "words": 11,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "2 days ago jeanne learned where her father is men were hiding",
+ "normalized_hypothesis": " .",
+ "errors": 12,
+ "words": 12,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0305",
+ "category": "phonetic_balance",
+ "text": "His previous wives had never lived long enough to bear him children.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0305.wav",
+ "render_seconds": 171.7069034690503,
+ "audio_seconds": 4.597551020408163,
+ "rtf": 37.34747101377603,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I swear I never left you alone, Satan.",
+ "raw": {
+ "normalized_reference": "his previous wives had never lived long enough to bear him children.",
+ "normalized_hypothesis": "i swear i never left you alone, satan.",
+ "errors": 11,
+ "words": 12,
+ "wer": 0.9166666666666666
+ },
+ "semantic": {
+ "normalized_reference": "his previous wives had never lived long enough to bear him children",
+ "normalized_hypothesis": "i swear i never left you alone satan",
+ "errors": 11,
+ "words": 12,
+ "wer": 0.9166666666666666
+ }
+ },
+ {
+ "id": "release500_0306",
+ "category": "phonetic_balance",
+ "text": "But this little defect did not worry him.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0306.wav",
+ "render_seconds": 130.60353156202473,
+ "audio_seconds": 2.8560544217687074,
+ "rtf": 45.72865648727524,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "But I say to fuck together.",
+ "raw": {
+ "normalized_reference": "but this little defect did not worry him.",
+ "normalized_hypothesis": "but i say to fuck together.",
+ "errors": 7,
+ "words": 8,
+ "wer": 0.875
+ },
+ "semantic": {
+ "normalized_reference": "but this little defect did not worry him",
+ "normalized_hypothesis": "but i say to fuck together",
+ "errors": 7,
+ "words": 8,
+ "wer": 0.875
+ }
+ },
+ {
+ "id": "release500_0307",
+ "category": "phonetic_balance",
+ "text": "Eli Harding asked, as Shunk started to follow.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0307.wav",
+ "render_seconds": 144.3118335370673,
+ "audio_seconds": 3.4829931972789114,
+ "rtf": 41.43328033193143,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you.",
+ "raw": {
+ "normalized_reference": "eli harding asked, as shunk started to follow.",
+ "normalized_hypothesis": "i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you.",
+ "errors": 333,
+ "words": 8,
+ "wer": 41.625
+ },
+ "semantic": {
+ "normalized_reference": "eli harding asked as shunk started to follow",
+ "normalized_hypothesis": "i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you",
+ "errors": 333,
+ "words": 8,
+ "wer": 41.625
+ }
+ },
+ {
+ "id": "release500_0308",
+ "category": "phonetic_balance",
+ "text": "And this was their sole conversation throughout the meal.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0308.wav",
+ "render_seconds": 109.88795409700833,
+ "audio_seconds": 3.900952380952381,
+ "rtf": 28.16951948287567,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "And that's what's all she told us after a long time.",
+ "raw": {
+ "normalized_reference": "and this was their sole conversation throughout the meal.",
+ "normalized_hypothesis": "and that's what's all she told us after a long time.",
+ "errors": 10,
+ "words": 9,
+ "wer": 1.1111111111111112
+ },
+ "semantic": {
+ "normalized_reference": "and this was their sole conversation throughout the meal",
+ "normalized_hypothesis": "and that is what is all she told us after a long time",
+ "errors": 12,
+ "words": 9,
+ "wer": 1.3333333333333333
+ }
+ },
+ {
+ "id": "release500_0309",
+ "category": "phonetic_balance",
+ "text": "Then, again, Tudor had such an irritating way about him.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0309.wav",
+ "render_seconds": 137.29976294899825,
+ "audio_seconds": 4.179591836734694,
+ "rtf": 32.85004093994587,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Thank you for watching.",
+ "raw": {
+ "normalized_reference": "then, again, tudor had such an irritating way about him.",
+ "normalized_hypothesis": "thank you for watching.",
+ "errors": 10,
+ "words": 10,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "then again tudor had such an irritating way about him",
+ "normalized_hypothesis": "thank you for watching",
+ "errors": 10,
+ "words": 10,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0310",
+ "category": "phonetic_balance",
+ "text": "In it there was something that was almost tragedy.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0310.wav",
+ "render_seconds": 144.90152485005092,
+ "audio_seconds": 3.4829931972789114,
+ "rtf": 41.60258623624509,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I don't know what's up. I don't know what's up. I don't know what's up.",
+ "raw": {
+ "normalized_reference": "in it there was something that was almost tragedy.",
+ "normalized_hypothesis": "i don't know what's up. i don't know what's up. i don't know what's up.",
+ "errors": 15,
+ "words": 9,
+ "wer": 1.6666666666666667
+ },
+ "semantic": {
+ "normalized_reference": "in it there was something that was almost tragedy",
+ "normalized_hypothesis": "i do not know what is up i do not know what is up i do not know what is up",
+ "errors": 21,
+ "words": 9,
+ "wer": 2.3333333333333335
+ }
+ },
+ {
+ "id": "release500_0311",
+ "category": "phonetic_balance",
+ "text": "Thirty pounds, said the captain with finality.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0311.wav",
+ "render_seconds": 128.39992312190589,
+ "audio_seconds": 3.274013605442177,
+ "rtf": 39.2178954016792,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "So I thought I should talk to you for a little bit.",
+ "raw": {
+ "normalized_reference": "thirty pounds, said the captain with finality.",
+ "normalized_hypothesis": "so i thought i should talk to you for a little bit.",
+ "errors": 12,
+ "words": 7,
+ "wer": 1.7142857142857142
+ },
+ "semantic": {
+ "normalized_reference": "\u00a330 said the captain with finality",
+ "normalized_hypothesis": "so i thought i should talk to you for a little bit",
+ "errors": 12,
+ "words": 6,
+ "wer": 2.0
+ }
+ },
+ {
+ "id": "release500_0312",
+ "category": "phonetic_balance",
+ "text": "It was over when he made his way through the ring of spectators.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0312.wav",
+ "render_seconds": 160.9956856520148,
+ "audio_seconds": 4.31891156462585,
+ "rtf": 37.27691184294068,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "And this is the one that is worth for the one of spectacles.",
+ "raw": {
+ "normalized_reference": "it was over when he made his way through the ring of spectators.",
+ "normalized_hypothesis": "and this is the one that is worth for the one of spectacles.",
+ "errors": 11,
+ "words": 13,
+ "wer": 0.8461538461538461
+ },
+ "semantic": {
+ "normalized_reference": "it was over when he made his way through the ring of spectators",
+ "normalized_hypothesis": "and this is the one that is worth for the one of spectacles",
+ "errors": 11,
+ "words": 13,
+ "wer": 0.8461538461538461
+ }
+ },
+ {
+ "id": "release500_0313",
+ "category": "phonetic_balance",
+ "text": "She had died from cold and starvation.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0313.wav",
+ "render_seconds": 138.3968471980188,
+ "audio_seconds": 2.9257142857142857,
+ "rtf": 47.30360988213533,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I don't want to do this to myself.",
+ "raw": {
+ "normalized_reference": "she had died from cold and starvation.",
+ "normalized_hypothesis": "i don't want to do this to myself.",
+ "errors": 8,
+ "words": 7,
+ "wer": 1.1428571428571428
+ },
+ "semantic": {
+ "normalized_reference": "she had died from cold and starvation",
+ "normalized_hypothesis": "i do not want to do this to myself",
+ "errors": 9,
+ "words": 7,
+ "wer": 1.2857142857142858
+ }
+ },
+ {
+ "id": "release500_0314",
+ "category": "phonetic_balance",
+ "text": "We fished sharks on Niihau together.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0314.wav",
+ "render_seconds": 84.799765813048,
+ "audio_seconds": 2.647074829931973,
+ "rtf": 32.03527337201392,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The first feature I tried.",
+ "raw": {
+ "normalized_reference": "we fished sharks on niihau together.",
+ "normalized_hypothesis": "the first feature i tried.",
+ "errors": 6,
+ "words": 6,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "we fished sharks on niihau together",
+ "normalized_hypothesis": "the 1st feature i tried",
+ "errors": 6,
+ "words": 6,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0315",
+ "category": "phonetic_balance",
+ "text": "His slim hands gripped the edges of the table.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0315.wav",
+ "render_seconds": 139.40758166799787,
+ "audio_seconds": 3.274013605442177,
+ "rtf": 42.580025152085454,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I still have quite the access of the top.",
+ "raw": {
+ "normalized_reference": "his slim hands gripped the edges of the table.",
+ "normalized_hypothesis": "i still have quite the access of the top.",
+ "errors": 6,
+ "words": 9,
+ "wer": 0.6666666666666666
+ },
+ "semantic": {
+ "normalized_reference": "his slim hands gripped the edges of the table",
+ "normalized_hypothesis": "i still have quite the access of the top",
+ "errors": 6,
+ "words": 9,
+ "wer": 0.6666666666666666
+ }
+ },
+ {
+ "id": "release500_0316",
+ "category": "phonetic_balance",
+ "text": "We had been chased by them ourselves, more than once.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0316.wav",
+ "render_seconds": 160.19796254707035,
+ "audio_seconds": 3.7616326530612243,
+ "rtf": 42.58734898440998,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "What happens when a brother or sister lets out a kiss?",
+ "raw": {
+ "normalized_reference": "we had been chased by them ourselves, more than once.",
+ "normalized_hypothesis": "what happens when a brother or sister lets out a kiss?",
+ "errors": 11,
+ "words": 10,
+ "wer": 1.1
+ },
+ "semantic": {
+ "normalized_reference": "we had been chased by them ourselves more than once",
+ "normalized_hypothesis": "what happens when a brother or sister lets out a kiss",
+ "errors": 11,
+ "words": 10,
+ "wer": 1.1
+ }
+ },
+ {
+ "id": "release500_0317",
+ "category": "phonetic_balance",
+ "text": "Or have they already devised one.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0317.wav",
+ "render_seconds": 118.99141736305319,
+ "audio_seconds": 2.5774149659863945,
+ "rtf": 46.1669614452302,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "What do you want to see us do?",
+ "raw": {
+ "normalized_reference": "or have they already devised one.",
+ "normalized_hypothesis": "what do you want to see us do?",
+ "errors": 8,
+ "words": 6,
+ "wer": 1.3333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "or have they already devised one",
+ "normalized_hypothesis": "what do you want to see us do",
+ "errors": 8,
+ "words": 6,
+ "wer": 1.3333333333333333
+ }
+ },
+ {
+ "id": "release500_0318",
+ "category": "phonetic_balance",
+ "text": "I cannot follow you, she said.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0318.wav",
+ "render_seconds": 110.09986936999485,
+ "audio_seconds": 2.2987755102040817,
+ "rtf": 47.89500709454676,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "What can I say? I'm shocked.",
+ "raw": {
+ "normalized_reference": "i cannot follow you, she said.",
+ "normalized_hypothesis": "what can i say? i'm shocked.",
+ "errors": 6,
+ "words": 6,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "i cannot follow you she said",
+ "normalized_hypothesis": "what can i say i am shocked",
+ "errors": 7,
+ "words": 6,
+ "wer": 1.1666666666666667
+ }
+ },
+ {
+ "id": "release500_0319",
+ "category": "phonetic_balance",
+ "text": "He went down in midstream, searching the shadows of both shores.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0319.wav",
+ "render_seconds": 160.31019487290177,
+ "audio_seconds": 4.388571428571429,
+ "rtf": 36.52901575879923,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I want to go to that school. I want to go to that school. I want to go to that school. I want to go to that school. I want to go to that school. I want to go to that school. I want to go to that school. I want to go to that school. I want to go to that school. I want to go to that school. I want to go to that school. I want to go to that school. I want to go to that school. I want to go to that school. I want to go to that school. I want to go to that school. I want to go to that school. I want to go to that school. I want to go to that school. I want to go to that school. I want to go to that school. I want to go to that school. I want to go to that school. I want to go to that school. I want to go to that school. I want to go to that school. I want to go to that school. I want to go to that school. I want to go to that school. I want to go to that school. I want to go to that school. I want to go to that school. I want to go to that school. I want to go to that school. I want to go to that school. I want to go to that school. I want to go to that school. I want to go to that school. I want to go to that school. I want to go to that school. I want to go to that school. I want to go to that school. I want to go to that school. I want to go to that school. I want to go to that school. I want to go to that school. I want to go to that school. I want to go to that school. I want to go to that school. I want to go to that school. I want to go to that school. I want to go to that school. I want to go to that school. I want to go to that school. I want to go to that school. I want to go",
+ "raw": {
+ "normalized_reference": "he went down in midstream, searching the shadows of both shores.",
+ "normalized_hypothesis": "i want to go to that school. i want to go to that school. i want to go to that school. i want to go to that school. i want to go to that school. i want to go to that school. i want to go to that school. i want to go to that school. i want to go to that school. i want to go to that school. i want to go to that school. i want to go to that school. i want to go to that school. i want to go to that school. i want to go to that school. i want to go to that school. i want to go to that school. i want to go to that school. i want to go to that school. i want to go to that school. i want to go to that school. i want to go to that school. i want to go to that school. i want to go to that school. i want to go to that school. i want to go to that school. i want to go to that school. i want to go to that school. i want to go to that school. i want to go to that school. i want to go to that school. i want to go to that school. i want to go to that school. i want to go to that school. i want to go to that school. i want to go to that school. i want to go to that school. i want to go to that school. i want to go to that school. i want to go to that school. i want to go to that school. i want to go to that school. i want to go to that school. i want to go to that school. i want to go to that school. i want to go to that school. i want to go to that school. i want to go to that school. i want to go to that school. i want to go to that school. i want to go to that school. i want to go to that school. i want to go to that school. i want to go to that school. i want to go to that school. i want to go",
+ "errors": 389,
+ "words": 11,
+ "wer": 35.36363636363637
+ },
+ "semantic": {
+ "normalized_reference": "he went down in midstream searching the shadows of both shores",
+ "normalized_hypothesis": "i want to go to that school i want to go to that school i want to go to that school i want to go to that school i want to go to that school i want to go to that school i want to go to that school i want to go to that school i want to go to that school i want to go to that school i want to go to that school i want to go to that school i want to go to that school i want to go to that school i want to go to that school i want to go to that school i want to go to that school i want to go to that school i want to go to that school i want to go to that school i want to go to that school i want to go to that school i want to go to that school i want to go to that school i want to go to that school i want to go to that school i want to go to that school i want to go to that school i want to go to that school i want to go to that school i want to go to that school i want to go to that school i want to go to that school i want to go to that school i want to go to that school i want to go to that school i want to go to that school i want to go to that school i want to go to that school i want to go to that school i want to go to that school i want to go to that school i want to go to that school i want to go to that school i want to go to that school i want to go to that school i want to go to that school i want to go to that school i want to go to that school i want to go to that school i want to go to that school i want to go to that school i want to go to that school i want to go to that school i want to go to that school i want to go",
+ "errors": 389,
+ "words": 11,
+ "wer": 35.36363636363637
+ }
+ },
+ {
+ "id": "release500_0320",
+ "category": "phonetic_balance",
+ "text": "I was sick once -- typhoid.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0320.wav",
+ "render_seconds": 98.59971724194475,
+ "audio_seconds": 2.2291156462585033,
+ "rtf": 44.23266123829919,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I will shut this stuff up.",
+ "raw": {
+ "normalized_reference": "i was sick once -- typhoid.",
+ "normalized_hypothesis": "i will shut this stuff up.",
+ "errors": 5,
+ "words": 5,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "i was sick once typhoid",
+ "normalized_hypothesis": "i will shut this stuff up",
+ "errors": 5,
+ "words": 5,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0321",
+ "category": "phonetic_balance",
+ "text": "Why not like any railroad station or ferry depot.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0321.wav",
+ "render_seconds": 150.5955464920262,
+ "audio_seconds": 3.622312925170068,
+ "rtf": 41.57441656837412,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "You are all welcome to watch the show for the book.",
+ "raw": {
+ "normalized_reference": "why not like any railroad station or ferry depot.",
+ "normalized_hypothesis": "you are all welcome to watch the show for the book.",
+ "errors": 11,
+ "words": 9,
+ "wer": 1.2222222222222223
+ },
+ "semantic": {
+ "normalized_reference": "why not like any railroad station or ferry depot",
+ "normalized_hypothesis": "you are all welcome to watch the show for the book",
+ "errors": 11,
+ "words": 9,
+ "wer": 1.2222222222222223
+ }
+ },
+ {
+ "id": "release500_0322",
+ "category": "phonetic_balance",
+ "text": "Red-Eye swung back and forth on the branch farther down.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0322.wav",
+ "render_seconds": 145.4036583720008,
+ "audio_seconds": 3.900952380952381,
+ "rtf": 37.27388703383809,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I'll see you back in the future. Bye for now.",
+ "raw": {
+ "normalized_reference": "red-eye swung back and forth on the branch farther down.",
+ "normalized_hypothesis": "i'll see you back in the future. bye for now.",
+ "errors": 10,
+ "words": 11,
+ "wer": 0.9090909090909091
+ },
+ "semantic": {
+ "normalized_reference": "red eye swung back and forth on the branch farther down",
+ "normalized_hypothesis": "i will see you back in the future bye for now",
+ "errors": 11,
+ "words": 11,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0323",
+ "category": "phonetic_balance",
+ "text": "Within himself he called it no longer his own.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0323.wav",
+ "render_seconds": 123.80804807099048,
+ "audio_seconds": 3.0650340136054424,
+ "rtf": 40.39369466094549,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Thank you for watching.",
+ "raw": {
+ "normalized_reference": "within himself he called it no longer his own.",
+ "normalized_hypothesis": "thank you for watching.",
+ "errors": 9,
+ "words": 9,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "within himself he called it no longer his own",
+ "normalized_hypothesis": "thank you for watching",
+ "errors": 9,
+ "words": 9,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0324",
+ "category": "phonetic_balance",
+ "text": "It lasted as a deterrent for two days.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0324.wav",
+ "render_seconds": 131.70378227403853,
+ "audio_seconds": 2.8560544217687074,
+ "rtf": 46.113890992482055,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "At last, I get to talk for two hours.",
+ "raw": {
+ "normalized_reference": "it lasted as a deterrent for two days.",
+ "normalized_hypothesis": "at last, i get to talk for two hours.",
+ "errors": 7,
+ "words": 8,
+ "wer": 0.875
+ },
+ "semantic": {
+ "normalized_reference": "it lasted as a deterrent for 2 days",
+ "normalized_hypothesis": "at last i get to talk for 2 hours",
+ "errors": 7,
+ "words": 8,
+ "wer": 0.875
+ }
+ },
+ {
+ "id": "release500_0325",
+ "category": "phonetic_balance",
+ "text": "He loved to play Chinese lottery.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0325.wav",
+ "render_seconds": 119.28722140903119,
+ "audio_seconds": 2.4380952380952383,
+ "rtf": 48.92639940604794,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I used to be a troll as you.",
+ "raw": {
+ "normalized_reference": "he loved to play chinese lottery.",
+ "normalized_hypothesis": "i used to be a troll as you.",
+ "errors": 7,
+ "words": 6,
+ "wer": 1.1666666666666667
+ },
+ "semantic": {
+ "normalized_reference": "he loved to play chinese lottery",
+ "normalized_hypothesis": "i used to be a troll as you",
+ "errors": 7,
+ "words": 6,
+ "wer": 1.1666666666666667
+ }
+ },
+ {
+ "id": "release500_0326",
+ "category": "phonetic_balance",
+ "text": "Lord, but I'm glad to see you again, Phil.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0326.wav",
+ "render_seconds": 99.2118411361007,
+ "audio_seconds": 2.9257142857142857,
+ "rtf": 33.91029726331567,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "You better get to see what I say.",
+ "raw": {
+ "normalized_reference": "lord, but i'm glad to see you again, phil.",
+ "normalized_hypothesis": "you better get to see what i say.",
+ "errors": 7,
+ "words": 9,
+ "wer": 0.7777777777777778
+ },
+ "semantic": {
+ "normalized_reference": "lord but i am glad to see you again phil",
+ "normalized_hypothesis": "you better get to see what i say",
+ "errors": 8,
+ "words": 10,
+ "wer": 0.8
+ }
+ },
+ {
+ "id": "release500_0327",
+ "category": "phonetic_balance",
+ "text": "There is not an iota of truth in it, certainly not.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0327.wav",
+ "render_seconds": 145.19193411397282,
+ "audio_seconds": 3.6919727891156464,
+ "rtf": 39.32638250802256,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I want you to believe me for a second. Shut your mouth.",
+ "raw": {
+ "normalized_reference": "there is not an iota of truth in it, certainly not.",
+ "normalized_hypothesis": "i want you to believe me for a second. shut your mouth.",
+ "errors": 12,
+ "words": 11,
+ "wer": 1.0909090909090908
+ },
+ "semantic": {
+ "normalized_reference": "there is not an iota of truth in it certainly not",
+ "normalized_hypothesis": "i want you to believe me for a 2nd shut your mouth",
+ "errors": 12,
+ "words": 11,
+ "wer": 1.0909090909090908
+ }
+ },
+ {
+ "id": "release500_0328",
+ "category": "phonetic_balance",
+ "text": "She added, with genuine sympathy in her face and voice.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0328.wav",
+ "render_seconds": 145.19768194598146,
+ "audio_seconds": 3.9706122448979593,
+ "rtf": 36.5680839604908,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "How about we call a little show for some old-fashioned fish?",
+ "raw": {
+ "normalized_reference": "she added, with genuine sympathy in her face and voice.",
+ "normalized_hypothesis": "how about we call a little show for some old-fashioned fish?",
+ "errors": 12,
+ "words": 10,
+ "wer": 1.2
+ },
+ "semantic": {
+ "normalized_reference": "she added with genuine sympathy in her face and voice",
+ "normalized_hypothesis": "how about we call a little show for some old fashioned fish",
+ "errors": 12,
+ "words": 10,
+ "wer": 1.2
+ }
+ },
+ {
+ "id": "release500_0329",
+ "category": "phonetic_balance",
+ "text": "We have plenty of capital ourselves, and yet we want more.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0329.wav",
+ "render_seconds": 150.00380608905107,
+ "audio_seconds": 4.179591836734694,
+ "rtf": 35.88958251154054,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Last time I was proud of yourself. I'm not mad at you.",
+ "raw": {
+ "normalized_reference": "we have plenty of capital ourselves, and yet we want more.",
+ "normalized_hypothesis": "last time i was proud of yourself. i'm not mad at you.",
+ "errors": 12,
+ "words": 11,
+ "wer": 1.0909090909090908
+ },
+ "semantic": {
+ "normalized_reference": "we have plenty of capital ourselves and yet we want more",
+ "normalized_hypothesis": "last time i was proud of yourself i am not mad at you",
+ "errors": 12,
+ "words": 11,
+ "wer": 1.0909090909090908
+ }
+ },
+ {
+ "id": "release500_0330",
+ "category": "phonetic_balance",
+ "text": "And now put yourself in my place for a moment.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0330.wav",
+ "render_seconds": 126.39591903507244,
+ "audio_seconds": 3.274013605442177,
+ "rtf": 38.60580140075559,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I hope that I'll show some of that trash for a moment.",
+ "raw": {
+ "normalized_reference": "and now put yourself in my place for a moment.",
+ "normalized_hypothesis": "i hope that i'll show some of that trash for a moment.",
+ "errors": 9,
+ "words": 10,
+ "wer": 0.9
+ },
+ "semantic": {
+ "normalized_reference": "and now put yourself in my place for a moment",
+ "normalized_hypothesis": "i hope that i will show some of that trash for a moment",
+ "errors": 10,
+ "words": 10,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0331",
+ "category": "phonetic_balance",
+ "text": "Fast, but endure.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0331.wav",
+ "render_seconds": 81.29378070903476,
+ "audio_seconds": 1.6718367346938776,
+ "rtf": 48.62543035574589,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Thanks for watching!",
+ "raw": {
+ "normalized_reference": "fast, but endure.",
+ "normalized_hypothesis": "thanks for watching!",
+ "errors": 3,
+ "words": 3,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "fast but endure",
+ "normalized_hypothesis": "thanks for watching",
+ "errors": 3,
+ "words": 3,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0332",
+ "category": "phonetic_balance",
+ "text": "You were looking squeamish this afternoon, he began.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0332.wav",
+ "render_seconds": 74.608209892991,
+ "audio_seconds": 3.4829931972789114,
+ "rtf": 21.420716512245463,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I won't let you spoil the show for me. Bye-bye.",
+ "raw": {
+ "normalized_reference": "you were looking squeamish this afternoon, he began.",
+ "normalized_hypothesis": "i won't let you spoil the show for me. bye-bye.",
+ "errors": 10,
+ "words": 8,
+ "wer": 1.25
+ },
+ "semantic": {
+ "normalized_reference": "you were looking squeamish this afternoon he began",
+ "normalized_hypothesis": "i will not let you spoil the show for me bye bye",
+ "errors": 11,
+ "words": 8,
+ "wer": 1.375
+ }
+ },
+ {
+ "id": "release500_0333",
+ "category": "phonetic_balance",
+ "text": "It is the nearest refuge.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0333.wav",
+ "render_seconds": 67.5107208150439,
+ "audio_seconds": 2.020136054421769,
+ "rtf": 33.418898032770244,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Arsenault Schofield.",
+ "raw": {
+ "normalized_reference": "it is the nearest refuge.",
+ "normalized_hypothesis": "arsenault schofield.",
+ "errors": 5,
+ "words": 5,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "it is the nearest refuge",
+ "normalized_hypothesis": "arsenault schofield",
+ "errors": 5,
+ "words": 5,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0334",
+ "category": "phonetic_balance",
+ "text": "I want to die in it.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0334.wav",
+ "render_seconds": 97.8909430099884,
+ "audio_seconds": 1.7414965986394557,
+ "rtf": 56.21081493151678,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you. Thank you.",
+ "raw": {
+ "normalized_reference": "i want to die in it.",
+ "normalized_hypothesis": "thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you. thank you.",
+ "errors": 296,
+ "words": 6,
+ "wer": 49.333333333333336
+ },
+ "semantic": {
+ "normalized_reference": "i want to die in it",
+ "normalized_hypothesis": "thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you thank you",
+ "errors": 296,
+ "words": 6,
+ "wer": 49.333333333333336
+ }
+ },
+ {
+ "id": "release500_0335",
+ "category": "phonetic_balance",
+ "text": "He stopped, and Philip nodded at the horrified question in his eyes.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0335.wav",
+ "render_seconds": 160.99368157691788,
+ "audio_seconds": 4.73687074829932,
+ "rtf": 33.98734948271905,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I still can't find the word I want to use for this conversation.",
+ "raw": {
+ "normalized_reference": "he stopped, and philip nodded at the horrified question in his eyes.",
+ "normalized_hypothesis": "i still can't find the word i want to use for this conversation.",
+ "errors": 13,
+ "words": 12,
+ "wer": 1.0833333333333333
+ },
+ "semantic": {
+ "normalized_reference": "he stopped and philip nodded at the horrified question in his eyes",
+ "normalized_hypothesis": "i still can not find the word i want to use for this conversation",
+ "errors": 14,
+ "words": 12,
+ "wer": 1.1666666666666667
+ }
+ },
+ {
+ "id": "release500_0336",
+ "category": "phonetic_balance",
+ "text": "She turned, fearing that Jacques might see what was in her face.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0336.wav",
+ "render_seconds": 168.80214889801573,
+ "audio_seconds": 4.249251700680272,
+ "rtf": 39.72514710554609,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I told someone not to watch a nutritional fast.",
+ "raw": {
+ "normalized_reference": "she turned, fearing that jacques might see what was in her face.",
+ "normalized_hypothesis": "i told someone not to watch a nutritional fast.",
+ "errors": 12,
+ "words": 12,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "she turned fearing that jacques might see what was in her face",
+ "normalized_hypothesis": "i told someone not to watch a nutritional fast",
+ "errors": 12,
+ "words": 12,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0337",
+ "category": "phonetic_balance",
+ "text": "It was a curious coincidence.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0337.wav",
+ "render_seconds": 115.40780701104086,
+ "audio_seconds": 2.4380952380952383,
+ "rtf": 47.33523334437222,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I wish I could last a lot of seconds.",
+ "raw": {
+ "normalized_reference": "it was a curious coincidence.",
+ "normalized_hypothesis": "i wish i could last a lot of seconds.",
+ "errors": 8,
+ "words": 5,
+ "wer": 1.6
+ },
+ "semantic": {
+ "normalized_reference": "it was a curious coincidence",
+ "normalized_hypothesis": "i wish i could last a lot of seconds",
+ "errors": 8,
+ "words": 5,
+ "wer": 1.6
+ }
+ },
+ {
+ "id": "release500_0338",
+ "category": "phonetic_balance",
+ "text": "Daylight made no answer, and the door closed behind him.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0338.wav",
+ "render_seconds": 88.88503494800534,
+ "audio_seconds": 3.9706122448979593,
+ "rtf": 22.385725290153484,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I look like a monster. I look like a monster. I look like a monster. I look like a monster. I look like a monster. I look like a monster. I look like a monster. I look like a monster. I look like a monster. I look like a monster. I look like a monster. I look like a monster. I look like a monster. I look like a monster. I look like a monster. I look like a monster. I look like a monster. I look like a monster. I look like a monster. I look like a monster. I look like a monster. I look like a monster. I look like a monster. I look like a monster. I look like a monster. I look like a monster. I look like a monster. I look like a monster. I look like a monster. I look like a monster. I look like a monster. I look like a monster. I look like a monster. I look like a monster. I look like a monster. I look like a monster. I look like a monster. I look like a monster. I look like a monster. I look like a monster. I look like a monster. I look like a monster. I look like a monster. I look like a monster. I look like a monster. I look like a monster. I look like a monster. I look like a monster. I look like a monster. I look like a monster. I look like a monster. I look like a monster. I look like a monster. I look like a monster. I look like a monster. I look like a monster. I look like a monster. I look like a monster. I look like a monster. I look like a monster. I look like a monster. I look like a monster. I look like a monster. I look like a monster. I look like a monster. I look like a monster. I look like a monster. I look like a monster. I look like a monster. I look like a monster. I look like a monster. I look like a monster. I look like a monster. I look like a monster.",
+ "raw": {
+ "normalized_reference": "daylight made no answer, and the door closed behind him.",
+ "normalized_hypothesis": "i look like a monster. i look like a monster. i look like a monster. i look like a monster. i look like a monster. i look like a monster. i look like a monster. i look like a monster. i look like a monster. i look like a monster. i look like a monster. i look like a monster. i look like a monster. i look like a monster. i look like a monster. i look like a monster. i look like a monster. i look like a monster. i look like a monster. i look like a monster. i look like a monster. i look like a monster. i look like a monster. i look like a monster. i look like a monster. i look like a monster. i look like a monster. i look like a monster. i look like a monster. i look like a monster. i look like a monster. i look like a monster. i look like a monster. i look like a monster. i look like a monster. i look like a monster. i look like a monster. i look like a monster. i look like a monster. i look like a monster. i look like a monster. i look like a monster. i look like a monster. i look like a monster. i look like a monster. i look like a monster. i look like a monster. i look like a monster. i look like a monster. i look like a monster. i look like a monster. i look like a monster. i look like a monster. i look like a monster. i look like a monster. i look like a monster. i look like a monster. i look like a monster. i look like a monster. i look like a monster. i look like a monster. i look like a monster. i look like a monster. i look like a monster. i look like a monster. i look like a monster. i look like a monster. i look like a monster. i look like a monster. i look like a monster. i look like a monster. i look like a monster. i look like a monster. i look like a monster.",
+ "errors": 370,
+ "words": 10,
+ "wer": 37.0
+ },
+ "semantic": {
+ "normalized_reference": "daylight made no answer and the door closed behind him",
+ "normalized_hypothesis": "i look like a monster i look like a monster i look like a monster i look like a monster i look like a monster i look like a monster i look like a monster i look like a monster i look like a monster i look like a monster i look like a monster i look like a monster i look like a monster i look like a monster i look like a monster i look like a monster i look like a monster i look like a monster i look like a monster i look like a monster i look like a monster i look like a monster i look like a monster i look like a monster i look like a monster i look like a monster i look like a monster i look like a monster i look like a monster i look like a monster i look like a monster i look like a monster i look like a monster i look like a monster i look like a monster i look like a monster i look like a monster i look like a monster i look like a monster i look like a monster i look like a monster i look like a monster i look like a monster i look like a monster i look like a monster i look like a monster i look like a monster i look like a monster i look like a monster i look like a monster i look like a monster i look like a monster i look like a monster i look like a monster i look like a monster i look like a monster i look like a monster i look like a monster i look like a monster i look like a monster i look like a monster i look like a monster i look like a monster i look like a monster i look like a monster i look like a monster i look like a monster i look like a monster i look like a monster i look like a monster i look like a monster i look like a monster i look like a monster i look like a monster",
+ "errors": 370,
+ "words": 10,
+ "wer": 37.0
+ }
+ },
+ {
+ "id": "release500_0339",
+ "category": "phonetic_balance",
+ "text": "And he did hurt my arm.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0339.wav",
+ "render_seconds": 72.90102104295511,
+ "audio_seconds": 1.9504761904761905,
+ "rtf": 37.376011765186945,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I'm not back home for you.",
+ "raw": {
+ "normalized_reference": "and he did hurt my arm.",
+ "normalized_hypothesis": "i'm not back home for you.",
+ "errors": 6,
+ "words": 6,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "and he did hurt my arm",
+ "normalized_hypothesis": "i am not back home for you",
+ "errors": 7,
+ "words": 6,
+ "wer": 1.1666666666666667
+ }
+ },
+ {
+ "id": "release500_0340",
+ "category": "phonetic_balance",
+ "text": "Dunham, can your boy go along with Jesse.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0340.wav",
+ "render_seconds": 123.90290090604685,
+ "audio_seconds": 2.9257142857142857,
+ "rtf": 42.34962433312148,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Hello, and welcome to a new video.",
+ "raw": {
+ "normalized_reference": "dunham, can your boy go along with jesse.",
+ "normalized_hypothesis": "hello, and welcome to a new video.",
+ "errors": 8,
+ "words": 8,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "dunham can your boy go along with jesse",
+ "normalized_hypothesis": "hello and welcome to a new video",
+ "errors": 8,
+ "words": 8,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0341",
+ "category": "phonetic_balance",
+ "text": "The creative joy, I murmured.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0341.wav",
+ "render_seconds": 124.90152400592342,
+ "audio_seconds": 2.36843537414966,
+ "rtf": 52.73588013807083,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "God bless. God bless.",
+ "raw": {
+ "normalized_reference": "the creative joy, i murmured.",
+ "normalized_hypothesis": "god bless. god bless.",
+ "errors": 5,
+ "words": 5,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "the creative joy i murmured",
+ "normalized_hypothesis": "god bless god bless",
+ "errors": 5,
+ "words": 5,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0342",
+ "category": "phonetic_balance",
+ "text": "He pulled, and the log crashed down to break his back.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0342.wav",
+ "render_seconds": 145.49746312305797,
+ "audio_seconds": 3.8312925170068026,
+ "rtf": 37.976077910315205,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I'm Pete, and I'll see you in the next video.",
+ "raw": {
+ "normalized_reference": "he pulled, and the log crashed down to break his back.",
+ "normalized_hypothesis": "i'm pete, and i'll see you in the next video.",
+ "errors": 10,
+ "words": 11,
+ "wer": 0.9090909090909091
+ },
+ "semantic": {
+ "normalized_reference": "he pulled and the log crashed down to break his back",
+ "normalized_hypothesis": "i am pete and i will see you in the next video",
+ "errors": 11,
+ "words": 11,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0343",
+ "category": "phonetic_balance",
+ "text": "By golly, the boy wins.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0343.wav",
+ "render_seconds": 99.89419081900269,
+ "audio_seconds": 2.020136054421769,
+ "rtf": 49.44923912443896,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Brrrr, a Brrrr, Brrrr, Brrrr, Brrrr, Brrrr, Brrrr, Brrrr, Brrrr, Brrrr, Brrrr, Brrrr, Brrrr, Brrrr, Brrrr, Brrrr, Brrrr, Brrrr, Brrrr, Brrrr, Brrrr, Brrrr, Brrrr, Brrrr, Brrrr, Brrrr, Brrrr, Brrrr, Brrrr, Brrrr, Brrrr, Brrrr, Brrrr, Brrrr, Brrrr, Brrrr, Brrrr, Brrrr, Brrrr, Brrrr, Brrrr, Brrrr, Brrrr, Brrrr, Brrrr, Brrrr, Brrrr, Brrrr, Brrrr, Brrrr, Brrrr, Brrrr, Brrrr, Brrrr, Brrrr, Brrrr, Brrrr, Brrrr, Brrrr, Brrrr, Brrrr, Brrrr, Brrrr, Brrrr, Brrrr, Brrrr, Brrrr, Brrrr, Brrrr, Brrrr, Brrrr, Brrrr, Brrrr, Brrrr, Brrrr, Brrrr, Brrrr, Brrrr, Brrrr, Brrrr, Brrrr, Brrrr, Brrrr, Brrrr, Brrrr, Brrrr, Brrrr, Brrrr, Brrrr, Brrrr, Brrrr, Brrrr, Brrrr, Brrrr, Brrrr, Brrrr, Brrrr, Brrrr, Brrrr, Brrrr, Brrrr, Brrrr, Brrrr, Brrrr, Brrrr, Brrrr, Brrr, Brrrr, Brrr, Brrrr, Brrr, Brrr",
+ "raw": {
+ "normalized_reference": "by golly, the boy wins.",
+ "normalized_hypothesis": "brrrr, a brrrr, brrrr, brrrr, brrrr, brrrr, brrrr, brrrr, brrrr, brrrr, brrrr, brrrr, brrrr, brrrr, brrrr, brrrr, brrrr, brrrr, brrrr, brrrr, brrrr, brrrr, brrrr, brrrr, brrrr, brrrr, brrrr, brrrr, brrrr, brrrr, brrrr, brrrr, brrrr, brrrr, brrrr, brrrr, brrrr, brrrr, brrrr, brrrr, brrrr, brrrr, brrrr, brrrr, brrrr, brrrr, brrrr, brrrr, brrrr, brrrr, brrrr, brrrr, brrrr, brrrr, brrrr, brrrr, brrrr, brrrr, brrrr, brrrr, brrrr, brrrr, brrrr, brrrr, brrrr, brrrr, brrrr, brrrr, brrrr, brrrr, brrrr, brrrr, brrrr, brrrr, brrrr, brrrr, brrrr, brrrr, brrrr, brrrr, brrrr, brrrr, brrrr, brrrr, brrrr, brrrr, brrrr, brrrr, brrrr, brrrr, brrrr, brrrr, brrrr, brrrr, brrrr, brrrr, brrrr, brrrr, brrrr, brrrr, brrrr, brrrr, brrrr, brrrr, brrrr, brrrr, brrr, brrrr, brrr, brrrr, brrr, brrr",
+ "errors": 113,
+ "words": 5,
+ "wer": 22.6
+ },
+ "semantic": {
+ "normalized_reference": "by golly the boy wins",
+ "normalized_hypothesis": "brrrr a brrrr brrrr brrrr brrrr brrrr brrrr brrrr brrrr brrrr brrrr brrrr brrrr brrrr brrrr brrrr brrrr brrrr brrrr brrrr brrrr brrrr brrrr brrrr brrrr brrrr brrrr brrrr brrrr brrrr brrrr brrrr brrrr brrrr brrrr brrrr brrrr brrrr brrrr brrrr brrrr brrrr brrrr brrrr brrrr brrrr brrrr brrrr brrrr brrrr brrrr brrrr brrrr brrrr brrrr brrrr brrrr brrrr brrrr brrrr brrrr brrrr brrrr brrrr brrrr brrrr brrrr brrrr brrrr brrrr brrrr brrrr brrrr brrrr brrrr brrrr brrrr brrrr brrrr brrrr brrrr brrrr brrrr brrrr brrrr brrrr brrrr brrrr brrrr brrrr brrrr brrrr brrrr brrrr brrrr brrrr brrrr brrrr brrrr brrrr brrrr brrrr brrrr brrrr brrrr brrrr brrr brrrr brrr brrrr brrr brrr",
+ "errors": 113,
+ "words": 5,
+ "wer": 22.6
+ }
+ },
+ {
+ "id": "release500_0344",
+ "category": "phonetic_balance",
+ "text": "The boy threw back his head with pride.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0344.wav",
+ "render_seconds": 55.69395537499804,
+ "audio_seconds": 2.716734693877551,
+ "rtf": 20.500329126914842,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The bear fell down and his head was torn.",
+ "raw": {
+ "normalized_reference": "the boy threw back his head with pride.",
+ "normalized_hypothesis": "the bear fell down and his head was torn.",
+ "errors": 6,
+ "words": 8,
+ "wer": 0.75
+ },
+ "semantic": {
+ "normalized_reference": "the boy threw back his head with pride",
+ "normalized_hypothesis": "the bear fell down and his head was torn",
+ "errors": 6,
+ "words": 8,
+ "wer": 0.75
+ }
+ },
+ {
+ "id": "release500_0345",
+ "category": "phonetic_balance",
+ "text": "The last one I knew was an overseer.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0345.wav",
+ "render_seconds": 121.69537639594637,
+ "audio_seconds": 2.647074829931973,
+ "rtf": 45.97353086503936,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Yes, we all know this is over, son.",
+ "raw": {
+ "normalized_reference": "the last one i knew was an overseer.",
+ "normalized_hypothesis": "yes, we all know this is over, son.",
+ "errors": 8,
+ "words": 8,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "the last one i knew was an overseer",
+ "normalized_hypothesis": "yes we all know this is over son",
+ "errors": 8,
+ "words": 8,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0346",
+ "category": "phonetic_balance",
+ "text": "Two years ago I gave up civilization for this.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0346.wav",
+ "render_seconds": 146.50045349705033,
+ "audio_seconds": 3.4133333333333336,
+ "rtf": 42.920054735463964,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "...........................................................................................................................................................................................................................................................................................................................................................................................",
+ "raw": {
+ "normalized_reference": "two years ago i gave up civilization for this.",
+ "normalized_hypothesis": "...........................................................................................................................................................................................................................................................................................................................................................................................",
+ "errors": 9,
+ "words": 9,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "2 years ago i gave up civilization for this",
+ "normalized_hypothesis": " .",
+ "errors": 9,
+ "words": 9,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0347",
+ "category": "phonetic_balance",
+ "text": "The river bared its bosom, and snorting steamboats challenged the wilderness.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0347.wav",
+ "render_seconds": 199.09577607398387,
+ "audio_seconds": 5.50312925170068,
+ "rtf": 36.1786479960389,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The weather is bad and special, and so are the starbursts, high and stilettos.",
+ "raw": {
+ "normalized_reference": "the river bared its bosom, and snorting steamboats challenged the wilderness.",
+ "normalized_hypothesis": "the weather is bad and special, and so are the starbursts, high and stilettos.",
+ "errors": 12,
+ "words": 11,
+ "wer": 1.0909090909090908
+ },
+ "semantic": {
+ "normalized_reference": "the river bared its bosom and snorting steamboats challenged the wilderness",
+ "normalized_hypothesis": "the weather is bad and special and so are the starbursts high and stilettos",
+ "errors": 12,
+ "words": 11,
+ "wer": 1.0909090909090908
+ }
+ },
+ {
+ "id": "release500_0348",
+ "category": "phonetic_balance",
+ "text": "Mab, she said.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0348.wav",
+ "render_seconds": 80.50041246297769,
+ "audio_seconds": 1.6021768707482993,
+ "rtf": 50.24439806410378,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Thanks for watching!",
+ "raw": {
+ "normalized_reference": "mab, she said.",
+ "normalized_hypothesis": "thanks for watching!",
+ "errors": 3,
+ "words": 3,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "mab she said",
+ "normalized_hypothesis": "thanks for watching",
+ "errors": 3,
+ "words": 3,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0349",
+ "category": "phonetic_balance",
+ "text": "The regret in his voice was provocative of a second burst of laughter.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0349.wav",
+ "render_seconds": 176.50368143501692,
+ "audio_seconds": 4.876190476190477,
+ "rtf": 36.19704404429058,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": ".....................................................................................................................................................................................................................................................................................................................................................................",
+ "raw": {
+ "normalized_reference": "the regret in his voice was provocative of a second burst of laughter.",
+ "normalized_hypothesis": ".....................................................................................................................................................................................................................................................................................................................................................................",
+ "errors": 13,
+ "words": 13,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "the regret in his voice was provocative of a 2nd burst of laughter",
+ "normalized_hypothesis": " .",
+ "errors": 13,
+ "words": 13,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0350",
+ "category": "phonetic_balance",
+ "text": "And each year something happened, and I did not go.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0350.wav",
+ "render_seconds": 103.49894928804133,
+ "audio_seconds": 3.622312925170068,
+ "rtf": 28.572614080044463,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "And that's how something happened. And that got me home.",
+ "raw": {
+ "normalized_reference": "and each year something happened, and i did not go.",
+ "normalized_hypothesis": "and that's how something happened. and that got me home.",
+ "errors": 6,
+ "words": 10,
+ "wer": 0.6
+ },
+ "semantic": {
+ "normalized_reference": "and each year something happened and i did not go",
+ "normalized_hypothesis": "and that is how something happened and that got me home",
+ "errors": 7,
+ "words": 10,
+ "wer": 0.7
+ }
+ },
+ {
+ "id": "release500_0351",
+ "category": "phonetic_balance",
+ "text": "Gad, your letter came just in time.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0351.wav",
+ "render_seconds": 114.69290540600196,
+ "audio_seconds": 2.647074829931973,
+ "rtf": 43.3281689316465,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Thank you for watching.",
+ "raw": {
+ "normalized_reference": "gad, your letter came just in time.",
+ "normalized_hypothesis": "thank you for watching.",
+ "errors": 7,
+ "words": 7,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "gad your letter came just in time",
+ "normalized_hypothesis": "thank you for watching",
+ "errors": 7,
+ "words": 7,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0352",
+ "category": "phonetic_balance",
+ "text": "But if Pierre did not return, until tomorrow.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0352.wav",
+ "render_seconds": 120.09822258690838,
+ "audio_seconds": 3.2043537414965986,
+ "rtf": 37.47970176688929,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "That's how I know what time it is.",
+ "raw": {
+ "normalized_reference": "but if pierre did not return, until tomorrow.",
+ "normalized_hypothesis": "that's how i know what time it is.",
+ "errors": 8,
+ "words": 8,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "but if pierre did not return until tomorrow",
+ "normalized_hypothesis": "that is how i know what time it is",
+ "errors": 9,
+ "words": 8,
+ "wer": 1.125
+ }
+ },
+ {
+ "id": "release500_0353",
+ "category": "phonetic_balance",
+ "text": "Billinger may arrive in time.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0353.wav",
+ "render_seconds": 91.80361258005723,
+ "audio_seconds": 2.2987755102040817,
+ "rtf": 39.93587550091268,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I'm in for a wild war for two.",
+ "raw": {
+ "normalized_reference": "billinger may arrive in time.",
+ "normalized_hypothesis": "i'm in for a wild war for two.",
+ "errors": 8,
+ "words": 5,
+ "wer": 1.6
+ },
+ "semantic": {
+ "normalized_reference": "billinger may arrive in time",
+ "normalized_hypothesis": "i am in for a wild war for 2",
+ "errors": 9,
+ "words": 5,
+ "wer": 1.8
+ }
+ },
+ {
+ "id": "release500_0354",
+ "category": "phonetic_balance",
+ "text": "In such a tumbling of values was no time to sell.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0354.wav",
+ "render_seconds": 124.19518921303097,
+ "audio_seconds": 3.4829931972789114,
+ "rtf": 35.657603152960064,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "And such a terrible thing as a firehouse was no time to say.",
+ "raw": {
+ "normalized_reference": "in such a tumbling of values was no time to sell.",
+ "normalized_hypothesis": "and such a terrible thing as a firehouse was no time to say.",
+ "errors": 7,
+ "words": 11,
+ "wer": 0.6363636363636364
+ },
+ "semantic": {
+ "normalized_reference": "in such a tumbling of values was no time to sell",
+ "normalized_hypothesis": "and such a terrible thing as a firehouse was no time to say",
+ "errors": 7,
+ "words": 11,
+ "wer": 0.6363636363636364
+ }
+ },
+ {
+ "id": "release500_0355",
+ "category": "phonetic_balance",
+ "text": "There's not an iota of truth in it.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0355.wav",
+ "render_seconds": 95.59949900896754,
+ "audio_seconds": 2.4380952380952383,
+ "rtf": 39.21073201539684,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you.",
+ "raw": {
+ "normalized_reference": "there's not an iota of truth in it.",
+ "normalized_hypothesis": "i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you.",
+ "errors": 333,
+ "words": 8,
+ "wer": 41.625
+ },
+ "semantic": {
+ "normalized_reference": "there is not an iota of truth in it",
+ "normalized_hypothesis": "i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you",
+ "errors": 333,
+ "words": 9,
+ "wer": 37.0
+ }
+ },
+ {
+ "id": "release500_0356",
+ "category": "phonetic_balance",
+ "text": "It is dog eat dog, and you ate them up.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0356.wav",
+ "render_seconds": 79.30646503996104,
+ "audio_seconds": 2.8560544217687074,
+ "rtf": 27.76784098912508,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "That's true. That's true. I know that too.",
+ "raw": {
+ "normalized_reference": "it is dog eat dog, and you ate them up.",
+ "normalized_hypothesis": "that's true. that's true. i know that too.",
+ "errors": 10,
+ "words": 10,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "it is dog eat dog and you ate them up",
+ "normalized_hypothesis": "that is true that is true i know that too",
+ "errors": 9,
+ "words": 10,
+ "wer": 0.9
+ }
+ },
+ {
+ "id": "release500_0357",
+ "category": "phonetic_balance",
+ "text": "You can take a vacation on pay.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0357.wav",
+ "render_seconds": 109.1108391280286,
+ "audio_seconds": 2.4380952380952383,
+ "rtf": 44.752492611105474,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I can't talk as I can't sleep.",
+ "raw": {
+ "normalized_reference": "you can take a vacation on pay.",
+ "normalized_hypothesis": "i can't talk as i can't sleep.",
+ "errors": 7,
+ "words": 7,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "you can take a vacation on pay",
+ "normalized_hypothesis": "i can not talk as i can not sleep",
+ "errors": 8,
+ "words": 7,
+ "wer": 1.1428571428571428
+ }
+ },
+ {
+ "id": "release500_0358",
+ "category": "phonetic_balance",
+ "text": "I have seen myself that one man contemplated by Pascal's philosophic eye.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0358.wav",
+ "render_seconds": 163.09745570202358,
+ "audio_seconds": 4.876190476190477,
+ "rtf": 33.447720407641555,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry. I'm sorry.",
+ "raw": {
+ "normalized_reference": "i have seen myself that one man contemplated by pascal's philosophic eye.",
+ "normalized_hypothesis": "i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry. i'm sorry.",
+ "errors": 222,
+ "words": 12,
+ "wer": 18.5
+ },
+ "semantic": {
+ "normalized_reference": "i have seen myself that one man contemplated by pascal is philosophic eye",
+ "normalized_hypothesis": "i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry i am sorry",
+ "errors": 332,
+ "words": 13,
+ "wer": 25.53846153846154
+ }
+ },
+ {
+ "id": "release500_0359",
+ "category": "phonetic_balance",
+ "text": "Ah, I had forgotten, he exclaimed.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0359.wav",
+ "render_seconds": 131.39737610006705,
+ "audio_seconds": 2.5774149659863945,
+ "rtf": 50.98029530909485,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Thanks for watching. I'll see you next time.",
+ "raw": {
+ "normalized_reference": "ah, i had forgotten, he exclaimed.",
+ "normalized_hypothesis": "thanks for watching. i'll see you next time.",
+ "errors": 8,
+ "words": 6,
+ "wer": 1.3333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "ah i had forgotten he exclaimed",
+ "normalized_hypothesis": "thanks for watching i will see you next time",
+ "errors": 8,
+ "words": 6,
+ "wer": 1.3333333333333333
+ }
+ },
+ {
+ "id": "release500_0360",
+ "category": "phonetic_balance",
+ "text": "Providence had delivered him through the maelstrom.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0360.wav",
+ "render_seconds": 115.09659487800673,
+ "audio_seconds": 3.4133333333333336,
+ "rtf": 33.71970553066603,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Brother has had to leave the house for the last call.",
+ "raw": {
+ "normalized_reference": "providence had delivered him through the maelstrom.",
+ "normalized_hypothesis": "brother has had to leave the house for the last call.",
+ "errors": 9,
+ "words": 7,
+ "wer": 1.2857142857142858
+ },
+ "semantic": {
+ "normalized_reference": "providence had delivered him through the maelstrom",
+ "normalized_hypothesis": "brother has had to leave the house for the last call",
+ "errors": 9,
+ "words": 7,
+ "wer": 1.2857142857142858
+ }
+ },
+ {
+ "id": "release500_0361",
+ "category": "phonetic_balance",
+ "text": "She was his now, forever.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0361.wav",
+ "render_seconds": 98.50019540800713,
+ "audio_seconds": 2.089795918367347,
+ "rtf": 47.133882568284655,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I wish us no further.",
+ "raw": {
+ "normalized_reference": "she was his now, forever.",
+ "normalized_hypothesis": "i wish us no further.",
+ "errors": 5,
+ "words": 5,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "she was his now forever",
+ "normalized_hypothesis": "i wish us no further",
+ "errors": 5,
+ "words": 5,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0362",
+ "category": "phonetic_balance",
+ "text": "It was unobtrusive, yet it was there.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0362.wav",
+ "render_seconds": 54.78923688700888,
+ "audio_seconds": 2.716734693877551,
+ "rtf": 20.167312255584697,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I wish I could have this now.",
+ "raw": {
+ "normalized_reference": "it was unobtrusive, yet it was there.",
+ "normalized_hypothesis": "i wish i could have this now.",
+ "errors": 7,
+ "words": 7,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "it was unobtrusive yet it was there",
+ "normalized_hypothesis": "i wish i could have this now",
+ "errors": 7,
+ "words": 7,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0363",
+ "category": "phonetic_balance",
+ "text": "Men like Joe Goose dated existence from drunk to drunk.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0363.wav",
+ "render_seconds": 150.6957912640646,
+ "audio_seconds": 3.9706122448979593,
+ "rtf": 37.95278460083864,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Thank you for watching.",
+ "raw": {
+ "normalized_reference": "men like joe goose dated existence from drunk to drunk.",
+ "normalized_hypothesis": "thank you for watching.",
+ "errors": 10,
+ "words": 10,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "men like joe goose dated existence from drunk to drunk",
+ "normalized_hypothesis": "thank you for watching",
+ "errors": 10,
+ "words": 10,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0364",
+ "category": "phonetic_balance",
+ "text": "I was completely lost in my work.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0364.wav",
+ "render_seconds": 113.29968268994708,
+ "audio_seconds": 2.5774149659863945,
+ "rtf": 43.95865011460679,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Oh, it's cold outside. I wish I were warm. I wish I were warm. I wish I were warm. I wish I were warm. I wish I were warm. I wish I were warm. I wish I were warm. I wish I were warm. I wish I were warm. I wish I were warm. I wish I were warm. I wish I were warm. I wish I were warm. I wish I were warm. I wish I were warm. I wish I were warm. I wish I were warm. I wish I were warm. I wish I were warm. I wish I was warm. I wish I was warm. I wish I was warm. I wish I was warm. I wish I was warm. I wish I was warm. I wish I was warm. I wish I was warm. I wish I was warm. I wish I was warm. I wish I was warm. I wish I was warm. I wish I was warm. I wish I was warm. I wish I was warm. I wish I was warm. I wish I was warm. I wish I was warm. I wish I was warm. I wish I was warm. I wish I was warm. I wish I was warm. I wish I was warm. I wish I was warm. I wish I was warm. I wish I was warm. I wish I was warm. I wish I was warm. I wish I was warm. I wish I was warm. I wish I was warm. I wish I was warm. I wish I was warm. I wish I was warm. I wish I was warm. I wish I was warm. I wish I was warm. I wish I was warm. I wish I was warm. I wish I was warm. I wish I was warm. I wish I was warm. I wish I was warm. I wish I was warm. I wish I was warm. I wish I was warm. I wish I was warm. I wish I was warm. I wish I was warm. I wish I was warm. I wish I was warm. I wish I was warm. I wish I was warm. I wish I was warm",
+ "raw": {
+ "normalized_reference": "i was completely lost in my work.",
+ "normalized_hypothesis": "oh, it's cold outside. i wish i were warm. i wish i were warm. i wish i were warm. i wish i were warm. i wish i were warm. i wish i were warm. i wish i were warm. i wish i were warm. i wish i were warm. i wish i were warm. i wish i were warm. i wish i were warm. i wish i were warm. i wish i were warm. i wish i were warm. i wish i were warm. i wish i were warm. i wish i were warm. i wish i were warm. i wish i was warm. i wish i was warm. i wish i was warm. i wish i was warm. i wish i was warm. i wish i was warm. i wish i was warm. i wish i was warm. i wish i was warm. i wish i was warm. i wish i was warm. i wish i was warm. i wish i was warm. i wish i was warm. i wish i was warm. i wish i was warm. i wish i was warm. i wish i was warm. i wish i was warm. i wish i was warm. i wish i was warm. i wish i was warm. i wish i was warm. i wish i was warm. i wish i was warm. i wish i was warm. i wish i was warm. i wish i was warm. i wish i was warm. i wish i was warm. i wish i was warm. i wish i was warm. i wish i was warm. i wish i was warm. i wish i was warm. i wish i was warm. i wish i was warm. i wish i was warm. i wish i was warm. i wish i was warm. i wish i was warm. i wish i was warm. i wish i was warm. i wish i was warm. i wish i was warm. i wish i was warm. i wish i was warm. i wish i was warm. i wish i was warm. i wish i was warm. i wish i was warm. i wish i was warm. i wish i was warm. i wish i was warm",
+ "errors": 367,
+ "words": 7,
+ "wer": 52.42857142857143
+ },
+ "semantic": {
+ "normalized_reference": "i was completely lost in my work",
+ "normalized_hypothesis": "0 it is cold outside i wish i were warm i wish i were warm i wish i were warm i wish i were warm i wish i were warm i wish i were warm i wish i were warm i wish i were warm i wish i were warm i wish i were warm i wish i were warm i wish i were warm i wish i were warm i wish i were warm i wish i were warm i wish i were warm i wish i were warm i wish i were warm i wish i were warm i wish i was warm i wish i was warm i wish i was warm i wish i was warm i wish i was warm i wish i was warm i wish i was warm i wish i was warm i wish i was warm i wish i was warm i wish i was warm i wish i was warm i wish i was warm i wish i was warm i wish i was warm i wish i was warm i wish i was warm i wish i was warm i wish i was warm i wish i was warm i wish i was warm i wish i was warm i wish i was warm i wish i was warm i wish i was warm i wish i was warm i wish i was warm i wish i was warm i wish i was warm i wish i was warm i wish i was warm i wish i was warm i wish i was warm i wish i was warm i wish i was warm i wish i was warm i wish i was warm i wish i was warm i wish i was warm i wish i was warm i wish i was warm i wish i was warm i wish i was warm i wish i was warm i wish i was warm i wish i was warm i wish i was warm i wish i was warm i wish i was warm i wish i was warm i wish i was warm i wish i was warm i wish i was warm i wish i was warm",
+ "errors": 368,
+ "words": 7,
+ "wer": 52.57142857142857
+ }
+ },
+ {
+ "id": "release500_0365",
+ "category": "phonetic_balance",
+ "text": "The very thing, Ernest agreed.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0365.wav",
+ "render_seconds": 113.60242960799951,
+ "audio_seconds": 2.2987755102040817,
+ "rtf": 49.4186705503549,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Before I found all of this to be.",
+ "raw": {
+ "normalized_reference": "the very thing, ernest agreed.",
+ "normalized_hypothesis": "before i found all of this to be.",
+ "errors": 8,
+ "words": 5,
+ "wer": 1.6
+ },
+ "semantic": {
+ "normalized_reference": "the very thing ernest agreed",
+ "normalized_hypothesis": "before i found all of this to be",
+ "errors": 8,
+ "words": 5,
+ "wer": 1.6
+ }
+ },
+ {
+ "id": "release500_0366",
+ "category": "phonetic_balance",
+ "text": "He was the leader, and Tudor was his lieutenant.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0366.wav",
+ "render_seconds": 144.5026921579847,
+ "audio_seconds": 3.622312925170068,
+ "rtf": 39.892382337784994,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I wish them battle, until all this has ended.",
+ "raw": {
+ "normalized_reference": "he was the leader, and tudor was his lieutenant.",
+ "normalized_hypothesis": "i wish them battle, until all this has ended.",
+ "errors": 9,
+ "words": 9,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "he was the leader and tudor was his lieutenant",
+ "normalized_hypothesis": "i wish them battle until all this has ended",
+ "errors": 9,
+ "words": 9,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0367",
+ "category": "phonetic_balance",
+ "text": "The Fire-Men wore animal skins around their waists and across their shoulders.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0367.wav",
+ "render_seconds": 188.79228968708776,
+ "audio_seconds": 5.712108843537415,
+ "rtf": 33.05124164443824,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Before I have your allies, I shall wear their masks and approach their hurdles.",
+ "raw": {
+ "normalized_reference": "the fire-men wore animal skins around their waists and across their shoulders.",
+ "normalized_hypothesis": "before i have your allies, i shall wear their masks and approach their hurdles.",
+ "errors": 11,
+ "words": 13,
+ "wer": 0.8461538461538461
+ },
+ "semantic": {
+ "normalized_reference": "the fire men wore animal skins around their waists and across their shoulders",
+ "normalized_hypothesis": "before i have your allies i shall wear their masks and approach their hurdles",
+ "errors": 11,
+ "words": 13,
+ "wer": 0.8461538461538461
+ }
+ },
+ {
+ "id": "release500_0368",
+ "category": "phonetic_balance",
+ "text": "We could throw stones with our feet.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0368.wav",
+ "render_seconds": 70.40826302394271,
+ "audio_seconds": 2.5774149659863945,
+ "rtf": 27.31739512383757,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Like a toy stays with all fat.",
+ "raw": {
+ "normalized_reference": "we could throw stones with our feet.",
+ "normalized_hypothesis": "like a toy stays with all fat.",
+ "errors": 6,
+ "words": 7,
+ "wer": 0.8571428571428571
+ },
+ "semantic": {
+ "normalized_reference": "we could throw stones with our feet",
+ "normalized_hypothesis": "like a toy stays with all fat",
+ "errors": 6,
+ "words": 7,
+ "wer": 0.8571428571428571
+ }
+ },
+ {
+ "id": "release500_0369",
+ "category": "phonetic_balance",
+ "text": "He turned sharply, and faced Gregson across the table.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0369.wav",
+ "render_seconds": 160.1924027170753,
+ "audio_seconds": 4.249251700680272,
+ "rtf": 37.698967724465405,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I can't stop you. I'll fast watch and I'll push the tab.",
+ "raw": {
+ "normalized_reference": "he turned sharply, and faced gregson across the table.",
+ "normalized_hypothesis": "i can't stop you. i'll fast watch and i'll push the tab.",
+ "errors": 11,
+ "words": 9,
+ "wer": 1.2222222222222223
+ },
+ "semantic": {
+ "normalized_reference": "he turned sharply and faced gregson across the table",
+ "normalized_hypothesis": "i can not stop you i will fast watch and i will push the tab",
+ "errors": 13,
+ "words": 9,
+ "wer": 1.4444444444444444
+ }
+ },
+ {
+ "id": "release500_0370",
+ "category": "phonetic_balance",
+ "text": "They are to attack your camp tomorrow night.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0370.wav",
+ "render_seconds": 120.7102004759945,
+ "audio_seconds": 3.0650340136054424,
+ "rtf": 39.382988880440315,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Thank you.",
+ "raw": {
+ "normalized_reference": "they are to attack your camp tomorrow night.",
+ "normalized_hypothesis": "thank you.",
+ "errors": 8,
+ "words": 8,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "they are to attack your camp tomorrow night",
+ "normalized_hypothesis": "thank you",
+ "errors": 8,
+ "words": 8,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0371",
+ "category": "phonetic_balance",
+ "text": "And watch out for wet feet, was his parting advice.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0371.wav",
+ "render_seconds": 139.09965348092373,
+ "audio_seconds": 3.5526530612244898,
+ "rtf": 39.15373977806332,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I look out for that fat, which is growing like fish.",
+ "raw": {
+ "normalized_reference": "and watch out for wet feet, was his parting advice.",
+ "normalized_hypothesis": "i look out for that fat, which is growing like fish.",
+ "errors": 9,
+ "words": 10,
+ "wer": 0.9
+ },
+ "semantic": {
+ "normalized_reference": "and watch out for wet feet was his parting advice",
+ "normalized_hypothesis": "i look out for that fat which is growing like fish",
+ "errors": 9,
+ "words": 10,
+ "wer": 0.9
+ }
+ },
+ {
+ "id": "release500_0372",
+ "category": "phonetic_balance",
+ "text": "Two weeks passed, and in that time Thorpe left camp three times.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0372.wav",
+ "render_seconds": 164.39656222204212,
+ "audio_seconds": 4.527891156462585,
+ "rtf": 36.307534024399324,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "To watch post, I may not tell, so just count to a two.",
+ "raw": {
+ "normalized_reference": "two weeks passed, and in that time thorpe left camp three times.",
+ "normalized_hypothesis": "to watch post, i may not tell, so just count to a two.",
+ "errors": 13,
+ "words": 12,
+ "wer": 1.0833333333333333
+ },
+ "semantic": {
+ "normalized_reference": "2 weeks passed and in that time thorpe left camp 3 times",
+ "normalized_hypothesis": "to watch post i may not tell so just count to a 2",
+ "errors": 13,
+ "words": 12,
+ "wer": 1.0833333333333333
+ }
+ },
+ {
+ "id": "release500_0373",
+ "category": "phonetic_balance",
+ "text": "But ever his gaze returned to that Crouched Venus on the piano.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0373.wav",
+ "render_seconds": 155.61082126491237,
+ "audio_seconds": 4.388571428571429,
+ "rtf": 35.45819494968706,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Thank you for watching. Please like, share, and subscribe.",
+ "raw": {
+ "normalized_reference": "but ever his gaze returned to that crouched venus on the piano.",
+ "normalized_hypothesis": "thank you for watching. please like, share, and subscribe.",
+ "errors": 12,
+ "words": 12,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "but ever his gaze returned to that crouched venus on the piano",
+ "normalized_hypothesis": "thank you for watching please like share and subscribe",
+ "errors": 12,
+ "words": 12,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0374",
+ "category": "phonetic_balance",
+ "text": "But she had become an automaton.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0374.wav",
+ "render_seconds": 59.1937680329429,
+ "audio_seconds": 2.36843537414966,
+ "rtf": 24.992773152695904,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Thank you for watching.",
+ "raw": {
+ "normalized_reference": "but she had become an automaton.",
+ "normalized_hypothesis": "thank you for watching.",
+ "errors": 6,
+ "words": 6,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "but she had become an automaton",
+ "normalized_hypothesis": "thank you for watching",
+ "errors": 6,
+ "words": 6,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0375",
+ "category": "phonetic_balance",
+ "text": "I do not blame you for anything; remember that.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0375.wav",
+ "render_seconds": 127.7072864070069,
+ "audio_seconds": 3.1346938775510202,
+ "rtf": 40.73995464806861,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Thank you for watching. See you next time.",
+ "raw": {
+ "normalized_reference": "i do not blame you for anything; remember that.",
+ "normalized_hypothesis": "thank you for watching. see you next time.",
+ "errors": 9,
+ "words": 9,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "i do not blame you for anything remember that",
+ "normalized_hypothesis": "thank you for watching see you next time",
+ "errors": 9,
+ "words": 9,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0376",
+ "category": "phonetic_balance",
+ "text": "The next thing to watch out for is bed sores.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0376.wav",
+ "render_seconds": 110.20190875290427,
+ "audio_seconds": 3.0650340136054424,
+ "rtf": 35.95454675665156,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The next song to look out for is Axios.",
+ "raw": {
+ "normalized_reference": "the next thing to watch out for is bed sores.",
+ "normalized_hypothesis": "the next song to look out for is axios.",
+ "errors": 4,
+ "words": 10,
+ "wer": 0.4
+ },
+ "semantic": {
+ "normalized_reference": "the next thing to watch out for is bed sores",
+ "normalized_hypothesis": "the next song to look out for is axios",
+ "errors": 4,
+ "words": 10,
+ "wer": 0.4
+ }
+ },
+ {
+ "id": "release500_0377",
+ "category": "phonetic_balance",
+ "text": "He will follow us soon.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0377.wav",
+ "render_seconds": 85.29273871390615,
+ "audio_seconds": 1.9504761904761905,
+ "rtf": 43.72918732890696,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I will see you in the shower.",
+ "raw": {
+ "normalized_reference": "he will follow us soon.",
+ "normalized_hypothesis": "i will see you in the shower.",
+ "errors": 6,
+ "words": 5,
+ "wer": 1.2
+ },
+ "semantic": {
+ "normalized_reference": "he will follow us soon",
+ "normalized_hypothesis": "i will see you in the shower",
+ "errors": 6,
+ "words": 5,
+ "wer": 1.2
+ }
+ },
+ {
+ "id": "release500_0378",
+ "category": "phonetic_balance",
+ "text": "His beady black eyes saw bargains where other men saw bankruptcy.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0378.wav",
+ "render_seconds": 128.69972406700253,
+ "audio_seconds": 4.527891156462585,
+ "rtf": 28.423767184268886,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Thank you for watching this video. I hope you enjoyed it.",
+ "raw": {
+ "normalized_reference": "his beady black eyes saw bargains where other men saw bankruptcy.",
+ "normalized_hypothesis": "thank you for watching this video. i hope you enjoyed it.",
+ "errors": 11,
+ "words": 11,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "his beady black eyes saw bargains where other men saw bankruptcy",
+ "normalized_hypothesis": "thank you for watching this video i hope you enjoyed it",
+ "errors": 11,
+ "words": 11,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0379",
+ "category": "phonetic_balance",
+ "text": "They robbed me a few years later.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0379.wav",
+ "render_seconds": 125.78808541595936,
+ "audio_seconds": 2.36843537414966,
+ "rtf": 53.11020380326868,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Thank you for watching.",
+ "raw": {
+ "normalized_reference": "they robbed me a few years later.",
+ "normalized_hypothesis": "thank you for watching.",
+ "errors": 7,
+ "words": 7,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "they robbed me a few years later",
+ "normalized_hypothesis": "thank you for watching",
+ "errors": 7,
+ "words": 7,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0380",
+ "category": "phonetic_balance",
+ "text": "Gregson held a lighted match until it burnt his fingertips.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0380.wav",
+ "render_seconds": 104.29416135197971,
+ "audio_seconds": 4.249251700680272,
+ "rtf": 24.544124165504957,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "That's a whole different world to tell about as far as I'm concerned.",
+ "raw": {
+ "normalized_reference": "gregson held a lighted match until it burnt his fingertips.",
+ "normalized_hypothesis": "that's a whole different world to tell about as far as i'm concerned.",
+ "errors": 13,
+ "words": 10,
+ "wer": 1.3
+ },
+ "semantic": {
+ "normalized_reference": "gregson held a lighted match until it burnt his fingertips",
+ "normalized_hypothesis": "that is a whole different world to tell about as far as i am concerned",
+ "errors": 14,
+ "words": 10,
+ "wer": 1.4
+ }
+ },
+ {
+ "id": "release500_0381",
+ "category": "phonetic_balance",
+ "text": "A burst of laughter was his reward.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0381.wav",
+ "render_seconds": 112.99254741205368,
+ "audio_seconds": 2.647074829931973,
+ "rtf": 42.68581535149026,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The past of the earth can't be changed for the new.",
+ "raw": {
+ "normalized_reference": "a burst of laughter was his reward.",
+ "normalized_hypothesis": "the past of the earth can't be changed for the new.",
+ "errors": 10,
+ "words": 7,
+ "wer": 1.4285714285714286
+ },
+ "semantic": {
+ "normalized_reference": "a burst of laughter was his reward",
+ "normalized_hypothesis": "the past of the earth can not be changed for the new",
+ "errors": 11,
+ "words": 7,
+ "wer": 1.5714285714285714
+ }
+ },
+ {
+ "id": "release500_0382",
+ "category": "phonetic_balance",
+ "text": "He cried, and swung the club wildly.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0382.wav",
+ "render_seconds": 109.38749186589848,
+ "audio_seconds": 2.716734693877551,
+ "rtf": 40.264326182609864,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I can't sleep. I can't sleep. I can't sleep. I can't sleep. I can't sleep. I can't sleep. I can't sleep. I can't sleep. I can't sleep. I can't sleep. I can't sleep. I can't sleep. I can't sleep. I can't sleep. I can't sleep. I can't sleep. I can't sleep. I can't sleep. I can't sleep. I can't sleep. I can't sleep. I can't sleep. I can't sleep. I can't sleep. I can't sleep. I can't sleep. I can't sleep. I can't sleep. I can't sleep. I can't sleep. I can't sleep. I can't sleep. I can't sleep. I can't sleep. I can't sleep. I can't sleep. I can't sleep. I can't sleep. I can't sleep. I can't sleep. I can't sleep. I can't sleep. I can't sleep. I can't sleep. I can't sleep. I can't sleep. I can't sleep. I can't sleep. I can't sleep. I can't sleep. I can't sleep. I can't sleep. I can't sleep. I can't sleep. I can't sleep. I can't sleep. I can't sleep. I can't sleep. I can't sleep. I can't sleep. I can't sleep. I can't sleep. I can't sleep. I can't sleep. I can't sleep. I can't sleep. I can't sleep. I can't sleep. I can't sleep. I can't sleep. I can't sleep. I can't sleep. I can't sleep. I can't sleep. I can't sleep. I can't sleep. I can't sleep. I can't sleep. I can't sleep. I can't sleep. I can't sleep. I can't sleep. I can't sleep. I can't sleep. I can't sleep. I can't sleep. I can't sleep. I can't sleep. I can't sleep",
+ "raw": {
+ "normalized_reference": "he cried, and swung the club wildly.",
+ "normalized_hypothesis": "i can't sleep. i can't sleep. i can't sleep. i can't sleep. i can't sleep. i can't sleep. i can't sleep. i can't sleep. i can't sleep. i can't sleep. i can't sleep. i can't sleep. i can't sleep. i can't sleep. i can't sleep. i can't sleep. i can't sleep. i can't sleep. i can't sleep. i can't sleep. i can't sleep. i can't sleep. i can't sleep. i can't sleep. i can't sleep. i can't sleep. i can't sleep. i can't sleep. i can't sleep. i can't sleep. i can't sleep. i can't sleep. i can't sleep. i can't sleep. i can't sleep. i can't sleep. i can't sleep. i can't sleep. i can't sleep. i can't sleep. i can't sleep. i can't sleep. i can't sleep. i can't sleep. i can't sleep. i can't sleep. i can't sleep. i can't sleep. i can't sleep. i can't sleep. i can't sleep. i can't sleep. i can't sleep. i can't sleep. i can't sleep. i can't sleep. i can't sleep. i can't sleep. i can't sleep. i can't sleep. i can't sleep. i can't sleep. i can't sleep. i can't sleep. i can't sleep. i can't sleep. i can't sleep. i can't sleep. i can't sleep. i can't sleep. i can't sleep. i can't sleep. i can't sleep. i can't sleep. i can't sleep. i can't sleep. i can't sleep. i can't sleep. i can't sleep. i can't sleep. i can't sleep. i can't sleep. i can't sleep. i can't sleep. i can't sleep. i can't sleep. i can't sleep. i can't sleep. i can't sleep",
+ "errors": 267,
+ "words": 7,
+ "wer": 38.142857142857146
+ },
+ "semantic": {
+ "normalized_reference": "he cried and swung the club wildly",
+ "normalized_hypothesis": "i can not sleep i can not sleep i can not sleep i can not sleep i can not sleep i can not sleep i can not sleep i can not sleep i can not sleep i can not sleep i can not sleep i can not sleep i can not sleep i can not sleep i can not sleep i can not sleep i can not sleep i can not sleep i can not sleep i can not sleep i can not sleep i can not sleep i can not sleep i can not sleep i can not sleep i can not sleep i can not sleep i can not sleep i can not sleep i can not sleep i can not sleep i can not sleep i can not sleep i can not sleep i can not sleep i can not sleep i can not sleep i can not sleep i can not sleep i can not sleep i can not sleep i can not sleep i can not sleep i can not sleep i can not sleep i can not sleep i can not sleep i can not sleep i can not sleep i can not sleep i can not sleep i can not sleep i can not sleep i can not sleep i can not sleep i can not sleep i can not sleep i can not sleep i can not sleep i can not sleep i can not sleep i can not sleep i can not sleep i can not sleep i can not sleep i can not sleep i can not sleep i can not sleep i can not sleep i can not sleep i can not sleep i can not sleep i can not sleep i can not sleep i can not sleep i can not sleep i can not sleep i can not sleep i can not sleep i can not sleep i can not sleep i can not sleep i can not sleep i can not sleep i can not sleep i can not sleep i can not sleep i can not sleep i can not sleep",
+ "errors": 356,
+ "words": 7,
+ "wer": 50.857142857142854
+ }
+ },
+ {
+ "id": "release500_0383",
+ "category": "phonetic_balance",
+ "text": "Sandel would never become a world champion.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0383.wav",
+ "render_seconds": 128.50690559204668,
+ "audio_seconds": 3.2043537414965986,
+ "rtf": 40.10384494317014,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Shabbat Shalom Shabbat Shalom Shabbat Shalom Shabbat Shalom Shabbat Shalom Shabbat Shalom Shabbat Shalom Shabbat Shalom Shabbat Shalom Shabbat Shalom Shabbat Shalom Shabbat Shalom Shabbat Shalom Shabbat Shalom Shabbat Shalom Shabbat Shalom Shabbat Shalom Shabbat Shalom Shabbat Shalom Shabbat Shalom Shabbat Shalom Shabbat Shalom Shabbat Shalom Shabbat Shalom Shabbat Shalom Shabbat Shalom Shabbat Shalom Shabbat Shalom Shabbat Shalom Shabbat Shalom Shabbat Shalom Shabbat Shalom Shabbat Shalom Shabbat Shalom Shabbat Shalom Shabbat Shalom Shabbat Shabbat Shalom Shabbat Shabbat Shabbat Shabbat Shabbat Shabbat Shabbat Shabbat Shabbat Shabbat Shabbat Shabbat Shabbat Shabbat Shabbat Shabbat Shabbat Shabbat Shabbat Shabbat Shabbat Shabbat Shabbat Shabbat Shabbat Shabbat Shabbat Shabbat Shabbat Shabbat Shabbat Shabbat Shabbat Shabbat Shabbat Shabbat Shabbat Shabbat Shabbat Shabbat Shabbat Shabbat Shabbat Shabbat Shabbat Shabbat Shabbat Shabbat Shabbat Shabbat Shabbat Shabbat Shabbat Shabbat Shabbat Shabbat Shabbat Shabbat Shabbat Shabbat Shabbat Shabbat Shabbat Shabbat Shabbat Shabbat Shabbat Shabbat Shabbat Shabbat Shabbat Shabbat Shabbat Shabbat Shabbat Shabbat Shabbat Shabbat Shabbat Shabbat Shabbat Shabbat Shabbat Shabbat Shabbat Sh",
+ "raw": {
+ "normalized_reference": "sandel would never become a world champion.",
+ "normalized_hypothesis": "shabbat shalom shabbat shalom shabbat shalom shabbat shalom shabbat shalom shabbat shalom shabbat shalom shabbat shalom shabbat shalom shabbat shalom shabbat shalom shabbat shalom shabbat shalom shabbat shalom shabbat shalom shabbat shalom shabbat shalom shabbat shalom shabbat shalom shabbat shalom shabbat shalom shabbat shalom shabbat shalom shabbat shalom shabbat shalom shabbat shalom shabbat shalom shabbat shalom shabbat shalom shabbat shalom shabbat shalom shabbat shalom shabbat shalom shabbat shalom shabbat shalom shabbat shalom shabbat shabbat shalom shabbat shabbat shabbat shabbat shabbat shabbat shabbat shabbat shabbat shabbat shabbat shabbat shabbat shabbat shabbat shabbat shabbat shabbat shabbat shabbat shabbat shabbat shabbat shabbat shabbat shabbat shabbat shabbat shabbat shabbat shabbat shabbat shabbat shabbat shabbat shabbat shabbat shabbat shabbat shabbat shabbat shabbat shabbat shabbat shabbat shabbat shabbat shabbat shabbat shabbat shabbat shabbat shabbat shabbat shabbat shabbat shabbat shabbat shabbat shabbat shabbat shabbat shabbat shabbat shabbat shabbat shabbat shabbat shabbat shabbat shabbat shabbat shabbat shabbat shabbat shabbat shabbat shabbat shabbat shabbat shabbat shabbat shabbat shabbat shabbat sh",
+ "errors": 161,
+ "words": 7,
+ "wer": 23.0
+ },
+ "semantic": {
+ "normalized_reference": "sandel would never become a world champion",
+ "normalized_hypothesis": "shabbat shalom shabbat shalom shabbat shalom shabbat shalom shabbat shalom shabbat shalom shabbat shalom shabbat shalom shabbat shalom shabbat shalom shabbat shalom shabbat shalom shabbat shalom shabbat shalom shabbat shalom shabbat shalom shabbat shalom shabbat shalom shabbat shalom shabbat shalom shabbat shalom shabbat shalom shabbat shalom shabbat shalom shabbat shalom shabbat shalom shabbat shalom shabbat shalom shabbat shalom shabbat shalom shabbat shalom shabbat shalom shabbat shalom shabbat shalom shabbat shalom shabbat shalom shabbat shabbat shalom shabbat shabbat shabbat shabbat shabbat shabbat shabbat shabbat shabbat shabbat shabbat shabbat shabbat shabbat shabbat shabbat shabbat shabbat shabbat shabbat shabbat shabbat shabbat shabbat shabbat shabbat shabbat shabbat shabbat shabbat shabbat shabbat shabbat shabbat shabbat shabbat shabbat shabbat shabbat shabbat shabbat shabbat shabbat shabbat shabbat shabbat shabbat shabbat shabbat shabbat shabbat shabbat shabbat shabbat shabbat shabbat shabbat shabbat shabbat shabbat shabbat shabbat shabbat shabbat shabbat shabbat shabbat shabbat shabbat shabbat shabbat shabbat shabbat shabbat shabbat shabbat shabbat shabbat shabbat shabbat shabbat shabbat shabbat shabbat shabbat sh",
+ "errors": 161,
+ "words": 7,
+ "wer": 23.0
+ }
+ },
+ {
+ "id": "release500_0384",
+ "category": "phonetic_balance",
+ "text": "And after the bath a shave would not be bad.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0384.wav",
+ "render_seconds": 125.00666798499878,
+ "audio_seconds": 3.0650340136054424,
+ "rtf": 40.78475717727898,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I have to go to class. I have to go to class. I have to go to class. I have to go to class. I have to go to class. I have to go to class. I have to go to class. I have to go to class. I have to go to class. I have to go to class. I have to go to class. I have to go to class. I have to go to class. I have to go to class. I have to go to class. I have to go to class. I have to go to class. I have to go to class. I have to go to class. I have to go to class. I have to go to class. I have to go to class. I have to go to class. I have to go to class. I have to go to class. I have to go to class. I have to go to class. I have to go to class. I have to go to class. I have to go to class. I have to go to class. I have to go to class. I have to go to class. I have to go to class. I have to go to class. I have to go to class. I have to go to class. I have to go to class. I have to go to class. I have to go to class. I have to go to class. I have to go to class. I have to go to class. I have to go to class. I have to go to class. I have to go to class. I have to go to class. I have to go to class. I have to go to class. I have to go to class. I have to go to class. I have to go to class. I have to go to class. I have to go to class. I have to go to class. I have to go to class. I have to go to class. I have to go to class. I have to go to class. I have to go to class. I have to go to class. I have to go to class. I have to go to class. I have to",
+ "raw": {
+ "normalized_reference": "and after the bath a shave would not be bad.",
+ "normalized_hypothesis": "i have to go to class. i have to go to class. i have to go to class. i have to go to class. i have to go to class. i have to go to class. i have to go to class. i have to go to class. i have to go to class. i have to go to class. i have to go to class. i have to go to class. i have to go to class. i have to go to class. i have to go to class. i have to go to class. i have to go to class. i have to go to class. i have to go to class. i have to go to class. i have to go to class. i have to go to class. i have to go to class. i have to go to class. i have to go to class. i have to go to class. i have to go to class. i have to go to class. i have to go to class. i have to go to class. i have to go to class. i have to go to class. i have to go to class. i have to go to class. i have to go to class. i have to go to class. i have to go to class. i have to go to class. i have to go to class. i have to go to class. i have to go to class. i have to go to class. i have to go to class. i have to go to class. i have to go to class. i have to go to class. i have to go to class. i have to go to class. i have to go to class. i have to go to class. i have to go to class. i have to go to class. i have to go to class. i have to go to class. i have to go to class. i have to go to class. i have to go to class. i have to go to class. i have to go to class. i have to go to class. i have to go to class. i have to go to class. i have to go to class. i have to",
+ "errors": 381,
+ "words": 10,
+ "wer": 38.1
+ },
+ "semantic": {
+ "normalized_reference": "and after the bath a shave would not be bad",
+ "normalized_hypothesis": "i have to go to class i have to go to class i have to go to class i have to go to class i have to go to class i have to go to class i have to go to class i have to go to class i have to go to class i have to go to class i have to go to class i have to go to class i have to go to class i have to go to class i have to go to class i have to go to class i have to go to class i have to go to class i have to go to class i have to go to class i have to go to class i have to go to class i have to go to class i have to go to class i have to go to class i have to go to class i have to go to class i have to go to class i have to go to class i have to go to class i have to go to class i have to go to class i have to go to class i have to go to class i have to go to class i have to go to class i have to go to class i have to go to class i have to go to class i have to go to class i have to go to class i have to go to class i have to go to class i have to go to class i have to go to class i have to go to class i have to go to class i have to go to class i have to go to class i have to go to class i have to go to class i have to go to class i have to go to class i have to go to class i have to go to class i have to go to class i have to go to class i have to go to class i have to go to class i have to go to class i have to go to class i have to go to class i have to go to class i have to",
+ "errors": 381,
+ "words": 10,
+ "wer": 38.1
+ }
+ },
+ {
+ "id": "release500_0385",
+ "category": "phonetic_balance",
+ "text": "They saw each other for the first time in Boston.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0385.wav",
+ "render_seconds": 127.90721304703038,
+ "audio_seconds": 3.3436734693877552,
+ "rtf": 38.253499995754936,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I show a channel for the first time on boost.",
+ "raw": {
+ "normalized_reference": "they saw each other for the first time in boston.",
+ "normalized_hypothesis": "i show a channel for the first time on boost.",
+ "errors": 6,
+ "words": 10,
+ "wer": 0.6
+ },
+ "semantic": {
+ "normalized_reference": "they saw each other for the 1st time in boston",
+ "normalized_hypothesis": "i show a channel for the 1st time on boost",
+ "errors": 6,
+ "words": 10,
+ "wer": 0.6
+ }
+ },
+ {
+ "id": "release500_0386",
+ "category": "phonetic_balance",
+ "text": "I had been born with no organic, chemical predisposition toward alcohol.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0386.wav",
+ "render_seconds": 165.60716925095767,
+ "audio_seconds": 5.224489795918367,
+ "rtf": 31.69824723944112,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Thank you for listening. I'll see you next time.",
+ "raw": {
+ "normalized_reference": "i had been born with no organic, chemical predisposition toward alcohol.",
+ "normalized_hypothesis": "thank you for listening. i'll see you next time.",
+ "errors": 11,
+ "words": 11,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "i had been born with no organic chemical predisposition toward alcohol",
+ "normalized_hypothesis": "thank you for listening i will see you next time",
+ "errors": 11,
+ "words": 11,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0387",
+ "category": "phonetic_balance",
+ "text": "They had no fixed values, to be altered by adjectives and adverbs.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0387.wav",
+ "render_seconds": 192.80699968105182,
+ "audio_seconds": 4.9458503401360545,
+ "rtf": 38.98358955919154,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I hope you've watched this. I hope you've watched this. I hope you've watched this. I hope you've watched this. I hope you've watched this. I hope you've watched this. I hope you've watched this. I hope you've watched this. I hope you've watched this. I hope you've watched this. I hope you've watched this. I hope you've watched this. I hope you've watched this. I hope you've watched this. I hope you've watched this. I hope you've watched this. I hope you've watched this. I hope you've watched this. I hope you've watched this. I hope you've watched this. I hope you've watched this. I hope you've watched this. I hope you've watched this. I hope you've watched this. I hope you've watched this. I hope you've watched this. I hope you've watched this. I hope you've watched this. I hope you've watched this. I hope you've watched this. I hope you've watched this. I hope you've watched this. I hope you've watched this. I hope you've watched this. I hope you've watched this. I hope you've watched this. I hope you've watched this. I hope you've watched this. I hope you've watched this. I hope you've watched this. I hope you've watched this. I hope you've watched this. I hope you've watched this. I hope you've watched this. I hope you've watched this. I hope you've watched this. I hope you've watched this. I hope you've watched this. I hope you've watched this. I hope you've watched this. I hope you've watched this. I hope you've watched this. I hope you've watched this. I hope you've watched this. I hope you've watched this. I hope you've watched this. I hope you've watched this. I hope you've watched this. I hope you've watched this. I hope you've watched this. I hope you've watched this. I hope you've watched this. I hope you've watched this. I hope you",
+ "raw": {
+ "normalized_reference": "they had no fixed values, to be altered by adjectives and adverbs.",
+ "normalized_hypothesis": "i hope you've watched this. i hope you've watched this. i hope you've watched this. i hope you've watched this. i hope you've watched this. i hope you've watched this. i hope you've watched this. i hope you've watched this. i hope you've watched this. i hope you've watched this. i hope you've watched this. i hope you've watched this. i hope you've watched this. i hope you've watched this. i hope you've watched this. i hope you've watched this. i hope you've watched this. i hope you've watched this. i hope you've watched this. i hope you've watched this. i hope you've watched this. i hope you've watched this. i hope you've watched this. i hope you've watched this. i hope you've watched this. i hope you've watched this. i hope you've watched this. i hope you've watched this. i hope you've watched this. i hope you've watched this. i hope you've watched this. i hope you've watched this. i hope you've watched this. i hope you've watched this. i hope you've watched this. i hope you've watched this. i hope you've watched this. i hope you've watched this. i hope you've watched this. i hope you've watched this. i hope you've watched this. i hope you've watched this. i hope you've watched this. i hope you've watched this. i hope you've watched this. i hope you've watched this. i hope you've watched this. i hope you've watched this. i hope you've watched this. i hope you've watched this. i hope you've watched this. i hope you've watched this. i hope you've watched this. i hope you've watched this. i hope you've watched this. i hope you've watched this. i hope you've watched this. i hope you've watched this. i hope you've watched this. i hope you've watched this. i hope you've watched this. i hope you've watched this. i hope you've watched this. i hope you",
+ "errors": 318,
+ "words": 12,
+ "wer": 26.5
+ },
+ "semantic": {
+ "normalized_reference": "they had no fixed values to be altered by adjectives and adverbs",
+ "normalized_hypothesis": "i hope you have watched this i hope you have watched this i hope you have watched this i hope you have watched this i hope you have watched this i hope you have watched this i hope you have watched this i hope you have watched this i hope you have watched this i hope you have watched this i hope you have watched this i hope you have watched this i hope you have watched this i hope you have watched this i hope you have watched this i hope you have watched this i hope you have watched this i hope you have watched this i hope you have watched this i hope you have watched this i hope you have watched this i hope you have watched this i hope you have watched this i hope you have watched this i hope you have watched this i hope you have watched this i hope you have watched this i hope you have watched this i hope you have watched this i hope you have watched this i hope you have watched this i hope you have watched this i hope you have watched this i hope you have watched this i hope you have watched this i hope you have watched this i hope you have watched this i hope you have watched this i hope you have watched this i hope you have watched this i hope you have watched this i hope you have watched this i hope you have watched this i hope you have watched this i hope you have watched this i hope you have watched this i hope you have watched this i hope you have watched this i hope you have watched this i hope you have watched this i hope you have watched this i hope you have watched this i hope you have watched this i hope you have watched this i hope you have watched this i hope you have watched this i hope you have watched this i hope you have watched this i hope you have watched this i hope you have watched this i hope you have watched this i hope you have watched this i hope you have watched this i hope you",
+ "errors": 381,
+ "words": 12,
+ "wer": 31.75
+ }
+ },
+ {
+ "id": "release500_0388",
+ "category": "phonetic_balance",
+ "text": "Nor was Elam Harnish an exception.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0388.wav",
+ "render_seconds": 111.40671193704475,
+ "audio_seconds": 2.716734693877551,
+ "rtf": 41.007578762884556,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Hello, guys. How are you? Welcome back to my channel.",
+ "raw": {
+ "normalized_reference": "nor was elam harnish an exception.",
+ "normalized_hypothesis": "hello, guys. how are you? welcome back to my channel.",
+ "errors": 10,
+ "words": 6,
+ "wer": 1.6666666666666667
+ },
+ "semantic": {
+ "normalized_reference": "nor was elam harnish an exception",
+ "normalized_hypothesis": "hello guys how are you welcome back to my channel",
+ "errors": 10,
+ "words": 6,
+ "wer": 1.6666666666666667
+ }
+ },
+ {
+ "id": "release500_0389",
+ "category": "phonetic_balance",
+ "text": "He was a wise hyena.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0389.wav",
+ "render_seconds": 77.29535092203878,
+ "audio_seconds": 1.8808163265306121,
+ "rtf": 41.09670350672635,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you.",
+ "raw": {
+ "normalized_reference": "he was a wise hyena.",
+ "normalized_hypothesis": "i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you.",
+ "errors": 333,
+ "words": 5,
+ "wer": 66.6
+ },
+ "semantic": {
+ "normalized_reference": "he was a wise hyena",
+ "normalized_hypothesis": "i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you",
+ "errors": 333,
+ "words": 5,
+ "wer": 66.6
+ }
+ },
+ {
+ "id": "release500_0390",
+ "category": "phonetic_balance",
+ "text": "And now behold the perversity of things.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0390.wav",
+ "render_seconds": 118.90010294201784,
+ "audio_seconds": 2.8560544217687074,
+ "rtf": 41.63089541843708,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I hope you enjoyed the professional space.",
+ "raw": {
+ "normalized_reference": "and now behold the perversity of things.",
+ "normalized_hypothesis": "i hope you enjoyed the professional space.",
+ "errors": 7,
+ "words": 7,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "and now behold the perversity of things",
+ "normalized_hypothesis": "i hope you enjoyed the professional space",
+ "errors": 7,
+ "words": 7,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0391",
+ "category": "phonetic_balance",
+ "text": "The Fire People, like ourselves, lived in caves.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0391.wav",
+ "render_seconds": 139.59252273198217,
+ "audio_seconds": 3.4829931972789114,
+ "rtf": 40.078321956252694,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Before Papi up our shoes, let's go crash.",
+ "raw": {
+ "normalized_reference": "the fire people, like ourselves, lived in caves.",
+ "normalized_hypothesis": "before papi up our shoes, let's go crash.",
+ "errors": 8,
+ "words": 8,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "the fire people like ourselves lived in caves",
+ "normalized_hypothesis": "before papi up our shoes let us go crash",
+ "errors": 9,
+ "words": 8,
+ "wer": 1.125
+ }
+ },
+ {
+ "id": "release500_0392",
+ "category": "phonetic_balance",
+ "text": "To her the bridge was tambo, which is the native for taboo.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0392.wav",
+ "render_seconds": 148.99898482393473,
+ "audio_seconds": 4.040272108843538,
+ "rtf": 36.87845293830551,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "To all the boxless children, what is the matter for today?",
+ "raw": {
+ "normalized_reference": "to her the bridge was tambo, which is the native for taboo.",
+ "normalized_hypothesis": "to all the boxless children, what is the matter for today?",
+ "errors": 7,
+ "words": 12,
+ "wer": 0.5833333333333334
+ },
+ "semantic": {
+ "normalized_reference": "to her the bridge was tambo which is the native for taboo",
+ "normalized_hypothesis": "to all the boxless children what is the matter for today",
+ "errors": 7,
+ "words": 12,
+ "wer": 0.5833333333333334
+ }
+ },
+ {
+ "id": "release500_0393",
+ "category": "phonetic_balance",
+ "text": "The voyage was our idea of a good time.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0393.wav",
+ "render_seconds": 117.19434719404671,
+ "audio_seconds": 2.8560544217687074,
+ "rtf": 41.033653385872874,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Thank you for watching. I'll see you next time.",
+ "raw": {
+ "normalized_reference": "the voyage was our idea of a good time.",
+ "normalized_hypothesis": "thank you for watching. i'll see you next time.",
+ "errors": 8,
+ "words": 9,
+ "wer": 0.8888888888888888
+ },
+ "semantic": {
+ "normalized_reference": "the voyage was our idea of a good time",
+ "normalized_hypothesis": "thank you for watching i will see you next time",
+ "errors": 9,
+ "words": 9,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0394",
+ "category": "phonetic_balance",
+ "text": "Of course much grumbling went on, and little outbursts were continually occurring.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0394.wav",
+ "render_seconds": 186.702436683001,
+ "audio_seconds": 5.712108843537415,
+ "rtf": 32.68537799209359,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Of course, let's play a bit of that, too. And let me know, boss, what you think of that, huh?",
+ "raw": {
+ "normalized_reference": "of course much grumbling went on, and little outbursts were continually occurring.",
+ "normalized_hypothesis": "of course, let's play a bit of that, too. and let me know, boss, what you think of that, huh?",
+ "errors": 17,
+ "words": 12,
+ "wer": 1.4166666666666667
+ },
+ "semantic": {
+ "normalized_reference": "of course much grumbling went on and little outbursts were continually occurring",
+ "normalized_hypothesis": "of course let us play a bit of that too and let me know boss what you think of that huh",
+ "errors": 18,
+ "words": 12,
+ "wer": 1.5
+ }
+ },
+ {
+ "id": "release500_0395",
+ "category": "phonetic_balance",
+ "text": "Get down and dig in.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0395.wav",
+ "render_seconds": 81.00418672303203,
+ "audio_seconds": 1.811156462585034,
+ "rtf": 44.72511812006351,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Uh-oh. Uh-oh. Uh-oh. Uh-oh. Uh-oh. Uh-oh. Uh-oh. Uh-oh. Uh-oh. Uh-oh. Uh-oh. Uh-oh. Uh-oh. Uh-oh. Uh-oh. Uh-oh. Uh-oh. Uh-oh. Uh-oh. Uh-oh. Uh-oh. Uh-oh. Uh-oh. Uh-oh. Uh-oh. Uh-oh. Uh-oh. Uh-oh. Uh-oh. Uh-oh. Uh-oh. Uh-oh. Uh-oh. Uh-oh. Uh-oh. Uh-oh. Uh-oh. Uh-oh. Uh-oh. Uh-oh. Uh-oh. Uh-oh. Uh-oh. Uh-oh. Uh-oh. Uh-oh. Uh-oh. Uh-oh. Uh-oh. Uh-oh. Uh-oh. Uh-oh. Uh-oh. Uh-oh. Uh-oh. Uh-oh. Uh-oh. Uh-oh. Uh-oh. Uh-oh. Uh-oh. Uh-oh. Uh-oh. Uh-oh. Uh-oh. Uh-oh. Uh-oh. Uh-oh. Uh-oh. Uh-oh. Uh-oh. Uh-oh. Uh-oh. Uh-oh. Uh-oh. Uh-oh. Uh-oh. Uh-oh. Uh-oh. Uh-oh. Uh-oh. Uh-oh. Uh-oh. Uh-oh. Uh-oh. Uh-oh. Uh-oh. Uh-oh. Uh-oh. Uh-oh. Uh-oh. Uh-oh. Uh-oh. Uh-oh. Uh-oh. Uh-oh. Uh-oh. Uh-oh. Uh-oh. Uh-oh. Uh-oh. Uh-oh. Uh-oh. Uh-oh. Uh-oh. Uh-oh. Uh-oh. Uh-oh. Uh-oh. Uh-oh. Uh-oh.",
+ "raw": {
+ "normalized_reference": "get down and dig in.",
+ "normalized_hypothesis": "uh-oh. uh-oh. uh-oh. uh-oh. uh-oh. uh-oh. uh-oh. uh-oh. uh-oh. uh-oh. uh-oh. uh-oh. uh-oh. uh-oh. uh-oh. uh-oh. uh-oh. uh-oh. uh-oh. uh-oh. uh-oh. uh-oh. uh-oh. uh-oh. uh-oh. uh-oh. uh-oh. uh-oh. uh-oh. uh-oh. uh-oh. uh-oh. uh-oh. uh-oh. uh-oh. uh-oh. uh-oh. uh-oh. uh-oh. uh-oh. uh-oh. uh-oh. uh-oh. uh-oh. uh-oh. uh-oh. uh-oh. uh-oh. uh-oh. uh-oh. uh-oh. uh-oh. uh-oh. uh-oh. uh-oh. uh-oh. uh-oh. uh-oh. uh-oh. uh-oh. uh-oh. uh-oh. uh-oh. uh-oh. uh-oh. uh-oh. uh-oh. uh-oh. uh-oh. uh-oh. uh-oh. uh-oh. uh-oh. uh-oh. uh-oh. uh-oh. uh-oh. uh-oh. uh-oh. uh-oh. uh-oh. uh-oh. uh-oh. uh-oh. uh-oh. uh-oh. uh-oh. uh-oh. uh-oh. uh-oh. uh-oh. uh-oh. uh-oh. uh-oh. uh-oh. uh-oh. uh-oh. uh-oh. uh-oh. uh-oh. uh-oh. uh-oh. uh-oh. uh-oh. uh-oh. uh-oh. uh-oh. uh-oh. uh-oh. uh-oh. uh-oh.",
+ "errors": 222,
+ "words": 5,
+ "wer": 44.4
+ },
+ "semantic": {
+ "normalized_reference": "get down and dig in",
+ "normalized_hypothesis": "000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000",
+ "errors": 5,
+ "words": 5,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0396",
+ "category": "phonetic_balance",
+ "text": "It was put together in a casual, helter-skelter sort of way.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0396.wav",
+ "render_seconds": 140.20012241892982,
+ "audio_seconds": 4.31891156462585,
+ "rtf": 32.46191090534067,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I was put to anger because of you. Haters, haters, all of you.",
+ "raw": {
+ "normalized_reference": "it was put together in a casual, helter-skelter sort of way.",
+ "normalized_hypothesis": "i was put to anger because of you. haters, haters, all of you.",
+ "errors": 10,
+ "words": 12,
+ "wer": 0.8333333333333334
+ },
+ "semantic": {
+ "normalized_reference": "it was put together in a casual helter skelter sort of way",
+ "normalized_hypothesis": "i was put to anger because of you haters haters all of you",
+ "errors": 10,
+ "words": 12,
+ "wer": 0.8333333333333334
+ }
+ },
+ {
+ "id": "release500_0397",
+ "category": "phonetic_balance",
+ "text": "He deluged me, overwhelmed me with argument.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0397.wav",
+ "render_seconds": 102.50742393604014,
+ "audio_seconds": 3.0650340136054424,
+ "rtf": 33.44413911265514,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Happy Holidays! Happy Holidays! Happy Holidays! Happy Holidays! Happy Holidays! Happy Holidays! Happy Holidays! Happy Holidays! Happy Holidays! Happy Holidays! Happy Holidays! Happy Holidays! Happy Holidays! Happy Holidays! Happy Holidays! Happy Holidays! Happy Holidays! Happy Holidays! Happy Holidays! Happy Holidays! Happy Holidays! Happy Holidays! Happy Holidays! Happy Holidays! Happy Holidays! Happy Holidays! Happy Holidays! Happy Holidays! Happy Holidays! Happy Holidays! Happy Holidays! Happy Holidays! Happy Holidays! Happy Holidays! Happy Holidays! Happy Holidays! Happy Holidays! Happy Holidays! Happy Holidays! Happy Holidays! Happy Holidays! Happy Holidays! Happy Holidays! Happy Holidays! Happy Holidays! Happy Holidays! Happy Holidays! Happy Holidays! Happy Holidays! Happy Holidays! Happy Holidays! Happy Holidays! Happy Holidays! Happy Holidays! Happy Holidays! Happy Holidays! Happy Holidays! Happy Holidays! Happy Holidays! Happy Holidays! Happy Holidays! Happy Holidays! Happy Holidays! Happy Holidays! Happy Holidays! Happy Holidays! Happy Holidays! Happy Holidays! Happy Holidays! Happy Holidays! Happy Holidays! Happy Holidays! Happy Holidays! Happy Holidays! Happy Holidays! Happy Holidays! Happy Holidays! Happy Holidays! Happy Holidays! Happy Holidays! Happy Holidays! Happy Holidays! Happy Holidays! Happy Holidays! Happy Holidays! Happy Holidays! Happy Holidays! Happy Holidays! Happy Holidays! Happy Holidays! Happy Holidays! Happy Holidays! Happy Holidays! Happy Holidays! Happy Holidays! Happy Holidays! Happy Holidays! Happy Holidays! Happy Holidays! Happy Holidays! Happy Holidays! Happy Holidays! Happy Holidays! Happy Holidays! Happy Holidays! Happy Holidays! Happy Holidays! Happy Holidays! Happy Holidays! Happy Holidays! Happy Holidays!",
+ "raw": {
+ "normalized_reference": "he deluged me, overwhelmed me with argument.",
+ "normalized_hypothesis": "happy holidays! happy holidays! happy holidays! happy holidays! happy holidays! happy holidays! happy holidays! happy holidays! happy holidays! happy holidays! happy holidays! happy holidays! happy holidays! happy holidays! happy holidays! happy holidays! happy holidays! happy holidays! happy holidays! happy holidays! happy holidays! happy holidays! happy holidays! happy holidays! happy holidays! happy holidays! happy holidays! happy holidays! happy holidays! happy holidays! happy holidays! happy holidays! happy holidays! happy holidays! happy holidays! happy holidays! happy holidays! happy holidays! happy holidays! happy holidays! happy holidays! happy holidays! happy holidays! happy holidays! happy holidays! happy holidays! happy holidays! happy holidays! happy holidays! happy holidays! happy holidays! happy holidays! happy holidays! happy holidays! happy holidays! happy holidays! happy holidays! happy holidays! happy holidays! happy holidays! happy holidays! happy holidays! happy holidays! happy holidays! happy holidays! happy holidays! happy holidays! happy holidays! happy holidays! happy holidays! happy holidays! happy holidays! happy holidays! happy holidays! happy holidays! happy holidays! happy holidays! happy holidays! happy holidays! happy holidays! happy holidays! happy holidays! happy holidays! happy holidays! happy holidays! happy holidays! happy holidays! happy holidays! happy holidays! happy holidays! happy holidays! happy holidays! happy holidays! happy holidays! happy holidays! happy holidays! happy holidays! happy holidays! happy holidays! happy holidays! happy holidays! happy holidays! happy holidays! happy holidays! happy holidays! happy holidays! happy holidays! happy holidays! happy holidays! happy holidays! happy holidays!",
+ "errors": 222,
+ "words": 7,
+ "wer": 31.714285714285715
+ },
+ "semantic": {
+ "normalized_reference": "he deluged me overwhelmed me with argument",
+ "normalized_hypothesis": "happy holidays happy holidays happy holidays happy holidays happy holidays happy holidays happy holidays happy holidays happy holidays happy holidays happy holidays happy holidays happy holidays happy holidays happy holidays happy holidays happy holidays happy holidays happy holidays happy holidays happy holidays happy holidays happy holidays happy holidays happy holidays happy holidays happy holidays happy holidays happy holidays happy holidays happy holidays happy holidays happy holidays happy holidays happy holidays happy holidays happy holidays happy holidays happy holidays happy holidays happy holidays happy holidays happy holidays happy holidays happy holidays happy holidays happy holidays happy holidays happy holidays happy holidays happy holidays happy holidays happy holidays happy holidays happy holidays happy holidays happy holidays happy holidays happy holidays happy holidays happy holidays happy holidays happy holidays happy holidays happy holidays happy holidays happy holidays happy holidays happy holidays happy holidays happy holidays happy holidays happy holidays happy holidays happy holidays happy holidays happy holidays happy holidays happy holidays happy holidays happy holidays happy holidays happy holidays happy holidays happy holidays happy holidays happy holidays happy holidays happy holidays happy holidays happy holidays happy holidays happy holidays happy holidays happy holidays happy holidays happy holidays happy holidays happy holidays happy holidays happy holidays happy holidays happy holidays happy holidays happy holidays happy holidays happy holidays happy holidays happy holidays happy holidays happy holidays",
+ "errors": 222,
+ "words": 7,
+ "wer": 31.714285714285715
+ }
+ },
+ {
+ "id": "release500_0398",
+ "category": "phonetic_balance",
+ "text": "The girl faced him, her eyes shining with sudden fear.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0398.wav",
+ "render_seconds": 87.59573250403628,
+ "audio_seconds": 3.6919727891156464,
+ "rtf": 23.72599623764249,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "To go fast, hold your ass short and you shouldn't fall.",
+ "raw": {
+ "normalized_reference": "the girl faced him, her eyes shining with sudden fear.",
+ "normalized_hypothesis": "to go fast, hold your ass short and you shouldn't fall.",
+ "errors": 11,
+ "words": 10,
+ "wer": 1.1
+ },
+ "semantic": {
+ "normalized_reference": "the girl faced him her eyes shining with sudden fear",
+ "normalized_hypothesis": "to go fast hold your ass short and you should not fall",
+ "errors": 12,
+ "words": 10,
+ "wer": 1.2
+ }
+ },
+ {
+ "id": "release500_0399",
+ "category": "phonetic_balance",
+ "text": "It will break our hearts and our backs to hoist anchor by hand.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0399.wav",
+ "render_seconds": 168.71729675191455,
+ "audio_seconds": 4.597551020408163,
+ "rtf": 36.69721031979675,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I need that too. I need that too. I need that too. I need that too. I need that too. I need that too. I need that too. I need that too. I need that too. I need that too. I need that too. I need that too. I need that too. I need that too. I need that too. I need that too. I need that too. I need that too. I need that too. I need that too. I need that too. I need that too. I need that too. I need that too. I need that too. I need that too. I need that too. I need that too. I need that too. I need that too. I need that too. I need that too. I need that too. I need that too. I need that too. I need that too. I need that too. I need that too. I need that too. I need that too. I need that too. I need that too. I need that too. I need that too. I need that too. I need that too. I need that too. I need that too. I need that too. I need that too. I need that too. I need that too. I need that too. I need that too. I need that too. I need that too. I need that too. I need that too. I need that too. I need that too. I need that too. I need that too. I need that too. I need that too. I need that too. I need that too. I need that too. I need that too. I need that too. I need that too. I need that too. I need that too. I need that too. I need that too. I need that too. I need that too. I need that too. I need that too. I need that too. I need that too. I need that too. I need that too. I need that too. I need that too. I need that too. I need that too. I need that too. I need that too. I need that too",
+ "raw": {
+ "normalized_reference": "it will break our hearts and our backs to hoist anchor by hand.",
+ "normalized_hypothesis": "i need that too. i need that too. i need that too. i need that too. i need that too. i need that too. i need that too. i need that too. i need that too. i need that too. i need that too. i need that too. i need that too. i need that too. i need that too. i need that too. i need that too. i need that too. i need that too. i need that too. i need that too. i need that too. i need that too. i need that too. i need that too. i need that too. i need that too. i need that too. i need that too. i need that too. i need that too. i need that too. i need that too. i need that too. i need that too. i need that too. i need that too. i need that too. i need that too. i need that too. i need that too. i need that too. i need that too. i need that too. i need that too. i need that too. i need that too. i need that too. i need that too. i need that too. i need that too. i need that too. i need that too. i need that too. i need that too. i need that too. i need that too. i need that too. i need that too. i need that too. i need that too. i need that too. i need that too. i need that too. i need that too. i need that too. i need that too. i need that too. i need that too. i need that too. i need that too. i need that too. i need that too. i need that too. i need that too. i need that too. i need that too. i need that too. i need that too. i need that too. i need that too. i need that too. i need that too. i need that too. i need that too. i need that too. i need that too. i need that too. i need that too",
+ "errors": 356,
+ "words": 13,
+ "wer": 27.384615384615383
+ },
+ "semantic": {
+ "normalized_reference": "it will break our hearts and our backs to hoist anchor by hand",
+ "normalized_hypothesis": "i need that too i need that too i need that too i need that too i need that too i need that too i need that too i need that too i need that too i need that too i need that too i need that too i need that too i need that too i need that too i need that too i need that too i need that too i need that too i need that too i need that too i need that too i need that too i need that too i need that too i need that too i need that too i need that too i need that too i need that too i need that too i need that too i need that too i need that too i need that too i need that too i need that too i need that too i need that too i need that too i need that too i need that too i need that too i need that too i need that too i need that too i need that too i need that too i need that too i need that too i need that too i need that too i need that too i need that too i need that too i need that too i need that too i need that too i need that too i need that too i need that too i need that too i need that too i need that too i need that too i need that too i need that too i need that too i need that too i need that too i need that too i need that too i need that too i need that too i need that too i need that too i need that too i need that too i need that too i need that too i need that too i need that too i need that too i need that too i need that too i need that too i need that too i need that too i need that too",
+ "errors": 356,
+ "words": 13,
+ "wer": 27.384615384615383
+ }
+ },
+ {
+ "id": "release500_0400",
+ "category": "phonetic_balance",
+ "text": "He made no reply as he waited for Whittemore to continue.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0400.wav",
+ "render_seconds": 142.1903632429894,
+ "audio_seconds": 4.040272108843538,
+ "rtf": 35.19326407044626,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I love you all so much. I love you all so much. I love you all so much. I love you all so much. I love you all so much. I love you all so much. I love you all so much. I love you all so much. I love you all so much. I love you all so much. I love you all so much. I love you all so much. I love you all so much. I love you all so much. I love you all so much. I love you all so much. I love you all so much. I love you all so much. I love you all so much. I love you all so much. I love you all so much. I love you all so much. I love you all so much. I love you all so much. I love you all so much. I love you all so much. I love you all so much. I love you all so much. I love you all so much. I love you all so much. I love you all so much. I love you all so much. I love you all so much. I love you all so much. I love you all so much. I love you all so much. I love you all so much. I love you all so much. I love you all so much. I love you all so much. I love you all so much. I love you all so much. I love you all so much. I love you all so much. I love you all so much. I love you all so much. I love you all so much. I love you all so much. I love you all so much. I love you all so much. I love you all so much. I love you all so much. I love you all so much. I love you all so much. I love you all so much. I love you all so much. I love you all so much. I love you all so much. I love you all so much. I love you all so much. I love you all so much. I love you all so much. I love you all so much. I love you",
+ "raw": {
+ "normalized_reference": "he made no reply as he waited for whittemore to continue.",
+ "normalized_hypothesis": "i love you all so much. i love you all so much. i love you all so much. i love you all so much. i love you all so much. i love you all so much. i love you all so much. i love you all so much. i love you all so much. i love you all so much. i love you all so much. i love you all so much. i love you all so much. i love you all so much. i love you all so much. i love you all so much. i love you all so much. i love you all so much. i love you all so much. i love you all so much. i love you all so much. i love you all so much. i love you all so much. i love you all so much. i love you all so much. i love you all so much. i love you all so much. i love you all so much. i love you all so much. i love you all so much. i love you all so much. i love you all so much. i love you all so much. i love you all so much. i love you all so much. i love you all so much. i love you all so much. i love you all so much. i love you all so much. i love you all so much. i love you all so much. i love you all so much. i love you all so much. i love you all so much. i love you all so much. i love you all so much. i love you all so much. i love you all so much. i love you all so much. i love you all so much. i love you all so much. i love you all so much. i love you all so much. i love you all so much. i love you all so much. i love you all so much. i love you all so much. i love you all so much. i love you all so much. i love you all so much. i love you all so much. i love you all so much. i love you all so much. i love you",
+ "errors": 381,
+ "words": 11,
+ "wer": 34.63636363636363
+ },
+ "semantic": {
+ "normalized_reference": "he made no reply as he waited for whittemore to continue",
+ "normalized_hypothesis": "i love you all so much i love you all so much i love you all so much i love you all so much i love you all so much i love you all so much i love you all so much i love you all so much i love you all so much i love you all so much i love you all so much i love you all so much i love you all so much i love you all so much i love you all so much i love you all so much i love you all so much i love you all so much i love you all so much i love you all so much i love you all so much i love you all so much i love you all so much i love you all so much i love you all so much i love you all so much i love you all so much i love you all so much i love you all so much i love you all so much i love you all so much i love you all so much i love you all so much i love you all so much i love you all so much i love you all so much i love you all so much i love you all so much i love you all so much i love you all so much i love you all so much i love you all so much i love you all so much i love you all so much i love you all so much i love you all so much i love you all so much i love you all so much i love you all so much i love you all so much i love you all so much i love you all so much i love you all so much i love you all so much i love you all so much i love you all so much i love you all so much i love you all so much i love you all so much i love you all so much i love you all so much i love you all so much i love you all so much i love you",
+ "errors": 381,
+ "words": 11,
+ "wer": 34.63636363636363
+ }
+ },
+ {
+ "id": "release500_0401",
+ "category": "phonetic_balance",
+ "text": "Who the devil gave it to you to be judge and jury.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0401.wav",
+ "render_seconds": 113.69965788710397,
+ "audio_seconds": 3.3436734693877552,
+ "rtf": 34.004414285083584,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Thank you for watching. Please like, share and subscribe. Thank you for watching. Please like, share and subscribe. Thank you for watching. Please like, share and subscribe. Thank you for watching. Please like, share and subscribe. Thank you for watching. Please like, share and subscribe. Thank you for watching. Please like, share and subscribe. Thank you for watching. Please like, share and subscribe. Thank you for watching. Please like, share and subscribe. Thank you for watching. Please like, share and subscribe. Please like, share and subscribe. Please like, share and subscribe. Please like, share and subscribe. Please like, share and subscribe. Please like, share and subscribe. Please like, share and subscribe. Please like, share and subscribe. Please like, share and subscribe. Please like, share and subscribe. Please like, share and subscribe. Please like, share and subscribe. Please like, share and subscribe. Please like, share and subscribe. Please like, share and subscribe. Please like, share and subscribe. Please like, share and subscribe. Please like, share and subscribe. Please like, share and subscribe. Please like, share and subscribe. Please like, share and subscribe. Please like, share and subscribe. Please like, share and subscribe. Please like, share and subscribe. Please like, share and subscribe. Please like, share and subscribe. Please like, share and subscribe. Please like, share and subscribe. Please like, share and subscribe. Please like, share and subscribe. Please like, share and subscribe. Please like, share and subscribe. Please like, share and subscribe. Please like, share and subscribe. Please like, share and subscribe. Please like, share and subscribe. Please like, share and subscribe. Please like, share and subscribe. Please like, share and subscribe. Please like, share and subscribe. Please like, share and subscribe. Please like, share and subscribe. Please like, share and subscribe. Please like, share and subscribe. Please like, share and subscribe. Please like, share and subscribe. Please like, share and subscribe. Please like, share and subscribe. Please like, subscribe and subscribe.",
+ "raw": {
+ "normalized_reference": "who the devil gave it to you to be judge and jury.",
+ "normalized_hypothesis": "thank you for watching. please like, share and subscribe. thank you for watching. please like, share and subscribe. thank you for watching. please like, share and subscribe. thank you for watching. please like, share and subscribe. thank you for watching. please like, share and subscribe. thank you for watching. please like, share and subscribe. thank you for watching. please like, share and subscribe. thank you for watching. please like, share and subscribe. thank you for watching. please like, share and subscribe. please like, share and subscribe. please like, share and subscribe. please like, share and subscribe. please like, share and subscribe. please like, share and subscribe. please like, share and subscribe. please like, share and subscribe. please like, share and subscribe. please like, share and subscribe. please like, share and subscribe. please like, share and subscribe. please like, share and subscribe. please like, share and subscribe. please like, share and subscribe. please like, share and subscribe. please like, share and subscribe. please like, share and subscribe. please like, share and subscribe. please like, share and subscribe. please like, share and subscribe. please like, share and subscribe. please like, share and subscribe. please like, share and subscribe. please like, share and subscribe. please like, share and subscribe. please like, share and subscribe. please like, share and subscribe. please like, share and subscribe. please like, share and subscribe. please like, share and subscribe. please like, share and subscribe. please like, share and subscribe. please like, share and subscribe. please like, share and subscribe. please like, share and subscribe. please like, share and subscribe. please like, share and subscribe. please like, share and subscribe. please like, share and subscribe. please like, share and subscribe. please like, share and subscribe. please like, share and subscribe. please like, share and subscribe. please like, share and subscribe. please like, share and subscribe. please like, share and subscribe. please like, share and subscribe. please like, subscribe and subscribe.",
+ "errors": 319,
+ "words": 12,
+ "wer": 26.583333333333332
+ },
+ "semantic": {
+ "normalized_reference": "who the devil gave it to you to be judge and jury",
+ "normalized_hypothesis": "thank you for watching please like share and subscribe thank you for watching please like share and subscribe thank you for watching please like share and subscribe thank you for watching please like share and subscribe thank you for watching please like share and subscribe thank you for watching please like share and subscribe thank you for watching please like share and subscribe thank you for watching please like share and subscribe thank you for watching please like share and subscribe please like share and subscribe please like share and subscribe please like share and subscribe please like share and subscribe please like share and subscribe please like share and subscribe please like share and subscribe please like share and subscribe please like share and subscribe please like share and subscribe please like share and subscribe please like share and subscribe please like share and subscribe please like share and subscribe please like share and subscribe please like share and subscribe please like share and subscribe please like share and subscribe please like share and subscribe please like share and subscribe please like share and subscribe please like share and subscribe please like share and subscribe please like share and subscribe please like share and subscribe please like share and subscribe please like share and subscribe please like share and subscribe please like share and subscribe please like share and subscribe please like share and subscribe please like share and subscribe please like share and subscribe please like share and subscribe please like share and subscribe please like share and subscribe please like share and subscribe please like share and subscribe please like share and subscribe please like share and subscribe please like share and subscribe please like share and subscribe please like share and subscribe please like share and subscribe please like share and subscribe please like share and subscribe please like share and subscribe please like subscribe and subscribe",
+ "errors": 319,
+ "words": 12,
+ "wer": 26.583333333333332
+ }
+ },
+ {
+ "id": "release500_0402",
+ "category": "phonetic_balance",
+ "text": "Daylight found himself charmed and made curious by the little man.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0402.wav",
+ "render_seconds": 166.39715103199705,
+ "audio_seconds": 4.527891156462585,
+ "rtf": 36.749370795828675,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I want to go to school. I want to go to school. I want to go to school. I want to go to school. I want to go to school. I want to go to school. I want to go to school. I want to go to school. I want to go to school. I want to go to school. I want to go to school. I want to go to school. I want to go to school. I want to go to school. I want to go to school. I want to go to school. I want to go to school. I want to go to school. I want to go to school. I want to go to school. I want to go to school. I want to go to school. I want to go to school. I want to go to school. I want to go to school. I want to go to school. I want to go to school. I want to go to school. I want to go to school. I want to go to school. I want to go to school. I want to go to school. I want to go to school. I want to go to school. I want to go to school. I want to go to school. I want to go to school. I want to go to school. I want to go to school. I want to go to school. I want to go to school. I want to go to school. I want to go to school. I want to go to school. I want to go to school. I want to go to school. I want to go to school. I want to go to school. I want to go to school. I want to go to school. I want to go to school. I want to go to school. I want to go to school. I want to go to school. I want to go to school. I want to go to school. I want to go to school. I want to go to school. I want to go to school. I want to go to school. I want to go to school. I want to go to school. I want to go to school. I want to",
+ "raw": {
+ "normalized_reference": "daylight found himself charmed and made curious by the little man.",
+ "normalized_hypothesis": "i want to go to school. i want to go to school. i want to go to school. i want to go to school. i want to go to school. i want to go to school. i want to go to school. i want to go to school. i want to go to school. i want to go to school. i want to go to school. i want to go to school. i want to go to school. i want to go to school. i want to go to school. i want to go to school. i want to go to school. i want to go to school. i want to go to school. i want to go to school. i want to go to school. i want to go to school. i want to go to school. i want to go to school. i want to go to school. i want to go to school. i want to go to school. i want to go to school. i want to go to school. i want to go to school. i want to go to school. i want to go to school. i want to go to school. i want to go to school. i want to go to school. i want to go to school. i want to go to school. i want to go to school. i want to go to school. i want to go to school. i want to go to school. i want to go to school. i want to go to school. i want to go to school. i want to go to school. i want to go to school. i want to go to school. i want to go to school. i want to go to school. i want to go to school. i want to go to school. i want to go to school. i want to go to school. i want to go to school. i want to go to school. i want to go to school. i want to go to school. i want to go to school. i want to go to school. i want to go to school. i want to go to school. i want to go to school. i want to go to school. i want to",
+ "errors": 381,
+ "words": 11,
+ "wer": 34.63636363636363
+ },
+ "semantic": {
+ "normalized_reference": "daylight found himself charmed and made curious by the little man",
+ "normalized_hypothesis": "i want to go to school i want to go to school i want to go to school i want to go to school i want to go to school i want to go to school i want to go to school i want to go to school i want to go to school i want to go to school i want to go to school i want to go to school i want to go to school i want to go to school i want to go to school i want to go to school i want to go to school i want to go to school i want to go to school i want to go to school i want to go to school i want to go to school i want to go to school i want to go to school i want to go to school i want to go to school i want to go to school i want to go to school i want to go to school i want to go to school i want to go to school i want to go to school i want to go to school i want to go to school i want to go to school i want to go to school i want to go to school i want to go to school i want to go to school i want to go to school i want to go to school i want to go to school i want to go to school i want to go to school i want to go to school i want to go to school i want to go to school i want to go to school i want to go to school i want to go to school i want to go to school i want to go to school i want to go to school i want to go to school i want to go to school i want to go to school i want to go to school i want to go to school i want to go to school i want to go to school i want to go to school i want to go to school i want to go to school i want to",
+ "errors": 381,
+ "words": 11,
+ "wer": 34.63636363636363
+ }
+ },
+ {
+ "id": "release500_0403",
+ "category": "phonetic_balance",
+ "text": "It's a Yankee, Joan cried.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0403.wav",
+ "render_seconds": 82.81564847100526,
+ "audio_seconds": 2.089795918367347,
+ "rtf": 39.62858178788337,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Achoo! Achoo! Achoo! Achoo! Achoo! Achoo! Achoo! Achoo! Achoo! Achoo! Achoo! Achoo! Achoo! Achoo! Achoo! Achoo! Achoo! Achoo! Achoo! Achoo! Achoo! Achoo! Achoo! Achoo! Achoo! Achoo! Achoo! Achoo! Achoo! Achoo! Achoo! Achoo! Achoo! Achoo! Achoo! Achoo! Achoo! Achoo! Achoo! Achoo! Achoo! Achoo! Achoo! Achoo! Achoo! Achoo! Achoo! Achoo! Achoo! Achoo! Achoo! Achoo! Achoo! Achoo! Achoo! Achoo! Achoo! Achoo! Achoo! Achoo! Achoo! Achoo! Achoo! Achoo! Achoo! Achoo! Achoo! Achoo! Achoo! Achoo! Achoo! Achoo! Achoo! Achoo! Achoo! Achoo! Achoo! Achoo! Achoo! Achoo! Achoo! Achoo! Achoo! Achoo! Achoo! Achoo! Achoo! Achoo! Achoo! Achoo! Achoo! Achoo! Achoo! Achoo! Achoo! Achoo! Achoo! Achoo! Achoo! Achoo! Achoo! Achoo! Achoo! Achoo! Achoo! Achoo! Achoo! Achoo! Achoo! Achoo! Achoo! Achoo! Achoo! Achoo! Achoo! Achoo! Achoo! Achoo! Achoo! Achoo! Achoo! Achoo! Achoo! Achoo! Achoo! Achoo! Achoo! Achoo! Achoo! Achoo! Achoo! Achoo! Achoo! Achoo! Achoo! Achoo! Achoo! Achoo! Achoo! Achoo! Achoo! Achoo! Achoo! Achoo! Achoo! Achoo! Achoo! Achoo!",
+ "raw": {
+ "normalized_reference": "it's a yankee, joan cried.",
+ "normalized_hypothesis": "achoo! achoo! achoo! achoo! achoo! achoo! achoo! achoo! achoo! achoo! achoo! achoo! achoo! achoo! achoo! achoo! achoo! achoo! achoo! achoo! achoo! achoo! achoo! achoo! achoo! achoo! achoo! achoo! achoo! achoo! achoo! achoo! achoo! achoo! achoo! achoo! achoo! achoo! achoo! achoo! achoo! achoo! achoo! achoo! achoo! achoo! achoo! achoo! achoo! achoo! achoo! achoo! achoo! achoo! achoo! achoo! achoo! achoo! achoo! achoo! achoo! achoo! achoo! achoo! achoo! achoo! achoo! achoo! achoo! achoo! achoo! achoo! achoo! achoo! achoo! achoo! achoo! achoo! achoo! achoo! achoo! achoo! achoo! achoo! achoo! achoo! achoo! achoo! achoo! achoo! achoo! achoo! achoo! achoo! achoo! achoo! achoo! achoo! achoo! achoo! achoo! achoo! achoo! achoo! achoo! achoo! achoo! achoo! achoo! achoo! achoo! achoo! achoo! achoo! achoo! achoo! achoo! achoo! achoo! achoo! achoo! achoo! achoo! achoo! achoo! achoo! achoo! achoo! achoo! achoo! achoo! achoo! achoo! achoo! achoo! achoo! achoo! achoo! achoo! achoo! achoo! achoo! achoo! achoo! achoo! achoo! achoo! achoo!",
+ "errors": 148,
+ "words": 5,
+ "wer": 29.6
+ },
+ "semantic": {
+ "normalized_reference": "it is a yankee joan cried",
+ "normalized_hypothesis": "achoo achoo achoo achoo achoo achoo achoo achoo achoo achoo achoo achoo achoo achoo achoo achoo achoo achoo achoo achoo achoo achoo achoo achoo achoo achoo achoo achoo achoo achoo achoo achoo achoo achoo achoo achoo achoo achoo achoo achoo achoo achoo achoo achoo achoo achoo achoo achoo achoo achoo achoo achoo achoo achoo achoo achoo achoo achoo achoo achoo achoo achoo achoo achoo achoo achoo achoo achoo achoo achoo achoo achoo achoo achoo achoo achoo achoo achoo achoo achoo achoo achoo achoo achoo achoo achoo achoo achoo achoo achoo achoo achoo achoo achoo achoo achoo achoo achoo achoo achoo achoo achoo achoo achoo achoo achoo achoo achoo achoo achoo achoo achoo achoo achoo achoo achoo achoo achoo achoo achoo achoo achoo achoo achoo achoo achoo achoo achoo achoo achoo achoo achoo achoo achoo achoo achoo achoo achoo achoo achoo achoo achoo achoo achoo achoo achoo achoo achoo",
+ "errors": 148,
+ "words": 6,
+ "wer": 24.666666666666668
+ }
+ },
+ {
+ "id": "release500_0404",
+ "category": "phonetic_balance",
+ "text": "But what they want with your toothbrush is more than I can imagine.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0404.wav",
+ "render_seconds": 135.29682943108492,
+ "audio_seconds": 4.388571428571429,
+ "rtf": 30.829355664635234,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "But what about work? What's your task force? What's your work and what can I learn from you?",
+ "raw": {
+ "normalized_reference": "but what they want with your toothbrush is more than i can imagine.",
+ "normalized_hypothesis": "but what about work? what's your task force? what's your work and what can i learn from you?",
+ "errors": 14,
+ "words": 13,
+ "wer": 1.0769230769230769
+ },
+ "semantic": {
+ "normalized_reference": "but what they want with your toothbrush is more than i can imagine",
+ "normalized_hypothesis": "but what about work what is your task force what is your work and what can i learn from you",
+ "errors": 15,
+ "words": 13,
+ "wer": 1.1538461538461537
+ }
+ },
+ {
+ "id": "release500_0405",
+ "category": "phonetic_balance",
+ "text": "They were babbling and chattering all together.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0405.wav",
+ "render_seconds": 107.5880376159912,
+ "audio_seconds": 3.2043537414965986,
+ "rtf": 33.57558069283014,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I will help you and travel on each other.",
+ "raw": {
+ "normalized_reference": "they were babbling and chattering all together.",
+ "normalized_hypothesis": "i will help you and travel on each other.",
+ "errors": 8,
+ "words": 7,
+ "wer": 1.1428571428571428
+ },
+ "semantic": {
+ "normalized_reference": "they were babbling and chattering all together",
+ "normalized_hypothesis": "i will help you and travel on each other",
+ "errors": 8,
+ "words": 7,
+ "wer": 1.1428571428571428
+ }
+ },
+ {
+ "id": "release500_0406",
+ "category": "phonetic_balance",
+ "text": "Men of Selden's stamp don't stop at women and children.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0406.wav",
+ "render_seconds": 131.00622504099738,
+ "audio_seconds": 3.4133333333333336,
+ "rtf": 38.3807299924797,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I have saved the sky, I have saved the world, I have saved the world.",
+ "raw": {
+ "normalized_reference": "men of selden's stamp don't stop at women and children.",
+ "normalized_hypothesis": "i have saved the sky, i have saved the world, i have saved the world.",
+ "errors": 15,
+ "words": 10,
+ "wer": 1.5
+ },
+ "semantic": {
+ "normalized_reference": "men of selden is stamp do not stop at women and children",
+ "normalized_hypothesis": "i have saved the sky i have saved the world i have saved the world",
+ "errors": 15,
+ "words": 12,
+ "wer": 1.25
+ }
+ },
+ {
+ "id": "release500_0407",
+ "category": "phonetic_balance",
+ "text": "Cameron looked at his hands with their long, sinewy fingers.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0407.wav",
+ "render_seconds": 137.7075223509455,
+ "audio_seconds": 4.179591836734694,
+ "rtf": 32.94760056248208,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "How I look at this house without you, sir, I'm not famous.",
+ "raw": {
+ "normalized_reference": "cameron looked at his hands with their long, sinewy fingers.",
+ "normalized_hypothesis": "how i look at this house without you, sir, i'm not famous.",
+ "errors": 11,
+ "words": 10,
+ "wer": 1.1
+ },
+ "semantic": {
+ "normalized_reference": "cameron looked at his hands with their long sinewy fingers",
+ "normalized_hypothesis": "how i look at this house without you sir i am not famous",
+ "errors": 12,
+ "words": 10,
+ "wer": 1.2
+ }
+ },
+ {
+ "id": "release500_0408",
+ "category": "phonetic_balance",
+ "text": "Saxon waited, for she knew a fresh idea had struck Billy.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0408.wav",
+ "render_seconds": 137.90217854199,
+ "audio_seconds": 4.179591836734694,
+ "rtf": 32.994173576940966,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Satsang with Mooji Satsang with Mooji Satsang with Mooji Satsang with Mooji Satsang with Mooji Satsang with Mooji Satsang with Mooji Satsang with Mooji Satsang with Mooji Satsang with Mooji Satsang with Mooji Satsang with Mooji Satsang with Mooji Satsang with Mooji Satsang with Mooji Satsang with Mooji Satsang with Mooji Satsang with Mooji Satsang with Mooji Satsang with Mooji Satsang with Mooji Satsang with Mooji Satsang with Mooji Satsang with Mooji Satsang with Mooji Satsang with Mooji Satsang with Mooji Satsang with Mooji Satsang with Mooji Satsang with Mooji Satsang with Mooji Satsang with Mooji Satsang with Mooji Satsang with Mooji Satsang with Mooji Satsang with Mooji Satsang with Mooji Satsang with Mooji Satsang with Mooji Satsang with Mooji Satsang with Mooji Satsang with Mooji Satsang with Mooji Satsang with Mooji Satsang with Mooji Satsang with Mooji Satsang with Mooji Satsang with Mooji Satsang with Mooji Satsang with Mooji Satsang with Mooji Satsang with Mooji Satsang with Mooji Satsang with Mooji Satsang with Mooji Satsang with Mooji Satsang with Mooji Satsang with Mooji Satsang with Mooji Satsang with Mooji Satsang with Mooji Satsang with Mooji Satsang with Mooji Satsang with Mooji Satsang with Mooji Satsang with Mooji Satsang with Mooji Satsang with Mooji Satsang with Mooji Satsang with Mooji Satsang with Mooji Satsang with Mooji Satsang with Mooji Satsang with Mooji",
+ "raw": {
+ "normalized_reference": "saxon waited, for she knew a fresh idea had struck billy.",
+ "normalized_hypothesis": "satsang with mooji satsang with mooji satsang with mooji satsang with mooji satsang with mooji satsang with mooji satsang with mooji satsang with mooji satsang with mooji satsang with mooji satsang with mooji satsang with mooji satsang with mooji satsang with mooji satsang with mooji satsang with mooji satsang with mooji satsang with mooji satsang with mooji satsang with mooji satsang with mooji satsang with mooji satsang with mooji satsang with mooji satsang with mooji satsang with mooji satsang with mooji satsang with mooji satsang with mooji satsang with mooji satsang with mooji satsang with mooji satsang with mooji satsang with mooji satsang with mooji satsang with mooji satsang with mooji satsang with mooji satsang with mooji satsang with mooji satsang with mooji satsang with mooji satsang with mooji satsang with mooji satsang with mooji satsang with mooji satsang with mooji satsang with mooji satsang with mooji satsang with mooji satsang with mooji satsang with mooji satsang with mooji satsang with mooji satsang with mooji satsang with mooji satsang with mooji satsang with mooji satsang with mooji satsang with mooji satsang with mooji satsang with mooji satsang with mooji satsang with mooji satsang with mooji satsang with mooji satsang with mooji satsang with mooji satsang with mooji satsang with mooji satsang with mooji satsang with mooji satsang with mooji satsang with mooji",
+ "errors": 222,
+ "words": 11,
+ "wer": 20.181818181818183
+ },
+ "semantic": {
+ "normalized_reference": "saxon waited for she knew a fresh idea had struck billy",
+ "normalized_hypothesis": "satsang with mooji satsang with mooji satsang with mooji satsang with mooji satsang with mooji satsang with mooji satsang with mooji satsang with mooji satsang with mooji satsang with mooji satsang with mooji satsang with mooji satsang with mooji satsang with mooji satsang with mooji satsang with mooji satsang with mooji satsang with mooji satsang with mooji satsang with mooji satsang with mooji satsang with mooji satsang with mooji satsang with mooji satsang with mooji satsang with mooji satsang with mooji satsang with mooji satsang with mooji satsang with mooji satsang with mooji satsang with mooji satsang with mooji satsang with mooji satsang with mooji satsang with mooji satsang with mooji satsang with mooji satsang with mooji satsang with mooji satsang with mooji satsang with mooji satsang with mooji satsang with mooji satsang with mooji satsang with mooji satsang with mooji satsang with mooji satsang with mooji satsang with mooji satsang with mooji satsang with mooji satsang with mooji satsang with mooji satsang with mooji satsang with mooji satsang with mooji satsang with mooji satsang with mooji satsang with mooji satsang with mooji satsang with mooji satsang with mooji satsang with mooji satsang with mooji satsang with mooji satsang with mooji satsang with mooji satsang with mooji satsang with mooji satsang with mooji satsang with mooji satsang with mooji satsang with mooji",
+ "errors": 222,
+ "words": 11,
+ "wer": 20.181818181818183
+ }
+ },
+ {
+ "id": "release500_0409",
+ "category": "phonetic_balance",
+ "text": "He saw only the effect in a general, sketchy way.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0409.wav",
+ "render_seconds": 128.47979144204874,
+ "audio_seconds": 3.4133333333333336,
+ "rtf": 37.640563899037716,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I'll see you at the factory town, Scottsdale.",
+ "raw": {
+ "normalized_reference": "he saw only the effect in a general, sketchy way.",
+ "normalized_hypothesis": "i'll see you at the factory town, scottsdale.",
+ "errors": 10,
+ "words": 10,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "he saw only the effect in a general sketchy way",
+ "normalized_hypothesis": "i will see you at the factory town scottsdale",
+ "errors": 10,
+ "words": 10,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0410",
+ "category": "phonetic_balance",
+ "text": "He began to follow the footprints of the dog.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0410.wav",
+ "render_seconds": 93.816970793996,
+ "audio_seconds": 2.995374149659864,
+ "rtf": 31.320618429136566,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I'll see you in the next video.",
+ "raw": {
+ "normalized_reference": "he began to follow the footprints of the dog.",
+ "normalized_hypothesis": "i'll see you in the next video.",
+ "errors": 8,
+ "words": 9,
+ "wer": 0.8888888888888888
+ },
+ "semantic": {
+ "normalized_reference": "he began to follow the footprints of the dog",
+ "normalized_hypothesis": "i will see you in the next video",
+ "errors": 9,
+ "words": 9,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0411",
+ "category": "phonetic_balance",
+ "text": "Do you value your hide.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0411.wav",
+ "render_seconds": 70.79543694294989,
+ "audio_seconds": 1.811156462585034,
+ "rtf": 39.088526256812024,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I'll see you all soon.",
+ "raw": {
+ "normalized_reference": "do you value your hide.",
+ "normalized_hypothesis": "i'll see you all soon.",
+ "errors": 5,
+ "words": 5,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "do you value your hide",
+ "normalized_hypothesis": "i will see you all soon",
+ "errors": 6,
+ "words": 5,
+ "wer": 1.2
+ }
+ },
+ {
+ "id": "release500_0412",
+ "category": "phonetic_balance",
+ "text": "A rising tide of fat had submerged them.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0412.wav",
+ "render_seconds": 116.99265528493561,
+ "audio_seconds": 2.8560544217687074,
+ "rtf": 40.96303431518086,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Welcome to the third semester.",
+ "raw": {
+ "normalized_reference": "a rising tide of fat had submerged them.",
+ "normalized_hypothesis": "welcome to the third semester.",
+ "errors": 8,
+ "words": 8,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "a rising tide of fat had submerged them",
+ "normalized_hypothesis": "welcome to the 3rd semester",
+ "errors": 8,
+ "words": 8,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0413",
+ "category": "phonetic_balance",
+ "text": "There was nothing on the rock.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0413.wav",
+ "render_seconds": 86.40459188702516,
+ "audio_seconds": 2.2291156462585033,
+ "rtf": 38.761825584084164,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I wish you all the best in the world.",
+ "raw": {
+ "normalized_reference": "there was nothing on the rock.",
+ "normalized_hypothesis": "i wish you all the best in the world.",
+ "errors": 8,
+ "words": 6,
+ "wer": 1.3333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "there was nothing on the rock",
+ "normalized_hypothesis": "i wish you all the best in the world",
+ "errors": 8,
+ "words": 6,
+ "wer": 1.3333333333333333
+ }
+ },
+ {
+ "id": "release500_0414",
+ "category": "phonetic_balance",
+ "text": "Whatever he guessed he locked away in the taboo room of Naomi.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0414.wav",
+ "render_seconds": 146.0921727620298,
+ "audio_seconds": 4.458231292517007,
+ "rtf": 32.769087823514376,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "You have all had a taste of the good, the raw, and the terrible wars now.",
+ "raw": {
+ "normalized_reference": "whatever he guessed he locked away in the taboo room of naomi.",
+ "normalized_hypothesis": "you have all had a taste of the good, the raw, and the terrible wars now.",
+ "errors": 15,
+ "words": 12,
+ "wer": 1.25
+ },
+ "semantic": {
+ "normalized_reference": "whatever he guessed he locked away in the taboo room of naomi",
+ "normalized_hypothesis": "you have all had a taste of the good the raw and the terrible wars now",
+ "errors": 15,
+ "words": 12,
+ "wer": 1.25
+ }
+ },
+ {
+ "id": "release500_0415",
+ "category": "phonetic_balance",
+ "text": "Will we ever forget it.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0415.wav",
+ "render_seconds": 78.00979957601521,
+ "audio_seconds": 1.9504761904761905,
+ "rtf": 39.995258571687486,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "You rather forget it.",
+ "raw": {
+ "normalized_reference": "will we ever forget it.",
+ "normalized_hypothesis": "you rather forget it.",
+ "errors": 3,
+ "words": 5,
+ "wer": 0.6
+ },
+ "semantic": {
+ "normalized_reference": "will we ever forget it",
+ "normalized_hypothesis": "you rather forget it",
+ "errors": 3,
+ "words": 5,
+ "wer": 0.6
+ }
+ },
+ {
+ "id": "release500_0416",
+ "category": "phonetic_balance",
+ "text": "Jeanne and Pierre both gazed toward the great rock.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0416.wav",
+ "render_seconds": 141.89520880999044,
+ "audio_seconds": 3.622312925170068,
+ "rtf": 39.17254299704889,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "How about this ice cream?",
+ "raw": {
+ "normalized_reference": "jeanne and pierre both gazed toward the great rock.",
+ "normalized_hypothesis": "how about this ice cream?",
+ "errors": 9,
+ "words": 9,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "jeanne and pierre both gazed toward the great rock",
+ "normalized_hypothesis": "how about this ice cream",
+ "errors": 9,
+ "words": 9,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0417",
+ "category": "phonetic_balance",
+ "text": "Does the old boy often go off at half-cock that way.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0417.wav",
+ "render_seconds": 118.60370925499592,
+ "audio_seconds": 3.4829931972789114,
+ "rtf": 34.052236836883594,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Just try to be honest and be as sarcastic as you can.",
+ "raw": {
+ "normalized_reference": "does the old boy often go off at half-cock that way.",
+ "normalized_hypothesis": "just try to be honest and be as sarcastic as you can.",
+ "errors": 12,
+ "words": 12,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "does the old boy often go off at half cock that way",
+ "normalized_hypothesis": "just try to be honest and be as sarcastic as you can",
+ "errors": 12,
+ "words": 12,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0418",
+ "category": "phonetic_balance",
+ "text": "The temperature dropped to fifty below zero and remained there the whole trip.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0418.wav",
+ "render_seconds": 172.6113193150377,
+ "audio_seconds": 5.363809523809524,
+ "rtf": 32.180732471731105,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Thank you for watching. Please subscribe to my channel for more videos.",
+ "raw": {
+ "normalized_reference": "the temperature dropped to fifty below zero and remained there the whole trip.",
+ "normalized_hypothesis": "thank you for watching. please subscribe to my channel for more videos.",
+ "errors": 13,
+ "words": 13,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "the temperature dropped to 50 below 0 and remained there the whole trip",
+ "normalized_hypothesis": "thank you for watching please subscribe to my channel for more videos",
+ "errors": 13,
+ "words": 13,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0419",
+ "category": "phonetic_balance",
+ "text": "It's worth eight dollars.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0419.wav",
+ "render_seconds": 82.20326712797396,
+ "audio_seconds": 2.020136054421769,
+ "rtf": 40.69194594494939,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Atreus Atreus",
+ "raw": {
+ "normalized_reference": "it's worth eight dollars.",
+ "normalized_hypothesis": "atreus atreus",
+ "errors": 4,
+ "words": 4,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "it is worth $8",
+ "normalized_hypothesis": "atreus atreus",
+ "errors": 4,
+ "words": 4,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0420",
+ "category": "phonetic_balance",
+ "text": "But I did not enjoy it long.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0420.wav",
+ "render_seconds": 87.99971777200699,
+ "audio_seconds": 2.159455782312925,
+ "rtf": 40.75087737047955,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you.",
+ "raw": {
+ "normalized_reference": "but i did not enjoy it long.",
+ "normalized_hypothesis": "i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you.",
+ "errors": 332,
+ "words": 7,
+ "wer": 47.42857142857143
+ },
+ "semantic": {
+ "normalized_reference": "but i did not enjoy it long",
+ "normalized_hypothesis": "i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you",
+ "errors": 332,
+ "words": 7,
+ "wer": 47.42857142857143
+ }
+ },
+ {
+ "id": "release500_0421",
+ "category": "phonetic_balance",
+ "text": "Shorty turned to their employers.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0421.wav",
+ "render_seconds": 108.18734872399364,
+ "audio_seconds": 2.4380952380952383,
+ "rtf": 44.37371725007551,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Hello and welcome to my channel.",
+ "raw": {
+ "normalized_reference": "shorty turned to their employers.",
+ "normalized_hypothesis": "hello and welcome to my channel.",
+ "errors": 5,
+ "words": 5,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "shorty turned to their employers",
+ "normalized_hypothesis": "hello and welcome to my channel",
+ "errors": 5,
+ "words": 5,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0422",
+ "category": "phonetic_balance",
+ "text": "It was the third or fourth time that Philip had heard MacDougall swear.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0422.wav",
+ "render_seconds": 119.79499838501215,
+ "audio_seconds": 5.015510204081632,
+ "rtf": 23.884907718365866,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I wish the thought of your first child had fallen apart like a loose straw.",
+ "raw": {
+ "normalized_reference": "it was the third or fourth time that philip had heard macdougall swear.",
+ "normalized_hypothesis": "i wish the thought of your first child had fallen apart like a loose straw.",
+ "errors": 14,
+ "words": 13,
+ "wer": 1.0769230769230769
+ },
+ "semantic": {
+ "normalized_reference": "it was the 3rd or 4th time that philip had heard macdougall swear",
+ "normalized_hypothesis": "i wish the thought of your 1st child had fallen apart like a loose straw",
+ "errors": 14,
+ "words": 13,
+ "wer": 1.0769230769230769
+ }
+ },
+ {
+ "id": "release500_0423",
+ "category": "phonetic_balance",
+ "text": "Ernest saw in the affair the most sinister import.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0423.wav",
+ "render_seconds": 127.59987312403973,
+ "audio_seconds": 3.6919727891156464,
+ "rtf": 34.56143379502108,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I should have found this telescope.",
+ "raw": {
+ "normalized_reference": "ernest saw in the affair the most sinister import.",
+ "normalized_hypothesis": "i should have found this telescope.",
+ "errors": 9,
+ "words": 9,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "ernest saw in the affair the most sinister import",
+ "normalized_hypothesis": "i should have found this telescope",
+ "errors": 9,
+ "words": 9,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0424",
+ "category": "phonetic_balance",
+ "text": "Also, at regular intervals, he would mutter.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0424.wav",
+ "render_seconds": 104.59484435198829,
+ "audio_seconds": 3.3436734693877552,
+ "rtf": 31.28141707304337,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It's a lot like what I know I call peace. I love it.",
+ "raw": {
+ "normalized_reference": "also, at regular intervals, he would mutter.",
+ "normalized_hypothesis": "it's a lot like what i know i call peace. i love it.",
+ "errors": 13,
+ "words": 7,
+ "wer": 1.8571428571428572
+ },
+ "semantic": {
+ "normalized_reference": "also at regular intervals he would mutter",
+ "normalized_hypothesis": "it is a lot like what i know i call peace i love it",
+ "errors": 14,
+ "words": 7,
+ "wer": 2.0
+ }
+ },
+ {
+ "id": "release500_0425",
+ "category": "phonetic_balance",
+ "text": "He may anticipate the day of his death.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0425.wav",
+ "render_seconds": 57.284214230952784,
+ "audio_seconds": 2.9257142857142857,
+ "rtf": 19.57956541097019,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I want to support the cause of death.",
+ "raw": {
+ "normalized_reference": "he may anticipate the day of his death.",
+ "normalized_hypothesis": "i want to support the cause of death.",
+ "errors": 6,
+ "words": 8,
+ "wer": 0.75
+ },
+ "semantic": {
+ "normalized_reference": "he may anticipate the day of his death",
+ "normalized_hypothesis": "i want to support the cause of death",
+ "errors": 6,
+ "words": 8,
+ "wer": 0.75
+ }
+ },
+ {
+ "id": "release500_0426",
+ "category": "phonetic_balance",
+ "text": "I defy any man to get a Solomon Island sore in California.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0426.wav",
+ "render_seconds": 124.20933483995032,
+ "audio_seconds": 4.31891156462585,
+ "rtf": 28.759406850857953,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "What the fuck am I allowed to ever say that I am sure I can't see?",
+ "raw": {
+ "normalized_reference": "i defy any man to get a solomon island sore in california.",
+ "normalized_hypothesis": "what the fuck am i allowed to ever say that i am sure i can't see?",
+ "errors": 15,
+ "words": 12,
+ "wer": 1.25
+ },
+ "semantic": {
+ "normalized_reference": "i defy any man to get a solomon island sore in california",
+ "normalized_hypothesis": "what the fuck am i allowed to ever say that i am sure i can not see",
+ "errors": 16,
+ "words": 12,
+ "wer": 1.3333333333333333
+ }
+ },
+ {
+ "id": "release500_0427",
+ "category": "phonetic_balance",
+ "text": "Now our figuring was all right.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0427.wav",
+ "render_seconds": 106.39812796097249,
+ "audio_seconds": 2.4380952380952383,
+ "rtf": 43.639857171492615,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Thank you for watching.",
+ "raw": {
+ "normalized_reference": "now our figuring was all right.",
+ "normalized_hypothesis": "thank you for watching.",
+ "errors": 6,
+ "words": 6,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "now our figuring was all right",
+ "normalized_hypothesis": "thank you for watching",
+ "errors": 6,
+ "words": 6,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0428",
+ "category": "phonetic_balance",
+ "text": "MacDougall tapped his forehead suspiciously with a stubby forefinger.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0428.wav",
+ "render_seconds": 152.4971202670131,
+ "audio_seconds": 5.085170068027211,
+ "rtf": 29.988597869289013,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "What are they talking about? What are they talking about? What are they talking about? What are they talking about? What are they talking about? What are they talking about? What are they talking about? What are they talking about? What are they talking about? What are they talking about? What are they talking about? What are they talking about? What are they talking about? What are they talking about? What are they talking about? What are they talking about? What are they talking about? What are they talking about? What are they talking about? What are they talking about? What are they talking about? What are they talking about? What are they talking about? What are they talking about? What are they talking about? What are they talking about? What are they talking about? What are they talking about? What are they talking about? What are they talking about? What are they talking about? What are they talking about? What are they talking about? What are they talking about? What are they talking about? What are they talking about? What are they talking about? What are they talking about? What are they talking about? What are they talking about? What are they talking about? What are they talking about? What are they talking about? What are they talking about? What are they talking about? What are they talking about? What are they talking about? What are they talking about? What are they talking about? What are they talking about? What are they talking about? What are they talking about? What are they talking about? What are they talking about? What are they talking about? What are they talking about? What are they talking about? What are they talking about? What are they talking about? What are they talking about? What are they talking about? What are they talking about? What are they talking about? What are they talking about? What are they talking about? What are they talking about? What are they talking about? What are they talking about? What are they talking about? What are they talking about? What are they talking about? What are they talking about? What are they talking about? What are they talking about?",
+ "raw": {
+ "normalized_reference": "macdougall tapped his forehead suspiciously with a stubby forefinger.",
+ "normalized_hypothesis": "what are they talking about? what are they talking about? what are they talking about? what are they talking about? what are they talking about? what are they talking about? what are they talking about? what are they talking about? what are they talking about? what are they talking about? what are they talking about? what are they talking about? what are they talking about? what are they talking about? what are they talking about? what are they talking about? what are they talking about? what are they talking about? what are they talking about? what are they talking about? what are they talking about? what are they talking about? what are they talking about? what are they talking about? what are they talking about? what are they talking about? what are they talking about? what are they talking about? what are they talking about? what are they talking about? what are they talking about? what are they talking about? what are they talking about? what are they talking about? what are they talking about? what are they talking about? what are they talking about? what are they talking about? what are they talking about? what are they talking about? what are they talking about? what are they talking about? what are they talking about? what are they talking about? what are they talking about? what are they talking about? what are they talking about? what are they talking about? what are they talking about? what are they talking about? what are they talking about? what are they talking about? what are they talking about? what are they talking about? what are they talking about? what are they talking about? what are they talking about? what are they talking about? what are they talking about? what are they talking about? what are they talking about? what are they talking about? what are they talking about? what are they talking about? what are they talking about? what are they talking about? what are they talking about? what are they talking about? what are they talking about? what are they talking about? what are they talking about? what are they talking about? what are they talking about? what are they talking about?",
+ "errors": 370,
+ "words": 9,
+ "wer": 41.111111111111114
+ },
+ "semantic": {
+ "normalized_reference": "macdougall tapped his forehead suspiciously with a stubby forefinger",
+ "normalized_hypothesis": "what are they talking about what are they talking about what are they talking about what are they talking about what are they talking about what are they talking about what are they talking about what are they talking about what are they talking about what are they talking about what are they talking about what are they talking about what are they talking about what are they talking about what are they talking about what are they talking about what are they talking about what are they talking about what are they talking about what are they talking about what are they talking about what are they talking about what are they talking about what are they talking about what are they talking about what are they talking about what are they talking about what are they talking about what are they talking about what are they talking about what are they talking about what are they talking about what are they talking about what are they talking about what are they talking about what are they talking about what are they talking about what are they talking about what are they talking about what are they talking about what are they talking about what are they talking about what are they talking about what are they talking about what are they talking about what are they talking about what are they talking about what are they talking about what are they talking about what are they talking about what are they talking about what are they talking about what are they talking about what are they talking about what are they talking about what are they talking about what are they talking about what are they talking about what are they talking about what are they talking about what are they talking about what are they talking about what are they talking about what are they talking about what are they talking about what are they talking about what are they talking about what are they talking about what are they talking about what are they talking about what are they talking about what are they talking about what are they talking about what are they talking about",
+ "errors": 370,
+ "words": 9,
+ "wer": 41.111111111111114
+ }
+ },
+ {
+ "id": "release500_0429",
+ "category": "phonetic_balance",
+ "text": "O'Brien took off his coat and bared his right arm.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0429.wav",
+ "render_seconds": 135.5034232470207,
+ "audio_seconds": 3.4133333333333336,
+ "rtf": 39.69826852940059,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "You're born to do this cool and badass wobble.",
+ "raw": {
+ "normalized_reference": "o'brien took off his coat and bared his right arm.",
+ "normalized_hypothesis": "you're born to do this cool and badass wobble.",
+ "errors": 10,
+ "words": 10,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "0 brien took off his coat and bared his right arm",
+ "normalized_hypothesis": "you are born to do this cool and badass wobble",
+ "errors": 11,
+ "words": 11,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0430",
+ "category": "phonetic_balance",
+ "text": "But it won't continue, she said with easy confidence.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0430.wav",
+ "render_seconds": 121.50087084504776,
+ "audio_seconds": 3.5526530612244898,
+ "rtf": 34.20003832380136,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "But I will continue to share with you two seconds.",
+ "raw": {
+ "normalized_reference": "but it won't continue, she said with easy confidence.",
+ "normalized_hypothesis": "but i will continue to share with you two seconds.",
+ "errors": 7,
+ "words": 9,
+ "wer": 0.7777777777777778
+ },
+ "semantic": {
+ "normalized_reference": "but it will not continue she said with easy confidence",
+ "normalized_hypothesis": "but i will continue to share with you 2 seconds",
+ "errors": 7,
+ "words": 10,
+ "wer": 0.7
+ }
+ },
+ {
+ "id": "release500_0431",
+ "category": "phonetic_balance",
+ "text": "I was still weak from my prolonged immersion.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0431.wav",
+ "render_seconds": 96.10618606605567,
+ "audio_seconds": 3.3436734693877552,
+ "rtf": 28.742694807353075,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I'll stay back for a period.",
+ "raw": {
+ "normalized_reference": "i was still weak from my prolonged immersion.",
+ "normalized_hypothesis": "i'll stay back for a period.",
+ "errors": 8,
+ "words": 8,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "i was still weak from my prolonged immersion",
+ "normalized_hypothesis": "i will stay back for a period",
+ "errors": 7,
+ "words": 8,
+ "wer": 0.875
+ }
+ },
+ {
+ "id": "release500_0432",
+ "category": "phonetic_balance",
+ "text": "Oolong was two hundred and fifty miles from the nearest land.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0432.wav",
+ "render_seconds": 126.4020027249353,
+ "audio_seconds": 4.249251700680272,
+ "rtf": 29.746885246806944,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Yeah, that's true. I got my first crush when I was young.",
+ "raw": {
+ "normalized_reference": "oolong was two hundred and fifty miles from the nearest land.",
+ "normalized_hypothesis": "yeah, that's true. i got my first crush when i was young.",
+ "errors": 12,
+ "words": 11,
+ "wer": 1.0909090909090908
+ },
+ "semantic": {
+ "normalized_reference": "oolong was 250 miles from the nearest land",
+ "normalized_hypothesis": "yeah that is true i got my 1st crush when i was young",
+ "errors": 13,
+ "words": 8,
+ "wer": 1.625
+ }
+ },
+ {
+ "id": "release500_0433",
+ "category": "phonetic_balance",
+ "text": "I just do appreciate it without being able to express my feelings.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0433.wav",
+ "render_seconds": 142.10778395505622,
+ "audio_seconds": 4.806530612244898,
+ "rtf": 29.565563068095088,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I just don't want to have it. Without that, I'll be too exposed to the silence.",
+ "raw": {
+ "normalized_reference": "i just do appreciate it without being able to express my feelings.",
+ "normalized_hypothesis": "i just don't want to have it. without that, i'll be too exposed to the silence.",
+ "errors": 12,
+ "words": 12,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "i just do appreciate it without being able to express my feelings",
+ "normalized_hypothesis": "i just do not want to have it without that i will be too exposed to the silence",
+ "errors": 13,
+ "words": 12,
+ "wer": 1.0833333333333333
+ }
+ },
+ {
+ "id": "release500_0434",
+ "category": "phonetic_balance",
+ "text": "Now just what do you want to know.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0434.wav",
+ "render_seconds": 76.20195149700157,
+ "audio_seconds": 2.36843537414966,
+ "rtf": 32.17396274718299,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I'll just look how it goes.",
+ "raw": {
+ "normalized_reference": "now just what do you want to know.",
+ "normalized_hypothesis": "i'll just look how it goes.",
+ "errors": 7,
+ "words": 8,
+ "wer": 0.875
+ },
+ "semantic": {
+ "normalized_reference": "now just what do you want to know",
+ "normalized_hypothesis": "i will just look how it goes",
+ "errors": 8,
+ "words": 8,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0435",
+ "category": "phonetic_balance",
+ "text": "But she swung obediently on her heel into the wind.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0435.wav",
+ "render_seconds": 129.58877530391328,
+ "audio_seconds": 3.4133333333333336,
+ "rtf": 37.965461514818344,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "But I swear to God that I know how much of the world...",
+ "raw": {
+ "normalized_reference": "but she swung obediently on her heel into the wind.",
+ "normalized_hypothesis": "but i swear to god that i know how much of the world...",
+ "errors": 11,
+ "words": 10,
+ "wer": 1.1
+ },
+ "semantic": {
+ "normalized_reference": "but she swung obediently on her heel into the wind",
+ "normalized_hypothesis": "but i swear to god that i know how much of the world .",
+ "errors": 11,
+ "words": 10,
+ "wer": 1.1
+ }
+ },
+ {
+ "id": "release500_0436",
+ "category": "phonetic_balance",
+ "text": "Ask him, I laughed, then turned to Pasquini.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0436.wav",
+ "render_seconds": 114.49201777507551,
+ "audio_seconds": 3.0650340136054424,
+ "rtf": 37.35424052942138,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Ask where I left and talk past them.",
+ "raw": {
+ "normalized_reference": "ask him, i laughed, then turned to pasquini.",
+ "normalized_hypothesis": "ask where i left and talk past them.",
+ "errors": 6,
+ "words": 8,
+ "wer": 0.75
+ },
+ "semantic": {
+ "normalized_reference": "ask him i laughed then turned to pasquini",
+ "normalized_hypothesis": "ask where i left and talk past them",
+ "errors": 6,
+ "words": 8,
+ "wer": 0.75
+ }
+ },
+ {
+ "id": "release500_0437",
+ "category": "phonetic_balance",
+ "text": "He did not rush in.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0437.wav",
+ "render_seconds": 82.10251486906782,
+ "audio_seconds": 1.7414965986394557,
+ "rtf": 47.144803459972536,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I'm a wolf.",
+ "raw": {
+ "normalized_reference": "he did not rush in.",
+ "normalized_hypothesis": "i'm a wolf.",
+ "errors": 5,
+ "words": 5,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "he did not rush in",
+ "normalized_hypothesis": "i am a wolf",
+ "errors": 5,
+ "words": 5,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0438",
+ "category": "phonetic_balance",
+ "text": "They obeyed him, and went here and there at his commands.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0438.wav",
+ "render_seconds": 116.09897710604127,
+ "audio_seconds": 3.900952380952381,
+ "rtf": 29.76170067415609,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I'm back. I'm back home. I'm back home. I'm back home. I'm back home. I'm back home. I'm back home. I'm back home. I'm back home. I'm back home. I'm back home. I'm back home. I'm back home. I'm back home. I'm back home. I'm back home. I'm back home. I'm back home. I'm back home. I'm back home. I'm back home. I'm back home. I'm back home. I'm back home. I'm back home. I'm back home. I'm back home. I'm back home. I'm back home. I'm back home. I'm back home. I'm back home. I'm back home. I'm back home. I'm back home. I'm back home. I'm back home. I'm back home. I'm back home. I'm back home. I'm back home. I'm back home. I'm back home. I'm back home. I'm back home. I'm back home. I'm back home. I'm back home. I'm back home. I'm back home. I'm back home. I'm back home. I'm back home. I'm back home. I'm back home. I'm back home. I'm back home. I'm back home. I'm back home. I'm back home. I'm back home. I'm back home. I'm back home. I'm back home. I'm back home. I'm back home. I'm back home. I'm back home. I'm back home. I'm back home. I'm back home. I'm back home. I'm back home. I'm back home. I'm back home. I'm back home. I'm back home. I'm back home. I'm back home. I'm back home. I'm back home. I'm back home. I'm back home. I'm back home. I'm back home. I'm back home. I'm back home. I'm back home. I'm back home.",
+ "raw": {
+ "normalized_reference": "they obeyed him, and went here and there at his commands.",
+ "normalized_hypothesis": "i'm back. i'm back home. i'm back home. i'm back home. i'm back home. i'm back home. i'm back home. i'm back home. i'm back home. i'm back home. i'm back home. i'm back home. i'm back home. i'm back home. i'm back home. i'm back home. i'm back home. i'm back home. i'm back home. i'm back home. i'm back home. i'm back home. i'm back home. i'm back home. i'm back home. i'm back home. i'm back home. i'm back home. i'm back home. i'm back home. i'm back home. i'm back home. i'm back home. i'm back home. i'm back home. i'm back home. i'm back home. i'm back home. i'm back home. i'm back home. i'm back home. i'm back home. i'm back home. i'm back home. i'm back home. i'm back home. i'm back home. i'm back home. i'm back home. i'm back home. i'm back home. i'm back home. i'm back home. i'm back home. i'm back home. i'm back home. i'm back home. i'm back home. i'm back home. i'm back home. i'm back home. i'm back home. i'm back home. i'm back home. i'm back home. i'm back home. i'm back home. i'm back home. i'm back home. i'm back home. i'm back home. i'm back home. i'm back home. i'm back home. i'm back home. i'm back home. i'm back home. i'm back home. i'm back home. i'm back home. i'm back home. i'm back home. i'm back home. i'm back home. i'm back home. i'm back home. i'm back home. i'm back home. i'm back home.",
+ "errors": 266,
+ "words": 11,
+ "wer": 24.181818181818183
+ },
+ "semantic": {
+ "normalized_reference": "they obeyed him and went here and there at his commands",
+ "normalized_hypothesis": "i am back i am back home i am back home i am back home i am back home i am back home i am back home i am back home i am back home i am back home i am back home i am back home i am back home i am back home i am back home i am back home i am back home i am back home i am back home i am back home i am back home i am back home i am back home i am back home i am back home i am back home i am back home i am back home i am back home i am back home i am back home i am back home i am back home i am back home i am back home i am back home i am back home i am back home i am back home i am back home i am back home i am back home i am back home i am back home i am back home i am back home i am back home i am back home i am back home i am back home i am back home i am back home i am back home i am back home i am back home i am back home i am back home i am back home i am back home i am back home i am back home i am back home i am back home i am back home i am back home i am back home i am back home i am back home i am back home i am back home i am back home i am back home i am back home i am back home i am back home i am back home i am back home i am back home i am back home i am back home i am back home i am back home i am back home i am back home i am back home i am back home i am back home i am back home i am back home",
+ "errors": 355,
+ "words": 11,
+ "wer": 32.27272727272727
+ }
+ },
+ {
+ "id": "release500_0439",
+ "category": "phonetic_balance",
+ "text": "Tom Spink has a harpoon.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0439.wav",
+ "render_seconds": 90.49203233397566,
+ "audio_seconds": 2.159455782312925,
+ "rtf": 41.905017493367005,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "TASTE OF HISTORY",
+ "raw": {
+ "normalized_reference": "tom spink has a harpoon.",
+ "normalized_hypothesis": "taste of history",
+ "errors": 5,
+ "words": 5,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "tom spink has a harpoon",
+ "normalized_hypothesis": "taste of history",
+ "errors": 5,
+ "words": 5,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0440",
+ "category": "phonetic_balance",
+ "text": "It was my reports from the north which chiefly induced people to buy.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0440.wav",
+ "render_seconds": 123.99391946999822,
+ "audio_seconds": 4.876190476190477,
+ "rtf": 25.428440516308225,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I wish you all the best for the years to come. I wish you all the best for the years to come. I wish you all the best for the years to come. I wish you all the best for the years to come. I wish you all the best for the years to come. I wish you all the best for the years to come. I wish you all the best for the years to come. I wish you all the best for the years to come. I wish you all the best for the years to come. I wish you all the best for the years to come. I wish you all the best for the years to come. I wish you all the best for the years to come. I wish you all the best for the years to come. I wish you all the best for the years to come. I wish you all the best for the years to come. I wish you all the best for the years to come. I wish you all the best for the years to come. I wish you all the best for the years to come. I wish you all the best for the years to come. I wish you all the best for the years to come. I wish you all the best for the years to come. I wish you all the best for the years to come. I wish you all the best for the years to come. I wish you all the best for the years to come. I wish you all the best for the years to come. I wish you all the best for the years to come. I wish you all the best for the years to come. I wish you all the best for the years to come. I wish you all the best for the years to come. I wish you all the best for the years to come. I wish you all the best for the years to come. I wish you all the best for the years to come. I wish you all the best for the years to come. I wish you all the best for the years to come. I wish you all the best for the years to come. I wish you all the best for the years to come. I wish you all the best for the years to come.",
+ "raw": {
+ "normalized_reference": "it was my reports from the north which chiefly induced people to buy.",
+ "normalized_hypothesis": "i wish you all the best for the years to come. i wish you all the best for the years to come. i wish you all the best for the years to come. i wish you all the best for the years to come. i wish you all the best for the years to come. i wish you all the best for the years to come. i wish you all the best for the years to come. i wish you all the best for the years to come. i wish you all the best for the years to come. i wish you all the best for the years to come. i wish you all the best for the years to come. i wish you all the best for the years to come. i wish you all the best for the years to come. i wish you all the best for the years to come. i wish you all the best for the years to come. i wish you all the best for the years to come. i wish you all the best for the years to come. i wish you all the best for the years to come. i wish you all the best for the years to come. i wish you all the best for the years to come. i wish you all the best for the years to come. i wish you all the best for the years to come. i wish you all the best for the years to come. i wish you all the best for the years to come. i wish you all the best for the years to come. i wish you all the best for the years to come. i wish you all the best for the years to come. i wish you all the best for the years to come. i wish you all the best for the years to come. i wish you all the best for the years to come. i wish you all the best for the years to come. i wish you all the best for the years to come. i wish you all the best for the years to come. i wish you all the best for the years to come. i wish you all the best for the years to come. i wish you all the best for the years to come. i wish you all the best for the years to come.",
+ "errors": 405,
+ "words": 13,
+ "wer": 31.153846153846153
+ },
+ "semantic": {
+ "normalized_reference": "it was my reports from the north which chiefly induced people to buy",
+ "normalized_hypothesis": "i wish you all the best for the years to come i wish you all the best for the years to come i wish you all the best for the years to come i wish you all the best for the years to come i wish you all the best for the years to come i wish you all the best for the years to come i wish you all the best for the years to come i wish you all the best for the years to come i wish you all the best for the years to come i wish you all the best for the years to come i wish you all the best for the years to come i wish you all the best for the years to come i wish you all the best for the years to come i wish you all the best for the years to come i wish you all the best for the years to come i wish you all the best for the years to come i wish you all the best for the years to come i wish you all the best for the years to come i wish you all the best for the years to come i wish you all the best for the years to come i wish you all the best for the years to come i wish you all the best for the years to come i wish you all the best for the years to come i wish you all the best for the years to come i wish you all the best for the years to come i wish you all the best for the years to come i wish you all the best for the years to come i wish you all the best for the years to come i wish you all the best for the years to come i wish you all the best for the years to come i wish you all the best for the years to come i wish you all the best for the years to come i wish you all the best for the years to come i wish you all the best for the years to come i wish you all the best for the years to come i wish you all the best for the years to come i wish you all the best for the years to come",
+ "errors": 405,
+ "words": 13,
+ "wer": 31.153846153846153
+ }
+ },
+ {
+ "id": "release500_0441",
+ "category": "phonetic_balance",
+ "text": "Cherokee identified himself with his instinct.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0441.wav",
+ "render_seconds": 116.09988292399794,
+ "audio_seconds": 3.2043537414965986,
+ "rtf": 36.2319182868285,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "How can I refer our shows with a hashtag?",
+ "raw": {
+ "normalized_reference": "cherokee identified himself with his instinct.",
+ "normalized_hypothesis": "how can i refer our shows with a hashtag?",
+ "errors": 8,
+ "words": 6,
+ "wer": 1.3333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "cherokee identified himself with his instinct",
+ "normalized_hypothesis": "how can i refer our shows with a hashtag",
+ "errors": 8,
+ "words": 6,
+ "wer": 1.3333333333333333
+ }
+ },
+ {
+ "id": "release500_0442",
+ "category": "phonetic_balance",
+ "text": "I graduated last of my class.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0442.wav",
+ "render_seconds": 89.70749325596262,
+ "audio_seconds": 2.4380952380952383,
+ "rtf": 36.79408903076592,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "God bless you. God bless you. God bless you.",
+ "raw": {
+ "normalized_reference": "i graduated last of my class.",
+ "normalized_hypothesis": "god bless you. god bless you. god bless you.",
+ "errors": 9,
+ "words": 6,
+ "wer": 1.5
+ },
+ "semantic": {
+ "normalized_reference": "i graduated last of my class",
+ "normalized_hypothesis": "god bless you god bless you god bless you",
+ "errors": 9,
+ "words": 6,
+ "wer": 1.5
+ }
+ },
+ {
+ "id": "release500_0443",
+ "category": "phonetic_balance",
+ "text": "It was a gigantic inadequacy.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0443.wav",
+ "render_seconds": 100.29291518195532,
+ "audio_seconds": 2.2987755102040817,
+ "rtf": 43.628842719423034,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I wish I could have talked to all of you, sir.",
+ "raw": {
+ "normalized_reference": "it was a gigantic inadequacy.",
+ "normalized_hypothesis": "i wish i could have talked to all of you, sir.",
+ "errors": 11,
+ "words": 5,
+ "wer": 2.2
+ },
+ "semantic": {
+ "normalized_reference": "it was a gigantic inadequacy",
+ "normalized_hypothesis": "i wish i could have talked to all of you sir",
+ "errors": 11,
+ "words": 5,
+ "wer": 2.2
+ }
+ },
+ {
+ "id": "release500_0444",
+ "category": "phonetic_balance",
+ "text": "Did I possess too much vitality.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0444.wav",
+ "render_seconds": 74.99495326098986,
+ "audio_seconds": 2.507755102040816,
+ "rtf": 29.90521410960696,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I want this to work for me.",
+ "raw": {
+ "normalized_reference": "did i possess too much vitality.",
+ "normalized_hypothesis": "i want this to work for me.",
+ "errors": 7,
+ "words": 6,
+ "wer": 1.1666666666666667
+ },
+ "semantic": {
+ "normalized_reference": "did i possess too much vitality",
+ "normalized_hypothesis": "i want this to work for me",
+ "errors": 7,
+ "words": 6,
+ "wer": 1.1666666666666667
+ }
+ },
+ {
+ "id": "release500_0445",
+ "category": "phonetic_balance",
+ "text": "Therefore, hurrah for the game.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0445.wav",
+ "render_seconds": 82.30921019101515,
+ "audio_seconds": 2.2291156462585033,
+ "rtf": 36.92460296044686,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I feel whole for the hour.",
+ "raw": {
+ "normalized_reference": "therefore, hurrah for the game.",
+ "normalized_hypothesis": "i feel whole for the hour.",
+ "errors": 4,
+ "words": 5,
+ "wer": 0.8
+ },
+ "semantic": {
+ "normalized_reference": "therefore hurrah for the game",
+ "normalized_hypothesis": "i feel whole for the hour",
+ "errors": 4,
+ "words": 5,
+ "wer": 0.8
+ }
+ },
+ {
+ "id": "release500_0446",
+ "category": "phonetic_balance",
+ "text": "He had fulfilled his duty and paid properly.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0446.wav",
+ "render_seconds": 115.70159818802495,
+ "audio_seconds": 3.1346938775510202,
+ "rtf": 36.910015046961085,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Oxytocin has turned out that popular.",
+ "raw": {
+ "normalized_reference": "he had fulfilled his duty and paid properly.",
+ "normalized_hypothesis": "oxytocin has turned out that popular.",
+ "errors": 8,
+ "words": 8,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "he had fulfilled his duty and paid properly",
+ "normalized_hypothesis": "oxytocin has turned out that popular",
+ "errors": 8,
+ "words": 8,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0447",
+ "category": "phonetic_balance",
+ "text": "How can you manage all alone, Mr Young.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0447.wav",
+ "render_seconds": 126.40151064202655,
+ "audio_seconds": 2.8560544217687074,
+ "rtf": 44.257388682302555,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": ".......................................................................................................................................................................................................................................................................................................................................................................................",
+ "raw": {
+ "normalized_reference": "how can you manage all alone, mr young.",
+ "normalized_hypothesis": ".......................................................................................................................................................................................................................................................................................................................................................................................",
+ "errors": 8,
+ "words": 8,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "how can you manage all alone mister young",
+ "normalized_hypothesis": " .",
+ "errors": 8,
+ "words": 8,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0448",
+ "category": "phonetic_balance",
+ "text": "Red-Eye never committed a more outrageous deed.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0448.wav",
+ "render_seconds": 104.1919863189105,
+ "audio_seconds": 3.5526530612244898,
+ "rtf": 29.327937325520534,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I love you. I love you. I love you. I love you. I love you. I love you.",
+ "raw": {
+ "normalized_reference": "red-eye never committed a more outrageous deed.",
+ "normalized_hypothesis": "i love you. i love you. i love you. i love you. i love you. i love you.",
+ "errors": 18,
+ "words": 8,
+ "wer": 2.25
+ },
+ "semantic": {
+ "normalized_reference": "red eye never committed a more outrageous deed",
+ "normalized_hypothesis": "i love you i love you i love you i love you i love you i love you",
+ "errors": 18,
+ "words": 8,
+ "wer": 2.25
+ }
+ },
+ {
+ "id": "release500_0449",
+ "category": "phonetic_balance",
+ "text": "He would keep his faith with Gregson for the promised day or two.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0449.wav",
+ "render_seconds": 114.20582774409559,
+ "audio_seconds": 4.597551020408163,
+ "rtf": 24.840578655276474,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I will cut this off with action for the poor style of tone.",
+ "raw": {
+ "normalized_reference": "he would keep his faith with gregson for the promised day or two.",
+ "normalized_hypothesis": "i will cut this off with action for the poor style of tone.",
+ "errors": 10,
+ "words": 13,
+ "wer": 0.7692307692307693
+ },
+ "semantic": {
+ "normalized_reference": "he would keep his faith with gregson for the promised day or 2",
+ "normalized_hypothesis": "i will cut this off with action for the poor style of tone",
+ "errors": 10,
+ "words": 13,
+ "wer": 0.7692307692307693
+ }
+ },
+ {
+ "id": "release500_0450",
+ "category": "phonetic_balance",
+ "text": "There was one chance, and only one, of saving Jeanne.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0450.wav",
+ "render_seconds": 77.39515878504608,
+ "audio_seconds": 3.8312925170068026,
+ "rtf": 20.20079606072758,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I wish you luck. I wish you luck. I wish you luck. I wish you luck. I wish you luck. I wish you luck. I wish you luck. I wish you luck. I wish you luck. I wish you luck. I wish you luck. I wish you luck. I wish you luck. I wish you luck. I wish you luck. I wish you luck. I wish you luck. I wish you luck. I wish you luck. I wish you luck. I wish you luck. I wish you luck. I wish you luck. I wish you luck. I wish you luck. I wish you luck. I wish you luck. I wish you luck. I wish you luck. I wish you luck. I wish you luck. I wish you luck. I wish you luck. I wish you luck. I wish you luck. I wish you luck. I wish you luck. I wish you luck. I wish you luck. I wish you luck. I wish you luck. I wish you luck. I wish you luck. I wish you luck. I wish you luck. I wish you luck. I wish you luck. I wish you luck. I wish you luck. I wish you luck. I wish you luck. I wish you luck. I wish you luck. I wish you luck. I wish you luck. I wish you luck. I wish you luck. I wish you luck. I wish you luck. I wish you luck. I wish you luck. I wish you luck. I wish you luck. I wish you luck. I wish you luck. I wish you luck. I wish you luck. I wish you luck. I wish you luck. I wish you luck. I wish you luck. I wish you luck. I wish you luck. I wish you luck. I wish you luck. I wish you luck. I wish you luck. I wish you luck. I wish you luck. I wish you luck. I wish you luck. I wish you luck. I wish you luck. I wish you luck. I wish you luck. I wish you luck. I wish you luck. I wish you luck. I wish you luck",
+ "raw": {
+ "normalized_reference": "there was one chance, and only one, of saving jeanne.",
+ "normalized_hypothesis": "i wish you luck. i wish you luck. i wish you luck. i wish you luck. i wish you luck. i wish you luck. i wish you luck. i wish you luck. i wish you luck. i wish you luck. i wish you luck. i wish you luck. i wish you luck. i wish you luck. i wish you luck. i wish you luck. i wish you luck. i wish you luck. i wish you luck. i wish you luck. i wish you luck. i wish you luck. i wish you luck. i wish you luck. i wish you luck. i wish you luck. i wish you luck. i wish you luck. i wish you luck. i wish you luck. i wish you luck. i wish you luck. i wish you luck. i wish you luck. i wish you luck. i wish you luck. i wish you luck. i wish you luck. i wish you luck. i wish you luck. i wish you luck. i wish you luck. i wish you luck. i wish you luck. i wish you luck. i wish you luck. i wish you luck. i wish you luck. i wish you luck. i wish you luck. i wish you luck. i wish you luck. i wish you luck. i wish you luck. i wish you luck. i wish you luck. i wish you luck. i wish you luck. i wish you luck. i wish you luck. i wish you luck. i wish you luck. i wish you luck. i wish you luck. i wish you luck. i wish you luck. i wish you luck. i wish you luck. i wish you luck. i wish you luck. i wish you luck. i wish you luck. i wish you luck. i wish you luck. i wish you luck. i wish you luck. i wish you luck. i wish you luck. i wish you luck. i wish you luck. i wish you luck. i wish you luck. i wish you luck. i wish you luck. i wish you luck. i wish you luck. i wish you luck. i wish you luck. i wish you luck",
+ "errors": 356,
+ "words": 10,
+ "wer": 35.6
+ },
+ "semantic": {
+ "normalized_reference": "there was one chance and only one of saving jeanne",
+ "normalized_hypothesis": "i wish you luck i wish you luck i wish you luck i wish you luck i wish you luck i wish you luck i wish you luck i wish you luck i wish you luck i wish you luck i wish you luck i wish you luck i wish you luck i wish you luck i wish you luck i wish you luck i wish you luck i wish you luck i wish you luck i wish you luck i wish you luck i wish you luck i wish you luck i wish you luck i wish you luck i wish you luck i wish you luck i wish you luck i wish you luck i wish you luck i wish you luck i wish you luck i wish you luck i wish you luck i wish you luck i wish you luck i wish you luck i wish you luck i wish you luck i wish you luck i wish you luck i wish you luck i wish you luck i wish you luck i wish you luck i wish you luck i wish you luck i wish you luck i wish you luck i wish you luck i wish you luck i wish you luck i wish you luck i wish you luck i wish you luck i wish you luck i wish you luck i wish you luck i wish you luck i wish you luck i wish you luck i wish you luck i wish you luck i wish you luck i wish you luck i wish you luck i wish you luck i wish you luck i wish you luck i wish you luck i wish you luck i wish you luck i wish you luck i wish you luck i wish you luck i wish you luck i wish you luck i wish you luck i wish you luck i wish you luck i wish you luck i wish you luck i wish you luck i wish you luck i wish you luck i wish you luck i wish you luck i wish you luck i wish you luck",
+ "errors": 356,
+ "words": 10,
+ "wer": 35.6
+ }
+ },
+ {
+ "id": "release500_0451",
+ "category": "phonetic_balance",
+ "text": "I was beneath the water, suffocating and drowning.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0451.wav",
+ "render_seconds": 133.20235759101342,
+ "audio_seconds": 3.4829931972789114,
+ "rtf": 38.24364563648237,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I wish the last of you should be crying and crying.",
+ "raw": {
+ "normalized_reference": "i was beneath the water, suffocating and drowning.",
+ "normalized_hypothesis": "i wish the last of you should be crying and crying.",
+ "errors": 9,
+ "words": 8,
+ "wer": 1.125
+ },
+ "semantic": {
+ "normalized_reference": "i was beneath the water suffocating and drowning",
+ "normalized_hypothesis": "i wish the last of you should be crying and crying",
+ "errors": 9,
+ "words": 8,
+ "wer": 1.125
+ }
+ },
+ {
+ "id": "release500_0452",
+ "category": "phonetic_balance",
+ "text": "Give them their choice between a fine or an official whipping.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0452.wav",
+ "render_seconds": 118.09779246896505,
+ "audio_seconds": 4.388571428571429,
+ "rtf": 26.910304273527192,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I don't know what you're expecting from me, sir. You're a sexy weapon.",
+ "raw": {
+ "normalized_reference": "give them their choice between a fine or an official whipping.",
+ "normalized_hypothesis": "i don't know what you're expecting from me, sir. you're a sexy weapon.",
+ "errors": 13,
+ "words": 11,
+ "wer": 1.1818181818181819
+ },
+ "semantic": {
+ "normalized_reference": "give them their choice between a fine or an official whipping",
+ "normalized_hypothesis": "i do not know what you are expecting from me sir you are a sexy weapon",
+ "errors": 16,
+ "words": 11,
+ "wer": 1.4545454545454546
+ }
+ },
+ {
+ "id": "release500_0453",
+ "category": "phonetic_balance",
+ "text": "They must have been swept away by the chaotic currents.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0453.wav",
+ "render_seconds": 134.59763296099845,
+ "audio_seconds": 3.900952380952381,
+ "rtf": 34.5037877463497,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I wish to have been stripped alive of a pair of horns.",
+ "raw": {
+ "normalized_reference": "they must have been swept away by the chaotic currents.",
+ "normalized_hypothesis": "i wish to have been stripped alive of a pair of horns.",
+ "errors": 10,
+ "words": 10,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "they must have been swept away by the chaotic currents",
+ "normalized_hypothesis": "i wish to have been stripped alive of a pair of horns",
+ "errors": 10,
+ "words": 10,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0454",
+ "category": "phonetic_balance",
+ "text": "You yellow giant thing of the frost.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0454.wav",
+ "render_seconds": 50.7071281389799,
+ "audio_seconds": 2.4380952380952383,
+ "rtf": 20.79784552575347,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I will try to have the first.",
+ "raw": {
+ "normalized_reference": "you yellow giant thing of the frost.",
+ "normalized_hypothesis": "i will try to have the first.",
+ "errors": 6,
+ "words": 7,
+ "wer": 0.8571428571428571
+ },
+ "semantic": {
+ "normalized_reference": "you yellow giant thing of the frost",
+ "normalized_hypothesis": "i will try to have the 1st",
+ "errors": 6,
+ "words": 7,
+ "wer": 0.8571428571428571
+ }
+ },
+ {
+ "id": "release500_0455",
+ "category": "phonetic_balance",
+ "text": "He had peeled off his shirt and was wildly waving it.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0455.wav",
+ "render_seconds": 113.69861823902465,
+ "audio_seconds": 3.6919727891156464,
+ "rtf": 30.79616907638676,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I can't believe the show. I wish you all the love I know.",
+ "raw": {
+ "normalized_reference": "he had peeled off his shirt and was wildly waving it.",
+ "normalized_hypothesis": "i can't believe the show. i wish you all the love i know.",
+ "errors": 13,
+ "words": 11,
+ "wer": 1.1818181818181819
+ },
+ "semantic": {
+ "normalized_reference": "he had peeled off his shirt and was wildly waving it",
+ "normalized_hypothesis": "i can not believe the show i wish you all the love i know",
+ "errors": 14,
+ "words": 11,
+ "wer": 1.2727272727272727
+ }
+ },
+ {
+ "id": "release500_0456",
+ "category": "phonetic_balance",
+ "text": "To these he gave castor oil.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0456.wav",
+ "render_seconds": 45.00009410397615,
+ "audio_seconds": 2.2291156462585033,
+ "rtf": 20.18742014552153,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Terrasse of Casteleon.",
+ "raw": {
+ "normalized_reference": "to these he gave castor oil.",
+ "normalized_hypothesis": "terrasse of casteleon.",
+ "errors": 6,
+ "words": 6,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "to these he gave castor oil",
+ "normalized_hypothesis": "terrasse of casteleon",
+ "errors": 6,
+ "words": 6,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0457",
+ "category": "phonetic_balance",
+ "text": "Tudor surveyed him with withering disgust.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0457.wav",
+ "render_seconds": 115.29798547306564,
+ "audio_seconds": 2.9257142857142857,
+ "rtf": 39.40849112848923,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Oh, so glad you look little and disposed.",
+ "raw": {
+ "normalized_reference": "tudor surveyed him with withering disgust.",
+ "normalized_hypothesis": "oh, so glad you look little and disposed.",
+ "errors": 8,
+ "words": 6,
+ "wer": 1.3333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "tudor surveyed him with withering disgust",
+ "normalized_hypothesis": "0 so glad you look little and disposed",
+ "errors": 8,
+ "words": 6,
+ "wer": 1.3333333333333333
+ }
+ },
+ {
+ "id": "release500_0458",
+ "category": "phonetic_balance",
+ "text": "After all, it was simply a mistake in judgment.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0458.wav",
+ "render_seconds": 82.90516074304469,
+ "audio_seconds": 3.4829931972789114,
+ "rtf": 23.802848885210096,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Of course, I will show my respect and courage to it.",
+ "raw": {
+ "normalized_reference": "after all, it was simply a mistake in judgment.",
+ "normalized_hypothesis": "of course, i will show my respect and courage to it.",
+ "errors": 11,
+ "words": 9,
+ "wer": 1.2222222222222223
+ },
+ "semantic": {
+ "normalized_reference": "after all it was simply a mistake in judgment",
+ "normalized_hypothesis": "of course i will show my respect and courage to it",
+ "errors": 11,
+ "words": 9,
+ "wer": 1.2222222222222223
+ }
+ },
+ {
+ "id": "release500_0459",
+ "category": "phonetic_balance",
+ "text": "That is the strange part of it.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0459.wav",
+ "render_seconds": 114.89998027298134,
+ "audio_seconds": 2.2987755102040817,
+ "rtf": 49.9831235207394,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this? What is this?",
+ "raw": {
+ "normalized_reference": "that is the strange part of it.",
+ "normalized_hypothesis": "what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this? what is this?",
+ "errors": 332,
+ "words": 7,
+ "wer": 47.42857142857143
+ },
+ "semantic": {
+ "normalized_reference": "that is the strange part of it",
+ "normalized_hypothesis": "what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this what is this",
+ "errors": 332,
+ "words": 7,
+ "wer": 47.42857142857143
+ }
+ },
+ {
+ "id": "release500_0460",
+ "category": "phonetic_balance",
+ "text": "Not a wheel moved in his empire.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0460.wav",
+ "render_seconds": 62.79256058996543,
+ "audio_seconds": 2.507755102040816,
+ "rtf": 25.039351146714733,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I'm a dying, loose, tiny, small pig.",
+ "raw": {
+ "normalized_reference": "not a wheel moved in his empire.",
+ "normalized_hypothesis": "i'm a dying, loose, tiny, small pig.",
+ "errors": 6,
+ "words": 7,
+ "wer": 0.8571428571428571
+ },
+ "semantic": {
+ "normalized_reference": "not a wheel moved in his empire",
+ "normalized_hypothesis": "i am a dying loose tiny small pig",
+ "errors": 7,
+ "words": 7,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0461",
+ "category": "phonetic_balance",
+ "text": "Its diameter was not more than two hundred yards.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0461.wav",
+ "render_seconds": 92.69938166497741,
+ "audio_seconds": 3.4133333333333336,
+ "rtf": 27.15802197216135,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It's still a long way to go before we get to the end of the course.",
+ "raw": {
+ "normalized_reference": "its diameter was not more than two hundred yards.",
+ "normalized_hypothesis": "it's still a long way to go before we get to the end of the course.",
+ "errors": 16,
+ "words": 9,
+ "wer": 1.7777777777777777
+ },
+ "semantic": {
+ "normalized_reference": "its diameter was not more than 200 yards",
+ "normalized_hypothesis": "it is still a long way to go before we get to the end of the course",
+ "errors": 17,
+ "words": 8,
+ "wer": 2.125
+ }
+ },
+ {
+ "id": "release500_0462",
+ "category": "phonetic_balance",
+ "text": "Nimrod replied, with a slight manifestation of sensitiveness.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0462.wav",
+ "render_seconds": 82.49773488100618,
+ "audio_seconds": 4.667210884353741,
+ "rtf": 17.676024701941333,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I will protect with the strength of the stars and the suns of the moon.",
+ "raw": {
+ "normalized_reference": "nimrod replied, with a slight manifestation of sensitiveness.",
+ "normalized_hypothesis": "i will protect with the strength of the stars and the suns of the moon.",
+ "errors": 13,
+ "words": 8,
+ "wer": 1.625
+ },
+ "semantic": {
+ "normalized_reference": "nimrod replied with a slight manifestation of sensitiveness",
+ "normalized_hypothesis": "i will protect with the strength of the stars and the suns of the moon",
+ "errors": 13,
+ "words": 8,
+ "wer": 1.625
+ }
+ },
+ {
+ "id": "release500_0463",
+ "category": "phonetic_balance",
+ "text": "I tried to read George Moore last night, and was dreadfully bored.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0463.wav",
+ "render_seconds": 156.9069790840149,
+ "audio_seconds": 4.667210884353741,
+ "rtf": 33.61900350593253,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I can't do it. I can't do it. I can't do it. I can't do it.",
+ "raw": {
+ "normalized_reference": "i tried to read george moore last night, and was dreadfully bored.",
+ "normalized_hypothesis": "i can't do it. i can't do it. i can't do it. i can't do it.",
+ "errors": 15,
+ "words": 12,
+ "wer": 1.25
+ },
+ "semantic": {
+ "normalized_reference": "i tried to read george moore last night and was dreadfully bored",
+ "normalized_hypothesis": "i can not do it i can not do it i can not do it i can not do it",
+ "errors": 19,
+ "words": 12,
+ "wer": 1.5833333333333333
+ }
+ },
+ {
+ "id": "release500_0464",
+ "category": "phonetic_balance",
+ "text": "Yes, it was a man who asked, a stranger.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0464.wav",
+ "render_seconds": 88.30820780899376,
+ "audio_seconds": 2.995374149659864,
+ "rtf": 29.48152831559339,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I'll see you next time.",
+ "raw": {
+ "normalized_reference": "yes, it was a man who asked, a stranger.",
+ "normalized_hypothesis": "i'll see you next time.",
+ "errors": 9,
+ "words": 9,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "yes it was a man who asked a stranger",
+ "normalized_hypothesis": "i will see you next time",
+ "errors": 9,
+ "words": 9,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0465",
+ "category": "phonetic_balance",
+ "text": "We were now good friends.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0465.wav",
+ "render_seconds": 81.29997687402647,
+ "audio_seconds": 2.020136054421769,
+ "rtf": 40.24480266864861,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you.",
+ "raw": {
+ "normalized_reference": "we were now good friends.",
+ "normalized_hypothesis": "i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you.",
+ "errors": 333,
+ "words": 5,
+ "wer": 66.6
+ },
+ "semantic": {
+ "normalized_reference": "we were now good friends",
+ "normalized_hypothesis": "i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you",
+ "errors": 333,
+ "words": 5,
+ "wer": 66.6
+ }
+ },
+ {
+ "id": "release500_0466",
+ "category": "phonetic_balance",
+ "text": "Yet, in accordance with Ernest's test of truth, it worked.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0466.wav",
+ "render_seconds": 67.91129141498823,
+ "audio_seconds": 3.900952380952381,
+ "rtf": 17.408900387142978,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I'm not sure I should always taste of coffee at work.",
+ "raw": {
+ "normalized_reference": "yet, in accordance with ernest's test of truth, it worked.",
+ "normalized_hypothesis": "i'm not sure i should always taste of coffee at work.",
+ "errors": 10,
+ "words": 10,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "yet in accordance with ernest is test of truth it worked",
+ "normalized_hypothesis": "i am not sure i should always taste of coffee at work",
+ "errors": 11,
+ "words": 11,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0467",
+ "category": "phonetic_balance",
+ "text": "It's the strap hangers that'll keep us from going under.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0467.wav",
+ "render_seconds": 99.0920732569648,
+ "audio_seconds": 3.4829931972789114,
+ "rtf": 28.450263220261377,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "It's a strong power so it can push you to the end.",
+ "raw": {
+ "normalized_reference": "it's the strap hangers that'll keep us from going under.",
+ "normalized_hypothesis": "it's a strong power so it can push you to the end.",
+ "errors": 11,
+ "words": 10,
+ "wer": 1.1
+ },
+ "semantic": {
+ "normalized_reference": "it is the strap hangers that will keep us from going under",
+ "normalized_hypothesis": "it is a strong power so it can push you to the end",
+ "errors": 11,
+ "words": 12,
+ "wer": 0.9166666666666666
+ }
+ },
+ {
+ "id": "release500_0468",
+ "category": "phonetic_balance",
+ "text": "I never allow what can't be changed to annoy me.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0468.wav",
+ "render_seconds": 64.5002737219911,
+ "audio_seconds": 3.0650340136054424,
+ "rtf": 21.043901449602032,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you. I love you.",
+ "raw": {
+ "normalized_reference": "i never allow what can't be changed to annoy me.",
+ "normalized_hypothesis": "i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you. i love you.",
+ "errors": 332,
+ "words": 10,
+ "wer": 33.2
+ },
+ "semantic": {
+ "normalized_reference": "i never allow what can not be changed to annoy me",
+ "normalized_hypothesis": "i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you i love you",
+ "errors": 332,
+ "words": 11,
+ "wer": 30.181818181818183
+ }
+ },
+ {
+ "id": "release500_0469",
+ "category": "phonetic_balance",
+ "text": "After all superfluous flesh is gone what is left is stringy and resistant.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0469.wav",
+ "render_seconds": 163.29407874599565,
+ "audio_seconds": 5.224489795918367,
+ "rtf": 31.255507259975733,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "After each and both years, yes, I still, yes, yes, I still resisted.",
+ "raw": {
+ "normalized_reference": "after all superfluous flesh is gone what is left is stringy and resistant.",
+ "normalized_hypothesis": "after each and both years, yes, i still, yes, yes, i still resisted.",
+ "errors": 12,
+ "words": 13,
+ "wer": 0.9230769230769231
+ },
+ "semantic": {
+ "normalized_reference": "after all superfluous flesh is gone what is left is stringy and resistant",
+ "normalized_hypothesis": "after each and both years yes i still yes yes i still resisted",
+ "errors": 12,
+ "words": 13,
+ "wer": 0.9230769230769231
+ }
+ },
+ {
+ "id": "release500_0470",
+ "category": "phonetic_balance",
+ "text": "Then he shouted, Shut up.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0470.wav",
+ "render_seconds": 45.892153688007966,
+ "audio_seconds": 2.089795918367347,
+ "rtf": 21.960112604613183,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Thank you.",
+ "raw": {
+ "normalized_reference": "then he shouted, shut up.",
+ "normalized_hypothesis": "thank you.",
+ "errors": 5,
+ "words": 5,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "then he shouted shut up",
+ "normalized_hypothesis": "thank you",
+ "errors": 5,
+ "words": 5,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0471",
+ "category": "phonetic_balance",
+ "text": "He had worshiped her, as Dante might have worshiped Beatrice.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0471.wav",
+ "render_seconds": 133.90137863601558,
+ "audio_seconds": 4.458231292517007,
+ "rtf": 30.034641509238114,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Hello, Haptor. It's time to run off your hook, Bacchus.",
+ "raw": {
+ "normalized_reference": "he had worshiped her, as dante might have worshiped beatrice.",
+ "normalized_hypothesis": "hello, haptor. it's time to run off your hook, bacchus.",
+ "errors": 10,
+ "words": 10,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "he had worshiped her as dante might have worshiped beatrice",
+ "normalized_hypothesis": "hello haptor it is time to run off your hook bacchus",
+ "errors": 11,
+ "words": 10,
+ "wer": 1.1
+ }
+ },
+ {
+ "id": "release500_0472",
+ "category": "phonetic_balance",
+ "text": "He looked like one who had passed through an uncomfortable hour or two.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0472.wav",
+ "render_seconds": 83.80051923799329,
+ "audio_seconds": 4.876190476190477,
+ "rtf": 17.18565335935409,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I don't care who I am. I'm just trying to be a little child.",
+ "raw": {
+ "normalized_reference": "he looked like one who had passed through an uncomfortable hour or two.",
+ "normalized_hypothesis": "i don't care who i am. i'm just trying to be a little child.",
+ "errors": 14,
+ "words": 13,
+ "wer": 1.0769230769230769
+ },
+ "semantic": {
+ "normalized_reference": "he looked like one who had passed through an uncomfortable hour or 2",
+ "normalized_hypothesis": "i do not care who i am i am just trying to be a little child",
+ "errors": 15,
+ "words": 13,
+ "wer": 1.1538461538461537
+ }
+ },
+ {
+ "id": "release500_0473",
+ "category": "phonetic_balance",
+ "text": "The history of our westward-faring race is written in it.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0473.wav",
+ "render_seconds": 139.7089644210646,
+ "audio_seconds": 4.179591836734694,
+ "rtf": 33.42646121402424,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "The hush cry, though, must look far less as what ran up.",
+ "raw": {
+ "normalized_reference": "the history of our westward-faring race is written in it.",
+ "normalized_hypothesis": "the hush cry, though, must look far less as what ran up.",
+ "errors": 11,
+ "words": 11,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "the history of our westward faring race is written in it",
+ "normalized_hypothesis": "the hush cry though must look far less as what ran up",
+ "errors": 11,
+ "words": 11,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0474",
+ "category": "phonetic_balance",
+ "text": "He may desire to escape pain, or to enjoy pleasure.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0474.wav",
+ "render_seconds": 76.1070360239828,
+ "audio_seconds": 3.8312925170068026,
+ "rtf": 19.86458504177108,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I want to go to the sky. I want to go to the sky.",
+ "raw": {
+ "normalized_reference": "he may desire to escape pain, or to enjoy pleasure.",
+ "normalized_hypothesis": "i want to go to the sky. i want to go to the sky.",
+ "errors": 12,
+ "words": 10,
+ "wer": 1.2
+ },
+ "semantic": {
+ "normalized_reference": "he may desire to escape pain or to enjoy pleasure",
+ "normalized_hypothesis": "i want to go to the sky i want to go to the sky",
+ "errors": 12,
+ "words": 10,
+ "wer": 1.2
+ }
+ },
+ {
+ "id": "release500_0475",
+ "category": "phonetic_balance",
+ "text": "And the air was growing chilly.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0475.wav",
+ "render_seconds": 101.69825918204151,
+ "audio_seconds": 2.2987755102040817,
+ "rtf": 44.24018732173326,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "And I was cool and chill.",
+ "raw": {
+ "normalized_reference": "and the air was growing chilly.",
+ "normalized_hypothesis": "and i was cool and chill.",
+ "errors": 5,
+ "words": 6,
+ "wer": 0.8333333333333334
+ },
+ "semantic": {
+ "normalized_reference": "and the air was growing chilly",
+ "normalized_hypothesis": "and i was cool and chill",
+ "errors": 5,
+ "words": 6,
+ "wer": 0.8333333333333334
+ }
+ },
+ {
+ "id": "release500_0476",
+ "category": "phonetic_balance",
+ "text": "Dig in; you're sure good, was Daylight's answer.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0476.wav",
+ "render_seconds": 117.79247821995523,
+ "audio_seconds": 2.9257142857142857,
+ "rtf": 40.26110095408626,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Bye-bye, I hope that was time-lapsing.",
+ "raw": {
+ "normalized_reference": "dig in; you're sure good, was daylight's answer.",
+ "normalized_hypothesis": "bye-bye, i hope that was time-lapsing.",
+ "errors": 7,
+ "words": 8,
+ "wer": 0.875
+ },
+ "semantic": {
+ "normalized_reference": "dig in you are sure good was daylight is answer",
+ "normalized_hypothesis": "bye bye i hope that was time lapsing",
+ "errors": 9,
+ "words": 10,
+ "wer": 0.9
+ }
+ },
+ {
+ "id": "release500_0477",
+ "category": "phonetic_balance",
+ "text": "My idea was that he would have more influence over the natives.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0477.wav",
+ "render_seconds": 115.8983665339183,
+ "audio_seconds": 4.458231292517007,
+ "rtf": 25.99649029615172,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I wish that I were able to ask you a truth or a lie.",
+ "raw": {
+ "normalized_reference": "my idea was that he would have more influence over the natives.",
+ "normalized_hypothesis": "i wish that i were able to ask you a truth or a lie.",
+ "errors": 14,
+ "words": 12,
+ "wer": 1.1666666666666667
+ },
+ "semantic": {
+ "normalized_reference": "my idea was that he would have more influence over the natives",
+ "normalized_hypothesis": "i wish that i were able to ask you a truth or a lie",
+ "errors": 14,
+ "words": 12,
+ "wer": 1.1666666666666667
+ }
+ },
+ {
+ "id": "release500_0478",
+ "category": "phonetic_balance",
+ "text": "Now run along, and tell them to hurry.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0478.wav",
+ "render_seconds": 58.08695649402216,
+ "audio_seconds": 2.786394557823129,
+ "rtf": 20.846637218313617,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Hey, what are you doing? I'm taking you home.",
+ "raw": {
+ "normalized_reference": "now run along, and tell them to hurry.",
+ "normalized_hypothesis": "hey, what are you doing? i'm taking you home.",
+ "errors": 9,
+ "words": 8,
+ "wer": 1.125
+ },
+ "semantic": {
+ "normalized_reference": "now run along and tell them to hurry",
+ "normalized_hypothesis": "hey what are you doing i am taking you home",
+ "errors": 10,
+ "words": 8,
+ "wer": 1.25
+ }
+ },
+ {
+ "id": "release500_0479",
+ "category": "phonetic_balance",
+ "text": "I have no idea, replied Philip.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0479.wav",
+ "render_seconds": 107.19098688301165,
+ "audio_seconds": 2.36843537414966,
+ "rtf": 45.25814301413922,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Thank you for watching.",
+ "raw": {
+ "normalized_reference": "i have no idea, replied philip.",
+ "normalized_hypothesis": "thank you for watching.",
+ "errors": 6,
+ "words": 6,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "i have no idea replied philip",
+ "normalized_hypothesis": "thank you for watching",
+ "errors": 6,
+ "words": 6,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0480",
+ "category": "phonetic_balance",
+ "text": "For the rest, he was a mere automaton.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0480.wav",
+ "render_seconds": 47.993807561928406,
+ "audio_seconds": 2.786394557823129,
+ "rtf": 17.224340116219423,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "For the worst, I wish you all the latter.",
+ "raw": {
+ "normalized_reference": "for the rest, he was a mere automaton.",
+ "normalized_hypothesis": "for the worst, i wish you all the latter.",
+ "errors": 7,
+ "words": 8,
+ "wer": 0.875
+ },
+ "semantic": {
+ "normalized_reference": "for the rest he was a mere automaton",
+ "normalized_hypothesis": "for the worst i wish you all the latter",
+ "errors": 7,
+ "words": 8,
+ "wer": 0.875
+ }
+ },
+ {
+ "id": "release500_0481",
+ "category": "phonetic_balance",
+ "text": "Why, the average review is more nauseating than cod liver oil.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0481.wav",
+ "render_seconds": 139.5997976079816,
+ "audio_seconds": 4.388571428571429,
+ "rtf": 31.80984971536039,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "You're out of your house. You're out of your house. You're out of your house. You're out of your house. You're out of your house. You're out of your house. You're out of your house. You're out of your house. You're out of your house. You're out of your house. You're out of your house. You're out of your house. You're out of your house. You're out of your house. You're out of your house. You're out of your house. You're out of your house. You're out of your house. You're out of your house. You're out of your house. You're out of your house. You're out of your house. You're out of your house. You're out of your house. You're out of your house. You're out of your house. You're out of your house. You're out of your house. You're out of your house. You're out of your house. You're out of your house. You're out of your house. You're out of your house. You're out of your house. You're out of your house. You're out of your house. You're out of your house. You're out of your house. You're out of your house. You're out of your house. You're out of your house. You're out of your house. You're out of your house. You're out of your house. You're out of your house. You're out of your house. You're out of your house. You're out of your house. You're out of your house. You're out of your house. You're out of your house. You're out of your house. You're out of your house. You're out of your house. You're out of your house. You're out of your house. You're out of your house. You're out of your house. You're out of your house. You're out of your house. You're out of your house. You're out of your house. You're out of your house. You're out",
+ "raw": {
+ "normalized_reference": "why, the average review is more nauseating than cod liver oil.",
+ "normalized_hypothesis": "you're out of your house. you're out of your house. you're out of your house. you're out of your house. you're out of your house. you're out of your house. you're out of your house. you're out of your house. you're out of your house. you're out of your house. you're out of your house. you're out of your house. you're out of your house. you're out of your house. you're out of your house. you're out of your house. you're out of your house. you're out of your house. you're out of your house. you're out of your house. you're out of your house. you're out of your house. you're out of your house. you're out of your house. you're out of your house. you're out of your house. you're out of your house. you're out of your house. you're out of your house. you're out of your house. you're out of your house. you're out of your house. you're out of your house. you're out of your house. you're out of your house. you're out of your house. you're out of your house. you're out of your house. you're out of your house. you're out of your house. you're out of your house. you're out of your house. you're out of your house. you're out of your house. you're out of your house. you're out of your house. you're out of your house. you're out of your house. you're out of your house. you're out of your house. you're out of your house. you're out of your house. you're out of your house. you're out of your house. you're out of your house. you're out of your house. you're out of your house. you're out of your house. you're out of your house. you're out of your house. you're out of your house. you're out of your house. you're out of your house. you're out",
+ "errors": 317,
+ "words": 11,
+ "wer": 28.818181818181817
+ },
+ "semantic": {
+ "normalized_reference": "why the average review is more nauseating than cod liver oil",
+ "normalized_hypothesis": "you are out of your house you are out of your house you are out of your house you are out of your house you are out of your house you are out of your house you are out of your house you are out of your house you are out of your house you are out of your house you are out of your house you are out of your house you are out of your house you are out of your house you are out of your house you are out of your house you are out of your house you are out of your house you are out of your house you are out of your house you are out of your house you are out of your house you are out of your house you are out of your house you are out of your house you are out of your house you are out of your house you are out of your house you are out of your house you are out of your house you are out of your house you are out of your house you are out of your house you are out of your house you are out of your house you are out of your house you are out of your house you are out of your house you are out of your house you are out of your house you are out of your house you are out of your house you are out of your house you are out of your house you are out of your house you are out of your house you are out of your house you are out of your house you are out of your house you are out of your house you are out of your house you are out of your house you are out of your house you are out of your house you are out of your house you are out of your house you are out of your house you are out of your house you are out of your house you are out of your house you are out of your house you are out of your house you are out of your house you are out",
+ "errors": 381,
+ "words": 11,
+ "wer": 34.63636363636363
+ }
+ },
+ {
+ "id": "release500_0482",
+ "category": "phonetic_balance",
+ "text": "I have been robbed, sir, I amended.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0482.wav",
+ "render_seconds": 68.40071363805328,
+ "audio_seconds": 2.716734693877551,
+ "rtf": 25.177546336122372,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Well, that looks a lot like it.",
+ "raw": {
+ "normalized_reference": "i have been robbed, sir, i amended.",
+ "normalized_hypothesis": "well, that looks a lot like it.",
+ "errors": 7,
+ "words": 7,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "i have been robbed sir i amended",
+ "normalized_hypothesis": "well that looks a lot like it",
+ "errors": 7,
+ "words": 7,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0483",
+ "category": "phonetic_balance",
+ "text": "He moved away as quietly as he had come.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0483.wav",
+ "render_seconds": 117.00307293399237,
+ "audio_seconds": 2.9257142857142857,
+ "rtf": 39.991284694235674,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I was a wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash",
+ "raw": {
+ "normalized_reference": "he moved away as quietly as he had come.",
+ "normalized_hypothesis": "i was a wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash, wash",
+ "errors": 224,
+ "words": 9,
+ "wer": 24.88888888888889
+ },
+ "semantic": {
+ "normalized_reference": "he moved away as quietly as he had come",
+ "normalized_hypothesis": "i was a wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash wash",
+ "errors": 224,
+ "words": 9,
+ "wer": 24.88888888888889
+ }
+ },
+ {
+ "id": "release500_0484",
+ "category": "phonetic_balance",
+ "text": "A trickle of fresh blood ran over his face.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0484.wav",
+ "render_seconds": 58.30075218703132,
+ "audio_seconds": 3.2043537414965986,
+ "rtf": 18.194231002661354,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Exactly the first bit while I wash my face.",
+ "raw": {
+ "normalized_reference": "a trickle of fresh blood ran over his face.",
+ "normalized_hypothesis": "exactly the first bit while i wash my face.",
+ "errors": 8,
+ "words": 9,
+ "wer": 0.8888888888888888
+ },
+ "semantic": {
+ "normalized_reference": "a trickle of fresh blood ran over his face",
+ "normalized_hypothesis": "exactly the 1st bit while i wash my face",
+ "errors": 8,
+ "words": 9,
+ "wer": 0.8888888888888888
+ }
+ },
+ {
+ "id": "release500_0485",
+ "category": "phonetic_balance",
+ "text": "Suppose you saw me at work through the window.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0485.wav",
+ "render_seconds": 105.9996848970186,
+ "audio_seconds": 3.0650340136054424,
+ "rtf": 34.58352645565903,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Subtitles by the Amara.org community",
+ "raw": {
+ "normalized_reference": "suppose you saw me at work through the window.",
+ "normalized_hypothesis": "subtitles by the amara.org community",
+ "errors": 9,
+ "words": 9,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "suppose you saw me at work through the window",
+ "normalized_hypothesis": "subtitles by the amara org community",
+ "errors": 9,
+ "words": 9,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0486",
+ "category": "phonetic_balance",
+ "text": "But who was Eileen's double.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0486.wav",
+ "render_seconds": 48.30594563600607,
+ "audio_seconds": 2.089795918367347,
+ "rtf": 23.115149767229465,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "But who was I last to be?",
+ "raw": {
+ "normalized_reference": "but who was eileen's double.",
+ "normalized_hypothesis": "but who was i last to be?",
+ "errors": 4,
+ "words": 5,
+ "wer": 0.8
+ },
+ "semantic": {
+ "normalized_reference": "but who was eileen is double",
+ "normalized_hypothesis": "but who was i last to be",
+ "errors": 4,
+ "words": 6,
+ "wer": 0.6666666666666666
+ }
+ },
+ {
+ "id": "release500_0487",
+ "category": "phonetic_balance",
+ "text": "Darkness hid him from Jeanne.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0487.wav",
+ "render_seconds": 79.11909808102064,
+ "audio_seconds": 2.2291156462585033,
+ "rtf": 35.49349187594615,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Take the shadow for a trial.",
+ "raw": {
+ "normalized_reference": "darkness hid him from jeanne.",
+ "normalized_hypothesis": "take the shadow for a trial.",
+ "errors": 6,
+ "words": 5,
+ "wer": 1.2
+ },
+ "semantic": {
+ "normalized_reference": "darkness hid him from jeanne",
+ "normalized_hypothesis": "take the shadow for a trial",
+ "errors": 6,
+ "words": 5,
+ "wer": 1.2
+ }
+ },
+ {
+ "id": "release500_0488",
+ "category": "phonetic_balance",
+ "text": "By this time Charley was as enraged as the Greek.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0488.wav",
+ "render_seconds": 82.11378037801478,
+ "audio_seconds": 3.5526530612244898,
+ "rtf": 23.113368787469692,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Bye. Bye. Bye. Bye. Bye. Bye. Bye. Goodbye. Goodbye. Goodbye. Goodbye. Goodbye. Goodbye. Goodbye. Goodbye. Goodbye. Goodbye. Goodbye. Goodbye. Goodbye. Goodbye. Goodbye. Goodbye. Goodbye. Goodbye. Goodbye. Goodbye. Goodbye. Goodbye. Goodbye. Goodbye. Goodbye. Goodbye. Goodbye. Goodbye. Goodbye. Goodbye. Goodbye. Goodbye. Goodbye. Goodbye. Goodbye. Goodbye. Goodbye. Goodbye. Goodbye. Goodbye. Goodbye. Goodbye. Goodbye. Goodbye. Goodbye. Goodbye. Goodbye. Goodbye. Goodbye. Goodbye. Goodbye. Goodbye. Goodbye. Goodbye. Goodbye. Goodbye. Goodbye. Goodbye. Goodbye. Goodbye. Goodbye. Goodbye. Goodbye. Goodbye. Goodbye. Goodbye. Goodbye. Goodbye. Goodbye. Goodbye. Goodbye. Goodbye. Goodbye. Goodbye. Goodbye. Goodbye. Goodbye. Goodbye. Goodbye. Goodbye. Goodbye. Goodbye. Goodbye. Goodbye. Goodbye. Goodbye. Goodbye. Goodbye. Goodbye. Goodbye. Goodbye. Goodbye. Goodbye. Goodbye. Goodbye. Goodbye. Goodbye. Goodbye. Goodbye. Goodbye. Goodbye. Goodbye. Goodbye. Goodbye. Goodbye. Goodbye. Goodbye. Goodbye. Goodbye. Goodbye. Goodbye. Goodbye. Goodbye. Goodbye. Goodbye. Goodbye.",
+ "raw": {
+ "normalized_reference": "by this time charley was as enraged as the greek.",
+ "normalized_hypothesis": "bye. bye. bye. bye. bye. bye. bye. goodbye. goodbye. goodbye. goodbye. goodbye. goodbye. goodbye. goodbye. goodbye. goodbye. goodbye. goodbye. goodbye. goodbye. goodbye. goodbye. goodbye. goodbye. goodbye. goodbye. goodbye. goodbye. goodbye. goodbye. goodbye. goodbye. goodbye. goodbye. goodbye. goodbye. goodbye. goodbye. goodbye. goodbye. goodbye. goodbye. goodbye. goodbye. goodbye. goodbye. goodbye. goodbye. goodbye. goodbye. goodbye. goodbye. goodbye. goodbye. goodbye. goodbye. goodbye. goodbye. goodbye. goodbye. goodbye. goodbye. goodbye. goodbye. goodbye. goodbye. goodbye. goodbye. goodbye. goodbye. goodbye. goodbye. goodbye. goodbye. goodbye. goodbye. goodbye. goodbye. goodbye. goodbye. goodbye. goodbye. goodbye. goodbye. goodbye. goodbye. goodbye. goodbye. goodbye. goodbye. goodbye. goodbye. goodbye. goodbye. goodbye. goodbye. goodbye. goodbye. goodbye. goodbye. goodbye. goodbye. goodbye. goodbye. goodbye. goodbye. goodbye. goodbye. goodbye. goodbye. goodbye. goodbye. goodbye. goodbye. goodbye. goodbye. goodbye. goodbye. goodbye. goodbye. goodbye. goodbye.",
+ "errors": 123,
+ "words": 10,
+ "wer": 12.3
+ },
+ "semantic": {
+ "normalized_reference": "by this time charley was as enraged as the greek",
+ "normalized_hypothesis": "bye bye bye bye bye bye bye goodbye goodbye goodbye goodbye goodbye goodbye goodbye goodbye goodbye goodbye goodbye goodbye goodbye goodbye goodbye goodbye goodbye goodbye goodbye goodbye goodbye goodbye goodbye goodbye goodbye goodbye goodbye goodbye goodbye goodbye goodbye goodbye goodbye goodbye goodbye goodbye goodbye goodbye goodbye goodbye goodbye goodbye goodbye goodbye goodbye goodbye goodbye goodbye goodbye goodbye goodbye goodbye goodbye goodbye goodbye goodbye goodbye goodbye goodbye goodbye goodbye goodbye goodbye goodbye goodbye goodbye goodbye goodbye goodbye goodbye goodbye goodbye goodbye goodbye goodbye goodbye goodbye goodbye goodbye goodbye goodbye goodbye goodbye goodbye goodbye goodbye goodbye goodbye goodbye goodbye goodbye goodbye goodbye goodbye goodbye goodbye goodbye goodbye goodbye goodbye goodbye goodbye goodbye goodbye goodbye goodbye goodbye goodbye goodbye goodbye goodbye goodbye goodbye goodbye goodbye goodbye",
+ "errors": 123,
+ "words": 10,
+ "wer": 12.3
+ }
+ },
+ {
+ "id": "release500_0489",
+ "category": "phonetic_balance",
+ "text": "I am writing these lines in Honolulu, Hawaii.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0489.wav",
+ "render_seconds": 115.39810844301246,
+ "audio_seconds": 3.274013605442177,
+ "rtf": 35.24667956516546,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Hello and welcome to Super Mario Bros. 3D.",
+ "raw": {
+ "normalized_reference": "i am writing these lines in honolulu, hawaii.",
+ "normalized_hypothesis": "hello and welcome to super mario bros. 3d.",
+ "errors": 8,
+ "words": 8,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "i am writing these lines in honolulu hawaii",
+ "normalized_hypothesis": "hello and welcome to super mario bros 3 d",
+ "errors": 9,
+ "words": 8,
+ "wer": 1.125
+ }
+ },
+ {
+ "id": "release500_0490",
+ "category": "phonetic_balance",
+ "text": "Nobody knows how the natives got them.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0490.wav",
+ "render_seconds": 48.70029347902164,
+ "audio_seconds": 2.716734693877551,
+ "rtf": 17.92603951676728,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Nobody knows who the last character is.",
+ "raw": {
+ "normalized_reference": "nobody knows how the natives got them.",
+ "normalized_hypothesis": "nobody knows who the last character is.",
+ "errors": 4,
+ "words": 7,
+ "wer": 0.5714285714285714
+ },
+ "semantic": {
+ "normalized_reference": "nobody knows how the natives got them",
+ "normalized_hypothesis": "nobody knows who the last character is",
+ "errors": 4,
+ "words": 7,
+ "wer": 0.5714285714285714
+ }
+ },
+ {
+ "id": "release500_0491",
+ "category": "phonetic_balance",
+ "text": "And how would we ever find ourselves.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0491.wav",
+ "render_seconds": 99.00526309898123,
+ "audio_seconds": 2.716734693877551,
+ "rtf": 36.442742576998796,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I can't look back before I go to sleep.",
+ "raw": {
+ "normalized_reference": "and how would we ever find ourselves.",
+ "normalized_hypothesis": "i can't look back before i go to sleep.",
+ "errors": 9,
+ "words": 7,
+ "wer": 1.2857142857142858
+ },
+ "semantic": {
+ "normalized_reference": "and how would we ever find ourselves",
+ "normalized_hypothesis": "i can not look back before i go to sleep",
+ "errors": 10,
+ "words": 7,
+ "wer": 1.4285714285714286
+ }
+ },
+ {
+ "id": "release500_0492",
+ "category": "phonetic_balance",
+ "text": "They are coming ashore, whoever they are.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0492.wav",
+ "render_seconds": 43.11082718800753,
+ "audio_seconds": 2.8560544217687074,
+ "rtf": 15.094539816685185,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Thank you for watching.",
+ "raw": {
+ "normalized_reference": "they are coming ashore, whoever they are.",
+ "normalized_hypothesis": "thank you for watching.",
+ "errors": 7,
+ "words": 7,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "they are coming ashore whoever they are",
+ "normalized_hypothesis": "thank you for watching",
+ "errors": 7,
+ "words": 7,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0493",
+ "category": "phonetic_balance",
+ "text": "This is my fifth voyage.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0493.wav",
+ "render_seconds": 71.49168569408357,
+ "audio_seconds": 2.159455782312925,
+ "rtf": 33.106343866652864,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "This is your first watch.",
+ "raw": {
+ "normalized_reference": "this is my fifth voyage.",
+ "normalized_hypothesis": "this is your first watch.",
+ "errors": 3,
+ "words": 5,
+ "wer": 0.6
+ },
+ "semantic": {
+ "normalized_reference": "this is my 5th voyage",
+ "normalized_hypothesis": "this is your 1st watch",
+ "errors": 3,
+ "words": 5,
+ "wer": 0.6
+ }
+ },
+ {
+ "id": "release500_0494",
+ "category": "phonetic_balance",
+ "text": "And yet if she came he had no words to say.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0494.wav",
+ "render_seconds": 79.69409804290626,
+ "audio_seconds": 2.9257142857142857,
+ "rtf": 27.239193667008976,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I don't know what you're talking about. I don't know what you're talking about. I don't know what you're talking about. I don't know what you're talking about. I don't know what you're talking about. I don't know what you're talking about. I don't know what you're talking about. I don't know what you're talking about. I don't know what you're talking about. I don't know what you're talking about. I don't know what you're talking about. I don't know what you're talking about. I don't know what you're talking about. I don't know what you're talking about. I don't know what you're talking about. I don't know what you're talking about. I don't know what you're talking about. I don't know what you're talking about. I don't know what you're talking about. I don't know what you're talking about. I don't know what you're talking about. I don't know what you're talking about. I don't know what you're talking about. I don't know what you're talking about. I don't know what you're talking about. I don't know what you're talking about. I don't know what you're talking about. I don't know what you're talking about. I don't know what you're talking about. I don't know what you're talking about. I don't know what you're talking about. I don't know what you're talking about. I don't know what you're talking about. I don't know what you're talking about. I don't know what you're talking about. I don't know what you're talking about. I don't know what you're talking about. I don't know what you're talking about. I don't know what you're talking about. I don't know what you're talking about. I don't know what you're talking about. I don't know what you're talking about. I don't know what you're talking about. I don't know what you're talking about. I don't know",
+ "raw": {
+ "normalized_reference": "and yet if she came he had no words to say.",
+ "normalized_hypothesis": "i don't know what you're talking about. i don't know what you're talking about. i don't know what you're talking about. i don't know what you're talking about. i don't know what you're talking about. i don't know what you're talking about. i don't know what you're talking about. i don't know what you're talking about. i don't know what you're talking about. i don't know what you're talking about. i don't know what you're talking about. i don't know what you're talking about. i don't know what you're talking about. i don't know what you're talking about. i don't know what you're talking about. i don't know what you're talking about. i don't know what you're talking about. i don't know what you're talking about. i don't know what you're talking about. i don't know what you're talking about. i don't know what you're talking about. i don't know what you're talking about. i don't know what you're talking about. i don't know what you're talking about. i don't know what you're talking about. i don't know what you're talking about. i don't know what you're talking about. i don't know what you're talking about. i don't know what you're talking about. i don't know what you're talking about. i don't know what you're talking about. i don't know what you're talking about. i don't know what you're talking about. i don't know what you're talking about. i don't know what you're talking about. i don't know what you're talking about. i don't know what you're talking about. i don't know what you're talking about. i don't know what you're talking about. i don't know what you're talking about. i don't know what you're talking about. i don't know what you're talking about. i don't know what you're talking about. i don't know what you're talking about. i don't know",
+ "errors": 311,
+ "words": 11,
+ "wer": 28.272727272727273
+ },
+ "semantic": {
+ "normalized_reference": "and yet if she came he had no words to say",
+ "normalized_hypothesis": "i do not know what you are talking about i do not know what you are talking about i do not know what you are talking about i do not know what you are talking about i do not know what you are talking about i do not know what you are talking about i do not know what you are talking about i do not know what you are talking about i do not know what you are talking about i do not know what you are talking about i do not know what you are talking about i do not know what you are talking about i do not know what you are talking about i do not know what you are talking about i do not know what you are talking about i do not know what you are talking about i do not know what you are talking about i do not know what you are talking about i do not know what you are talking about i do not know what you are talking about i do not know what you are talking about i do not know what you are talking about i do not know what you are talking about i do not know what you are talking about i do not know what you are talking about i do not know what you are talking about i do not know what you are talking about i do not know what you are talking about i do not know what you are talking about i do not know what you are talking about i do not know what you are talking about i do not know what you are talking about i do not know what you are talking about i do not know what you are talking about i do not know what you are talking about i do not know what you are talking about i do not know what you are talking about i do not know what you are talking about i do not know what you are talking about i do not know what you are talking about i do not know what you are talking about i do not know what you are talking about i do not know what you are talking about i do not know what you are talking about i do not know",
+ "errors": 400,
+ "words": 11,
+ "wer": 36.36363636363637
+ }
+ },
+ {
+ "id": "release500_0495",
+ "category": "phonetic_balance",
+ "text": "You must sleep, he urged.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0495.wav",
+ "render_seconds": 63.923034836072475,
+ "audio_seconds": 2.020136054421769,
+ "rtf": 31.64293548256551,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Oh, shut up.",
+ "raw": {
+ "normalized_reference": "you must sleep, he urged.",
+ "normalized_hypothesis": "oh, shut up.",
+ "errors": 5,
+ "words": 5,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "you must sleep he urged",
+ "normalized_hypothesis": "0 shut up",
+ "errors": 5,
+ "words": 5,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0496",
+ "category": "phonetic_balance",
+ "text": "Could the incident have anything to do with Jeanne and Pierre.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0496.wav",
+ "render_seconds": 64.49907698098104,
+ "audio_seconds": 4.179591836734694,
+ "rtf": 15.431908066738627,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "Goodbye, sugar-coated elephant. Go with Colin, pal.",
+ "raw": {
+ "normalized_reference": "could the incident have anything to do with jeanne and pierre.",
+ "normalized_hypothesis": "goodbye, sugar-coated elephant. go with colin, pal.",
+ "errors": 10,
+ "words": 11,
+ "wer": 0.9090909090909091
+ },
+ "semantic": {
+ "normalized_reference": "could the incident have anything to do with jeanne and pierre",
+ "normalized_hypothesis": "goodbye sugar coated elephant go with colin pal",
+ "errors": 10,
+ "words": 11,
+ "wer": 0.9090909090909091
+ }
+ },
+ {
+ "id": "release500_0497",
+ "category": "phonetic_balance",
+ "text": "Each day she became a more vital part of him.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0497.wav",
+ "render_seconds": 77.30156217201147,
+ "audio_seconds": 3.274013605442177,
+ "rtf": 23.610641703967932,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": ".......................................................................................................................................................................................................................................................................................................................................................................................",
+ "raw": {
+ "normalized_reference": "each day she became a more vital part of him.",
+ "normalized_hypothesis": ".......................................................................................................................................................................................................................................................................................................................................................................................",
+ "errors": 10,
+ "words": 10,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "each day she became a more vital part of him",
+ "normalized_hypothesis": " .",
+ "errors": 10,
+ "words": 10,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0498",
+ "category": "phonetic_balance",
+ "text": "But why continue the tirade, for tirade it was.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0498.wav",
+ "render_seconds": 29.597576294094324,
+ "audio_seconds": 3.3436734693877552,
+ "rtf": 8.85181419928358,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I'm not going to tell you what I want. I want to tell you what I wish.",
+ "raw": {
+ "normalized_reference": "but why continue the tirade, for tirade it was.",
+ "normalized_hypothesis": "i'm not going to tell you what i want. i want to tell you what i wish.",
+ "errors": 17,
+ "words": 9,
+ "wer": 1.8888888888888888
+ },
+ "semantic": {
+ "normalized_reference": "but why continue the tirade for tirade it was",
+ "normalized_hypothesis": "i am not going to tell you what i want i want to tell you what i wish",
+ "errors": 18,
+ "words": 9,
+ "wer": 2.0
+ }
+ },
+ {
+ "id": "release500_0499",
+ "category": "phonetic_balance",
+ "text": "That is what distinguishes all of us from the lower animals.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0499.wav",
+ "render_seconds": 111.79130288900342,
+ "audio_seconds": 4.31891156462585,
+ "rtf": 25.884137986207637,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I just want to stay alive. I just want to live my life.",
+ "raw": {
+ "normalized_reference": "that is what distinguishes all of us from the lower animals.",
+ "normalized_hypothesis": "i just want to stay alive. i just want to live my life.",
+ "errors": 13,
+ "words": 11,
+ "wer": 1.1818181818181819
+ },
+ "semantic": {
+ "normalized_reference": "that is what distinguishes all of us from the lower animals",
+ "normalized_hypothesis": "i just want to stay alive i just want to live my life",
+ "errors": 13,
+ "words": 11,
+ "wer": 1.1818181818181819
+ }
+ },
+ {
+ "id": "release500_0500",
+ "category": "phonetic_balance",
+ "text": "There was something pathetic in the girl's attitude now.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0500.wav",
+ "render_seconds": 109.39057729102205,
+ "audio_seconds": 3.622312925170068,
+ "rtf": 30.19909642011013,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "whisper-large-v3",
+ "hypothesis": "I will show you the fact and the cause of the trouble.",
+ "raw": {
+ "normalized_reference": "there was something pathetic in the girl's attitude now.",
+ "normalized_hypothesis": "i will show you the fact and the cause of the trouble.",
+ "errors": 11,
+ "words": 9,
+ "wer": 1.2222222222222223
+ },
+ "semantic": {
+ "normalized_reference": "there was something pathetic in the girl is attitude now",
+ "normalized_hypothesis": "i will show you the fact and the cause of the trouble",
+ "errors": 11,
+ "words": 10,
+ "wer": 1.1
+ }
+ },
+ {
+ "id": "release500_0001",
+ "category": "conversational",
+ "text": "It wasn't until later that I realized what had actually happened.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0001.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT WASN'T UNTIL LATER THAT I REALIZED WHAT HAD ACTUALLY HAPPENED",
+ "raw": {
+ "normalized_reference": "it wasn't until later that i realized what had actually happened.",
+ "normalized_hypothesis": "it wasn't until later that i realized what had actually happened",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it was not until later that i realized what had actually happened",
+ "normalized_hypothesis": "it was not until later that i realized what had actually happened",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0002",
+ "category": "descriptive",
+ "text": "The light at that time of afternoon is perfect for about twenty minutes.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0002.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE LIGHT AT THAT TIME OF AFTERNOON IS PERFECT FOR ABOUT TWENTY MINUTES",
+ "raw": {
+ "normalized_reference": "the light at that time of afternoon is perfect for about twenty minutes.",
+ "normalized_hypothesis": "the light at that time of afternoon is perfect for about twenty minutes",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the light at that time of afternoon is perfect for about 20 minutes",
+ "normalized_hypothesis": "the light at that time of afternoon is perfect for about 20 minutes",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0003",
+ "category": "emotional",
+ "text": "That was the scariest thing I've ever done, and also maybe the best.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0003.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THAT WAS THE SCARIEST THING I'VE EVER DONE AND ALSO MAYBE THE BEST",
+ "raw": {
+ "normalized_reference": "that was the scariest thing i've ever done, and also maybe the best.",
+ "normalized_hypothesis": "that was the scariest thing i've ever done and also maybe the best",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "that was the scariest thing i have ever done and also maybe the best",
+ "normalized_hypothesis": "that was the scariest thing i have ever done and also maybe the best",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0004",
+ "category": "homographs",
+ "text": "After I record the record, please present the present to the bass player.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0004.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "AFTER I RECORD THE RECORD PLEASE PRESENT THE PRESENT TO THE BASE PLAYER",
+ "raw": {
+ "normalized_reference": "after i record the record, please present the present to the bass player.",
+ "normalized_hypothesis": "after i record the record please present the present to the base player",
+ "errors": 1,
+ "words": 13,
+ "wer": 0.07692307692307693
+ },
+ "semantic": {
+ "normalized_reference": "after i record the record please present the present to the bass player",
+ "normalized_hypothesis": "after i record the record please present the present to the base player",
+ "errors": 1,
+ "words": 13,
+ "wer": 0.07692307692307693
+ }
+ },
+ {
+ "id": "release500_0005",
+ "category": "long",
+ "text": "If I'm being completely honest, I had no idea what I was doing for most of it, and I think that might be the most normal thing about the whole experience.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0005.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IF I AM BEING COMPLETELY HONEST I HAD NO IDEA WHAT I WAS DOING FOR MOST OF IT AND I THINK THAT MIGHT BE THE MOST NORMAL THING ABOUT THE WHOLE EXPERIENCE",
+ "raw": {
+ "normalized_reference": "if i'm being completely honest, i had no idea what i was doing for most of it, and i think that might be the most normal thing about the whole experience.",
+ "normalized_hypothesis": "if i am being completely honest i had no idea what i was doing for most of it and i think that might be the most normal thing about the whole experience",
+ "errors": 2,
+ "words": 31,
+ "wer": 0.06451612903225806
+ },
+ "semantic": {
+ "normalized_reference": "if i am being completely honest i had no idea what i was doing for most of it and i think that might be the most normal thing about the whole experience",
+ "normalized_hypothesis": "if i am being completely honest i had no idea what i was doing for most of it and i think that might be the most normal thing about the whole experience",
+ "errors": 0,
+ "words": 32,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0006",
+ "category": "long_range",
+ "text": "The committee postponed its decision after discovering that three pages of the report had been printed in the wrong order.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0006.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE COMMITTEE POSTPONED ITS DECISION AFTER DISCOVERING THAT THREE PAGES OF THE REPORT HAD BEEN PRINTED IN THE WRONG ORDER",
+ "raw": {
+ "normalized_reference": "the committee postponed its decision after discovering that three pages of the report had been printed in the wrong order.",
+ "normalized_hypothesis": "the committee postponed its decision after discovering that three pages of the report had been printed in the wrong order",
+ "errors": 0,
+ "words": 20,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the committee postponed its decision after discovering that 3 pages of the report had been printed in the wrong order",
+ "normalized_hypothesis": "the committee postponed its decision after discovering that 3 pages of the report had been printed in the wrong order",
+ "errors": 0,
+ "words": 20,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0007",
+ "category": "names_places",
+ "text": "Gwendolyn photographed the eucalyptus trees outside Ljubljana.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0007.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "GWENDOLEN PHOTOGRAPHED THE EUCALYPTIS TREES OUTSIDE LIUBLIANA",
+ "raw": {
+ "normalized_reference": "gwendolyn photographed the eucalyptus trees outside ljubljana.",
+ "normalized_hypothesis": "gwendolen photographed the eucalyptis trees outside liubliana",
+ "errors": 3,
+ "words": 7,
+ "wer": 0.42857142857142855
+ },
+ "semantic": {
+ "normalized_reference": "gwendolyn photographed the eucalyptus trees outside ljubljana",
+ "normalized_hypothesis": "gwendolen photographed the eucalyptis trees outside liubliana",
+ "errors": 3,
+ "words": 7,
+ "wer": 0.42857142857142855
+ }
+ },
+ {
+ "id": "release500_0008",
+ "category": "narrative",
+ "text": "He had this theory, and the more time passes, the more I think he was right.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0008.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE HAD THIS THEORY AND THE MORE TIME PASSES THE MORE I THINK HE WAS RIGHT",
+ "raw": {
+ "normalized_reference": "he had this theory, and the more time passes, the more i think he was right.",
+ "normalized_hypothesis": "he had this theory and the more time passes the more i think he was right",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he had this theory and the more time passes the more i think he was right",
+ "normalized_hypothesis": "he had this theory and the more time passes the more i think he was right",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0009",
+ "category": "numbers",
+ "text": "The package weighs twelve point six kilograms and arrived on July twenty-first.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0009.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE PACKAGE WEIGHS TWELVE POINT SIX KILOGRAMS AND ARRIVED ON JULY TWENTY FIRST",
+ "raw": {
+ "normalized_reference": "the package weighs twelve point six kilograms and arrived on july twenty-first.",
+ "normalized_hypothesis": "the package weighs twelve point six kilograms and arrived on july twenty first",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the package weighs 12.6 kilograms and arrived on july 21st",
+ "normalized_hypothesis": "the package weighs 12.6 kilograms and arrived on july 21st",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0010",
+ "category": "punctuation",
+ "text": "First, close the window; second, turn off the lamp; finally, lock the door.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0010.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "FIRST CLOSE THE WINDOW SECOND TURN OFF THE LAMP FINALLY LOCK THE DOOR",
+ "raw": {
+ "normalized_reference": "first, close the window; second, turn off the lamp; finally, lock the door.",
+ "normalized_hypothesis": "first close the window second turn off the lamp finally lock the door",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "1st close the window 2nd turn off the lamp finally lock the door",
+ "normalized_hypothesis": "1st close the window 2nd turn off the lamp finally lock the door",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0011",
+ "category": "question",
+ "text": "When did you first realize something was wrong?",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0011.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WHEN DID YOU FIRST REALIZE SOMETHING WAS WRONG",
+ "raw": {
+ "normalized_reference": "when did you first realize something was wrong?",
+ "normalized_hypothesis": "when did you first realize something was wrong",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "when did you 1st realize something was wrong",
+ "normalized_hypothesis": "when did you 1st realize something was wrong",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0012",
+ "category": "rare_words",
+ "text": "The archaeologist catalogued a Byzantine reliquary and several anachronistic fragments.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0012.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE ARCHAEOLOGIST CATALOGUED A BIZANTINE RILLIQUARY AND SEVERAL ANACHRENISTIC FRAGMENTS",
+ "raw": {
+ "normalized_reference": "the archaeologist catalogued a byzantine reliquary and several anachronistic fragments.",
+ "normalized_hypothesis": "the archaeologist catalogued a bizantine rilliquary and several anachrenistic fragments",
+ "errors": 3,
+ "words": 10,
+ "wer": 0.3
+ },
+ "semantic": {
+ "normalized_reference": "the archaeologist catalogued a byzantine reliquary and several anachronistic fragments",
+ "normalized_hypothesis": "the archaeologist catalogued a bizantine rilliquary and several anachrenistic fragments",
+ "errors": 3,
+ "words": 10,
+ "wer": 0.3
+ }
+ },
+ {
+ "id": "release500_0013",
+ "category": "short",
+ "text": "I have no idea.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0013.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I HAVE NO IDEA",
+ "raw": {
+ "normalized_reference": "i have no idea.",
+ "normalized_hypothesis": "i have no idea",
+ "errors": 0,
+ "words": 4,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i have no idea",
+ "normalized_hypothesis": "i have no idea",
+ "errors": 0,
+ "words": 4,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0014",
+ "category": "technical",
+ "text": "The system runs on three core components that all have to stay in sync.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0014.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE SYSTEM RUNS ON THREE CORPS COMPONENTS THAT ALL HAVE TO STAY AND SINK",
+ "raw": {
+ "normalized_reference": "the system runs on three core components that all have to stay in sync.",
+ "normalized_hypothesis": "the system runs on three corps components that all have to stay and sink",
+ "errors": 3,
+ "words": 14,
+ "wer": 0.21428571428571427
+ },
+ "semantic": {
+ "normalized_reference": "the system runs on 3 core components that all have to stay in sync",
+ "normalized_hypothesis": "the system runs on 3 corps components that all have to stay and sink",
+ "errors": 3,
+ "words": 14,
+ "wer": 0.21428571428571427
+ }
+ },
+ {
+ "id": "release500_0015",
+ "category": "conversational",
+ "text": "We improvised the entire second half and somehow nobody noticed.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0015.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WE IMPROVISED THE ENTIRE SECOND HALF AND SOMEHOW NOBODY NOTICED",
+ "raw": {
+ "normalized_reference": "we improvised the entire second half and somehow nobody noticed.",
+ "normalized_hypothesis": "we improvised the entire second half and somehow nobody noticed",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "we improvised the entire 2nd half and somehow nobody noticed",
+ "normalized_hypothesis": "we improvised the entire 2nd half and somehow nobody noticed",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0016",
+ "category": "descriptive",
+ "text": "There's a smell in old buildings that's impossible to describe but immediately familiar.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0016.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THERE'S A SMELL IN OLD BUILDINGS THAT'S IMPOSSIBLE TO DESCRIBE BUT IMMEDIATELY FAMILIAR",
+ "raw": {
+ "normalized_reference": "there's a smell in old buildings that's impossible to describe but immediately familiar.",
+ "normalized_hypothesis": "there's a smell in old buildings that's impossible to describe but immediately familiar",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "there is a smell in old buildings that is impossible to describe but immediately familiar",
+ "normalized_hypothesis": "there is a smell in old buildings that is impossible to describe but immediately familiar",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0017",
+ "category": "emotional",
+ "text": "I specifically asked for one thing. One thing. And it still didn't happen.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0017.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I SPECIFICALLY ASKED FOR ONE THING ONE THING AND IT STILL DIDN'T HAPPEN",
+ "raw": {
+ "normalized_reference": "i specifically asked for one thing. one thing. and it still didn't happen.",
+ "normalized_hypothesis": "i specifically asked for one thing one thing and it still didn't happen",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i specifically asked for one thing one thing and it still did not happen",
+ "normalized_hypothesis": "i specifically asked for one thing one thing and it still did not happen",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0018",
+ "category": "homographs",
+ "text": "They refuse to place the refuse beside the produce aisle.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0018.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THEY REFUSED TO PLACE THE REFUSE BESIDE THE PRODUSALE",
+ "raw": {
+ "normalized_reference": "they refuse to place the refuse beside the produce aisle.",
+ "normalized_hypothesis": "they refused to place the refuse beside the produsale",
+ "errors": 3,
+ "words": 10,
+ "wer": 0.3
+ },
+ "semantic": {
+ "normalized_reference": "they refuse to place the refuse beside the produce aisle",
+ "normalized_hypothesis": "they refused to place the refuse beside the produsale",
+ "errors": 3,
+ "words": 10,
+ "wer": 0.3
+ }
+ },
+ {
+ "id": "release500_0019",
+ "category": "long",
+ "text": "There were days last year when I genuinely didn't know how I was going to get through it, and somehow I did, and I still don't fully understand how.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0019.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THERE WERE DAYS LAST YEAR WHEN I GENUINELY DIDN'T KNOW HOW I WAS GOING TO GET THROUGH IT AND SOMEHOW I DID AND I STILL DON'T FULLY UNDERSTAND HOW",
+ "raw": {
+ "normalized_reference": "there were days last year when i genuinely didn't know how i was going to get through it, and somehow i did, and i still don't fully understand how.",
+ "normalized_hypothesis": "there were days last year when i genuinely didn't know how i was going to get through it and somehow i did and i still don't fully understand how",
+ "errors": 0,
+ "words": 29,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "there were days last year when i genuinely did not know how i was going to get through it and somehow i did and i still do not fully understand how",
+ "normalized_hypothesis": "there were days last year when i genuinely did not know how i was going to get through it and somehow i did and i still do not fully understand how",
+ "errors": 0,
+ "words": 31,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0020",
+ "category": "long_range",
+ "text": "Before announcing the winner, the judge thanked the volunteers who had worked through the storm to repair the damaged stage.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0020.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "BEFORE ANNOUNCING THE WINNER THE JUDGE FANK THE VOLUNTEERS WHO HAD WORKED THROUGH THE STORM TO REPAIR THE DAMAGE STAGE",
+ "raw": {
+ "normalized_reference": "before announcing the winner, the judge thanked the volunteers who had worked through the storm to repair the damaged stage.",
+ "normalized_hypothesis": "before announcing the winner the judge fank the volunteers who had worked through the storm to repair the damage stage",
+ "errors": 2,
+ "words": 20,
+ "wer": 0.1
+ },
+ "semantic": {
+ "normalized_reference": "before announcing the winner the judge thanked the volunteers who had worked through the storm to repair the damaged stage",
+ "normalized_hypothesis": "before announcing the winner the judge fank the volunteers who had worked through the storm to repair the damage stage",
+ "errors": 2,
+ "words": 20,
+ "wer": 0.1
+ }
+ },
+ {
+ "id": "release500_0021",
+ "category": "names_places",
+ "text": "Professor Nguyen mailed the fluorescent poster to Reykjavik on Thursday.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0021.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "PROFESSOR NEWYAN MAILED THE FLORIS AND POSTER TO RAK JEVIC ON THURSDAY",
+ "raw": {
+ "normalized_reference": "professor nguyen mailed the fluorescent poster to reykjavik on thursday.",
+ "normalized_hypothesis": "professor newyan mailed the floris and poster to rak jevic on thursday",
+ "errors": 5,
+ "words": 10,
+ "wer": 0.5
+ },
+ "semantic": {
+ "normalized_reference": "professor nguyen mailed the fluorescent poster to reykjavik on thursday",
+ "normalized_hypothesis": "professor newyan mailed the floris and poster to rak jevic on thursday",
+ "errors": 5,
+ "words": 10,
+ "wer": 0.5
+ }
+ },
+ {
+ "id": "release500_0022",
+ "category": "narrative",
+ "text": "The plan was simple. The execution was anything but.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0022.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE PLAN WAS SIMPLE THE EXECUTION WAS ANYTHING BUT",
+ "raw": {
+ "normalized_reference": "the plan was simple. the execution was anything but.",
+ "normalized_hypothesis": "the plan was simple the execution was anything but",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the plan was simple the execution was anything but",
+ "normalized_hypothesis": "the plan was simple the execution was anything but",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0023",
+ "category": "numbers",
+ "text": "Her score improved from eighty-two percent to ninety-seven percent in two weeks.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0023.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "GERSCORE IMPROVED FROM EIGHTY TWO PER CENT TO NINETY SEVEN PER CENT IN TWO WEEKS",
+ "raw": {
+ "normalized_reference": "her score improved from eighty-two percent to ninety-seven percent in two weeks.",
+ "normalized_hypothesis": "gerscore improved from eighty two per cent to ninety seven per cent in two weeks",
+ "errors": 6,
+ "words": 14,
+ "wer": 0.42857142857142855
+ },
+ "semantic": {
+ "normalized_reference": "her score improved from 82% to 97% in 2 weeks",
+ "normalized_hypothesis": "gerscore improved from 82% to 97% in 2 weeks",
+ "errors": 2,
+ "words": 10,
+ "wer": 0.2
+ }
+ },
+ {
+ "id": "release500_0024",
+ "category": "punctuation",
+ "text": "Why would anyone pack spoons, batteries, and a violin for one short trip?",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0024.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WHY WOULD ANYONE PACK SPOONS BATTERIES AND A VIOLIN FOR ONE SHORT TRIP",
+ "raw": {
+ "normalized_reference": "why would anyone pack spoons, batteries, and a violin for one short trip?",
+ "normalized_hypothesis": "why would anyone pack spoons batteries and a violin for one short trip",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "why would anyone pack spoons batteries and a violin for one short trip",
+ "normalized_hypothesis": "why would anyone pack spoons batteries and a violin for one short trip",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0025",
+ "category": "question",
+ "text": "How long have you been sitting on this?",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0025.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HOW LONG HAVE YOU BEEN SITTING ON THIS",
+ "raw": {
+ "normalized_reference": "how long have you been sitting on this?",
+ "normalized_hypothesis": "how long have you been sitting on this",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "how long have you been sitting on this",
+ "normalized_hypothesis": "how long have you been sitting on this",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0026",
+ "category": "rare_words",
+ "text": "A mischievous chameleon rested beside the chrysanthemum and the bougainvillea.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0026.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "A MISCHIEVOUS CAMELIAN RESTED BESIDE THE CHRYSANTHEMUM AND THE BAUGAN VILLIA",
+ "raw": {
+ "normalized_reference": "a mischievous chameleon rested beside the chrysanthemum and the bougainvillea.",
+ "normalized_hypothesis": "a mischievous camelian rested beside the chrysanthemum and the baugan villia",
+ "errors": 3,
+ "words": 10,
+ "wer": 0.3
+ },
+ "semantic": {
+ "normalized_reference": "a mischievous chameleon rested beside the chrysanthemum and the bougainvillea",
+ "normalized_hypothesis": "a mischievous camelian rested beside the chrysanthemum and the baugan villia",
+ "errors": 3,
+ "words": 10,
+ "wer": 0.3
+ }
+ },
+ {
+ "id": "release500_0027",
+ "category": "short",
+ "text": "We don't talk about that.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0027.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WE DON'T TALK ABOUT THAT",
+ "raw": {
+ "normalized_reference": "we don't talk about that.",
+ "normalized_hypothesis": "we don't talk about that",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "we do not talk about that",
+ "normalized_hypothesis": "we do not talk about that",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0028",
+ "category": "technical",
+ "text": "Hash the UTF eight payload with S H A two fifty-six before opening the checkpoint.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0028.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "CASH THE U T F AEIGHT PAYLOAD WITH S H AT TWO FIFTY SIX BEFORE OPENING THE CHECK POINT",
+ "raw": {
+ "normalized_reference": "hash the utf eight payload with s h a two fifty-six before opening the checkpoint.",
+ "normalized_hypothesis": "cash the u t f aeight payload with s h at two fifty six before opening the check point",
+ "errors": 8,
+ "words": 16,
+ "wer": 0.5
+ },
+ "semantic": {
+ "normalized_reference": "hash the utf 8 payload with s h a 256 before opening the checkpoint",
+ "normalized_hypothesis": "cash the u t f aeight payload with s h at 256 before opening the check point",
+ "errors": 8,
+ "words": 14,
+ "wer": 0.5714285714285714
+ }
+ },
+ {
+ "id": "release500_0029",
+ "category": "conversational",
+ "text": "I thought I had it all figured out, and then this happened.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0029.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I THOUGHT I HAD IT ALL FIGURED OUT AND THEN THIS HAPPENED",
+ "raw": {
+ "normalized_reference": "i thought i had it all figured out, and then this happened.",
+ "normalized_hypothesis": "i thought i had it all figured out and then this happened",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i thought i had it all figured out and then this happened",
+ "normalized_hypothesis": "i thought i had it all figured out and then this happened",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0030",
+ "category": "descriptive",
+ "text": "The kind of quiet you only get at three in the morning in a city.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0030.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE KIND OF QUIET YOU ONLY GET AT THREE IN THE MORNING IN A CITY",
+ "raw": {
+ "normalized_reference": "the kind of quiet you only get at three in the morning in a city.",
+ "normalized_hypothesis": "the kind of quiet you only get at three in the morning in a city",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the kind of quiet you only get at 3 in the morning in a city",
+ "normalized_hypothesis": "the kind of quiet you only get at 3 in the morning in a city",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0031",
+ "category": "emotional",
+ "text": "The joy of it was almost overwhelming, like too much good all at once.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0031.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE JOY OF IT WAS ALMOST OVERWHELMING LIKE TOO MUCH GOOD ALL AT ONCE",
+ "raw": {
+ "normalized_reference": "the joy of it was almost overwhelming, like too much good all at once.",
+ "normalized_hypothesis": "the joy of it was almost overwhelming like too much good all at once",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the joy of it was almost overwhelming like too much good all at once",
+ "normalized_hypothesis": "the joy of it was almost overwhelming like too much good all at once",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0032",
+ "category": "homographs",
+ "text": "Please close the door while I stand close to the microphone.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0032.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "PLEASE CLOSE THE DOOR WHILE I STAND CLOSE TO THE MICROPHONE",
+ "raw": {
+ "normalized_reference": "please close the door while i stand close to the microphone.",
+ "normalized_hypothesis": "please close the door while i stand close to the microphone",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "please close the door while i stand close to the microphone",
+ "normalized_hypothesis": "please close the door while i stand close to the microphone",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0033",
+ "category": "long",
+ "text": "I think the most underrated skill in the world is knowing when to stop explaining yourself and just let people draw their own conclusions.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0033.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I THINK THE MOST UNDERRATED SKILL IN THE WORLD IS KNOWING WHEN TO STOP EXPLAINING YOURSELF AND JUST LET PEOPLE DRAW THEIR OWN CONCLUSIONS",
+ "raw": {
+ "normalized_reference": "i think the most underrated skill in the world is knowing when to stop explaining yourself and just let people draw their own conclusions.",
+ "normalized_hypothesis": "i think the most underrated skill in the world is knowing when to stop explaining yourself and just let people draw their own conclusions",
+ "errors": 0,
+ "words": 24,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i think the most underrated skill in the world is knowing when to stop explaining yourself and just let people draw their own conclusions",
+ "normalized_hypothesis": "i think the most underrated skill in the world is knowing when to stop explaining yourself and just let people draw their own conclusions",
+ "errors": 0,
+ "words": 24,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0034",
+ "category": "long_range",
+ "text": "When the final bus disappeared around the corner, Nadia remembered that her umbrella, notebook, and lunch were still inside the quiet office.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0034.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WHEN THE FINAL BUST DISAPPEARED AROUND THE CORNER NAUTYA REMEMBERED THAT HER UMBRELLA NOTEBOOK AND LUNCH WERE STILL INSIDE THE QUIET OFFICE",
+ "raw": {
+ "normalized_reference": "when the final bus disappeared around the corner, nadia remembered that her umbrella, notebook, and lunch were still inside the quiet office.",
+ "normalized_hypothesis": "when the final bust disappeared around the corner nautya remembered that her umbrella notebook and lunch were still inside the quiet office",
+ "errors": 2,
+ "words": 22,
+ "wer": 0.09090909090909091
+ },
+ "semantic": {
+ "normalized_reference": "when the final bus disappeared around the corner nadia remembered that her umbrella notebook and lunch were still inside the quiet office",
+ "normalized_hypothesis": "when the final bust disappeared around the corner nautya remembered that her umbrella notebook and lunch were still inside the quiet office",
+ "errors": 2,
+ "words": 22,
+ "wer": 0.09090909090909091
+ }
+ },
+ {
+ "id": "release500_0035",
+ "category": "names_places",
+ "text": "Aurelia and Joaquin crossed the Schuylkill River before noon.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0035.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "ORLIAH AND JOKING CROSSD THE SHALKO RIVER BEFORE NOON",
+ "raw": {
+ "normalized_reference": "aurelia and joaquin crossed the schuylkill river before noon.",
+ "normalized_hypothesis": "orliah and joking crossd the shalko river before noon",
+ "errors": 4,
+ "words": 9,
+ "wer": 0.4444444444444444
+ },
+ "semantic": {
+ "normalized_reference": "aurelia and joaquin crossed the schuylkill river before noon",
+ "normalized_hypothesis": "orliah and joking crossd the shalko river before noon",
+ "errors": 4,
+ "words": 9,
+ "wer": 0.4444444444444444
+ }
+ },
+ {
+ "id": "release500_0036",
+ "category": "narrative",
+ "text": "The village had a name, but nobody used it. Everyone just called it home.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0036.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE VILLAGE HAD A NAME BUT NOBODY USED IT AN EVERYONE JUST CALLED IT HOME",
+ "raw": {
+ "normalized_reference": "the village had a name, but nobody used it. everyone just called it home.",
+ "normalized_hypothesis": "the village had a name but nobody used it an everyone just called it home",
+ "errors": 1,
+ "words": 14,
+ "wer": 0.07142857142857142
+ },
+ "semantic": {
+ "normalized_reference": "the village had a name but nobody used it everyone just called it home",
+ "normalized_hypothesis": "the village had a name but nobody used it an everyone just called it home",
+ "errors": 1,
+ "words": 14,
+ "wer": 0.07142857142857142
+ }
+ },
+ {
+ "id": "release500_0037",
+ "category": "numbers",
+ "text": "Increase the gain by three decibels, wait five seconds, then reduce it by one.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0037.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "INCREASE THE GAIN BY THREE DESIBALES WAIT FIVE SECONDS THEN REDUCE IT BY ONE",
+ "raw": {
+ "normalized_reference": "increase the gain by three decibels, wait five seconds, then reduce it by one.",
+ "normalized_hypothesis": "increase the gain by three desibales wait five seconds then reduce it by one",
+ "errors": 1,
+ "words": 14,
+ "wer": 0.07142857142857142
+ },
+ "semantic": {
+ "normalized_reference": "increase the gain by 3 decibels wait 5 seconds then reduce it by one",
+ "normalized_hypothesis": "increase the gain by 3 desibales wait 5 seconds then reduce it by one",
+ "errors": 1,
+ "words": 14,
+ "wer": 0.07142857142857142
+ }
+ },
+ {
+ "id": "release500_0038",
+ "category": "punctuation",
+ "text": "The result was unexpected: clear, stable, and surprisingly natural.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0038.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE RESULT WAS UNEXPECTED CLEAR STABLE AND SURPRISINGLY NATURAL",
+ "raw": {
+ "normalized_reference": "the result was unexpected: clear, stable, and surprisingly natural.",
+ "normalized_hypothesis": "the result was unexpected clear stable and surprisingly natural",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the result was unexpected clear stable and surprisingly natural",
+ "normalized_hypothesis": "the result was unexpected clear stable and surprisingly natural",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0039",
+ "category": "question",
+ "text": "Who told you that was the right way to do it?",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0039.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WHO TOLD YOU THAT WAS THE RIGHT WAY TO DO IT",
+ "raw": {
+ "normalized_reference": "who told you that was the right way to do it?",
+ "normalized_hypothesis": "who told you that was the right way to do it",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "who told you that was the right way to do it",
+ "normalized_hypothesis": "who told you that was the right way to do it",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0040",
+ "category": "rare_words",
+ "text": "The meticulous cartographer examined an iridescent archipelago through binoculars.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0040.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE METICULOUS CARTOGRAPHER EXAMINED AN IRIDESCENT ARCHIPELAGO THROUGH BONOCULARS",
+ "raw": {
+ "normalized_reference": "the meticulous cartographer examined an iridescent archipelago through binoculars.",
+ "normalized_hypothesis": "the meticulous cartographer examined an iridescent archipelago through bonoculars",
+ "errors": 1,
+ "words": 9,
+ "wer": 0.1111111111111111
+ },
+ "semantic": {
+ "normalized_reference": "the meticulous cartographer examined an iridescent archipelago through binoculars",
+ "normalized_hypothesis": "the meticulous cartographer examined an iridescent archipelago through bonoculars",
+ "errors": 1,
+ "words": 9,
+ "wer": 0.1111111111111111
+ }
+ },
+ {
+ "id": "release500_0041",
+ "category": "short",
+ "text": "I mean. Yeah. Kind of.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0041.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I MEAN YE A KIND OF",
+ "raw": {
+ "normalized_reference": "i mean. yeah. kind of.",
+ "normalized_hypothesis": "i mean ye a kind of",
+ "errors": 2,
+ "words": 5,
+ "wer": 0.4
+ },
+ "semantic": {
+ "normalized_reference": "i mean yeah kind of",
+ "normalized_hypothesis": "i mean ye a kind of",
+ "errors": 2,
+ "words": 5,
+ "wer": 0.4
+ }
+ },
+ {
+ "id": "release500_0042",
+ "category": "technical",
+ "text": "The logs are your best friend when something behaves unexpectedly.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0042.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE LOGS ARE YOUR BEST FRIEND WHEN SOMETHING BEHAVES UNEXPECTEDLY",
+ "raw": {
+ "normalized_reference": "the logs are your best friend when something behaves unexpectedly.",
+ "normalized_hypothesis": "the logs are your best friend when something behaves unexpectedly",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the logs are your best friend when something behaves unexpectedly",
+ "normalized_hypothesis": "the logs are your best friend when something behaves unexpectedly",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0043",
+ "category": "conversational",
+ "text": "He kept talking and I kept nodding and neither of us was listening.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0043.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE KEPT TALKING AND I KEPT NODDING AND NEITHER OF US WAS LISTENING",
+ "raw": {
+ "normalized_reference": "he kept talking and i kept nodding and neither of us was listening.",
+ "normalized_hypothesis": "he kept talking and i kept nodding and neither of us was listening",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he kept talking and i kept nodding and neither of us was listening",
+ "normalized_hypothesis": "he kept talking and i kept nodding and neither of us was listening",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0044",
+ "category": "descriptive",
+ "text": "It's one of those places where everyone seems to know each other already.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0044.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT'S ONE OF THOSE PLACES WHERE EVERYONE SEEMS TO KNOW EACH OTHER ALREADY",
+ "raw": {
+ "normalized_reference": "it's one of those places where everyone seems to know each other already.",
+ "normalized_hypothesis": "it's one of those places where everyone seems to know each other already",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it is one of those places where everyone seems to know each other already",
+ "normalized_hypothesis": "it is one of those places where everyone seems to know each other already",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0045",
+ "category": "emotional",
+ "text": "He said he was proud of me and nobody had ever said that to me before.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0045.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE SAID HE WAS PROUD OF ME AND NOBODY HAD EVER SAID THAT TO ME BEFORE",
+ "raw": {
+ "normalized_reference": "he said he was proud of me and nobody had ever said that to me before.",
+ "normalized_hypothesis": "he said he was proud of me and nobody had ever said that to me before",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he said he was proud of me and nobody had ever said that to me before",
+ "normalized_hypothesis": "he said he was proud of me and nobody had ever said that to me before",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0046",
+ "category": "homographs",
+ "text": "The minute detail took a minute to explain clearly.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0046.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE MINUTE DETAIL TOOK A MINUTE TO EXPLAIN CLEARLY",
+ "raw": {
+ "normalized_reference": "the minute detail took a minute to explain clearly.",
+ "normalized_hypothesis": "the minute detail took a minute to explain clearly",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the minute detail took a minute to explain clearly",
+ "normalized_hypothesis": "the minute detail took a minute to explain clearly",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0047",
+ "category": "long",
+ "text": "The strange thing about really wanting something is that getting it never quite feels the way you imagined it would, and not getting it never quite feels as bad.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0047.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE STRANGE THING ABOUT REALLY WANTING SOMETHING IS THAT GETTING IT NEVER QUITE FEELS THE WAY YOU IMAGINED IT WOULD AND NOT GETTING IT NEVER QUITE FEELS AS BAD",
+ "raw": {
+ "normalized_reference": "the strange thing about really wanting something is that getting it never quite feels the way you imagined it would, and not getting it never quite feels as bad.",
+ "normalized_hypothesis": "the strange thing about really wanting something is that getting it never quite feels the way you imagined it would and not getting it never quite feels as bad",
+ "errors": 0,
+ "words": 29,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the strange thing about really wanting something is that getting it never quite feels the way you imagined it would and not getting it never quite feels as bad",
+ "normalized_hypothesis": "the strange thing about really wanting something is that getting it never quite feels the way you imagined it would and not getting it never quite feels as bad",
+ "errors": 0,
+ "words": 29,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0048",
+ "category": "long_range",
+ "text": "Even though the recipe looked simple at first, preparing the sauce required patience, careful timing, and a much larger pan.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0048.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "EVEN THOUGH THE RECIPE LOOKED SIMPLE AT FIRST PREPARING THE SAUCE REQUIRED PATIENT'S CAREFUL TIMING AND A MUCH LARGER PAN",
+ "raw": {
+ "normalized_reference": "even though the recipe looked simple at first, preparing the sauce required patience, careful timing, and a much larger pan.",
+ "normalized_hypothesis": "even though the recipe looked simple at first preparing the sauce required patient's careful timing and a much larger pan",
+ "errors": 1,
+ "words": 20,
+ "wer": 0.05
+ },
+ "semantic": {
+ "normalized_reference": "even though the recipe looked simple at 1st preparing the sauce required patience careful timing and a much larger pan",
+ "normalized_hypothesis": "even though the recipe looked simple at 1st preparing the sauce required patient is careful timing and a much larger pan",
+ "errors": 2,
+ "words": 20,
+ "wer": 0.1
+ }
+ },
+ {
+ "id": "release500_0049",
+ "category": "names_places",
+ "text": "Xiomara met Siobhan near the old library in Worcestershire.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0049.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "ZAMARA MET CHIVONNE NEAR THE OLD LIBRARY IN WORCESTERSHIRE",
+ "raw": {
+ "normalized_reference": "xiomara met siobhan near the old library in worcestershire.",
+ "normalized_hypothesis": "zamara met chivonne near the old library in worcestershire",
+ "errors": 2,
+ "words": 9,
+ "wer": 0.2222222222222222
+ },
+ "semantic": {
+ "normalized_reference": "xiomara met siobhan near the old library in worcestershire",
+ "normalized_hypothesis": "zamara met chivonne near the old library in worcestershire",
+ "errors": 2,
+ "words": 9,
+ "wer": 0.2222222222222222
+ }
+ },
+ {
+ "id": "release500_0050",
+ "category": "narrative",
+ "text": "She walked into the room and everyone went quiet, not because she was loud, but because she wasn't.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0050.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SHE WALKED INTO THE ROOM AND EVERYONE WENT QUIET NOT BECAUSE SHE WAS LOUD BUT BECAUSE SHE WASN'T",
+ "raw": {
+ "normalized_reference": "she walked into the room and everyone went quiet, not because she was loud, but because she wasn't.",
+ "normalized_hypothesis": "she walked into the room and everyone went quiet not because she was loud but because she wasn't",
+ "errors": 0,
+ "words": 18,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "she walked into the room and everyone went quiet not because she was loud but because she was not",
+ "normalized_hypothesis": "she walked into the room and everyone went quiet not because she was loud but because she was not",
+ "errors": 0,
+ "words": 19,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0051",
+ "category": "numbers",
+ "text": "At exactly seven forty-five, Morgan placed eighteen dollars and fifty cents beside the notebook.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0051.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "AT EXACTLY SEVEN FORTY FIVE MORGAN PLACED EIGHTEEN DOLLARS AND FIFTY CENTS BESIDE THE NOTE BOOK",
+ "raw": {
+ "normalized_reference": "at exactly seven forty-five, morgan placed eighteen dollars and fifty cents beside the notebook.",
+ "normalized_hypothesis": "at exactly seven forty five morgan placed eighteen dollars and fifty cents beside the note book",
+ "errors": 2,
+ "words": 15,
+ "wer": 0.13333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "at exactly 745 morgan placed $18.50 beside the notebook",
+ "normalized_hypothesis": "at exactly 745 morgan placed $18.50 beside the note book",
+ "errors": 2,
+ "words": 10,
+ "wer": 0.2
+ }
+ },
+ {
+ "id": "release500_0052",
+ "category": "punctuation",
+ "text": "Does punctuation work? Yes: commas, semicolons, and periods should create sensible boundaries.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0052.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "DOES PUNCTUATION WORK YES CAMMAS SEMICOLANS AND PERIODS SHOULD CREATE SENSIBLE BOUNDARIES",
+ "raw": {
+ "normalized_reference": "does punctuation work? yes: commas, semicolons, and periods should create sensible boundaries.",
+ "normalized_hypothesis": "does punctuation work yes cammas semicolans and periods should create sensible boundaries",
+ "errors": 2,
+ "words": 12,
+ "wer": 0.16666666666666666
+ },
+ "semantic": {
+ "normalized_reference": "does punctuation work yes commas semicolons and periods should create sensible boundaries",
+ "normalized_hypothesis": "does punctuation work yes cammas semicolans and periods should create sensible boundaries",
+ "errors": 2,
+ "words": 12,
+ "wer": 0.16666666666666666
+ }
+ },
+ {
+ "id": "release500_0053",
+ "category": "question",
+ "text": "Have you ever just stopped and thought about how strange all of this is?",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0053.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HAVE YOU EVER JUST STOPPED AND THOUGHT ABOUT HOW STRANGE ALL OF THIS IS",
+ "raw": {
+ "normalized_reference": "have you ever just stopped and thought about how strange all of this is?",
+ "normalized_hypothesis": "have you ever just stopped and thought about how strange all of this is",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "have you ever just stopped and thought about how strange all of this is",
+ "normalized_hypothesis": "have you ever just stopped and thought about how strange all of this is",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0054",
+ "category": "rare_words",
+ "text": "Her idiosyncratic explanation sounded plausible but remained fundamentally ambiguous.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0054.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HER IDIOSYNCRATIC EXPLANATION SOUNDED PLAUSIBLE BUT REMAINED FUNDAMENTALLY AMBIGUOUS",
+ "raw": {
+ "normalized_reference": "her idiosyncratic explanation sounded plausible but remained fundamentally ambiguous.",
+ "normalized_hypothesis": "her idiosyncratic explanation sounded plausible but remained fundamentally ambiguous",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "her idiosyncratic explanation sounded plausible but remained fundamentally ambiguous",
+ "normalized_hypothesis": "her idiosyncratic explanation sounded plausible but remained fundamentally ambiguous",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0055",
+ "category": "short",
+ "text": "That tracks.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0055.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THAT TRACKS",
+ "raw": {
+ "normalized_reference": "that tracks.",
+ "normalized_hypothesis": "that tracks",
+ "errors": 0,
+ "words": 2,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "that tracks",
+ "normalized_hypothesis": "that tracks",
+ "errors": 0,
+ "words": 2,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0056",
+ "category": "technical",
+ "text": "The key thing to remember is that order matters here more than you'd think.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0056.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE KEY THING TO REMEMBER IS THAT ORDER MATTERS HERE MORE THAN YOU'D THINK",
+ "raw": {
+ "normalized_reference": "the key thing to remember is that order matters here more than you'd think.",
+ "normalized_hypothesis": "the key thing to remember is that order matters here more than you'd think",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the key thing to remember is that order matters here more than you would think",
+ "normalized_hypothesis": "the key thing to remember is that order matters here more than you would think",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0057",
+ "category": "conversational",
+ "text": "I would not say that again in this room if I were you.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0057.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I WOULD NOT SAY THAT AGAIN IN THIS ROOM IF I WERE YOU",
+ "raw": {
+ "normalized_reference": "i would not say that again in this room if i were you.",
+ "normalized_hypothesis": "i would not say that again in this room if i were you",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i would not say that again in this room if i were you",
+ "normalized_hypothesis": "i would not say that again in this room if i were you",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0058",
+ "category": "descriptive",
+ "text": "The room was filled with the quiet that means something is about to change.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0058.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE ROOM WAS FILLED WITH THE QUIET THAT MEANS SOMETHING IS ABOUT TO CHANGE",
+ "raw": {
+ "normalized_reference": "the room was filled with the quiet that means something is about to change.",
+ "normalized_hypothesis": "the room was filled with the quiet that means something is about to change",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the room was filled with the quiet that means something is about to change",
+ "normalized_hypothesis": "the room was filled with the quiet that means something is about to change",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0059",
+ "category": "emotional",
+ "text": "I didn't expect to feel that way about it, but I did.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0059.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I DIDN'T EXPECT TO FEEL THAT WAY ABOUT IT BUT I DID",
+ "raw": {
+ "normalized_reference": "i didn't expect to feel that way about it, but i did.",
+ "normalized_hypothesis": "i didn't expect to feel that way about it but i did",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i did not expect to feel that way about it but i did",
+ "normalized_hypothesis": "i did not expect to feel that way about it but i did",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0060",
+ "category": "homographs",
+ "text": "I object to the strange object that was left near the entrance.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0060.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I OBJECT TO THE STRANGE OBJECT THAT WAS LEFT NEAR THE ENTRANCE",
+ "raw": {
+ "normalized_reference": "i object to the strange object that was left near the entrance.",
+ "normalized_hypothesis": "i object to the strange object that was left near the entrance",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i object to the strange object that was left near the entrance",
+ "normalized_hypothesis": "i object to the strange object that was left near the entrance",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0061",
+ "category": "long",
+ "text": "I've learned more from the things that went wrong than from the things that worked, which is a frustrating way to have to learn things but apparently the only way that stuck.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0061.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I'VE LEARNED MORE FROM THE THINGS THAT WENT WRUN THAN FROM THE THINGS THAT WORKED WHICH IS A FRUSTRATING WAY TO HAVE TO LEARN THINGS BUT APPARENTLY THE ONLY WAY THAT STUT",
+ "raw": {
+ "normalized_reference": "i've learned more from the things that went wrong than from the things that worked, which is a frustrating way to have to learn things but apparently the only way that stuck.",
+ "normalized_hypothesis": "i've learned more from the things that went wrun than from the things that worked which is a frustrating way to have to learn things but apparently the only way that stut",
+ "errors": 2,
+ "words": 32,
+ "wer": 0.0625
+ },
+ "semantic": {
+ "normalized_reference": "i have learned more from the things that went wrong than from the things that worked which is a frustrating way to have to learn things but apparently the only way that stuck",
+ "normalized_hypothesis": "i have learned more from the things that went wrun than from the things that worked which is a frustrating way to have to learn things but apparently the only way that stut",
+ "errors": 2,
+ "words": 33,
+ "wer": 0.06060606060606061
+ }
+ },
+ {
+ "id": "release500_0062",
+ "category": "long_range",
+ "text": "Because every nearby hotel was full, the exhausted hikers followed the narrow road until they found a farmhouse with one empty room.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0062.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "BECAUSE EVERY NEARBY HOTEL WAS FULL THE EXHAUSTED HIKERS FOLLOWED THE NARROW ROAD UNTIL THEY FOUND A FARMHOUSE WITH ONE EMPTY ROOM",
+ "raw": {
+ "normalized_reference": "because every nearby hotel was full, the exhausted hikers followed the narrow road until they found a farmhouse with one empty room.",
+ "normalized_hypothesis": "because every nearby hotel was full the exhausted hikers followed the narrow road until they found a farmhouse with one empty room",
+ "errors": 0,
+ "words": 22,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "because every nearby hotel was full the exhausted hikers followed the narrow road until they found a farmhouse with one empty room",
+ "normalized_hypothesis": "because every nearby hotel was full the exhausted hikers followed the narrow road until they found a farmhouse with one empty room",
+ "errors": 0,
+ "words": 22,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0063",
+ "category": "names_places",
+ "text": "The itinerary sends Beatrice from Albuquerque to Des Moines and then to Boise.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0063.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE ITINERARY SENDS BEATRICE FROM ALBAKIRKI TO DEMOIN AND THEN TO BOESI",
+ "raw": {
+ "normalized_reference": "the itinerary sends beatrice from albuquerque to des moines and then to boise.",
+ "normalized_hypothesis": "the itinerary sends beatrice from albakirki to demoin and then to boesi",
+ "errors": 4,
+ "words": 13,
+ "wer": 0.3076923076923077
+ },
+ "semantic": {
+ "normalized_reference": "the itinerary sends beatrice from albuquerque to des moines and then to boise",
+ "normalized_hypothesis": "the itinerary sends beatrice from albakirki to demoin and then to boesi",
+ "errors": 4,
+ "words": 13,
+ "wer": 0.3076923076923077
+ }
+ },
+ {
+ "id": "release500_0064",
+ "category": "narrative",
+ "text": "We had been driving for six hours when we finally saw the sign for the town.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0064.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WE HAD BEEN DRIVING FOR SIX HOURS WHEN WE FINALLY SAW THE SIGN FOR THE TOWN",
+ "raw": {
+ "normalized_reference": "we had been driving for six hours when we finally saw the sign for the town.",
+ "normalized_hypothesis": "we had been driving for six hours when we finally saw the sign for the town",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "we had been driving for 6 hours when we finally saw the sign for the town",
+ "normalized_hypothesis": "we had been driving for 6 hours when we finally saw the sign for the town",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0065",
+ "category": "numbers",
+ "text": "Call extension zero four nine after eight thirty in the morning.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0065.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "CALL EXTENSION ZERO FOUR NINE AFTER EIGHT THIRTY IN THE MORNING",
+ "raw": {
+ "normalized_reference": "call extension zero four nine after eight thirty in the morning.",
+ "normalized_hypothesis": "call extension zero four nine after eight thirty in the morning",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "call extension 049 after 830 in the morning",
+ "normalized_hypothesis": "call extension 049 after 830 in the morning",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0066",
+ "category": "punctuation",
+ "text": "Stop - listen carefully - then answer the question without guessing.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0066.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "STOP LISTEN CAREFULLY THEN ANSWER THE QUESTION WITHOUT GUESSING",
+ "raw": {
+ "normalized_reference": "stop - listen carefully - then answer the question without guessing.",
+ "normalized_hypothesis": "stop listen carefully then answer the question without guessing",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "stop listen carefully then answer the question without guessing",
+ "normalized_hypothesis": "stop listen carefully then answer the question without guessing",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0067",
+ "category": "question",
+ "text": "Does it matter at this point, honestly?",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0067.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "DOES IT MATTER AT THIS POINT HONESTLY",
+ "raw": {
+ "normalized_reference": "does it matter at this point, honestly?",
+ "normalized_hypothesis": "does it matter at this point honestly",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "does it matter at this point honestly",
+ "normalized_hypothesis": "does it matter at this point honestly",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0068",
+ "category": "rare_words",
+ "text": "An entrepreneurial meteorologist described the unprecedented atmospheric phenomenon.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0068.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "AN ONTROPRENURIAL METEOROLOGIST DESCRIBED THE UNPRECEDENTED ATMOSPHERIC PHENOMENON",
+ "raw": {
+ "normalized_reference": "an entrepreneurial meteorologist described the unprecedented atmospheric phenomenon.",
+ "normalized_hypothesis": "an ontroprenurial meteorologist described the unprecedented atmospheric phenomenon",
+ "errors": 1,
+ "words": 8,
+ "wer": 0.125
+ },
+ "semantic": {
+ "normalized_reference": "an entrepreneurial meteorologist described the unprecedented atmospheric phenomenon",
+ "normalized_hypothesis": "an ontroprenurial meteorologist described the unprecedented atmospheric phenomenon",
+ "errors": 1,
+ "words": 8,
+ "wer": 0.125
+ }
+ },
+ {
+ "id": "release500_0069",
+ "category": "short",
+ "text": "That's not what I expected.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0069.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THAT'S NOT WHAT I EXPECTED",
+ "raw": {
+ "normalized_reference": "that's not what i expected.",
+ "normalized_hypothesis": "that's not what i expected",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "that is not what i expected",
+ "normalized_hypothesis": "that is not what i expected",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0070",
+ "category": "technical",
+ "text": "There's a known issue with that version that causes failures under load.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0070.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THERE'S A KNOWN ISSUE WITH THAT VERSION THAT CAUSES FAILURES UNDER LOAD",
+ "raw": {
+ "normalized_reference": "there's a known issue with that version that causes failures under load.",
+ "normalized_hypothesis": "there's a known issue with that version that causes failures under load",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "there is a known issue with that version that causes failures under load",
+ "normalized_hypothesis": "there is a known issue with that version that causes failures under load",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0071",
+ "category": "conversational",
+ "text": "Don't look now but I think that's him over there.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0071.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "DON'T LOOK NOW BUT I THINK THAT'S HIM OVER THERE",
+ "raw": {
+ "normalized_reference": "don't look now but i think that's him over there.",
+ "normalized_hypothesis": "don't look now but i think that's him over there",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "do not look now but i think that is him over there",
+ "normalized_hypothesis": "do not look now but i think that is him over there",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0072",
+ "category": "descriptive",
+ "text": "The city at night looks completely different from how it feels during the day.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0072.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE CITY AT NIGHT LOOKS COMPLETELY DIFFERENT FROM HOW IT FEELS DURING THE DAY",
+ "raw": {
+ "normalized_reference": "the city at night looks completely different from how it feels during the day.",
+ "normalized_hypothesis": "the city at night looks completely different from how it feels during the day",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the city at night looks completely different from how it feels during the day",
+ "normalized_hypothesis": "the city at night looks completely different from how it feels during the day",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0073",
+ "category": "emotional",
+ "text": "It's the kind of happy that makes you nervous because you know it can't last.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0073.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT'S THE KIND OF HAPPY THAT MAKES YOU NERVOUS BECAUSE YOU KNOW IT CAN'T LAST",
+ "raw": {
+ "normalized_reference": "it's the kind of happy that makes you nervous because you know it can't last.",
+ "normalized_hypothesis": "it's the kind of happy that makes you nervous because you know it can't last",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it is the kind of happy that makes you nervous because you know it can not last",
+ "normalized_hypothesis": "it is the kind of happy that makes you nervous because you know it can not last",
+ "errors": 0,
+ "words": 17,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0074",
+ "category": "homographs",
+ "text": "The dove dove beneath the branch while the wind began to wind the loose cable.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0074.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE DOVE DUP BENEATH THE BRANCH WHILE THE WIND BEGAN TO WIND THE LOOSE CABLE",
+ "raw": {
+ "normalized_reference": "the dove dove beneath the branch while the wind began to wind the loose cable.",
+ "normalized_hypothesis": "the dove dup beneath the branch while the wind began to wind the loose cable",
+ "errors": 1,
+ "words": 15,
+ "wer": 0.06666666666666667
+ },
+ "semantic": {
+ "normalized_reference": "the dove dove beneath the branch while the wind began to wind the loose cable",
+ "normalized_hypothesis": "the dove dup beneath the branch while the wind began to wind the loose cable",
+ "errors": 1,
+ "words": 15,
+ "wer": 0.06666666666666667
+ }
+ },
+ {
+ "id": "release500_0075",
+ "category": "long",
+ "text": "I've thought a lot about what I would do differently if I could go back, and the honest answer is probably not as much as I'd like to think.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0075.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I'VE THOUGHT A LOT ABOUT WHAT I WOULD DO DIFFERENTLY IF I COULD GO BACK AND THE HONEST ANSWER IS PROBABLY NOT AS MUCH AS I'D LIKE TO THINK",
+ "raw": {
+ "normalized_reference": "i've thought a lot about what i would do differently if i could go back, and the honest answer is probably not as much as i'd like to think.",
+ "normalized_hypothesis": "i've thought a lot about what i would do differently if i could go back and the honest answer is probably not as much as i'd like to think",
+ "errors": 0,
+ "words": 29,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i have thought a lot about what i would do differently if i could go back and the honest answer is probably not as much as i would like to think",
+ "normalized_hypothesis": "i have thought a lot about what i would do differently if i could go back and the honest answer is probably not as much as i would like to think",
+ "errors": 0,
+ "words": 31,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0076",
+ "category": "long_range",
+ "text": "Although the weather changed without warning, the musicians finished packing before anyone opened the enormous glass door.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0076.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "ALTHOUGH THE WEATHER CHANGED WITHOUT WARNING THE MUSICIANS FINISHED PACKING BEFORE ANYONE OPENED THE ENORMOUS GLASS DOOR",
+ "raw": {
+ "normalized_reference": "although the weather changed without warning, the musicians finished packing before anyone opened the enormous glass door.",
+ "normalized_hypothesis": "although the weather changed without warning the musicians finished packing before anyone opened the enormous glass door",
+ "errors": 0,
+ "words": 17,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "although the weather changed without warning the musicians finished packing before anyone opened the enormous glass door",
+ "normalized_hypothesis": "although the weather changed without warning the musicians finished packing before anyone opened the enormous glass door",
+ "errors": 0,
+ "words": 17,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0077",
+ "category": "names_places",
+ "text": "Saskatchewan looked unusually bright beneath the pale February sunrise.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0077.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SISCACHUAN LOOKED UNUSUALLY BRIGHT BENEATH THE PALE FEBRUARY SUNRISE",
+ "raw": {
+ "normalized_reference": "saskatchewan looked unusually bright beneath the pale february sunrise.",
+ "normalized_hypothesis": "siscachuan looked unusually bright beneath the pale february sunrise",
+ "errors": 1,
+ "words": 9,
+ "wer": 0.1111111111111111
+ },
+ "semantic": {
+ "normalized_reference": "saskatchewan looked unusually bright beneath the pale february sunrise",
+ "normalized_hypothesis": "siscachuan looked unusually bright beneath the pale february sunrise",
+ "errors": 1,
+ "words": 9,
+ "wer": 0.1111111111111111
+ }
+ },
+ {
+ "id": "release500_0078",
+ "category": "narrative",
+ "text": "She handled it with a grace I genuinely did not expect.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0078.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SHE HANDLED IT WITH A GRACE I GENUINELY DID NOT EXPECT",
+ "raw": {
+ "normalized_reference": "she handled it with a grace i genuinely did not expect.",
+ "normalized_hypothesis": "she handled it with a grace i genuinely did not expect",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "she handled it with a grace i genuinely did not expect",
+ "normalized_hypothesis": "she handled it with a grace i genuinely did not expect",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0079",
+ "category": "numbers",
+ "text": "Apartment four hundred seven is across from room twenty-nine on the third floor.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0079.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "APARTMENT FOUR SEVEN IS ACROSS FROM ROOM TWENTY NINE ON THE THIRD FLOOR",
+ "raw": {
+ "normalized_reference": "apartment four hundred seven is across from room twenty-nine on the third floor.",
+ "normalized_hypothesis": "apartment four seven is across from room twenty nine on the third floor",
+ "errors": 1,
+ "words": 14,
+ "wer": 0.07142857142857142
+ },
+ "semantic": {
+ "normalized_reference": "apartment 407 is across from room 29 on the 3rd floor",
+ "normalized_hypothesis": "apartment 47 is across from room 29 on the 3rd floor",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ }
+ },
+ {
+ "id": "release500_0080",
+ "category": "punctuation",
+ "text": "\"Not yet,\" Rowan whispered. \"Give me another minute... please.\"",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0080.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "NOT YET ROWAN WHISPERED GIVE ME ANOTHER MINUTE PLEASE",
+ "raw": {
+ "normalized_reference": "\"not yet,\" rowan whispered. \"give me another minute... please.\"",
+ "normalized_hypothesis": "not yet rowan whispered give me another minute please",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "not yet rowan whispered give me another minute please",
+ "normalized_hypothesis": "not yet rowan whispered give me another minute please",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0081",
+ "category": "question",
+ "text": "What are we actually trying to accomplish here?",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0081.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WHAT ARE WE ACTUALLY TRYING TO ACCOMPLISH HERE",
+ "raw": {
+ "normalized_reference": "what are we actually trying to accomplish here?",
+ "normalized_hypothesis": "what are we actually trying to accomplish here",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "what are we actually trying to accomplish here",
+ "normalized_hypothesis": "what are we actually trying to accomplish here",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0082",
+ "category": "rare_words",
+ "text": "The physician recommended a noninvasive procedure for the recurring neurological symptoms.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0082.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE PHYSICIAN RECOMMENDED A NON ENVASIVE PROCEDURE FOR THE RECURRING NEROLOGICAL SYMPTOMS",
+ "raw": {
+ "normalized_reference": "the physician recommended a noninvasive procedure for the recurring neurological symptoms.",
+ "normalized_hypothesis": "the physician recommended a non envasive procedure for the recurring nerological symptoms",
+ "errors": 3,
+ "words": 11,
+ "wer": 0.2727272727272727
+ },
+ "semantic": {
+ "normalized_reference": "the physician recommended a noninvasive procedure for the recurring neurological symptoms",
+ "normalized_hypothesis": "the physician recommended a non envasive procedure for the recurring nerological symptoms",
+ "errors": 3,
+ "words": 11,
+ "wer": 0.2727272727272727
+ }
+ },
+ {
+ "id": "release500_0083",
+ "category": "short",
+ "text": "Are you serious right now?",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0083.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "ARE YOU SERIOUS RIGHT NOW",
+ "raw": {
+ "normalized_reference": "are you serious right now?",
+ "normalized_hypothesis": "are you serious right now",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "are you serious right now",
+ "normalized_hypothesis": "are you serious right now",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0084",
+ "category": "technical",
+ "text": "The performance improvement is real, but it comes with tradeoffs.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0084.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE PERFORMANCE IMPROVEMENT IS REAL BUT IT COMES WITH TRADEUFFS",
+ "raw": {
+ "normalized_reference": "the performance improvement is real, but it comes with tradeoffs.",
+ "normalized_hypothesis": "the performance improvement is real but it comes with tradeuffs",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ },
+ "semantic": {
+ "normalized_reference": "the performance improvement is real but it comes with tradeoffs",
+ "normalized_hypothesis": "the performance improvement is real but it comes with tradeuffs",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ }
+ },
+ {
+ "id": "release500_0085",
+ "category": "conversational",
+ "text": "I don't know. It's just been one of those weeks, you know?",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0085.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I DON'T KNOW IT'S JUST BEEN ONE OF THOSE WEEKS YOU KNOW",
+ "raw": {
+ "normalized_reference": "i don't know. it's just been one of those weeks, you know?",
+ "normalized_hypothesis": "i don't know it's just been one of those weeks you know",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i do not know it is just been one of those weeks you know",
+ "normalized_hypothesis": "i do not know it is just been one of those weeks you know",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0086",
+ "category": "descriptive",
+ "text": "The moment you stop expecting perfection, things get a lot more interesting.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0086.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE MOMENT YOU STOP EXPECTING PERFECTION THINGS GET A LOT MORE INTERESTING",
+ "raw": {
+ "normalized_reference": "the moment you stop expecting perfection, things get a lot more interesting.",
+ "normalized_hypothesis": "the moment you stop expecting perfection things get a lot more interesting",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the moment you stop expecting perfection things get a lot more interesting",
+ "normalized_hypothesis": "the moment you stop expecting perfection things get a lot more interesting",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0087",
+ "category": "emotional",
+ "text": "It's the kind of thing that stays with you even years later.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0087.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT'S THE KIND OF THING THAT STAYS WITH YOU EVEN YEARS LATER",
+ "raw": {
+ "normalized_reference": "it's the kind of thing that stays with you even years later.",
+ "normalized_hypothesis": "it's the kind of thing that stays with you even years later",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it is the kind of thing that stays with you even years later",
+ "normalized_hypothesis": "it is the kind of thing that stays with you even years later",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0088",
+ "category": "long",
+ "text": "She told me once that the most important skill she'd developed was learning to disagree with someone without making them feel they were wrong.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0088.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SHE TOLD ME ONCE THAT THE MOST IMPORTANT SKILL SHE DEVELOPED WAS LEARNING TO DISAGREE WITH SOMEONE WITHOUT MAKING THEM FEEL THEY WERE WRONG",
+ "raw": {
+ "normalized_reference": "she told me once that the most important skill she'd developed was learning to disagree with someone without making them feel they were wrong.",
+ "normalized_hypothesis": "she told me once that the most important skill she developed was learning to disagree with someone without making them feel they were wrong",
+ "errors": 1,
+ "words": 24,
+ "wer": 0.041666666666666664
+ },
+ "semantic": {
+ "normalized_reference": "she told me once that the most important skill she would developed was learning to disagree with someone without making them feel they were wrong",
+ "normalized_hypothesis": "she told me once that the most important skill she developed was learning to disagree with someone without making them feel they were wrong",
+ "errors": 1,
+ "words": 25,
+ "wer": 0.04
+ }
+ },
+ {
+ "id": "release500_0089",
+ "category": "narrative",
+ "text": "He stood at the edge of the cliff and wondered, not for the first time, how he'd gotten here.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0089.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE STOOD AT THE EDGE OF THE CLIFF AND WONDERED NOT FOR THE FIRST TIME HOW HE'D GOTTEN HERE",
+ "raw": {
+ "normalized_reference": "he stood at the edge of the cliff and wondered, not for the first time, how he'd gotten here.",
+ "normalized_hypothesis": "he stood at the edge of the cliff and wondered not for the first time how he'd gotten here",
+ "errors": 0,
+ "words": 19,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he stood at the edge of the cliff and wondered not for the 1st time how he would gotten here",
+ "normalized_hypothesis": "he stood at the edge of the cliff and wondered not for the 1st time how he would gotten here",
+ "errors": 0,
+ "words": 20,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0090",
+ "category": "question",
+ "text": "Why does asking for help feel so hard when you actually need it?",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0090.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WHY DOES ASKING FOR HELP FIELD SO HARD WHEN YOU ACTUALLY NEED IT",
+ "raw": {
+ "normalized_reference": "why does asking for help feel so hard when you actually need it?",
+ "normalized_hypothesis": "why does asking for help field so hard when you actually need it",
+ "errors": 1,
+ "words": 13,
+ "wer": 0.07692307692307693
+ },
+ "semantic": {
+ "normalized_reference": "why does asking for help feel so hard when you actually need it",
+ "normalized_hypothesis": "why does asking for help field so hard when you actually need it",
+ "errors": 1,
+ "words": 13,
+ "wer": 0.07692307692307693
+ }
+ },
+ {
+ "id": "release500_0091",
+ "category": "short",
+ "text": "That's fair.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0091.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THAT'S FAIR",
+ "raw": {
+ "normalized_reference": "that's fair.",
+ "normalized_hypothesis": "that's fair",
+ "errors": 0,
+ "words": 2,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "that is fair",
+ "normalized_hypothesis": "that is fair",
+ "errors": 0,
+ "words": 3,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0092",
+ "category": "technical",
+ "text": "If you run into that error, the fix is usually just restarting the service.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0092.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IF YOU RUN INTO THAT ERROR THE FIX IS USUALLY JUST RESTARTING THIS SERVICE",
+ "raw": {
+ "normalized_reference": "if you run into that error, the fix is usually just restarting the service.",
+ "normalized_hypothesis": "if you run into that error the fix is usually just restarting this service",
+ "errors": 1,
+ "words": 14,
+ "wer": 0.07142857142857142
+ },
+ "semantic": {
+ "normalized_reference": "if you run into that error the fix is usually just restarting the service",
+ "normalized_hypothesis": "if you run into that error the fix is usually just restarting this service",
+ "errors": 1,
+ "words": 14,
+ "wer": 0.07142857142857142
+ }
+ },
+ {
+ "id": "release500_0093",
+ "category": "conversational",
+ "text": "It was a disaster in the best possible way.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0093.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT WAS A DISASTER IN THE BEST POSSIBLE WAY",
+ "raw": {
+ "normalized_reference": "it was a disaster in the best possible way.",
+ "normalized_hypothesis": "it was a disaster in the best possible way",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it was a disaster in the best possible way",
+ "normalized_hypothesis": "it was a disaster in the best possible way",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0094",
+ "category": "descriptive",
+ "text": "It had the look of a place that used to mean something to someone.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0094.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT HAD THE LOOK OF A PLACE THAT USED TO MEAN SOMETHING TO SOMEONE",
+ "raw": {
+ "normalized_reference": "it had the look of a place that used to mean something to someone.",
+ "normalized_hypothesis": "it had the look of a place that used to mean something to someone",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it had the look of a place that used to mean something to someone",
+ "normalized_hypothesis": "it had the look of a place that used to mean something to someone",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0095",
+ "category": "emotional",
+ "text": "There's this feeling you get when something finally works after weeks of failure.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0095.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THERE'S THIS FEELING YOU GET WHEN SOMETHING FINALLY WORKS AFTER WEEKS OF FAILURE",
+ "raw": {
+ "normalized_reference": "there's this feeling you get when something finally works after weeks of failure.",
+ "normalized_hypothesis": "there's this feeling you get when something finally works after weeks of failure",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "there is this feeling you get when something finally works after weeks of failure",
+ "normalized_hypothesis": "there is this feeling you get when something finally works after weeks of failure",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0096",
+ "category": "long",
+ "text": "He had a theory that most arguments aren't actually about what they seem to be about, and the longer I live the more I think he was onto something.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0096.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE HAD A THEORY THAT MOST ARGUMENTS AREN'T ACTUALLY ABOUT WHAT THEY SEEM TO BE ABOUT AND THE LONGER I LIVE THE MORE I THINK HE WAS ON TO SOMETHING",
+ "raw": {
+ "normalized_reference": "he had a theory that most arguments aren't actually about what they seem to be about, and the longer i live the more i think he was onto something.",
+ "normalized_hypothesis": "he had a theory that most arguments aren't actually about what they seem to be about and the longer i live the more i think he was on to something",
+ "errors": 2,
+ "words": 29,
+ "wer": 0.06896551724137931
+ },
+ "semantic": {
+ "normalized_reference": "he had a theory that most arguments are not actually about what they seem to be about and the longer i live the more i think he was onto something",
+ "normalized_hypothesis": "he had a theory that most arguments are not actually about what they seem to be about and the longer i live the more i think he was on to something",
+ "errors": 2,
+ "words": 30,
+ "wer": 0.06666666666666667
+ }
+ },
+ {
+ "id": "release500_0097",
+ "category": "narrative",
+ "text": "The thing nobody tells you about starting over is how ordinary most of it feels.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0097.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE THING NOBODY TELLS YOU ABOUT STARTING OVER IS HOW ORDINARY MOST OF IT FEELS",
+ "raw": {
+ "normalized_reference": "the thing nobody tells you about starting over is how ordinary most of it feels.",
+ "normalized_hypothesis": "the thing nobody tells you about starting over is how ordinary most of it feels",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the thing nobody tells you about starting over is how ordinary most of it feels",
+ "normalized_hypothesis": "the thing nobody tells you about starting over is how ordinary most of it feels",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0098",
+ "category": "question",
+ "text": "What exactly are we waiting for at this point?",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0098.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WHAT EXACTLY ARE WE WAITING FOR AT THIS POINT",
+ "raw": {
+ "normalized_reference": "what exactly are we waiting for at this point?",
+ "normalized_hypothesis": "what exactly are we waiting for at this point",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "what exactly are we waiting for at this point",
+ "normalized_hypothesis": "what exactly are we waiting for at this point",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0099",
+ "category": "short",
+ "text": "I stand by what I said.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0099.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I STAND BY WHAT I SAID",
+ "raw": {
+ "normalized_reference": "i stand by what i said.",
+ "normalized_hypothesis": "i stand by what i said",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i stand by what i said",
+ "normalized_hypothesis": "i stand by what i said",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0100",
+ "category": "technical",
+ "text": "Double-check that value before you push anything to production.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0100.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "DOUBLE CHECK THAT VALUE BEFORE YOU PUSH ANYTHING TO PRODUCTION",
+ "raw": {
+ "normalized_reference": "double-check that value before you push anything to production.",
+ "normalized_hypothesis": "double check that value before you push anything to production",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "double check that value before you push anything to production",
+ "normalized_hypothesis": "double check that value before you push anything to production",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0101",
+ "category": "conversational",
+ "text": "I understand your point; I just do not agree with the conclusion.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0101.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I UNDERSTAND YOUR POINT I JUST DO NOT AGREE WITH THE CONCLUSION",
+ "raw": {
+ "normalized_reference": "i understand your point; i just do not agree with the conclusion.",
+ "normalized_hypothesis": "i understand your point i just do not agree with the conclusion",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i understand your point i just do not agree with the conclusion",
+ "normalized_hypothesis": "i understand your point i just do not agree with the conclusion",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0102",
+ "category": "descriptive",
+ "text": "The sky that day was a strange shade of orange I'd never seen before.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0102.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE SKY THAT DAY WAS A STRANGE SHADE OF ORANGE I'D NEVER SEEN BEFORE",
+ "raw": {
+ "normalized_reference": "the sky that day was a strange shade of orange i'd never seen before.",
+ "normalized_hypothesis": "the sky that day was a strange shade of orange i'd never seen before",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the sky that day was a strange shade of orange i would never seen before",
+ "normalized_hypothesis": "the sky that day was a strange shade of orange i would never seen before",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0103",
+ "category": "emotional",
+ "text": "It hurt more than I expected, and I thought I was prepared.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0103.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT HURT MORE THAN I EXPECTED AND I THOUGHT I WAS PREPARED",
+ "raw": {
+ "normalized_reference": "it hurt more than i expected, and i thought i was prepared.",
+ "normalized_hypothesis": "it hurt more than i expected and i thought i was prepared",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it hurt more than i expected and i thought i was prepared",
+ "normalized_hypothesis": "it hurt more than i expected and i thought i was prepared",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0104",
+ "category": "long",
+ "text": "The hardest part wasn't making the decision, it was sitting with the uncertainty of not knowing whether it was the right one for months afterward.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0104.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE HARDEST PART WASN'T MAKING THE DECISION IT WAS SITTING WITH THE UNCERTAINTY OF NOT KNOWING WHETHER IT WAS THE RIGHT ONE FOR MONTHS AFTERWARD",
+ "raw": {
+ "normalized_reference": "the hardest part wasn't making the decision, it was sitting with the uncertainty of not knowing whether it was the right one for months afterward.",
+ "normalized_hypothesis": "the hardest part wasn't making the decision it was sitting with the uncertainty of not knowing whether it was the right one for months afterward",
+ "errors": 0,
+ "words": 25,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the hardest part was not making the decision it was sitting with the uncertainty of not knowing whether it was the right one for months afterward",
+ "normalized_hypothesis": "the hardest part was not making the decision it was sitting with the uncertainty of not knowing whether it was the right one for months afterward",
+ "errors": 0,
+ "words": 26,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0105",
+ "category": "narrative",
+ "text": "There was this one day last summer when everything just clicked into place.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0105.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THERE WAS THIS ONE DAY LAST SUMMER WHEN EVERYTHING JUST CLICKED INTO PLACE",
+ "raw": {
+ "normalized_reference": "there was this one day last summer when everything just clicked into place.",
+ "normalized_hypothesis": "there was this one day last summer when everything just clicked into place",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "there was this one day last summer when everything just clicked into place",
+ "normalized_hypothesis": "there was this one day last summer when everything just clicked into place",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0106",
+ "category": "question",
+ "text": "Have you ever thought about why we keep doing things the same way when they stop working?",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0106.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HAVE YOU EVER THOUGHT ABOUT WHY WE KEEP DOING THINGS THE SAME WAY WHEN THEY STOP WORKING",
+ "raw": {
+ "normalized_reference": "have you ever thought about why we keep doing things the same way when they stop working?",
+ "normalized_hypothesis": "have you ever thought about why we keep doing things the same way when they stop working",
+ "errors": 0,
+ "words": 17,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "have you ever thought about why we keep doing things the same way when they stop working",
+ "normalized_hypothesis": "have you ever thought about why we keep doing things the same way when they stop working",
+ "errors": 0,
+ "words": 17,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0107",
+ "category": "short",
+ "text": "I can't believe you just said that.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0107.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I CAN'T BELIEVE YOU JUST SAID THAT",
+ "raw": {
+ "normalized_reference": "i can't believe you just said that.",
+ "normalized_hypothesis": "i can't believe you just said that",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i can not believe you just said that",
+ "normalized_hypothesis": "i can not believe you just said that",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0108",
+ "category": "technical",
+ "text": "You'll want to isolate that variable before drawing any conclusions.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0108.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "YOU'LL WANT TO ISOLATE THAT VARIABLE BEFORE DRAWING ANY CONCLUSIONS",
+ "raw": {
+ "normalized_reference": "you'll want to isolate that variable before drawing any conclusions.",
+ "normalized_hypothesis": "you'll want to isolate that variable before drawing any conclusions",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "you will want to isolate that variable before drawing any conclusions",
+ "normalized_hypothesis": "you will want to isolate that variable before drawing any conclusions",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0109",
+ "category": "conversational",
+ "text": "The meeting that could have been an email lasted two hours.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0109.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE MEETING THAT COULD HAVE BEEN AN EMALE LASTED TWO HOURS",
+ "raw": {
+ "normalized_reference": "the meeting that could have been an email lasted two hours.",
+ "normalized_hypothesis": "the meeting that could have been an emale lasted two hours",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ },
+ "semantic": {
+ "normalized_reference": "the meeting that could have been an email lasted 2 hours",
+ "normalized_hypothesis": "the meeting that could have been an emale lasted 2 hours",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ }
+ },
+ {
+ "id": "release500_0110",
+ "category": "descriptive",
+ "text": "There's a specific kind of tired that sleep doesn't fix.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0110.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THERE'S A SPECIFIC KIND OF TIRE THAT SLEEP DOESN'T FIX",
+ "raw": {
+ "normalized_reference": "there's a specific kind of tired that sleep doesn't fix.",
+ "normalized_hypothesis": "there's a specific kind of tire that sleep doesn't fix",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ },
+ "semantic": {
+ "normalized_reference": "there is a specific kind of tired that sleep does not fix",
+ "normalized_hypothesis": "there is a specific kind of tire that sleep does not fix",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ }
+ },
+ {
+ "id": "release500_0111",
+ "category": "emotional",
+ "text": "Something about that moment broke something open in me that I hadn't realized was closed.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0111.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SOMETHING ABOUT THAT MOMENT BROKE SOMETHING OPEN IN ME THAT I HADN'T REALIZED WAS CLOSED",
+ "raw": {
+ "normalized_reference": "something about that moment broke something open in me that i hadn't realized was closed.",
+ "normalized_hypothesis": "something about that moment broke something open in me that i hadn't realized was closed",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "something about that moment broke something open in me that i had not realized was closed",
+ "normalized_hypothesis": "something about that moment broke something open in me that i had not realized was closed",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0112",
+ "category": "long",
+ "text": "The thing about making something good is that you don't usually know it's good while you're making it. You just know you're not done yet.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0112.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE THING ABOUT MAKING SOMETHING GOOD IS THAT YOU DON'T USUALLY KNOW IT'S GOOD WHILE YOU'RE MAKING IT YOU JUST KNOW YOU'RE NOT DONE YET",
+ "raw": {
+ "normalized_reference": "the thing about making something good is that you don't usually know it's good while you're making it. you just know you're not done yet.",
+ "normalized_hypothesis": "the thing about making something good is that you don't usually know it's good while you're making it you just know you're not done yet",
+ "errors": 0,
+ "words": 25,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the thing about making something good is that you do not usually know it is good while you are making it you just know you are not done yet",
+ "normalized_hypothesis": "the thing about making something good is that you do not usually know it is good while you are making it you just know you are not done yet",
+ "errors": 0,
+ "words": 29,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0113",
+ "category": "narrative",
+ "text": "She opened the door slowly, as if she expected to find something she wasn't ready for.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0113.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SHE OPENED THE DOOR SLOWLY AS IF SHE EXPECTED TO FIND SOMETHING SHE WASN'T READY FOR",
+ "raw": {
+ "normalized_reference": "she opened the door slowly, as if she expected to find something she wasn't ready for.",
+ "normalized_hypothesis": "she opened the door slowly as if she expected to find something she wasn't ready for",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "she opened the door slowly as if she expected to find something she was not ready for",
+ "normalized_hypothesis": "she opened the door slowly as if she expected to find something she was not ready for",
+ "errors": 0,
+ "words": 17,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0114",
+ "category": "question",
+ "text": "What would you do differently if you could go back?",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0114.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WHAT WOULD YOU DO DIFFERENTLY IF YOU COULD GO BACK",
+ "raw": {
+ "normalized_reference": "what would you do differently if you could go back?",
+ "normalized_hypothesis": "what would you do differently if you could go back",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "what would you do differently if you could go back",
+ "normalized_hypothesis": "what would you do differently if you could go back",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0115",
+ "category": "short",
+ "text": "Okay. Okay okay okay.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0115.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "OK OK OK O K",
+ "raw": {
+ "normalized_reference": "okay. okay okay okay.",
+ "normalized_hypothesis": "ok ok ok o k",
+ "errors": 5,
+ "words": 4,
+ "wer": 1.25
+ },
+ "semantic": {
+ "normalized_reference": "okay okay okay okay",
+ "normalized_hypothesis": "ok ok ok 0 k",
+ "errors": 5,
+ "words": 4,
+ "wer": 1.25
+ }
+ },
+ {
+ "id": "release500_0116",
+ "category": "technical",
+ "text": "The latency goes up significantly at scale, which is expected.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0116.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE LATENCY GOES UP SIGNIFICANTLY AT SCALE WHICH IS EXPECTED",
+ "raw": {
+ "normalized_reference": "the latency goes up significantly at scale, which is expected.",
+ "normalized_hypothesis": "the latency goes up significantly at scale which is expected",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the latency goes up significantly at scale which is expected",
+ "normalized_hypothesis": "the latency goes up significantly at scale which is expected",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0117",
+ "category": "conversational",
+ "text": "We ended up staying until two in the morning talking about nothing.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0117.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WE ENDED UP STAYING UNTIL TWO IN THE MORNING TALKING ABOUT NOTHING",
+ "raw": {
+ "normalized_reference": "we ended up staying until two in the morning talking about nothing.",
+ "normalized_hypothesis": "we ended up staying until two in the morning talking about nothing",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "we ended up staying until 2 in the morning talking about nothing",
+ "normalized_hypothesis": "we ended up staying until 2 in the morning talking about nothing",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0118",
+ "category": "descriptive",
+ "text": "The rain started slowly, and then all at once, the way it usually does.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0118.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE RAIN STARTED SLOWLY AND THEN ALL AT ONCE THE WAY IT USUALLY DOES",
+ "raw": {
+ "normalized_reference": "the rain started slowly, and then all at once, the way it usually does.",
+ "normalized_hypothesis": "the rain started slowly and then all at once the way it usually does",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the rain started slowly and then all at once the way it usually does",
+ "normalized_hypothesis": "the rain started slowly and then all at once the way it usually does",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0119",
+ "category": "emotional",
+ "text": "I honestly did not think I was going to make it through that year.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0119.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I HONESTLY DID NOT THINK I WAS GOING TO MAKE IT THROUGH THAT YEAR",
+ "raw": {
+ "normalized_reference": "i honestly did not think i was going to make it through that year.",
+ "normalized_hypothesis": "i honestly did not think i was going to make it through that year",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i honestly did not think i was going to make it through that year",
+ "normalized_hypothesis": "i honestly did not think i was going to make it through that year",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0120",
+ "category": "long",
+ "text": "There's an assumption that people who seem confident all the time actually feel confident, and in my experience that's almost never true.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0120.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THERE'S AN ASSUMPTION THAT PEOPLE WHO SEEM CONFIDENT ALL THE TIME ACTUALLY FEEL CONFIDENT AND IN MY EXPERIENCE THAT'S ALMOST NEVER TRUE",
+ "raw": {
+ "normalized_reference": "there's an assumption that people who seem confident all the time actually feel confident, and in my experience that's almost never true.",
+ "normalized_hypothesis": "there's an assumption that people who seem confident all the time actually feel confident and in my experience that's almost never true",
+ "errors": 0,
+ "words": 22,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "there is an assumption that people who seem confident all the time actually feel confident and in my experience that is almost never true",
+ "normalized_hypothesis": "there is an assumption that people who seem confident all the time actually feel confident and in my experience that is almost never true",
+ "errors": 0,
+ "words": 24,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0121",
+ "category": "narrative",
+ "text": "The moment I opened the door I knew something was different, even if I couldn't say what.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0121.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE MOMENT I OPENED THE DOOR I KNEW SOMETHING WAS DIFFERENT EVEN IF I COULDN'T SAY WHAT",
+ "raw": {
+ "normalized_reference": "the moment i opened the door i knew something was different, even if i couldn't say what.",
+ "normalized_hypothesis": "the moment i opened the door i knew something was different even if i couldn't say what",
+ "errors": 0,
+ "words": 17,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the moment i opened the door i knew something was different even if i could not say what",
+ "normalized_hypothesis": "the moment i opened the door i knew something was different even if i could not say what",
+ "errors": 0,
+ "words": 18,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0122",
+ "category": "question",
+ "text": "How do you explain something that doesn't make sense even to you?",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0122.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HOW DO YOU EXPLAIN SOMETHING THAT DOESN'T MAKE SENSE EVEN TO YOU",
+ "raw": {
+ "normalized_reference": "how do you explain something that doesn't make sense even to you?",
+ "normalized_hypothesis": "how do you explain something that doesn't make sense even to you",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "how do you explain something that does not make sense even to you",
+ "normalized_hypothesis": "how do you explain something that does not make sense even to you",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0123",
+ "category": "short",
+ "text": "Hold on a second.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0123.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HOLD ON A SECOND",
+ "raw": {
+ "normalized_reference": "hold on a second.",
+ "normalized_hypothesis": "hold on a second",
+ "errors": 0,
+ "words": 4,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "hold on a 2nd",
+ "normalized_hypothesis": "hold on a 2nd",
+ "errors": 0,
+ "words": 4,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0124",
+ "category": "technical",
+ "text": "A deterministic split prevents duplicate normalized text from leaking into evaluation.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0124.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "A DETERMINISTIC SPLIT PREVENTS DUPLICATE NORMALIZED TEXT FROM LEAKING INTO EVALUATION",
+ "raw": {
+ "normalized_reference": "a deterministic split prevents duplicate normalized text from leaking into evaluation.",
+ "normalized_hypothesis": "a deterministic split prevents duplicate normalized text from leaking into evaluation",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "a deterministic split prevents duplicate normalized text from leaking into evaluation",
+ "normalized_hypothesis": "a deterministic split prevents duplicate normalized text from leaking into evaluation",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0125",
+ "category": "conversational",
+ "text": "Honestly, that was much better than I expected, but the ending still felt rushed.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0125.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HONESTLY THAT WAS MUCH BETTER THAN I EXPECTED BUT THE ENDING STILL FELT RUSH",
+ "raw": {
+ "normalized_reference": "honestly, that was much better than i expected, but the ending still felt rushed.",
+ "normalized_hypothesis": "honestly that was much better than i expected but the ending still felt rush",
+ "errors": 1,
+ "words": 14,
+ "wer": 0.07142857142857142
+ },
+ "semantic": {
+ "normalized_reference": "honestly that was much better than i expected but the ending still felt rushed",
+ "normalized_hypothesis": "honestly that was much better than i expected but the ending still felt rush",
+ "errors": 1,
+ "words": 14,
+ "wer": 0.07142857142857142
+ }
+ },
+ {
+ "id": "release500_0126",
+ "category": "descriptive",
+ "text": "She had a habit of tilting her head slightly when thinking hard about something.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0126.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SHE HAD A HABIT OF TILTING HER HEAD SLIGHTLY WHEN THINKING HARD ABOUT SOMETHING",
+ "raw": {
+ "normalized_reference": "she had a habit of tilting her head slightly when thinking hard about something.",
+ "normalized_hypothesis": "she had a habit of tilting her head slightly when thinking hard about something",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "she had a habit of tilting her head slightly when thinking hard about something",
+ "normalized_hypothesis": "she had a habit of tilting her head slightly when thinking hard about something",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0127",
+ "category": "emotional",
+ "text": "I cannot stress enough how much I did not want to be right about this.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0127.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I CANNOT STRESS ENOUGH HOW MUCH I DID NOT WANT TO BE RIGHT ABOUT THIS",
+ "raw": {
+ "normalized_reference": "i cannot stress enough how much i did not want to be right about this.",
+ "normalized_hypothesis": "i cannot stress enough how much i did not want to be right about this",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i cannot stress enough how much i did not want to be right about this",
+ "normalized_hypothesis": "i cannot stress enough how much i did not want to be right about this",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0128",
+ "category": "long",
+ "text": "I used to think the goal was to get to a point where nothing was hard anymore, and now I think that point doesn't exist and I'm not sure I'd want it to.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0128.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I USED TO THINK THE GOAL WAS TO GET TO A POINT WHERE NOTHING WAS HARD ANIMORE AND NOW I THINK THAT POINT DOESN'T EXIST AND I'M NOT SURE I WANT IT TO",
+ "raw": {
+ "normalized_reference": "i used to think the goal was to get to a point where nothing was hard anymore, and now i think that point doesn't exist and i'm not sure i'd want it to.",
+ "normalized_hypothesis": "i used to think the goal was to get to a point where nothing was hard animore and now i think that point doesn't exist and i'm not sure i want it to",
+ "errors": 2,
+ "words": 33,
+ "wer": 0.06060606060606061
+ },
+ "semantic": {
+ "normalized_reference": "i used to think the goal was to get to a point where nothing was hard anymore and now i think that point does not exist and i am not sure i would want it to",
+ "normalized_hypothesis": "i used to think the goal was to get to a point where nothing was hard animore and now i think that point does not exist and i am not sure i want it to",
+ "errors": 2,
+ "words": 36,
+ "wer": 0.05555555555555555
+ }
+ },
+ {
+ "id": "release500_0129",
+ "category": "narrative",
+ "text": "So this is the part of the story where everything goes completely sideways.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0129.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SO THIS IS THE PART OF THE STORY WHEREEVERYTHING GOES COMPLETELY SIDEWAYS",
+ "raw": {
+ "normalized_reference": "so this is the part of the story where everything goes completely sideways.",
+ "normalized_hypothesis": "so this is the part of the story whereeverything goes completely sideways",
+ "errors": 2,
+ "words": 13,
+ "wer": 0.15384615384615385
+ },
+ "semantic": {
+ "normalized_reference": "so this is the part of the story where everything goes completely sideways",
+ "normalized_hypothesis": "so this is the part of the story whereeverything goes completely sideways",
+ "errors": 2,
+ "words": 13,
+ "wer": 0.15384615384615385
+ }
+ },
+ {
+ "id": "release500_0130",
+ "category": "question",
+ "text": "Is there a version of this where it actually works out okay?",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0130.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IS THERE A VERSION OF THIS WHERE IT ACTUALLY WORKS OUT OQUET",
+ "raw": {
+ "normalized_reference": "is there a version of this where it actually works out okay?",
+ "normalized_hypothesis": "is there a version of this where it actually works out oquet",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "is there a version of this where it actually works out okay",
+ "normalized_hypothesis": "is there a version of this where it actually works out oquet",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ }
+ },
+ {
+ "id": "release500_0131",
+ "category": "short",
+ "text": "Oh no. Oh no no no.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0131.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "OH NO OH NO NO NO",
+ "raw": {
+ "normalized_reference": "oh no. oh no no no.",
+ "normalized_hypothesis": "oh no oh no no no",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "0 no 0 no no no",
+ "normalized_hypothesis": "0 no 0 no no no",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0132",
+ "category": "technical",
+ "text": "Always test on a copy before modifying the original.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0132.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "ALWAYS TEST ON A COPY BEFORE MODIFYING THE ORIGINAL",
+ "raw": {
+ "normalized_reference": "always test on a copy before modifying the original.",
+ "normalized_hypothesis": "always test on a copy before modifying the original",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "always test on a copy before modifying the original",
+ "normalized_hypothesis": "always test on a copy before modifying the original",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0133",
+ "category": "conversational",
+ "text": "I wasn't even supposed to be there. That's the wild part.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0133.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I WASN'T EVEN SUPPOSED TO BE THERE THAT'S THE WILD PART",
+ "raw": {
+ "normalized_reference": "i wasn't even supposed to be there. that's the wild part.",
+ "normalized_hypothesis": "i wasn't even supposed to be there that's the wild part",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i was not even supposed to be there that is the wild part",
+ "normalized_hypothesis": "i was not even supposed to be there that is the wild part",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0134",
+ "category": "descriptive",
+ "text": "It's a small town. Not the kind you leave. The kind you leave and come back to.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0134.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT'S A SMALL TOWN NOT THE KIND YOU LEAVE THE KIND YOU LEAVE AND COME BACK TO",
+ "raw": {
+ "normalized_reference": "it's a small town. not the kind you leave. the kind you leave and come back to.",
+ "normalized_hypothesis": "it's a small town not the kind you leave the kind you leave and come back to",
+ "errors": 0,
+ "words": 17,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it is a small town not the kind you leave the kind you leave and come back to",
+ "normalized_hypothesis": "it is a small town not the kind you leave the kind you leave and come back to",
+ "errors": 0,
+ "words": 18,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0135",
+ "category": "emotional",
+ "text": "She cried. I cried. We all just sort of stood there crying.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0135.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SHE CRIED I CRIED WE ALL JUST SORT OF STOOD THERE CRYING",
+ "raw": {
+ "normalized_reference": "she cried. i cried. we all just sort of stood there crying.",
+ "normalized_hypothesis": "she cried i cried we all just sort of stood there crying",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "she cried i cried we all just sort of stood there crying",
+ "normalized_hypothesis": "she cried i cried we all just sort of stood there crying",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0136",
+ "category": "long",
+ "text": "There's a version of being busy that's actually just a very elaborate way of avoiding the one thing you know you should be doing.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0136.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THERE'S A VERSION OF BEING BUSY THAT'S ACTUALLY JUST A VERY ELABORATE WAY OF AVOIDING THE ONE THING YOU KNOW YOU SHOULD BE DOING",
+ "raw": {
+ "normalized_reference": "there's a version of being busy that's actually just a very elaborate way of avoiding the one thing you know you should be doing.",
+ "normalized_hypothesis": "there's a version of being busy that's actually just a very elaborate way of avoiding the one thing you know you should be doing",
+ "errors": 0,
+ "words": 24,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "there is a version of being busy that is actually just a very elaborate way of avoiding the one thing you know you should be doing",
+ "normalized_hypothesis": "there is a version of being busy that is actually just a very elaborate way of avoiding the one thing you know you should be doing",
+ "errors": 0,
+ "words": 26,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0137",
+ "category": "narrative",
+ "text": "The letter arrived on a Tuesday, which felt appropriate somehow.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0137.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE LETTER ARRIVED ON A TUESDAY WHICH FELT APPROPRIATE SOMEHOW",
+ "raw": {
+ "normalized_reference": "the letter arrived on a tuesday, which felt appropriate somehow.",
+ "normalized_hypothesis": "the letter arrived on a tuesday which felt appropriate somehow",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the letter arrived on a tuesday which felt appropriate somehow",
+ "normalized_hypothesis": "the letter arrived on a tuesday which felt appropriate somehow",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0138",
+ "category": "question",
+ "text": "What exactly did she say when you told her?",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0138.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WHAT EXACTLY DID SHE SAY WHEN YOU TOLD HER",
+ "raw": {
+ "normalized_reference": "what exactly did she say when you told her?",
+ "normalized_hypothesis": "what exactly did she say when you told her",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "what exactly did she say when you told her",
+ "normalized_hypothesis": "what exactly did she say when you told her",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0139",
+ "category": "short",
+ "text": "No way. Absolutely not.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0139.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "NO WEIT ABSOLUTELY NOT",
+ "raw": {
+ "normalized_reference": "no way. absolutely not.",
+ "normalized_hypothesis": "no weit absolutely not",
+ "errors": 1,
+ "words": 4,
+ "wer": 0.25
+ },
+ "semantic": {
+ "normalized_reference": "no way absolutely not",
+ "normalized_hypothesis": "no weit absolutely not",
+ "errors": 1,
+ "words": 4,
+ "wer": 0.25
+ }
+ },
+ {
+ "id": "release500_0140",
+ "category": "technical",
+ "text": "The first thing you want to do is make sure everything is backed up.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0140.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE FIRST THING YOU WANT TO DO IS MAKE SURE EVERYTHING IS BACKED UP",
+ "raw": {
+ "normalized_reference": "the first thing you want to do is make sure everything is backed up.",
+ "normalized_hypothesis": "the first thing you want to do is make sure everything is backed up",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the 1st thing you want to do is make sure everything is backed up",
+ "normalized_hypothesis": "the 1st thing you want to do is make sure everything is backed up",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0141",
+ "category": "conversational",
+ "text": "I have the most insane idea and you need to just hear me out.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0141.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I HAVE THE MOST INSANE IDEA AND YOU NEED TO JUST HEAR ME OUT",
+ "raw": {
+ "normalized_reference": "i have the most insane idea and you need to just hear me out.",
+ "normalized_hypothesis": "i have the most insane idea and you need to just hear me out",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i have the most insane idea and you need to just hear me out",
+ "normalized_hypothesis": "i have the most insane idea and you need to just hear me out",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0142",
+ "category": "descriptive",
+ "text": "There's a coffee shop on the corner that's been there longer than anyone can remember.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0142.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THERE'S A COFFEE SHOP ON THE CORNER THAT'S BEEN THERE LONGER THAN ANYONE CAN REMEMBER",
+ "raw": {
+ "normalized_reference": "there's a coffee shop on the corner that's been there longer than anyone can remember.",
+ "normalized_hypothesis": "there's a coffee shop on the corner that's been there longer than anyone can remember",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "there is a coffee shop on the corner that has been there longer than anyone can remember",
+ "normalized_hypothesis": "there is a coffee shop on the corner that has been there longer than anyone can remember",
+ "errors": 0,
+ "words": 17,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0143",
+ "category": "emotional",
+ "text": "I kept it together until I got to the car, and then I completely fell apart.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0143.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I KEPT IT TOGETHER UNTIL I GOT TO THE CAR AND THEN I COMPLETELY FELL APART",
+ "raw": {
+ "normalized_reference": "i kept it together until i got to the car, and then i completely fell apart.",
+ "normalized_hypothesis": "i kept it together until i got to the car and then i completely fell apart",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i kept it together until i got to the car and then i completely fell apart",
+ "normalized_hypothesis": "i kept it together until i got to the car and then i completely fell apart",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0144",
+ "category": "long",
+ "text": "The older I get, the more I think that most of what we call wisdom is just pattern recognition built from having been wrong enough times to start noticing.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0144.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE OLDER I GET THE MORE I THINK THAT MOST OF WHAT WE CALL WISDOM IS JUST PATTERN RECOGNITION BUILT FROM HAVING BEEN WRONG ENOUGH TIMES TO START NOTICING",
+ "raw": {
+ "normalized_reference": "the older i get, the more i think that most of what we call wisdom is just pattern recognition built from having been wrong enough times to start noticing.",
+ "normalized_hypothesis": "the older i get the more i think that most of what we call wisdom is just pattern recognition built from having been wrong enough times to start noticing",
+ "errors": 0,
+ "words": 29,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the older i get the more i think that most of what we call wisdom is just pattern recognition built from having been wrong enough times to start noticing",
+ "normalized_hypothesis": "the older i get the more i think that most of what we call wisdom is just pattern recognition built from having been wrong enough times to start noticing",
+ "errors": 0,
+ "words": 29,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0145",
+ "category": "narrative",
+ "text": "I remember the exact moment I realized I had made a huge mistake.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0145.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I REMEMBER THE EXACT MOMENT I REALIZED I HAD MADE A HUGE MISTAKE",
+ "raw": {
+ "normalized_reference": "i remember the exact moment i realized i had made a huge mistake.",
+ "normalized_hypothesis": "i remember the exact moment i realized i had made a huge mistake",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i remember the exact moment i realized i had made a huge mistake",
+ "normalized_hypothesis": "i remember the exact moment i realized i had made a huge mistake",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0146",
+ "category": "question",
+ "text": "Do you think it would have gone differently if we'd left earlier?",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0146.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "DO YOU THINK IT WOULD HAVE GONE DIFFERENTLY IF WE'D LEFT EARLIER",
+ "raw": {
+ "normalized_reference": "do you think it would have gone differently if we'd left earlier?",
+ "normalized_hypothesis": "do you think it would have gone differently if we'd left earlier",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "do you think it would have gone differently if we would left earlier",
+ "normalized_hypothesis": "do you think it would have gone differently if we would left earlier",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0147",
+ "category": "short",
+ "text": "You have to be kidding me.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0147.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "YOU HAVE TO BE KITTING ME",
+ "raw": {
+ "normalized_reference": "you have to be kidding me.",
+ "normalized_hypothesis": "you have to be kitting me",
+ "errors": 1,
+ "words": 6,
+ "wer": 0.16666666666666666
+ },
+ "semantic": {
+ "normalized_reference": "you have to be kidding me",
+ "normalized_hypothesis": "you have to be kitting me",
+ "errors": 1,
+ "words": 6,
+ "wer": 0.16666666666666666
+ }
+ },
+ {
+ "id": "release500_0148",
+ "category": "technical",
+ "text": "If the checksum fails, restart the worker process without deleting the cached tensors.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0148.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IF THE CHEXSUM FAILS RESTART THE WORKER PROCESS WITHOUT DELEDING THE CASH TENSERS",
+ "raw": {
+ "normalized_reference": "if the checksum fails, restart the worker process without deleting the cached tensors.",
+ "normalized_hypothesis": "if the chexsum fails restart the worker process without deleding the cash tensers",
+ "errors": 4,
+ "words": 13,
+ "wer": 0.3076923076923077
+ },
+ "semantic": {
+ "normalized_reference": "if the checksum fails restart the worker process without deleting the cached tensors",
+ "normalized_hypothesis": "if the chexsum fails restart the worker process without deleding the cash tensers",
+ "errors": 4,
+ "words": 13,
+ "wer": 0.3076923076923077
+ }
+ },
+ {
+ "id": "release500_0149",
+ "category": "conversational",
+ "text": "I need you to understand how close that was to not working.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0149.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I NEED YOU TO UNDERSTAND HOW CLOSE THAT WAS TO NOT WORKING",
+ "raw": {
+ "normalized_reference": "i need you to understand how close that was to not working.",
+ "normalized_hypothesis": "i need you to understand how close that was to not working",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i need you to understand how close that was to not working",
+ "normalized_hypothesis": "i need you to understand how close that was to not working",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0150",
+ "category": "descriptive",
+ "text": "It was the sort of morning where the whole world looks rinsed clean.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0150.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT WAS THE SORT OF MORNING WHERE THE WHOLE WORLD LOOKS RINCED CLEAN",
+ "raw": {
+ "normalized_reference": "it was the sort of morning where the whole world looks rinsed clean.",
+ "normalized_hypothesis": "it was the sort of morning where the whole world looks rinced clean",
+ "errors": 1,
+ "words": 13,
+ "wer": 0.07692307692307693
+ },
+ "semantic": {
+ "normalized_reference": "it was the sort of morning where the whole world looks rinsed clean",
+ "normalized_hypothesis": "it was the sort of morning where the whole world looks rinced clean",
+ "errors": 1,
+ "words": 13,
+ "wer": 0.07692307692307693
+ }
+ },
+ {
+ "id": "release500_0151",
+ "category": "emotional",
+ "text": "She started laughing and I started laughing and we couldn't stop for a full minute.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0151.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SHE STARTED LAUGHING AND I STARTED LAUGHING AND WE COULDN'T STOP FOR A FULL MINUTE",
+ "raw": {
+ "normalized_reference": "she started laughing and i started laughing and we couldn't stop for a full minute.",
+ "normalized_hypothesis": "she started laughing and i started laughing and we couldn't stop for a full minute",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "she started laughing and i started laughing and we could not stop for a full minute",
+ "normalized_hypothesis": "she started laughing and i started laughing and we could not stop for a full minute",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0152",
+ "category": "long",
+ "text": "We spent three years building something we were genuinely proud of, and then the market shifted and none of it mattered the way we thought it would.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0152.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WE SPENT THREE YEARS BUILDING SOMETHING WE WERE GENUINELY PROUD OF AND THEN THE MARKET SHIFTED AND NONE OF IT MATTERED THE WAY WE THOUGHT IT WOULD",
+ "raw": {
+ "normalized_reference": "we spent three years building something we were genuinely proud of, and then the market shifted and none of it mattered the way we thought it would.",
+ "normalized_hypothesis": "we spent three years building something we were genuinely proud of and then the market shifted and none of it mattered the way we thought it would",
+ "errors": 0,
+ "words": 27,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "we spent 3 years building something we were genuinely proud of and then the market shifted and none of it mattered the way we thought it would",
+ "normalized_hypothesis": "we spent 3 years building something we were genuinely proud of and then the market shifted and none of it mattered the way we thought it would",
+ "errors": 0,
+ "words": 27,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0153",
+ "category": "narrative",
+ "text": "And somehow, despite all of that, it actually turned out okay.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0153.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "AND SOMEHOW DESPITE ALL OF THAT IT ACTUALLY TURNED OUT OQUET",
+ "raw": {
+ "normalized_reference": "and somehow, despite all of that, it actually turned out okay.",
+ "normalized_hypothesis": "and somehow despite all of that it actually turned out oquet",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ },
+ "semantic": {
+ "normalized_reference": "and somehow despite all of that it actually turned out okay",
+ "normalized_hypothesis": "and somehow despite all of that it actually turned out oquet",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ }
+ },
+ {
+ "id": "release500_0154",
+ "category": "question",
+ "text": "Is that something you actually want, or something you think you should want?",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0154.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IS THAT SOMETHING YOU ACTUALLY WANT OR SOMETHING YOU THINK YOU SHOULD WANT",
+ "raw": {
+ "normalized_reference": "is that something you actually want, or something you think you should want?",
+ "normalized_hypothesis": "is that something you actually want or something you think you should want",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "is that something you actually want or something you think you should want",
+ "normalized_hypothesis": "is that something you actually want or something you think you should want",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0155",
+ "category": "short",
+ "text": "This is fine.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0155.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THIS IS FINE",
+ "raw": {
+ "normalized_reference": "this is fine.",
+ "normalized_hypothesis": "this is fine",
+ "errors": 0,
+ "words": 3,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "this is fine",
+ "normalized_hypothesis": "this is fine",
+ "errors": 0,
+ "words": 3,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0156",
+ "category": "technical",
+ "text": "The tricky part isn't the setup, it's making sure all dependencies line up.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0156.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE TRICKY PART ISN'T THE SET UP IT'S MAKING SURE ALL DEPENDENCIES LINE OUT",
+ "raw": {
+ "normalized_reference": "the tricky part isn't the setup, it's making sure all dependencies line up.",
+ "normalized_hypothesis": "the tricky part isn't the set up it's making sure all dependencies line out",
+ "errors": 3,
+ "words": 13,
+ "wer": 0.23076923076923078
+ },
+ "semantic": {
+ "normalized_reference": "the tricky part is not the setup it is making sure all dependencies line up",
+ "normalized_hypothesis": "the tricky part is not the set up it is making sure all dependencies line out",
+ "errors": 3,
+ "words": 15,
+ "wer": 0.2
+ }
+ },
+ {
+ "id": "release500_0157",
+ "category": "conversational",
+ "text": "He called it a minor setback. I called it a disaster.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0157.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE CALLED IT A MINOR SETBACK I CALLED IT A DISASTER",
+ "raw": {
+ "normalized_reference": "he called it a minor setback. i called it a disaster.",
+ "normalized_hypothesis": "he called it a minor setback i called it a disaster",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he called it a minor setback i called it a disaster",
+ "normalized_hypothesis": "he called it a minor setback i called it a disaster",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0158",
+ "category": "descriptive",
+ "text": "The whole place smelled like pine and old wood and cold air.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0158.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE WHOLE PLACE SMELLED LIKE PINE AND OLD WOOD AND COLD AIR",
+ "raw": {
+ "normalized_reference": "the whole place smelled like pine and old wood and cold air.",
+ "normalized_hypothesis": "the whole place smelled like pine and old wood and cold air",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the whole place smelled like pine and old wood and cold air",
+ "normalized_hypothesis": "the whole place smelled like pine and old wood and cold air",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0159",
+ "category": "emotional",
+ "text": "I've never been so relieved in my entire life.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0159.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I'VE NEVER BEEN SO RELIEVED IN MY ENTIRE LIFE",
+ "raw": {
+ "normalized_reference": "i've never been so relieved in my entire life.",
+ "normalized_hypothesis": "i've never been so relieved in my entire life",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i have never been so relieved in my entire life",
+ "normalized_hypothesis": "i have never been so relieved in my entire life",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0160",
+ "category": "narrative",
+ "text": "I wrote it all down that night so I wouldn't forget, and I still haven't looked at it.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0160.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I WROTE IT ALL DOWN THAT NIGHT SO I WOULDN'T FORGET AND I STILL HAVEN'T LOOKED AT IT",
+ "raw": {
+ "normalized_reference": "i wrote it all down that night so i wouldn't forget, and i still haven't looked at it.",
+ "normalized_hypothesis": "i wrote it all down that night so i wouldn't forget and i still haven't looked at it",
+ "errors": 0,
+ "words": 18,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i wrote it all down that night so i would not forget and i still have not looked at it",
+ "normalized_hypothesis": "i wrote it all down that night so i would not forget and i still have not looked at it",
+ "errors": 0,
+ "words": 20,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0161",
+ "category": "short",
+ "text": "That's kind of wild actually.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0161.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THAT'S KIND OF WILD ACTUALLY",
+ "raw": {
+ "normalized_reference": "that's kind of wild actually.",
+ "normalized_hypothesis": "that's kind of wild actually",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "that is kind of wild actually",
+ "normalized_hypothesis": "that is kind of wild actually",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0162",
+ "category": "technical",
+ "text": "Version pinning will save you a lot of pain later down the line.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0162.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "VERSION PINNING WILL SAVE YOU A LOT OF PAIN LATER DOWN THE LINE",
+ "raw": {
+ "normalized_reference": "version pinning will save you a lot of pain later down the line.",
+ "normalized_hypothesis": "version pinning will save you a lot of pain later down the line",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "version pinning will save you a lot of pain later down the line",
+ "normalized_hypothesis": "version pinning will save you a lot of pain later down the line",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0163",
+ "category": "conversational",
+ "text": "At some point you just have to commit and see what happens.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0163.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "AT SOME POINT YOU JUST HAVE TO COMMIT AND SEE WHAT HAPPENS",
+ "raw": {
+ "normalized_reference": "at some point you just have to commit and see what happens.",
+ "normalized_hypothesis": "at some point you just have to commit and see what happens",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "at some point you just have to commit and see what happens",
+ "normalized_hypothesis": "at some point you just have to commit and see what happens",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0164",
+ "category": "emotional",
+ "text": "Nobody told me it was going to be this hard, and honestly, I'm grateful.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0164.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "NOBODY TOLD ME IT WAS GOING TO BE THIS HARD AND HONESTLY I'M GRATEFUL",
+ "raw": {
+ "normalized_reference": "nobody told me it was going to be this hard, and honestly, i'm grateful.",
+ "normalized_hypothesis": "nobody told me it was going to be this hard and honestly i'm grateful",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "nobody told me it was going to be this hard and honestly i am grateful",
+ "normalized_hypothesis": "nobody told me it was going to be this hard and honestly i am grateful",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0165",
+ "category": "narrative",
+ "text": "Chapter seven. In which everything that could go wrong did, and a few things that couldn't.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0165.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "CHAPTER SEVEN IN WHICH EVERYTHING THAT COULD GO WRONG DID AND A FEW THINGS THAT COULDN'T",
+ "raw": {
+ "normalized_reference": "chapter seven. in which everything that could go wrong did, and a few things that couldn't.",
+ "normalized_hypothesis": "chapter seven in which everything that could go wrong did and a few things that couldn't",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "chapter 7 in which everything that could go wrong did and a few things that could not",
+ "normalized_hypothesis": "chapter 7 in which everything that could go wrong did and a few things that could not",
+ "errors": 0,
+ "words": 17,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0166",
+ "category": "short",
+ "text": "Wait, what?",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0166.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WAIT WHAT",
+ "raw": {
+ "normalized_reference": "wait, what?",
+ "normalized_hypothesis": "wait what",
+ "errors": 0,
+ "words": 2,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "wait what",
+ "normalized_hypothesis": "wait what",
+ "errors": 0,
+ "words": 2,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0167",
+ "category": "technical",
+ "text": "The graphics processor remained idle because the data loader exhausted every central processing core.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0167.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE GRAPHICS PROCESSOR REMAINED IDLE BECAUSE THE DATA LODER EXHAUSTED EVERY CENTRAL PROCESSING CORE",
+ "raw": {
+ "normalized_reference": "the graphics processor remained idle because the data loader exhausted every central processing core.",
+ "normalized_hypothesis": "the graphics processor remained idle because the data loder exhausted every central processing core",
+ "errors": 1,
+ "words": 14,
+ "wer": 0.07142857142857142
+ },
+ "semantic": {
+ "normalized_reference": "the graphics processor remained idle because the data loader exhausted every central processing core",
+ "normalized_hypothesis": "the graphics processor remained idle because the data loder exhausted every central processing core",
+ "errors": 1,
+ "words": 14,
+ "wer": 0.07142857142857142
+ }
+ },
+ {
+ "id": "release500_0168",
+ "category": "conversational",
+ "text": "Could you speak a little more slowly? I am trying to write every detail down.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0168.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "COULD YOU SPEAK A LITTLE MORE SLOWLY I AM TRYING TO WRITE EVERY DETAIL DOWN",
+ "raw": {
+ "normalized_reference": "could you speak a little more slowly? i am trying to write every detail down.",
+ "normalized_hypothesis": "could you speak a little more slowly i am trying to write every detail down",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "could you speak a little more slowly i am trying to write every detail down",
+ "normalized_hypothesis": "could you speak a little more slowly i am trying to write every detail down",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0169",
+ "category": "emotional",
+ "text": "I was so excited I could barely form a complete sentence.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0169.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I WAS SO EXCITED I COULD BARELY FORM A COMPLETE SENTENCE",
+ "raw": {
+ "normalized_reference": "i was so excited i could barely form a complete sentence.",
+ "normalized_hypothesis": "i was so excited i could barely form a complete sentence",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i was so excited i could barely form a complete sentence",
+ "normalized_hypothesis": "i was so excited i could barely form a complete sentence",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0170",
+ "category": "narrative",
+ "text": "He looked at me like I had said something in a language he didn't speak.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0170.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE LOOKED AT ME LIKE I HAD SAID SOMETHING IN A LANGUAGE HE DIDN'T SPEAK",
+ "raw": {
+ "normalized_reference": "he looked at me like i had said something in a language he didn't speak.",
+ "normalized_hypothesis": "he looked at me like i had said something in a language he didn't speak",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he looked at me like i had said something in a language he did not speak",
+ "normalized_hypothesis": "he looked at me like i had said something in a language he did not speak",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0171",
+ "category": "short",
+ "text": "That's actually hilarious.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0171.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THAT'S ACTUALLY HILLARIOUS",
+ "raw": {
+ "normalized_reference": "that's actually hilarious.",
+ "normalized_hypothesis": "that's actually hillarious",
+ "errors": 1,
+ "words": 3,
+ "wer": 0.3333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "that is actually hilarious",
+ "normalized_hypothesis": "that is actually hillarious",
+ "errors": 1,
+ "words": 4,
+ "wer": 0.25
+ }
+ },
+ {
+ "id": "release500_0172",
+ "category": "technical",
+ "text": "The encoder maps phonemes, stress, and phrase boundaries into a shared hidden representation.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0172.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE INKODER MAPS PHONIUMS STRESS AND PHRASE BOUNDARIES INTO A SHARED HIDEN REPRESENTATION",
+ "raw": {
+ "normalized_reference": "the encoder maps phonemes, stress, and phrase boundaries into a shared hidden representation.",
+ "normalized_hypothesis": "the inkoder maps phoniums stress and phrase boundaries into a shared hiden representation",
+ "errors": 3,
+ "words": 13,
+ "wer": 0.23076923076923078
+ },
+ "semantic": {
+ "normalized_reference": "the encoder maps phonemes stress and phrase boundaries into a shared hidden representation",
+ "normalized_hypothesis": "the inkoder maps phoniums stress and phrase boundaries into a shared hiden representation",
+ "errors": 3,
+ "words": 13,
+ "wer": 0.23076923076923078
+ }
+ },
+ {
+ "id": "release500_0173",
+ "category": "conversational",
+ "text": "She showed up three hours late and acted like nothing had happened.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0173.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SHE SHOWED UP THREE HOURS LATE AND ACTED LIKE NOTHING HAD HAPPENED",
+ "raw": {
+ "normalized_reference": "she showed up three hours late and acted like nothing had happened.",
+ "normalized_hypothesis": "she showed up three hours late and acted like nothing had happened",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "she showed up 3 hours late and acted like nothing had happened",
+ "normalized_hypothesis": "she showed up 3 hours late and acted like nothing had happened",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0174",
+ "category": "emotional",
+ "text": "Oh my god. This is the best thing I've ever tasted.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0174.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "O MY GOD THIS IS THE BEST THING I'VE EVER TASTED",
+ "raw": {
+ "normalized_reference": "oh my god. this is the best thing i've ever tasted.",
+ "normalized_hypothesis": "o my god this is the best thing i've ever tasted",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ },
+ "semantic": {
+ "normalized_reference": "0 my god this is the best thing i have ever tasted",
+ "normalized_hypothesis": "0 my god this is the best thing i have ever tasted",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0175",
+ "category": "narrative",
+ "text": "The note was short, which was somehow more frightening than a long one would have been.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0175.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE NOTE WAS SHORT WHICH WAS SOMEHOW MORE FRIGHTENING THAN A LONG ONE WOULD HAVE BEEN",
+ "raw": {
+ "normalized_reference": "the note was short, which was somehow more frightening than a long one would have been.",
+ "normalized_hypothesis": "the note was short which was somehow more frightening than a long one would have been",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the note was short which was somehow more frightening than a long one would have been",
+ "normalized_hypothesis": "the note was short which was somehow more frightening than a long one would have been",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0176",
+ "category": "short",
+ "text": "Noted.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0176.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "NOID",
+ "raw": {
+ "normalized_reference": "noted.",
+ "normalized_hypothesis": "noid",
+ "errors": 1,
+ "words": 1,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "noted",
+ "normalized_hypothesis": "noid",
+ "errors": 1,
+ "words": 1,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0177",
+ "category": "technical",
+ "text": "It's a simple process once you understand the underlying structure.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0177.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT'S A SIMPLE PROCESS ONCE YOU UNDERSTAND THE UNDERLYING STRUCTURE",
+ "raw": {
+ "normalized_reference": "it's a simple process once you understand the underlying structure.",
+ "normalized_hypothesis": "it's a simple process once you understand the underlying structure",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it is a simple process once you understand the underlying structure",
+ "normalized_hypothesis": "it is a simple process once you understand the underlying structure",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0178",
+ "category": "conversational",
+ "text": "You have to hear this story because it's almost too good to be true.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0178.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "YOU HAVE TO HEAR THIS STORY BECAUSE IT'S ALMOST TOO GOOD TO BE TRUE",
+ "raw": {
+ "normalized_reference": "you have to hear this story because it's almost too good to be true.",
+ "normalized_hypothesis": "you have to hear this story because it's almost too good to be true",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "you have to hear this story because it is almost too good to be true",
+ "normalized_hypothesis": "you have to hear this story because it is almost too good to be true",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0179",
+ "category": "emotional",
+ "text": "I have never felt more seen in my entire life than in that moment.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0179.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I HAVE NEVER FELT MORE SCEEN IN MY ENTIRE LIFE THAN IN THAT MOMENT",
+ "raw": {
+ "normalized_reference": "i have never felt more seen in my entire life than in that moment.",
+ "normalized_hypothesis": "i have never felt more sceen in my entire life than in that moment",
+ "errors": 1,
+ "words": 14,
+ "wer": 0.07142857142857142
+ },
+ "semantic": {
+ "normalized_reference": "i have never felt more seen in my entire life than in that moment",
+ "normalized_hypothesis": "i have never felt more sceen in my entire life than in that moment",
+ "errors": 1,
+ "words": 14,
+ "wer": 0.07142857142857142
+ }
+ },
+ {
+ "id": "release500_0180",
+ "category": "narrative",
+ "text": "She knew. She had always known. She just waited for the right moment.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0180.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SHE KNEW SHE HAD ALWAYS KNOWN SHE JUST WAITED FOR THE RIGHT MOMENT",
+ "raw": {
+ "normalized_reference": "she knew. she had always known. she just waited for the right moment.",
+ "normalized_hypothesis": "she knew she had always known she just waited for the right moment",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "she knew she had always known she just waited for the right moment",
+ "normalized_hypothesis": "she knew she had always known she just waited for the right moment",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0181",
+ "category": "short",
+ "text": "Yeah, no. Hard pass.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0181.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "YE NOB A HARD PASS",
+ "raw": {
+ "normalized_reference": "yeah, no. hard pass.",
+ "normalized_hypothesis": "ye nob a hard pass",
+ "errors": 3,
+ "words": 4,
+ "wer": 0.75
+ },
+ "semantic": {
+ "normalized_reference": "yeah no hard pass",
+ "normalized_hypothesis": "ye nob a hard pass",
+ "errors": 3,
+ "words": 4,
+ "wer": 0.75
+ }
+ },
+ {
+ "id": "release500_0182",
+ "category": "technical",
+ "text": "Please verify the sample rate, hop length, and mel filter configuration before inference.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0182.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "PLEASE VERIFY THE SAMPLE RATE HOP LENGTH AND MELL FILTER CONFIGURATION BEFORE INFERENCE",
+ "raw": {
+ "normalized_reference": "please verify the sample rate, hop length, and mel filter configuration before inference.",
+ "normalized_hypothesis": "please verify the sample rate hop length and mell filter configuration before inference",
+ "errors": 1,
+ "words": 13,
+ "wer": 0.07692307692307693
+ },
+ "semantic": {
+ "normalized_reference": "please verify the sample rate hop length and mel filter configuration before inference",
+ "normalized_hypothesis": "please verify the sample rate hop length and mell filter configuration before inference",
+ "errors": 1,
+ "words": 13,
+ "wer": 0.07692307692307693
+ }
+ },
+ {
+ "id": "release500_0183",
+ "category": "conversational",
+ "text": "That's the worst idea I've ever heard and I'm completely in.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0183.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THAT'S THE WORST IDEA I'VE EVER HEARD AND I'M COMPLETELY IN",
+ "raw": {
+ "normalized_reference": "that's the worst idea i've ever heard and i'm completely in.",
+ "normalized_hypothesis": "that's the worst idea i've ever heard and i'm completely in",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "that is the worst idea i have ever heard and i am completely in",
+ "normalized_hypothesis": "that is the worst idea i have ever heard and i am completely in",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0184",
+ "category": "emotional",
+ "text": "I remember thinking: this is it. This is the thing I've been waiting for.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0184.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I REMEMBER THINKING THIS IS IT THIS IS THE THING I'VE BEEN WAITING FOR",
+ "raw": {
+ "normalized_reference": "i remember thinking: this is it. this is the thing i've been waiting for.",
+ "normalized_hypothesis": "i remember thinking this is it this is the thing i've been waiting for",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i remember thinking this is it this is the thing i have been waiting for",
+ "normalized_hypothesis": "i remember thinking this is it this is the thing i have been waiting for",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0185",
+ "category": "narrative",
+ "text": "We never talked about it again, which I think was the right call.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0185.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WE NEVER TALKED ABOUT IT AGAIN WHICH I THINK WAS THE RIGHT CALL",
+ "raw": {
+ "normalized_reference": "we never talked about it again, which i think was the right call.",
+ "normalized_hypothesis": "we never talked about it again which i think was the right call",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "we never talked about it again which i think was the right call",
+ "normalized_hypothesis": "we never talked about it again which i think was the right call",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0186",
+ "category": "short",
+ "text": "In hindsight, obvious.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0186.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IN HINDSIGHT OBVIOUS",
+ "raw": {
+ "normalized_reference": "in hindsight, obvious.",
+ "normalized_hypothesis": "in hindsight obvious",
+ "errors": 0,
+ "words": 3,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "in hindsight obvious",
+ "normalized_hypothesis": "in hindsight obvious",
+ "errors": 0,
+ "words": 3,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0187",
+ "category": "technical",
+ "text": "Once it's configured correctly, it basically runs itself.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0187.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "ONCE IT'S CONFIGURED CORRECTLY IT BASICALLY RUNS ITSELF",
+ "raw": {
+ "normalized_reference": "once it's configured correctly, it basically runs itself.",
+ "normalized_hypothesis": "once it's configured correctly it basically runs itself",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "once it is configured correctly it basically runs itself",
+ "normalized_hypothesis": "once it is configured correctly it basically runs itself",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0188",
+ "category": "conversational",
+ "text": "You got engaged? When? How? Tell me everything right now.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0188.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "YOUD GOT ENGAGED WHEN HOW TELL ME EVERYTHING RIGHT NOW",
+ "raw": {
+ "normalized_reference": "you got engaged? when? how? tell me everything right now.",
+ "normalized_hypothesis": "youd got engaged when how tell me everything right now",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ },
+ "semantic": {
+ "normalized_reference": "you got engaged when how tell me everything right now",
+ "normalized_hypothesis": "youd got engaged when how tell me everything right now",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ }
+ },
+ {
+ "id": "release500_0189",
+ "category": "emotional",
+ "text": "It's hard to explain how much that moment meant to me.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0189.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT'S HARD TO EXPLAIN HOW MUCH THAT MOMENT MEANT TO ME",
+ "raw": {
+ "normalized_reference": "it's hard to explain how much that moment meant to me.",
+ "normalized_hypothesis": "it's hard to explain how much that moment meant to me",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it is hard to explain how much that moment meant to me",
+ "normalized_hypothesis": "it is hard to explain how much that moment meant to me",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0190",
+ "category": "narrative",
+ "text": "He told the story three different times that night and somehow it got better each time.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0190.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE TOLD THE STORY THREE DIFFERENT TIMES THAT NIGHT AND SOMEHOW IT GOT BETTER EACH TIME",
+ "raw": {
+ "normalized_reference": "he told the story three different times that night and somehow it got better each time.",
+ "normalized_hypothesis": "he told the story three different times that night and somehow it got better each time",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he told the story 3 different times that night and somehow it got better each time",
+ "normalized_hypothesis": "he told the story 3 different times that night and somehow it got better each time",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0191",
+ "category": "short",
+ "text": "Right. Moving on.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0191.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "RIGHT MOVING ON",
+ "raw": {
+ "normalized_reference": "right. moving on.",
+ "normalized_hypothesis": "right moving on",
+ "errors": 0,
+ "words": 3,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "right moving on",
+ "normalized_hypothesis": "right moving on",
+ "errors": 0,
+ "words": 3,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0192",
+ "category": "conversational",
+ "text": "No, I did not move your keys; check beside the small ceramic bowl.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0192.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "NO I DID NOT MOVE YOUR KEYS CHECK BESIDE THE SMALL SERAMIC BOWL",
+ "raw": {
+ "normalized_reference": "no, i did not move your keys; check beside the small ceramic bowl.",
+ "normalized_hypothesis": "no i did not move your keys check beside the small seramic bowl",
+ "errors": 1,
+ "words": 13,
+ "wer": 0.07692307692307693
+ },
+ "semantic": {
+ "normalized_reference": "no i did not move your keys check beside the small ceramic bowl",
+ "normalized_hypothesis": "no i did not move your keys check beside the small seramic bowl",
+ "errors": 1,
+ "words": 13,
+ "wer": 0.07692307692307693
+ }
+ },
+ {
+ "id": "release500_0193",
+ "category": "emotional",
+ "text": "I was devastated. I'm not going to pretend otherwise.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0193.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I WAS DEVASTATED AM NOT GOING TO PRETEND OTHERWISE",
+ "raw": {
+ "normalized_reference": "i was devastated. i'm not going to pretend otherwise.",
+ "normalized_hypothesis": "i was devastated am not going to pretend otherwise",
+ "errors": 1,
+ "words": 9,
+ "wer": 0.1111111111111111
+ },
+ "semantic": {
+ "normalized_reference": "i was devastated i am not going to pretend otherwise",
+ "normalized_hypothesis": "i was devastated am not going to pretend otherwise",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ }
+ },
+ {
+ "id": "release500_0194",
+ "category": "narrative",
+ "text": "That was the last summer before everything changed, though we didn't know it yet.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0194.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THAT WAS THE LAST SUMMER BEFORE EVERYTHING CHANGED THOUGH WE DIDN'T KNOW IT YET",
+ "raw": {
+ "normalized_reference": "that was the last summer before everything changed, though we didn't know it yet.",
+ "normalized_hypothesis": "that was the last summer before everything changed though we didn't know it yet",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "that was the last summer before everything changed though we did not know it yet",
+ "normalized_hypothesis": "that was the last summer before everything changed though we did not know it yet",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0195",
+ "category": "short",
+ "text": "Bold of you to assume.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0195.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "BOLD OF YOU TO ASSUME",
+ "raw": {
+ "normalized_reference": "bold of you to assume.",
+ "normalized_hypothesis": "bold of you to assume",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "bold of you to assume",
+ "normalized_hypothesis": "bold of you to assume",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0196",
+ "category": "conversational",
+ "text": "It's one of those decisions that makes total sense in the moment.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0196.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT'S ONE OF THOSE DECISIONS THAT MAKES TOTAL SENSE IN THE MOMENT",
+ "raw": {
+ "normalized_reference": "it's one of those decisions that makes total sense in the moment.",
+ "normalized_hypothesis": "it's one of those decisions that makes total sense in the moment",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it is one of those decisions that makes total sense in the moment",
+ "normalized_hypothesis": "it is one of those decisions that makes total sense in the moment",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0197",
+ "category": "emotional",
+ "text": "I was furious. Not the loud kind. The quiet kind that lasts for days.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0197.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I WAS FURIOUS NOT THE LOUD KIND THE QUIET KIND THAT LASTS FOR DAYS",
+ "raw": {
+ "normalized_reference": "i was furious. not the loud kind. the quiet kind that lasts for days.",
+ "normalized_hypothesis": "i was furious not the loud kind the quiet kind that lasts for days",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i was furious not the loud kind the quiet kind that lasts for days",
+ "normalized_hypothesis": "i was furious not the loud kind the quiet kind that lasts for days",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0198",
+ "category": "narrative",
+ "text": "That's the part of the story I usually leave out.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0198.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THAT'S THE PART OF THE STORY I USUALLY LEAVE OUT",
+ "raw": {
+ "normalized_reference": "that's the part of the story i usually leave out.",
+ "normalized_hypothesis": "that's the part of the story i usually leave out",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "that is the part of the story i usually leave out",
+ "normalized_hypothesis": "that is the part of the story i usually leave out",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0199",
+ "category": "short",
+ "text": "I've said too much.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0199.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I'VE SAID TOO MUCH",
+ "raw": {
+ "normalized_reference": "i've said too much.",
+ "normalized_hypothesis": "i've said too much",
+ "errors": 0,
+ "words": 4,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i have said too much",
+ "normalized_hypothesis": "i have said too much",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0200",
+ "category": "conversational",
+ "text": "I've been faking that I understood this the entire time.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0200.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I'VE BEEN FAKING THAT I UNDERSTOOD THIS THE ENTIRE TIME",
+ "raw": {
+ "normalized_reference": "i've been faking that i understood this the entire time.",
+ "normalized_hypothesis": "i've been faking that i understood this the entire time",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i have been faking that i understood this the entire time",
+ "normalized_hypothesis": "i have been faking that i understood this the entire time",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0201",
+ "category": "phonetic_balance",
+ "text": "Peace be unto you and grace before the Lord.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0201.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "PEACE BE UNTO YOU AND GRACE BEFORE THE LORD",
+ "raw": {
+ "normalized_reference": "peace be unto you and grace before the lord.",
+ "normalized_hypothesis": "peace be unto you and grace before the lord",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "peace be unto you and grace before the lord",
+ "normalized_hypothesis": "peace be unto you and grace before the lord",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0202",
+ "category": "phonetic_balance",
+ "text": "From now on we're pals.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0202.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "FROM NOW ON REAR PALS",
+ "raw": {
+ "normalized_reference": "from now on we're pals.",
+ "normalized_hypothesis": "from now on rear pals",
+ "errors": 1,
+ "words": 5,
+ "wer": 0.2
+ },
+ "semantic": {
+ "normalized_reference": "from now on we are pals",
+ "normalized_hypothesis": "from now on rear pals",
+ "errors": 2,
+ "words": 6,
+ "wer": 0.3333333333333333
+ }
+ },
+ {
+ "id": "release500_0203",
+ "category": "phonetic_balance",
+ "text": "Here we allow our solicitors to look after our legal work.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0203.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HERE WE ALLOW OUR SOLICITORS TO LOOK AFTER OUR LEGAL WORK",
+ "raw": {
+ "normalized_reference": "here we allow our solicitors to look after our legal work.",
+ "normalized_hypothesis": "here we allow our solicitors to look after our legal work",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "here we allow our solicitors to look after our legal work",
+ "normalized_hypothesis": "here we allow our solicitors to look after our legal work",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0204",
+ "category": "phonetic_balance",
+ "text": "Her words sent a strange chill through Philip.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0204.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HER WORDS SENT A STRANGE CHILL THROUGH PHILIP",
+ "raw": {
+ "normalized_reference": "her words sent a strange chill through philip.",
+ "normalized_hypothesis": "her words sent a strange chill through philip",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "her words sent a strange chill through philip",
+ "normalized_hypothesis": "her words sent a strange chill through philip",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0205",
+ "category": "phonetic_balance",
+ "text": "It was introduced by Representative Dick of Ohio.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0205.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT WAS INTRODUCED BY REPRESENTATIVE DICK OF OHIO",
+ "raw": {
+ "normalized_reference": "it was introduced by representative dick of ohio.",
+ "normalized_hypothesis": "it was introduced by representative dick of ohio",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it was introduced by representative dick of ohio",
+ "normalized_hypothesis": "it was introduced by representative dick of ohio",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0206",
+ "category": "phonetic_balance",
+ "text": "It's the aurora borealis.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0206.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT'S THE AURORA BORIALIS",
+ "raw": {
+ "normalized_reference": "it's the aurora borealis.",
+ "normalized_hypothesis": "it's the aurora borialis",
+ "errors": 1,
+ "words": 4,
+ "wer": 0.25
+ },
+ "semantic": {
+ "normalized_reference": "it is the aurora borealis",
+ "normalized_hypothesis": "it is the aurora borialis",
+ "errors": 1,
+ "words": 5,
+ "wer": 0.2
+ }
+ },
+ {
+ "id": "release500_0207",
+ "category": "phonetic_balance",
+ "text": "The boy grew and prospered.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0207.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE BOY GREW AND PROSPERED",
+ "raw": {
+ "normalized_reference": "the boy grew and prospered.",
+ "normalized_hypothesis": "the boy grew and prospered",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the boy grew and prospered",
+ "normalized_hypothesis": "the boy grew and prospered",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0208",
+ "category": "phonetic_balance",
+ "text": "His voice was passionately rebellious.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0208.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HIS VOICE WAS PASSIONATELY REBELLIOUS",
+ "raw": {
+ "normalized_reference": "his voice was passionately rebellious.",
+ "normalized_hypothesis": "his voice was passionately rebellious",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "his voice was passionately rebellious",
+ "normalized_hypothesis": "his voice was passionately rebellious",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0209",
+ "category": "phonetic_balance",
+ "text": "That's where they cut off the Scottish Chiefs and killed all hands.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0209.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THAT'S WHERE THEY CUT OFF THE SCOTTISH CHIEFS AND KILLED ALL HANDS",
+ "raw": {
+ "normalized_reference": "that's where they cut off the scottish chiefs and killed all hands.",
+ "normalized_hypothesis": "that's where they cut off the scottish chiefs and killed all hands",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "that is where they cut off the scottish chiefs and killed all hands",
+ "normalized_hypothesis": "that is where they cut off the scottish chiefs and killed all hands",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0210",
+ "category": "phonetic_balance",
+ "text": "And Tom King patiently endured.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0210.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "AND TOM KING PATIENTLY ENDURED",
+ "raw": {
+ "normalized_reference": "and tom king patiently endured.",
+ "normalized_hypothesis": "and tom king patiently endured",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "and tom king patiently endured",
+ "normalized_hypothesis": "and tom king patiently endured",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0211",
+ "category": "phonetic_balance",
+ "text": "From that moment his friendship for Belize turns to hatred and jealousy.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0211.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "FROM THAT MOMENT HIS FRIENDSHIP FOR BELIS TURNS TO HATRED AND JEALOUSY",
+ "raw": {
+ "normalized_reference": "from that moment his friendship for belize turns to hatred and jealousy.",
+ "normalized_hypothesis": "from that moment his friendship for belis turns to hatred and jealousy",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "from that moment his friendship for belize turns to hatred and jealousy",
+ "normalized_hypothesis": "from that moment his friendship for belis turns to hatred and jealousy",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ }
+ },
+ {
+ "id": "release500_0212",
+ "category": "phonetic_balance",
+ "text": "Then you don't believe in altruism.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0212.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THEN YOU DON'T BELIEVE IN ALTRUISM",
+ "raw": {
+ "normalized_reference": "then you don't believe in altruism.",
+ "normalized_hypothesis": "then you don't believe in altruism",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "then you do not believe in altruism",
+ "normalized_hypothesis": "then you do not believe in altruism",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0213",
+ "category": "phonetic_balance",
+ "text": "What Pascal glimpsed with the vision of a seer, I have lived.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0213.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WHAT PASCAL GLIMPSED WITH THE VISION OF A SEER I HAVE LIVED",
+ "raw": {
+ "normalized_reference": "what pascal glimpsed with the vision of a seer, i have lived.",
+ "normalized_hypothesis": "what pascal glimpsed with the vision of a seer i have lived",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "what pascal glimpsed with the vision of a seer i have lived",
+ "normalized_hypothesis": "what pascal glimpsed with the vision of a seer i have lived",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0214",
+ "category": "phonetic_balance",
+ "text": "You were making them talk shop, Ruth charged him.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0214.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "YOU WERE MAKING THEM TALK SHOP RUTH CHARGED HIM",
+ "raw": {
+ "normalized_reference": "you were making them talk shop, ruth charged him.",
+ "normalized_hypothesis": "you were making them talk shop ruth charged him",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "you were making them talk shop ruth charged him",
+ "normalized_hypothesis": "you were making them talk shop ruth charged him",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0215",
+ "category": "phonetic_balance",
+ "text": "Philip snatched at the letter which Gregson held out to him.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0215.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "PHILIP SNATCHED AT THE LETTER WHICH GREGSON HELD OUT TO HIM",
+ "raw": {
+ "normalized_reference": "philip snatched at the letter which gregson held out to him.",
+ "normalized_hypothesis": "philip snatched at the letter which gregson held out to him",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "philip snatched at the letter which gregson held out to him",
+ "normalized_hypothesis": "philip snatched at the letter which gregson held out to him",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0216",
+ "category": "phonetic_balance",
+ "text": "For a time the exciting thrill of his adventure was gone.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0216.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "FOR A TIME THE EXCITING THRILL OF HIS ADVENTURE WAS GONE",
+ "raw": {
+ "normalized_reference": "for a time the exciting thrill of his adventure was gone.",
+ "normalized_hypothesis": "for a time the exciting thrill of his adventure was gone",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "for a time the exciting thrill of his adventure was gone",
+ "normalized_hypothesis": "for a time the exciting thrill of his adventure was gone",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0217",
+ "category": "phonetic_balance",
+ "text": "Obviously, it was a disease that could be contracted by contact.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0217.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "OBVIOUSLY IT WAS A DISEASE THAT COULD BE CONTRACTED BY CONTACT",
+ "raw": {
+ "normalized_reference": "obviously, it was a disease that could be contracted by contact.",
+ "normalized_hypothesis": "obviously it was a disease that could be contracted by contact",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "obviously it was a disease that could be contracted by contact",
+ "normalized_hypothesis": "obviously it was a disease that could be contracted by contact",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0218",
+ "category": "phonetic_balance",
+ "text": "Blind with rage, he darted in.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0218.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "BLIND WITH RAGE HE DARTED IN",
+ "raw": {
+ "normalized_reference": "blind with rage, he darted in.",
+ "normalized_hypothesis": "blind with rage he darted in",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "blind with rage he darted in",
+ "normalized_hypothesis": "blind with rage he darted in",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0219",
+ "category": "phonetic_balance",
+ "text": "You live on an income which your father earned.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0219.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "YOU LIVE ON AN INCOME WHICH YOUR FATHER EARNED",
+ "raw": {
+ "normalized_reference": "you live on an income which your father earned.",
+ "normalized_hypothesis": "you live on an income which your father earned",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "you live on an income which your father earned",
+ "normalized_hypothesis": "you live on an income which your father earned",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0220",
+ "category": "phonetic_balance",
+ "text": "Something about it seemed to fascinate him, to challenge his presence.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0220.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SOMETHING ABOUT IT SEEMED TO FASCINATE HIM TO CHALLENGE HIS PRESENCE",
+ "raw": {
+ "normalized_reference": "something about it seemed to fascinate him, to challenge his presence.",
+ "normalized_hypothesis": "something about it seemed to fascinate him to challenge his presence",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "something about it seemed to fascinate him to challenge his presence",
+ "normalized_hypothesis": "something about it seemed to fascinate him to challenge his presence",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0221",
+ "category": "phonetic_balance",
+ "text": "Her beautiful hair was done up in shining coils.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0221.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HER BEAUTIFUL HAIR WAS DONE UP IN SHINING COILS",
+ "raw": {
+ "normalized_reference": "her beautiful hair was done up in shining coils.",
+ "normalized_hypothesis": "her beautiful hair was done up in shining coils",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "her beautiful hair was done up in shining coils",
+ "normalized_hypothesis": "her beautiful hair was done up in shining coils",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0222",
+ "category": "phonetic_balance",
+ "text": "It does, was her audacious answer.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0222.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT DES WAS HER AUDACIOUS ANSWER",
+ "raw": {
+ "normalized_reference": "it does, was her audacious answer.",
+ "normalized_hypothesis": "it des was her audacious answer",
+ "errors": 1,
+ "words": 6,
+ "wer": 0.16666666666666666
+ },
+ "semantic": {
+ "normalized_reference": "it does was her audacious answer",
+ "normalized_hypothesis": "it des was her audacious answer",
+ "errors": 1,
+ "words": 6,
+ "wer": 0.16666666666666666
+ }
+ },
+ {
+ "id": "release500_0223",
+ "category": "phonetic_balance",
+ "text": "He caught himself with a jerk.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0223.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE CAUGHT HIMSELF WITH A JERK",
+ "raw": {
+ "normalized_reference": "he caught himself with a jerk.",
+ "normalized_hypothesis": "he caught himself with a jerk",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he caught himself with a jerk",
+ "normalized_hypothesis": "he caught himself with a jerk",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0224",
+ "category": "phonetic_balance",
+ "text": "He did not believe in the burning of daylight for such a luxury.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0224.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE DID NOT BELIEVE IN THE BURNING OF DAYLIGHT FOR SUCH A LUXURY",
+ "raw": {
+ "normalized_reference": "he did not believe in the burning of daylight for such a luxury.",
+ "normalized_hypothesis": "he did not believe in the burning of daylight for such a luxury",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he did not believe in the burning of daylight for such a luxury",
+ "normalized_hypothesis": "he did not believe in the burning of daylight for such a luxury",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0225",
+ "category": "phonetic_balance",
+ "text": "He had comparatively no advantages at first.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0225.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE HAD COMPARATIVELY NO ADVANTAGES AT FIRST",
+ "raw": {
+ "normalized_reference": "he had comparatively no advantages at first.",
+ "normalized_hypothesis": "he had comparatively no advantages at first",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he had comparatively no advantages at 1st",
+ "normalized_hypothesis": "he had comparatively no advantages at 1st",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0226",
+ "category": "phonetic_balance",
+ "text": "Such men believe, when they come together.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0226.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SUCH MEN BELIEVE WHEN THEY COME TOGETHER",
+ "raw": {
+ "normalized_reference": "such men believe, when they come together.",
+ "normalized_hypothesis": "such men believe when they come together",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "such men believe when they come together",
+ "normalized_hypothesis": "such men believe when they come together",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0227",
+ "category": "phonetic_balance",
+ "text": "The ship should be in within a week or ten days.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0227.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE SHIP SHOULD BE IN WITHIN A WEEK OR TEN DAYS",
+ "raw": {
+ "normalized_reference": "the ship should be in within a week or ten days.",
+ "normalized_hypothesis": "the ship should be in within a week or ten days",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the ship should be in within a week or 10 days",
+ "normalized_hypothesis": "the ship should be in within a week or 10 days",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0228",
+ "category": "phonetic_balance",
+ "text": "Such a dog the wise driver kills, or turns loose.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0228.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SUCH A DOG THE WISE DRIVER KILLS OR TURNS LOOSE",
+ "raw": {
+ "normalized_reference": "such a dog the wise driver kills, or turns loose.",
+ "normalized_hypothesis": "such a dog the wise driver kills or turns loose",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "such a dog the wise driver kills or turns loose",
+ "normalized_hypothesis": "such a dog the wise driver kills or turns loose",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0229",
+ "category": "phonetic_balance",
+ "text": "You have heard always how he was the lover of the Princess Naomi.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0229.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "YOU HAVE HEARD ALWAYS HOW HE WAS THE LOVER OF THE PRINCESS NAOMI",
+ "raw": {
+ "normalized_reference": "you have heard always how he was the lover of the princess naomi.",
+ "normalized_hypothesis": "you have heard always how he was the lover of the princess naomi",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "you have heard always how he was the lover of the princess naomi",
+ "normalized_hypothesis": "you have heard always how he was the lover of the princess naomi",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0230",
+ "category": "phonetic_balance",
+ "text": "I want to know how all this is possible.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0230.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I WANT TO KNOW HOW ALL THIS IS POSSIBLE",
+ "raw": {
+ "normalized_reference": "i want to know how all this is possible.",
+ "normalized_hypothesis": "i want to know how all this is possible",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i want to know how all this is possible",
+ "normalized_hypothesis": "i want to know how all this is possible",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0231",
+ "category": "phonetic_balance",
+ "text": "It was my idea to a tee.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0231.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT WAS MY IDEA TU ATI",
+ "raw": {
+ "normalized_reference": "it was my idea to a tee.",
+ "normalized_hypothesis": "it was my idea tu ati",
+ "errors": 3,
+ "words": 7,
+ "wer": 0.42857142857142855
+ },
+ "semantic": {
+ "normalized_reference": "it was my idea to a tee",
+ "normalized_hypothesis": "it was my idea tu ati",
+ "errors": 3,
+ "words": 7,
+ "wer": 0.42857142857142855
+ }
+ },
+ {
+ "id": "release500_0232",
+ "category": "phonetic_balance",
+ "text": "They likewise are disinclined to being eaten.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0232.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THEY LIKEWISE ARE DISINCLINED TO BEING EATEN",
+ "raw": {
+ "normalized_reference": "they likewise are disinclined to being eaten.",
+ "normalized_hypothesis": "they likewise are disinclined to being eaten",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "they likewise are disinclined to being eaten",
+ "normalized_hypothesis": "they likewise are disinclined to being eaten",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0233",
+ "category": "phonetic_balance",
+ "text": "This state of mind comes of an undue prominence of the ego.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0233.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THIS STATE OF MIND COMES OF AN UNDUE PROMINENCE OF THE EGO",
+ "raw": {
+ "normalized_reference": "this state of mind comes of an undue prominence of the ego.",
+ "normalized_hypothesis": "this state of mind comes of an undue prominence of the ego",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "this state of mind comes of an undue prominence of the ego",
+ "normalized_hypothesis": "this state of mind comes of an undue prominence of the ego",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0234",
+ "category": "phonetic_balance",
+ "text": "Then it is as I said, Womble announced with finality.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0234.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THEN IT IS AS I SAID WAMBELLE ANNOUNCED WITH FINALITY",
+ "raw": {
+ "normalized_reference": "then it is as i said, womble announced with finality.",
+ "normalized_hypothesis": "then it is as i said wambelle announced with finality",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ },
+ "semantic": {
+ "normalized_reference": "then it is as i said womble announced with finality",
+ "normalized_hypothesis": "then it is as i said wambelle announced with finality",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ }
+ },
+ {
+ "id": "release500_0235",
+ "category": "phonetic_balance",
+ "text": "I can see that knife now.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0235.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I CAN SEE THAT KNIFE NOW",
+ "raw": {
+ "normalized_reference": "i can see that knife now.",
+ "normalized_hypothesis": "i can see that knife now",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i can see that knife now",
+ "normalized_hypothesis": "i can see that knife now",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0236",
+ "category": "phonetic_balance",
+ "text": "If I meet her again I shall apologize, said Eileen.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0236.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IF I MEET HER AGAIN I SHALL APOLOGIZE SAID ILEENE",
+ "raw": {
+ "normalized_reference": "if i meet her again i shall apologize, said eileen.",
+ "normalized_hypothesis": "if i meet her again i shall apologize said ileene",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ },
+ "semantic": {
+ "normalized_reference": "if i meet her again i shall apologize said eileen",
+ "normalized_hypothesis": "if i meet her again i shall apologize said ileene",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ }
+ },
+ {
+ "id": "release500_0237",
+ "category": "phonetic_balance",
+ "text": "A wildly exciting time was his during the week preceding Thursday the eighteenth.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0237.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "A WILDLY EXCITING TIME WAS HIS DURING THE WEEK PRECEDING THURSDAY THE EIGHTEENTH",
+ "raw": {
+ "normalized_reference": "a wildly exciting time was his during the week preceding thursday the eighteenth.",
+ "normalized_hypothesis": "a wildly exciting time was his during the week preceding thursday the eighteenth",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "a wildly exciting time was his during the week preceding thursday the 18th",
+ "normalized_hypothesis": "a wildly exciting time was his during the week preceding thursday the 18th",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0238",
+ "category": "phonetic_balance",
+ "text": "In what bucolic school of fence he had been taught was beyond imagining.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0238.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IN WHAT BUCALIK SCHOOL OF FENCE HE HAD BEEN TAUGHT WAS BEYOND IMAGINING",
+ "raw": {
+ "normalized_reference": "in what bucolic school of fence he had been taught was beyond imagining.",
+ "normalized_hypothesis": "in what bucalik school of fence he had been taught was beyond imagining",
+ "errors": 1,
+ "words": 13,
+ "wer": 0.07692307692307693
+ },
+ "semantic": {
+ "normalized_reference": "in what bucolic school of fence he had been taught was beyond imagining",
+ "normalized_hypothesis": "in what bucalik school of fence he had been taught was beyond imagining",
+ "errors": 1,
+ "words": 13,
+ "wer": 0.07692307692307693
+ }
+ },
+ {
+ "id": "release500_0239",
+ "category": "phonetic_balance",
+ "text": "So she said, the irate skipper dashed on.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0239.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SO SHE SAID THE IRATE SKIPPER DASHED ON",
+ "raw": {
+ "normalized_reference": "so she said, the irate skipper dashed on.",
+ "normalized_hypothesis": "so she said the irate skipper dashed on",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "so she said the irate skipper dashed on",
+ "normalized_hypothesis": "so she said the irate skipper dashed on",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0240",
+ "category": "phonetic_balance",
+ "text": "It is not an attempt to smash the market.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0240.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT IS NOT AN ATTEMPT TO SMASH THE MARKET",
+ "raw": {
+ "normalized_reference": "it is not an attempt to smash the market.",
+ "normalized_hypothesis": "it is not an attempt to smash the market",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it is not an attempt to smash the market",
+ "normalized_hypothesis": "it is not an attempt to smash the market",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0241",
+ "category": "phonetic_balance",
+ "text": "He destroyed everything that had belonged to the woman.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0241.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE DESTROYED EVERYTHING THAT HAD BELONGED TO THE WOMAN",
+ "raw": {
+ "normalized_reference": "he destroyed everything that had belonged to the woman.",
+ "normalized_hypothesis": "he destroyed everything that had belonged to the woman",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he destroyed everything that had belonged to the woman",
+ "normalized_hypothesis": "he destroyed everything that had belonged to the woman",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0242",
+ "category": "phonetic_balance",
+ "text": "Did Thorpe go to see any one in Churchill.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0242.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "DIDTHORPE GO TO SEE ANY ONE IN CHURCHILL",
+ "raw": {
+ "normalized_reference": "did thorpe go to see any one in churchill.",
+ "normalized_hypothesis": "didthorpe go to see any one in churchill",
+ "errors": 2,
+ "words": 9,
+ "wer": 0.2222222222222222
+ },
+ "semantic": {
+ "normalized_reference": "did thorpe go to see any one in churchill",
+ "normalized_hypothesis": "didthorpe go to see any one in churchill",
+ "errors": 2,
+ "words": 9,
+ "wer": 0.2222222222222222
+ }
+ },
+ {
+ "id": "release500_0243",
+ "category": "phonetic_balance",
+ "text": "Many other similar disconcerting omissions will be noticed in the Manuscript.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0243.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "MANY OTHER SIMILAR DISCONCERTING OMISSIONS WILL BE NOTICED IN THE MANUSCRIPT",
+ "raw": {
+ "normalized_reference": "many other similar disconcerting omissions will be noticed in the manuscript.",
+ "normalized_hypothesis": "many other similar disconcerting omissions will be noticed in the manuscript",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "many other similar disconcerting omissions will be noticed in the manuscript",
+ "normalized_hypothesis": "many other similar disconcerting omissions will be noticed in the manuscript",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0244",
+ "category": "phonetic_balance",
+ "text": "The ringing of the big bell aroused him.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0244.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE RINGING OF THE BIG BELL AROUSED HIM",
+ "raw": {
+ "normalized_reference": "the ringing of the big bell aroused him.",
+ "normalized_hypothesis": "the ringing of the big bell aroused him",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the ringing of the big bell aroused him",
+ "normalized_hypothesis": "the ringing of the big bell aroused him",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0245",
+ "category": "phonetic_balance",
+ "text": "A sinewy hand, dripping with water, was clutching the rail.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0245.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "A SINEWY HAND DRIPPING WITH WATER WAS CLUTCHING THE RAIL",
+ "raw": {
+ "normalized_reference": "a sinewy hand, dripping with water, was clutching the rail.",
+ "normalized_hypothesis": "a sinewy hand dripping with water was clutching the rail",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "a sinewy hand dripping with water was clutching the rail",
+ "normalized_hypothesis": "a sinewy hand dripping with water was clutching the rail",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0246",
+ "category": "phonetic_balance",
+ "text": "Enters now the psychology of the situation.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0246.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "ENTERS NOW THE PSYCHOLOGY OF THE SITUATION",
+ "raw": {
+ "normalized_reference": "enters now the psychology of the situation.",
+ "normalized_hypothesis": "enters now the psychology of the situation",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "enters now the psychology of the situation",
+ "normalized_hypothesis": "enters now the psychology of the situation",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0247",
+ "category": "phonetic_balance",
+ "text": "Wash your hands of me.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0247.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WASH YOUR HANDS OF ME",
+ "raw": {
+ "normalized_reference": "wash your hands of me.",
+ "normalized_hypothesis": "wash your hands of me",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "wash your hands of me",
+ "normalized_hypothesis": "wash your hands of me",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0248",
+ "category": "phonetic_balance",
+ "text": "He confessed that the sketch had startled him.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0248.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE CONFESSED THAT THE SKETCH HAD STARTLED HIM",
+ "raw": {
+ "normalized_reference": "he confessed that the sketch had startled him.",
+ "normalized_hypothesis": "he confessed that the sketch had startled him",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he confessed that the sketch had startled him",
+ "normalized_hypothesis": "he confessed that the sketch had startled him",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0249",
+ "category": "phonetic_balance",
+ "text": "Her face was against his breast.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0249.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HER FACE WAS AGAINST HIS BREAST",
+ "raw": {
+ "normalized_reference": "her face was against his breast.",
+ "normalized_hypothesis": "her face was against his breast",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "her face was against his breast",
+ "normalized_hypothesis": "her face was against his breast",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0250",
+ "category": "phonetic_balance",
+ "text": "He was a pariah; a wanderer without a friend or a home.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0250.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE WAS A PERIAH A WANDERER WITHOUT A FRIEND OR A HOME",
+ "raw": {
+ "normalized_reference": "he was a pariah; a wanderer without a friend or a home.",
+ "normalized_hypothesis": "he was a periah a wanderer without a friend or a home",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "he was a pariah a wanderer without a friend or a home",
+ "normalized_hypothesis": "he was a periah a wanderer without a friend or a home",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ }
+ },
+ {
+ "id": "release500_0251",
+ "category": "phonetic_balance",
+ "text": "A cry of joy burst from Philip's lips.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0251.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "A CRY OF JOY BURST FROM PHILIP'S LIPS",
+ "raw": {
+ "normalized_reference": "a cry of joy burst from philip's lips.",
+ "normalized_hypothesis": "a cry of joy burst from philip's lips",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "a cry of joy burst from philip is lips",
+ "normalized_hypothesis": "a cry of joy burst from philip is lips",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0252",
+ "category": "phonetic_balance",
+ "text": "And the big Persian knew of his existence before he did of hers.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0252.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "AND THE BIG PERSIAN KNEW OF HIS EXISTENCE BEFORE HE DID OF HERS",
+ "raw": {
+ "normalized_reference": "and the big persian knew of his existence before he did of hers.",
+ "normalized_hypothesis": "and the big persian knew of his existence before he did of hers",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "and the big persian knew of his existence before he did of hers",
+ "normalized_hypothesis": "and the big persian knew of his existence before he did of hers",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0253",
+ "category": "phonetic_balance",
+ "text": "The men stared into each other's face.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0253.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE MEN STARED INTO EACH OTHER'S FACE",
+ "raw": {
+ "normalized_reference": "the men stared into each other's face.",
+ "normalized_hypothesis": "the men stared into each other's face",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the men stared into each other is face",
+ "normalized_hypothesis": "the men stared into each other is face",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0254",
+ "category": "phonetic_balance",
+ "text": "It is the fire, partly, she said.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0254.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT IS THE FIRE PARTLY SHE SAID",
+ "raw": {
+ "normalized_reference": "it is the fire, partly, she said.",
+ "normalized_hypothesis": "it is the fire partly she said",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it is the fire partly she said",
+ "normalized_hypothesis": "it is the fire partly she said",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0255",
+ "category": "phonetic_balance",
+ "text": "It seemed nearer to him since he had seen and talked with Gregson.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0255.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT SEEMED NEARER TO HIM SINCE HE HAD SEEN AND TALKED WITH GREGSON",
+ "raw": {
+ "normalized_reference": "it seemed nearer to him since he had seen and talked with gregson.",
+ "normalized_hypothesis": "it seemed nearer to him since he had seen and talked with gregson",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it seemed nearer to him since he had seen and talked with gregson",
+ "normalized_hypothesis": "it seemed nearer to him since he had seen and talked with gregson",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0256",
+ "category": "phonetic_balance",
+ "text": "Among my minor afflictions, I may mention a new and mysterious one.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0256.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "AMONG MY MINOR AFFLICTIONS I MAY MENTION A NEW AND MYSTERIOUS ONE",
+ "raw": {
+ "normalized_reference": "among my minor afflictions, i may mention a new and mysterious one.",
+ "normalized_hypothesis": "among my minor afflictions i may mention a new and mysterious one",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "among my minor afflictions i may mention a new and mysterious one",
+ "normalized_hypothesis": "among my minor afflictions i may mention a new and mysterious one",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0257",
+ "category": "phonetic_balance",
+ "text": "Between him and all domestic animals there must be no hostilities.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0257.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "BETWEEN HIM AND ALL DOMESTIC ANIMALS THERE MUST BE NO HOSTILITIES",
+ "raw": {
+ "normalized_reference": "between him and all domestic animals there must be no hostilities.",
+ "normalized_hypothesis": "between him and all domestic animals there must be no hostilities",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "between him and all domestic animals there must be no hostilities",
+ "normalized_hypothesis": "between him and all domestic animals there must be no hostilities",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0258",
+ "category": "phonetic_balance",
+ "text": "With these arguments he convinced himself that he should go on alone.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0258.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WITH THESE ARGUMENTS HE CONVINCED HIMSELF THAT HE SHOULD GO ON ALONE",
+ "raw": {
+ "normalized_reference": "with these arguments he convinced himself that he should go on alone.",
+ "normalized_hypothesis": "with these arguments he convinced himself that he should go on alone",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "with these arguments he convinced himself that he should go on alone",
+ "normalized_hypothesis": "with these arguments he convinced himself that he should go on alone",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0259",
+ "category": "phonetic_balance",
+ "text": "We are consumed in our own flesh-pots.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0259.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WE ARE CONSUMED IN OUR OWN FLESH POTS",
+ "raw": {
+ "normalized_reference": "we are consumed in our own flesh-pots.",
+ "normalized_hypothesis": "we are consumed in our own flesh pots",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "we are consumed in our own flesh pots",
+ "normalized_hypothesis": "we are consumed in our own flesh pots",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0260",
+ "category": "phonetic_balance",
+ "text": "Mercedes screamed, cried, laughed, and manifested the chaotic abandonment of hysteria.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0260.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "MERCEDES SCREAMED CRIED LAUGHED AND MANIFESTED THE CHAOTIC ABANDONMENT OF HYSTERIA",
+ "raw": {
+ "normalized_reference": "mercedes screamed, cried, laughed, and manifested the chaotic abandonment of hysteria.",
+ "normalized_hypothesis": "mercedes screamed cried laughed and manifested the chaotic abandonment of hysteria",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "mercedes screamed cried laughed and manifested the chaotic abandonment of hysteria",
+ "normalized_hypothesis": "mercedes screamed cried laughed and manifested the chaotic abandonment of hysteria",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0261",
+ "category": "phonetic_balance",
+ "text": "Philip made no effort to follow.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0261.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "PHILIP MADE NO EFFORT TO FOLLOW",
+ "raw": {
+ "normalized_reference": "philip made no effort to follow.",
+ "normalized_hypothesis": "philip made no effort to follow",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "philip made no effort to follow",
+ "normalized_hypothesis": "philip made no effort to follow",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0262",
+ "category": "phonetic_balance",
+ "text": "She said with chattering teeth.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0262.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SHE SAID WITH CHATTERING TEETH",
+ "raw": {
+ "normalized_reference": "she said with chattering teeth.",
+ "normalized_hypothesis": "she said with chattering teeth",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "she said with chattering teeth",
+ "normalized_hypothesis": "she said with chattering teeth",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0263",
+ "category": "phonetic_balance",
+ "text": "Halfway around the track one donkey got into an argument with its rider.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0263.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HALF WAY AROUND THE TRACK ONE DONKEY GOT INTO AN ARGUMENT WITH ITS RIDER",
+ "raw": {
+ "normalized_reference": "halfway around the track one donkey got into an argument with its rider.",
+ "normalized_hypothesis": "half way around the track one donkey got into an argument with its rider",
+ "errors": 2,
+ "words": 13,
+ "wer": 0.15384615384615385
+ },
+ "semantic": {
+ "normalized_reference": "halfway around the track one donkey got into an argument with its rider",
+ "normalized_hypothesis": "half way around the track one donkey got into an argument with its rider",
+ "errors": 2,
+ "words": 13,
+ "wer": 0.15384615384615385
+ }
+ },
+ {
+ "id": "release500_0264",
+ "category": "phonetic_balance",
+ "text": "Earth and gravel seemed to fill the pan.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0264.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "EARTH AND GRAVEL SEEMED TO FILL THE PAN",
+ "raw": {
+ "normalized_reference": "earth and gravel seemed to fill the pan.",
+ "normalized_hypothesis": "earth and gravel seemed to fill the pan",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "earth and gravel seemed to fill the pan",
+ "normalized_hypothesis": "earth and gravel seemed to fill the pan",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0265",
+ "category": "phonetic_balance",
+ "text": "Her achievements with cocoanuts were a revelation.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0265.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HER ACHIEVEMENTS WITH COCOA NUTS WERE A REVELATION",
+ "raw": {
+ "normalized_reference": "her achievements with cocoanuts were a revelation.",
+ "normalized_hypothesis": "her achievements with cocoa nuts were a revelation",
+ "errors": 2,
+ "words": 7,
+ "wer": 0.2857142857142857
+ },
+ "semantic": {
+ "normalized_reference": "her achievements with cocoanuts were a revelation",
+ "normalized_hypothesis": "her achievements with cocoa nuts were a revelation",
+ "errors": 2,
+ "words": 7,
+ "wer": 0.2857142857142857
+ }
+ },
+ {
+ "id": "release500_0266",
+ "category": "phonetic_balance",
+ "text": "For a full minute he crouched and listened.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0266.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "FOR A FULL MINUTE HE CROUCHED AND LISTENED",
+ "raw": {
+ "normalized_reference": "for a full minute he crouched and listened.",
+ "normalized_hypothesis": "for a full minute he crouched and listened",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "for a full minute he crouched and listened",
+ "normalized_hypothesis": "for a full minute he crouched and listened",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0267",
+ "category": "phonetic_balance",
+ "text": "It was edged with ice.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0267.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT WAS EDGED WITH ICE",
+ "raw": {
+ "normalized_reference": "it was edged with ice.",
+ "normalized_hypothesis": "it was edged with ice",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it was edged with ice",
+ "normalized_hypothesis": "it was edged with ice",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0268",
+ "category": "phonetic_balance",
+ "text": "The Portuguese boy passed the Hawaiian.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0268.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE PORTUGUESE BOY PASSED THE HAWAIAN",
+ "raw": {
+ "normalized_reference": "the portuguese boy passed the hawaiian.",
+ "normalized_hypothesis": "the portuguese boy passed the hawaian",
+ "errors": 1,
+ "words": 6,
+ "wer": 0.16666666666666666
+ },
+ "semantic": {
+ "normalized_reference": "the portuguese boy passed the hawaiian",
+ "normalized_hypothesis": "the portuguese boy passed the hawaian",
+ "errors": 1,
+ "words": 6,
+ "wer": 0.16666666666666666
+ }
+ },
+ {
+ "id": "release500_0269",
+ "category": "phonetic_balance",
+ "text": "But he did not broach it, preferring to mature it carefully.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0269.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "BUT HE DID NOT BROOCH IT PREFERRING TO MATURE IT CAREFULLY",
+ "raw": {
+ "normalized_reference": "but he did not broach it, preferring to mature it carefully.",
+ "normalized_hypothesis": "but he did not brooch it preferring to mature it carefully",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ },
+ "semantic": {
+ "normalized_reference": "but he did not broach it preferring to mature it carefully",
+ "normalized_hypothesis": "but he did not brooch it preferring to mature it carefully",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ }
+ },
+ {
+ "id": "release500_0270",
+ "category": "phonetic_balance",
+ "text": "Come on, Del Mar challenged.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0270.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "COME ON DELMAR CHALLENGED",
+ "raw": {
+ "normalized_reference": "come on, del mar challenged.",
+ "normalized_hypothesis": "come on delmar challenged",
+ "errors": 2,
+ "words": 5,
+ "wer": 0.4
+ },
+ "semantic": {
+ "normalized_reference": "come on del mar challenged",
+ "normalized_hypothesis": "come on delmar challenged",
+ "errors": 2,
+ "words": 5,
+ "wer": 0.4
+ }
+ },
+ {
+ "id": "release500_0271",
+ "category": "phonetic_balance",
+ "text": "For a full half minute Jeanne looked at him without speaking.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0271.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "FOR A FULL HALF MINUTE JEAN LOOKED AT HIM WITHOUT SPEAKING",
+ "raw": {
+ "normalized_reference": "for a full half minute jeanne looked at him without speaking.",
+ "normalized_hypothesis": "for a full half minute jean looked at him without speaking",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ },
+ "semantic": {
+ "normalized_reference": "for a full half minute jeanne looked at him without speaking",
+ "normalized_hypothesis": "for a full half minute jean looked at him without speaking",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ }
+ },
+ {
+ "id": "release500_0272",
+ "category": "phonetic_balance",
+ "text": "He bore no grudges and had few enemies.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0272.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE BORE NO GRUDGES AND HAD FEW ENEMIES",
+ "raw": {
+ "normalized_reference": "he bore no grudges and had few enemies.",
+ "normalized_hypothesis": "he bore no grudges and had few enemies",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he bore no grudges and had few enemies",
+ "normalized_hypothesis": "he bore no grudges and had few enemies",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0273",
+ "category": "phonetic_balance",
+ "text": "I'll go over tomorrow afternoon.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0273.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I'LL GO OVER TO MORROW AFTERNOON",
+ "raw": {
+ "normalized_reference": "i'll go over tomorrow afternoon.",
+ "normalized_hypothesis": "i'll go over to morrow afternoon",
+ "errors": 2,
+ "words": 5,
+ "wer": 0.4
+ },
+ "semantic": {
+ "normalized_reference": "i will go over tomorrow afternoon",
+ "normalized_hypothesis": "i will go over to morrow afternoon",
+ "errors": 2,
+ "words": 6,
+ "wer": 0.3333333333333333
+ }
+ },
+ {
+ "id": "release500_0274",
+ "category": "phonetic_balance",
+ "text": "Well, there are better men in Hawaii, that's all.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0274.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WELL THERE ARE BETTER MEN IN HAWAII THAT'S ALL",
+ "raw": {
+ "normalized_reference": "well, there are better men in hawaii, that's all.",
+ "normalized_hypothesis": "well there are better men in hawaii that's all",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "well there are better men in hawaii that is all",
+ "normalized_hypothesis": "well there are better men in hawaii that is all",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0275",
+ "category": "phonetic_balance",
+ "text": "He was an athlete and a giant.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0275.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE WAS AN ATHLETE AND A GIANT",
+ "raw": {
+ "normalized_reference": "he was an athlete and a giant.",
+ "normalized_hypothesis": "he was an athlete and a giant",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he was an athlete and a giant",
+ "normalized_hypothesis": "he was an athlete and a giant",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0276",
+ "category": "phonetic_balance",
+ "text": "McVeigh when he returns from a trip to Honolulu.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0276.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "MC THAYWAY RETURNS FROM A TRIT AT THE HONOLULU",
+ "raw": {
+ "normalized_reference": "mcveigh when he returns from a trip to honolulu.",
+ "normalized_hypothesis": "mc thayway returns from a trit at the honolulu",
+ "errors": 6,
+ "words": 9,
+ "wer": 0.6666666666666666
+ },
+ "semantic": {
+ "normalized_reference": "mcveigh when he returns from a trip to honolulu",
+ "normalized_hypothesis": "mc thayway returns from a trit at the honolulu",
+ "errors": 6,
+ "words": 9,
+ "wer": 0.6666666666666666
+ }
+ },
+ {
+ "id": "release500_0277",
+ "category": "phonetic_balance",
+ "text": "You're a devil for fighting, and will surely win.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0277.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "YOU'RE A DEVIL FOR FIGHTING AND WILL SURELY WIN",
+ "raw": {
+ "normalized_reference": "you're a devil for fighting, and will surely win.",
+ "normalized_hypothesis": "you're a devil for fighting and will surely win",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "you are a devil for fighting and will surely win",
+ "normalized_hypothesis": "you are a devil for fighting and will surely win",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0278",
+ "category": "phonetic_balance",
+ "text": "All an appearance can know is mirage.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0278.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "ALL IN APPEARANCE CAN KNOW HIS MIRAGE",
+ "raw": {
+ "normalized_reference": "all an appearance can know is mirage.",
+ "normalized_hypothesis": "all in appearance can know his mirage",
+ "errors": 2,
+ "words": 7,
+ "wer": 0.2857142857142857
+ },
+ "semantic": {
+ "normalized_reference": "all an appearance can know is mirage",
+ "normalized_hypothesis": "all in appearance can know his mirage",
+ "errors": 2,
+ "words": 7,
+ "wer": 0.2857142857142857
+ }
+ },
+ {
+ "id": "release500_0279",
+ "category": "phonetic_balance",
+ "text": "You fired me out of your house, in short.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0279.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "YOU FIRED ME OUT OF YOUR HOUSE IN SHORT",
+ "raw": {
+ "normalized_reference": "you fired me out of your house, in short.",
+ "normalized_hypothesis": "you fired me out of your house in short",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "you fired me out of your house in short",
+ "normalized_hypothesis": "you fired me out of your house in short",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0280",
+ "category": "phonetic_balance",
+ "text": "He made sure that the magazine was loaded, and resumed his paddling.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0280.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE MADE SURE THAT THE MAGAZINE WAS LOADED AND RESUMED HIS PADDLING",
+ "raw": {
+ "normalized_reference": "he made sure that the magazine was loaded, and resumed his paddling.",
+ "normalized_hypothesis": "he made sure that the magazine was loaded and resumed his paddling",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he made sure that the magazine was loaded and resumed his paddling",
+ "normalized_hypothesis": "he made sure that the magazine was loaded and resumed his paddling",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0281",
+ "category": "phonetic_balance",
+ "text": "Change chairs, Daylight commanded.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0281.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "CHANGED CHERS DAYLIGHT COMMANDED",
+ "raw": {
+ "normalized_reference": "change chairs, daylight commanded.",
+ "normalized_hypothesis": "changed chers daylight commanded",
+ "errors": 2,
+ "words": 4,
+ "wer": 0.5
+ },
+ "semantic": {
+ "normalized_reference": "change chairs daylight commanded",
+ "normalized_hypothesis": "changed chers daylight commanded",
+ "errors": 2,
+ "words": 4,
+ "wer": 0.5
+ }
+ },
+ {
+ "id": "release500_0282",
+ "category": "phonetic_balance",
+ "text": "By that answer my professional medical prestige stood or fell.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0282.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "BY THAT ANSWER MY PROFESSIONAL MEDICAL PRESTIGE STOOD O'ERFELL",
+ "raw": {
+ "normalized_reference": "by that answer my professional medical prestige stood or fell.",
+ "normalized_hypothesis": "by that answer my professional medical prestige stood o'erfell",
+ "errors": 2,
+ "words": 10,
+ "wer": 0.2
+ },
+ "semantic": {
+ "normalized_reference": "by that answer my professional medical prestige stood or fell",
+ "normalized_hypothesis": "by that answer my professional medical prestige stood 0 erfell",
+ "errors": 2,
+ "words": 10,
+ "wer": 0.2
+ }
+ },
+ {
+ "id": "release500_0283",
+ "category": "phonetic_balance",
+ "text": "The moon had already begun its westward decline.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0283.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE MOON HAD ALREADY BEGUN ITS WESTWARD DECLINE",
+ "raw": {
+ "normalized_reference": "the moon had already begun its westward decline.",
+ "normalized_hypothesis": "the moon had already begun its westward decline",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the moon had already begun its westward decline",
+ "normalized_hypothesis": "the moon had already begun its westward decline",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0284",
+ "category": "phonetic_balance",
+ "text": "At sea, Monday, March 16, 1908.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0284.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "AT SE MONDAY MARCH SIXTEENTH ONE THOUSAND EIN HUNDRED EIGHT",
+ "raw": {
+ "normalized_reference": "at sea, monday, march 16, 1908.",
+ "normalized_hypothesis": "at se monday march sixteenth one thousand ein hundred eight",
+ "errors": 7,
+ "words": 6,
+ "wer": 1.1666666666666667
+ },
+ "semantic": {
+ "normalized_reference": "at sea monday march 16 1908",
+ "normalized_hypothesis": "at se monday march 16th 1000 ein 108",
+ "errors": 5,
+ "words": 6,
+ "wer": 0.8333333333333334
+ }
+ },
+ {
+ "id": "release500_0285",
+ "category": "phonetic_balance",
+ "text": "And wherever I ranged, the way lay along alcohol-drenched roads.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0285.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "AND WHEREVER I RANGED THE WAY LAY ALONG ALCOHOL DRENCHED ROADS",
+ "raw": {
+ "normalized_reference": "and wherever i ranged, the way lay along alcohol-drenched roads.",
+ "normalized_hypothesis": "and wherever i ranged the way lay along alcohol drenched roads",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "and wherever i ranged the way lay along alcohol drenched roads",
+ "normalized_hypothesis": "and wherever i ranged the way lay along alcohol drenched roads",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0286",
+ "category": "phonetic_balance",
+ "text": "The date was nearly eighteen years old.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0286.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE DATE WAS NEARLY EIGHTEEN YEARS OLD",
+ "raw": {
+ "normalized_reference": "the date was nearly eighteen years old.",
+ "normalized_hypothesis": "the date was nearly eighteen years old",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the date was nearly 18 years old",
+ "normalized_hypothesis": "the date was nearly 18 years old",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0287",
+ "category": "phonetic_balance",
+ "text": "Very early in my life, I separated from my mother.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0287.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "VERY EARLY IN MY LIFE I SEPARATED FROM MY MOTHER",
+ "raw": {
+ "normalized_reference": "very early in my life, i separated from my mother.",
+ "normalized_hypothesis": "very early in my life i separated from my mother",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "very early in my life i separated from my mother",
+ "normalized_hypothesis": "very early in my life i separated from my mother",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0288",
+ "category": "phonetic_balance",
+ "text": "There was pride and strength, the ring of triumph in his voice.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0288.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THERE WAS PRIDE AND STRENGTH THE RING OF TRIUMPH IN HIS VOICE",
+ "raw": {
+ "normalized_reference": "there was pride and strength, the ring of triumph in his voice.",
+ "normalized_hypothesis": "there was pride and strength the ring of triumph in his voice",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "there was pride and strength the ring of triumph in his voice",
+ "normalized_hypothesis": "there was pride and strength the ring of triumph in his voice",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0289",
+ "category": "phonetic_balance",
+ "text": "From my earliest recollection my sleep was a period of terror.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0289.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "FROM MY EARLIEST RECOLLECTION MY SLEEP WAS A PERIOD OF TERROR",
+ "raw": {
+ "normalized_reference": "from my earliest recollection my sleep was a period of terror.",
+ "normalized_hypothesis": "from my earliest recollection my sleep was a period of terror",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "from my earliest recollection my sleep was a period of terror",
+ "normalized_hypothesis": "from my earliest recollection my sleep was a period of terror",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0290",
+ "category": "phonetic_balance",
+ "text": "We'll have to watch our chances.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0290.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WE'LL HAVE TO WATCH OUR CHANCES",
+ "raw": {
+ "normalized_reference": "we'll have to watch our chances.",
+ "normalized_hypothesis": "we'll have to watch our chances",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "we will have to watch our chances",
+ "normalized_hypothesis": "we will have to watch our chances",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0291",
+ "category": "phonetic_balance",
+ "text": "Close beside him gleamed the white fangs of the wolf-dog.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0291.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "CLOSE BESIDE HIM BLEAMED THE WHITE FANGS OF THE WOLF DARD",
+ "raw": {
+ "normalized_reference": "close beside him gleamed the white fangs of the wolf-dog.",
+ "normalized_hypothesis": "close beside him bleamed the white fangs of the wolf dard",
+ "errors": 2,
+ "words": 11,
+ "wer": 0.18181818181818182
+ },
+ "semantic": {
+ "normalized_reference": "close beside him gleamed the white fangs of the wolf dog",
+ "normalized_hypothesis": "close beside him bleamed the white fangs of the wolf dard",
+ "errors": 2,
+ "words": 11,
+ "wer": 0.18181818181818182
+ }
+ },
+ {
+ "id": "release500_0292",
+ "category": "phonetic_balance",
+ "text": "Would you be satisfied with that one hundredth part of me.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0292.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WOULD YOU BE SATISFIED WITH THAT ONE HUNDREDTH PART OF ME",
+ "raw": {
+ "normalized_reference": "would you be satisfied with that one hundredth part of me.",
+ "normalized_hypothesis": "would you be satisfied with that one hundredth part of me",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "would you be satisfied with that 100th part of me",
+ "normalized_hypothesis": "would you be satisfied with that 100th part of me",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0293",
+ "category": "phonetic_balance",
+ "text": "Two gigantic owls were tearing at the carcass.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0293.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "TWO GIGANTIC OWLS WERE TEARING AT THE CARCASE",
+ "raw": {
+ "normalized_reference": "two gigantic owls were tearing at the carcass.",
+ "normalized_hypothesis": "two gigantic owls were tearing at the carcase",
+ "errors": 1,
+ "words": 8,
+ "wer": 0.125
+ },
+ "semantic": {
+ "normalized_reference": "2 gigantic owls were tearing at the carcass",
+ "normalized_hypothesis": "2 gigantic owls were tearing at the carcase",
+ "errors": 1,
+ "words": 8,
+ "wer": 0.125
+ }
+ },
+ {
+ "id": "release500_0294",
+ "category": "phonetic_balance",
+ "text": "The lush vegetation of that sheltered spot make a natural shield.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0294.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE LUSH VEGETATION OF THAT SHELTERED SPOT MAKE A NATURAL SHIELD",
+ "raw": {
+ "normalized_reference": "the lush vegetation of that sheltered spot make a natural shield.",
+ "normalized_hypothesis": "the lush vegetation of that sheltered spot make a natural shield",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the lush vegetation of that sheltered spot make a natural shield",
+ "normalized_hypothesis": "the lush vegetation of that sheltered spot make a natural shield",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0295",
+ "category": "phonetic_balance",
+ "text": "Below him the shadow was broken into a pool of rippling starlight.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0295.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "BELOW HIM THE SHADOW WAS BROKEN INTO A POOL OF RIPPLING STARLIGHT",
+ "raw": {
+ "normalized_reference": "below him the shadow was broken into a pool of rippling starlight.",
+ "normalized_hypothesis": "below him the shadow was broken into a pool of rippling starlight",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "below him the shadow was broken into a pool of rippling starlight",
+ "normalized_hypothesis": "below him the shadow was broken into a pool of rippling starlight",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0296",
+ "category": "phonetic_balance",
+ "text": "Lots of men take women buggy riding.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0296.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "LOTS OF MEN TAKE WOMEN BUGGY RIDING",
+ "raw": {
+ "normalized_reference": "lots of men take women buggy riding.",
+ "normalized_hypothesis": "lots of men take women buggy riding",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "lots of men take women buggy riding",
+ "normalized_hypothesis": "lots of men take women buggy riding",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0297",
+ "category": "phonetic_balance",
+ "text": "But it contributed to the smash.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0297.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "BUT IT CONTRIBUTED TO THE SPASH",
+ "raw": {
+ "normalized_reference": "but it contributed to the smash.",
+ "normalized_hypothesis": "but it contributed to the spash",
+ "errors": 1,
+ "words": 6,
+ "wer": 0.16666666666666666
+ },
+ "semantic": {
+ "normalized_reference": "but it contributed to the smash",
+ "normalized_hypothesis": "but it contributed to the spash",
+ "errors": 1,
+ "words": 6,
+ "wer": 0.16666666666666666
+ }
+ },
+ {
+ "id": "release500_0298",
+ "category": "phonetic_balance",
+ "text": "No man ate of the seal meat or the oil.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0298.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "NO MAN ATE OF THE SEAL MEAT OR THE OIL",
+ "raw": {
+ "normalized_reference": "no man ate of the seal meat or the oil.",
+ "normalized_hypothesis": "no man ate of the seal meat or the oil",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "no man ate of the seal meat or the oil",
+ "normalized_hypothesis": "no man ate of the seal meat or the oil",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0299",
+ "category": "phonetic_balance",
+ "text": "You're going in for grab sharing.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0299.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "YOU'RE GOING IN FOR A GRAB SHARING",
+ "raw": {
+ "normalized_reference": "you're going in for grab sharing.",
+ "normalized_hypothesis": "you're going in for a grab sharing",
+ "errors": 1,
+ "words": 6,
+ "wer": 0.16666666666666666
+ },
+ "semantic": {
+ "normalized_reference": "you are going in for grab sharing",
+ "normalized_hypothesis": "you are going in for a grab sharing",
+ "errors": 1,
+ "words": 7,
+ "wer": 0.14285714285714285
+ }
+ },
+ {
+ "id": "release500_0300",
+ "category": "phonetic_balance",
+ "text": "He it was that lived to found the family of the Patino.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0300.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "EIT WAS THAT LIVED TO FOUND THE FAMILY OF THE PETINO",
+ "raw": {
+ "normalized_reference": "he it was that lived to found the family of the patino.",
+ "normalized_hypothesis": "eit was that lived to found the family of the petino",
+ "errors": 3,
+ "words": 12,
+ "wer": 0.25
+ },
+ "semantic": {
+ "normalized_reference": "he it was that lived to found the family of the patino",
+ "normalized_hypothesis": "eit was that lived to found the family of the petino",
+ "errors": 3,
+ "words": 12,
+ "wer": 0.25
+ }
+ },
+ {
+ "id": "release500_0301",
+ "category": "phonetic_balance",
+ "text": "For such countries nothing remained but reorganization.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0301.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "FOR SUCH COUNTRIES NOTHING REMAINED BUT REORGANIZATION",
+ "raw": {
+ "normalized_reference": "for such countries nothing remained but reorganization.",
+ "normalized_hypothesis": "for such countries nothing remained but reorganization",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "for such countries nothing remained but reorganization",
+ "normalized_hypothesis": "for such countries nothing remained but reorganization",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0302",
+ "category": "phonetic_balance",
+ "text": "Stand off butcher and baker and all the rest.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0302.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "STAND OFF BUTCHER AND BAKER AND ALL THE REST",
+ "raw": {
+ "normalized_reference": "stand off butcher and baker and all the rest.",
+ "normalized_hypothesis": "stand off butcher and baker and all the rest",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "stand off butcher and baker and all the rest",
+ "normalized_hypothesis": "stand off butcher and baker and all the rest",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0303",
+ "category": "phonetic_balance",
+ "text": "All eyes, however, were staring at him in certitude of expectancy.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0303.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "ALL ICE HOWEVER WERE STARING AT HIM IN CERTITUDE OF EXPECTANCY",
+ "raw": {
+ "normalized_reference": "all eyes, however, were staring at him in certitude of expectancy.",
+ "normalized_hypothesis": "all ice however were staring at him in certitude of expectancy",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ },
+ "semantic": {
+ "normalized_reference": "all eyes however were staring at him in certitude of expectancy",
+ "normalized_hypothesis": "all ice however were staring at him in certitude of expectancy",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ }
+ },
+ {
+ "id": "release500_0304",
+ "category": "phonetic_balance",
+ "text": "Two days ago Jeanne learned where her father's men were hiding.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0304.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "TWO DAYS AGO JEAN LEARNED WHERE HER FATHER'S MEN WERE HIDING",
+ "raw": {
+ "normalized_reference": "two days ago jeanne learned where her father's men were hiding.",
+ "normalized_hypothesis": "two days ago jean learned where her father's men were hiding",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ },
+ "semantic": {
+ "normalized_reference": "2 days ago jeanne learned where her father is men were hiding",
+ "normalized_hypothesis": "2 days ago jean learned where her father is men were hiding",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ }
+ },
+ {
+ "id": "release500_0305",
+ "category": "phonetic_balance",
+ "text": "His previous wives had never lived long enough to bear him children.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0305.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "AS PREVIOUS WIVES HAD NEVER LIVED LONG ENOUGH TO BEAR HIM CHILDREN",
+ "raw": {
+ "normalized_reference": "his previous wives had never lived long enough to bear him children.",
+ "normalized_hypothesis": "as previous wives had never lived long enough to bear him children",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "his previous wives had never lived long enough to bear him children",
+ "normalized_hypothesis": "as previous wives had never lived long enough to bear him children",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ }
+ },
+ {
+ "id": "release500_0306",
+ "category": "phonetic_balance",
+ "text": "But this little defect did not worry him.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0306.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "BUT THIS LITTLE DEFICT DID NOT WORRY HIM",
+ "raw": {
+ "normalized_reference": "but this little defect did not worry him.",
+ "normalized_hypothesis": "but this little defict did not worry him",
+ "errors": 1,
+ "words": 8,
+ "wer": 0.125
+ },
+ "semantic": {
+ "normalized_reference": "but this little defect did not worry him",
+ "normalized_hypothesis": "but this little defict did not worry him",
+ "errors": 1,
+ "words": 8,
+ "wer": 0.125
+ }
+ },
+ {
+ "id": "release500_0307",
+ "category": "phonetic_balance",
+ "text": "Eli Harding asked, as Shunk started to follow.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0307.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "ELI HARDING ASKED AS SHUNK STARTED TO FOLLOW",
+ "raw": {
+ "normalized_reference": "eli harding asked, as shunk started to follow.",
+ "normalized_hypothesis": "eli harding asked as shunk started to follow",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "eli harding asked as shunk started to follow",
+ "normalized_hypothesis": "eli harding asked as shunk started to follow",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0308",
+ "category": "phonetic_balance",
+ "text": "And this was their sole conversation throughout the meal.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0308.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "AND THIS WAS THEIR SOLE CONVERSATION THROUGHOUT THE MEAL",
+ "raw": {
+ "normalized_reference": "and this was their sole conversation throughout the meal.",
+ "normalized_hypothesis": "and this was their sole conversation throughout the meal",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "and this was their sole conversation throughout the meal",
+ "normalized_hypothesis": "and this was their sole conversation throughout the meal",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0309",
+ "category": "phonetic_balance",
+ "text": "Then, again, Tudor had such an irritating way about him.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0309.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THEN AGAIN TUDER HAD SUCH AN IRRITATING WAY ABOUT HIM",
+ "raw": {
+ "normalized_reference": "then, again, tudor had such an irritating way about him.",
+ "normalized_hypothesis": "then again tuder had such an irritating way about him",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ },
+ "semantic": {
+ "normalized_reference": "then again tudor had such an irritating way about him",
+ "normalized_hypothesis": "then again tuder had such an irritating way about him",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ }
+ },
+ {
+ "id": "release500_0310",
+ "category": "phonetic_balance",
+ "text": "In it there was something that was almost tragedy.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0310.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IN IT THERE WAS SOMETHING THAT WAS ALMOST TRAGEDY",
+ "raw": {
+ "normalized_reference": "in it there was something that was almost tragedy.",
+ "normalized_hypothesis": "in it there was something that was almost tragedy",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "in it there was something that was almost tragedy",
+ "normalized_hypothesis": "in it there was something that was almost tragedy",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0311",
+ "category": "phonetic_balance",
+ "text": "Thirty pounds, said the captain with finality.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0311.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THIRTY POUNDS SAID THE CAPTAIN WITH FINALITY",
+ "raw": {
+ "normalized_reference": "thirty pounds, said the captain with finality.",
+ "normalized_hypothesis": "thirty pounds said the captain with finality",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "\u00a330 said the captain with finality",
+ "normalized_hypothesis": "\u00a330 said the captain with finality",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0312",
+ "category": "phonetic_balance",
+ "text": "It was over when he made his way through the ring of spectators.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0312.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT WAS OVER WHEN HE MADE HIS WAY THROUGH THE RING OF SPECTATORS",
+ "raw": {
+ "normalized_reference": "it was over when he made his way through the ring of spectators.",
+ "normalized_hypothesis": "it was over when he made his way through the ring of spectators",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it was over when he made his way through the ring of spectators",
+ "normalized_hypothesis": "it was over when he made his way through the ring of spectators",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0313",
+ "category": "phonetic_balance",
+ "text": "She had died from cold and starvation.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0313.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SHE HAD DIED FROM COLD AND STARVATION",
+ "raw": {
+ "normalized_reference": "she had died from cold and starvation.",
+ "normalized_hypothesis": "she had died from cold and starvation",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "she had died from cold and starvation",
+ "normalized_hypothesis": "she had died from cold and starvation",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0314",
+ "category": "phonetic_balance",
+ "text": "We fished sharks on Niihau together.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0314.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WE FISHED SHARKS ON NAHOW TOGETHER",
+ "raw": {
+ "normalized_reference": "we fished sharks on niihau together.",
+ "normalized_hypothesis": "we fished sharks on nahow together",
+ "errors": 1,
+ "words": 6,
+ "wer": 0.16666666666666666
+ },
+ "semantic": {
+ "normalized_reference": "we fished sharks on niihau together",
+ "normalized_hypothesis": "we fished sharks on nahow together",
+ "errors": 1,
+ "words": 6,
+ "wer": 0.16666666666666666
+ }
+ },
+ {
+ "id": "release500_0315",
+ "category": "phonetic_balance",
+ "text": "His slim hands gripped the edges of the table.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0315.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HIS SLIM HANDS GRIPPED THE EDGES OF THE TABLE",
+ "raw": {
+ "normalized_reference": "his slim hands gripped the edges of the table.",
+ "normalized_hypothesis": "his slim hands gripped the edges of the table",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "his slim hands gripped the edges of the table",
+ "normalized_hypothesis": "his slim hands gripped the edges of the table",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0316",
+ "category": "phonetic_balance",
+ "text": "We had been chased by them ourselves, more than once.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0316.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WE HAD BEEN CHASED BY THEM OURSELVES MORE THAN ONCE",
+ "raw": {
+ "normalized_reference": "we had been chased by them ourselves, more than once.",
+ "normalized_hypothesis": "we had been chased by them ourselves more than once",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "we had been chased by them ourselves more than once",
+ "normalized_hypothesis": "we had been chased by them ourselves more than once",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0317",
+ "category": "phonetic_balance",
+ "text": "Or have they already devised one.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0317.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "OR HAVE THEY ALREADY DEVISED ONE",
+ "raw": {
+ "normalized_reference": "or have they already devised one.",
+ "normalized_hypothesis": "or have they already devised one",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "or have they already devised one",
+ "normalized_hypothesis": "or have they already devised one",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0318",
+ "category": "phonetic_balance",
+ "text": "I cannot follow you, she said.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0318.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I CANNOT FOLLOW YOU SHE SAID",
+ "raw": {
+ "normalized_reference": "i cannot follow you, she said.",
+ "normalized_hypothesis": "i cannot follow you she said",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i cannot follow you she said",
+ "normalized_hypothesis": "i cannot follow you she said",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0319",
+ "category": "phonetic_balance",
+ "text": "He went down in midstream, searching the shadows of both shores.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0319.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE WENT DOWN IN MIDSTREAM SEARCHING THE SHADOWS OF BOTH SHORES",
+ "raw": {
+ "normalized_reference": "he went down in midstream, searching the shadows of both shores.",
+ "normalized_hypothesis": "he went down in midstream searching the shadows of both shores",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he went down in midstream searching the shadows of both shores",
+ "normalized_hypothesis": "he went down in midstream searching the shadows of both shores",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0320",
+ "category": "phonetic_balance",
+ "text": "I was sick once -- typhoid.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0320.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I WAS SICK ONCE TYPHOID",
+ "raw": {
+ "normalized_reference": "i was sick once -- typhoid.",
+ "normalized_hypothesis": "i was sick once typhoid",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i was sick once typhoid",
+ "normalized_hypothesis": "i was sick once typhoid",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0321",
+ "category": "phonetic_balance",
+ "text": "Why not like any railroad station or ferry depot.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0321.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WHY NOT LIKE ANY RAILROAD STATION OR FERRY DEPOT",
+ "raw": {
+ "normalized_reference": "why not like any railroad station or ferry depot.",
+ "normalized_hypothesis": "why not like any railroad station or ferry depot",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "why not like any railroad station or ferry depot",
+ "normalized_hypothesis": "why not like any railroad station or ferry depot",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0322",
+ "category": "phonetic_balance",
+ "text": "Red-Eye swung back and forth on the branch farther down.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0322.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "RED EYE SWUNG BACK AND FORTH ON THE BRANCH FARTHER DOWN",
+ "raw": {
+ "normalized_reference": "red-eye swung back and forth on the branch farther down.",
+ "normalized_hypothesis": "red eye swung back and forth on the branch farther down",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "red eye swung back and forth on the branch farther down",
+ "normalized_hypothesis": "red eye swung back and forth on the branch farther down",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0323",
+ "category": "phonetic_balance",
+ "text": "Within himself he called it no longer his own.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0323.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WITHIN HIMSELF HE CALLED IT NO LANDRE ESON",
+ "raw": {
+ "normalized_reference": "within himself he called it no longer his own.",
+ "normalized_hypothesis": "within himself he called it no landre eson",
+ "errors": 3,
+ "words": 9,
+ "wer": 0.3333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "within himself he called it no longer his own",
+ "normalized_hypothesis": "within himself he called it no landre eson",
+ "errors": 3,
+ "words": 9,
+ "wer": 0.3333333333333333
+ }
+ },
+ {
+ "id": "release500_0324",
+ "category": "phonetic_balance",
+ "text": "It lasted as a deterrent for two days.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0324.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT LASTED AS A DETERRANT FOR TWO DAYS",
+ "raw": {
+ "normalized_reference": "it lasted as a deterrent for two days.",
+ "normalized_hypothesis": "it lasted as a deterrant for two days",
+ "errors": 1,
+ "words": 8,
+ "wer": 0.125
+ },
+ "semantic": {
+ "normalized_reference": "it lasted as a deterrent for 2 days",
+ "normalized_hypothesis": "it lasted as a deterrant for 2 days",
+ "errors": 1,
+ "words": 8,
+ "wer": 0.125
+ }
+ },
+ {
+ "id": "release500_0325",
+ "category": "phonetic_balance",
+ "text": "He loved to play Chinese lottery.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0325.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE LOVED TO PLAY CHINESE LOTTERY",
+ "raw": {
+ "normalized_reference": "he loved to play chinese lottery.",
+ "normalized_hypothesis": "he loved to play chinese lottery",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he loved to play chinese lottery",
+ "normalized_hypothesis": "he loved to play chinese lottery",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0326",
+ "category": "phonetic_balance",
+ "text": "Lord, but I'm glad to see you again, Phil.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0326.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "LORD BUT I'M GLAD TO SEE YOU AGAIN PHIL",
+ "raw": {
+ "normalized_reference": "lord, but i'm glad to see you again, phil.",
+ "normalized_hypothesis": "lord but i'm glad to see you again phil",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "lord but i am glad to see you again phil",
+ "normalized_hypothesis": "lord but i am glad to see you again phil",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0327",
+ "category": "phonetic_balance",
+ "text": "There is not an iota of truth in it, certainly not.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0327.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THERE IS NOT AN IOTA OF TRUTH IN IT CERTAINLY NOT",
+ "raw": {
+ "normalized_reference": "there is not an iota of truth in it, certainly not.",
+ "normalized_hypothesis": "there is not an iota of truth in it certainly not",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "there is not an iota of truth in it certainly not",
+ "normalized_hypothesis": "there is not an iota of truth in it certainly not",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0328",
+ "category": "phonetic_balance",
+ "text": "She added, with genuine sympathy in her face and voice.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0328.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SHE ADDED WITH GENUINE SYMPATHY IN HER FACE AND VOICE",
+ "raw": {
+ "normalized_reference": "she added, with genuine sympathy in her face and voice.",
+ "normalized_hypothesis": "she added with genuine sympathy in her face and voice",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "she added with genuine sympathy in her face and voice",
+ "normalized_hypothesis": "she added with genuine sympathy in her face and voice",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0329",
+ "category": "phonetic_balance",
+ "text": "We have plenty of capital ourselves, and yet we want more.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0329.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WE HAVE PLENTY OF CAPITAL OURSELVES AND YET WE WANT MORE",
+ "raw": {
+ "normalized_reference": "we have plenty of capital ourselves, and yet we want more.",
+ "normalized_hypothesis": "we have plenty of capital ourselves and yet we want more",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "we have plenty of capital ourselves and yet we want more",
+ "normalized_hypothesis": "we have plenty of capital ourselves and yet we want more",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0330",
+ "category": "phonetic_balance",
+ "text": "And now put yourself in my place for a moment.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0330.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "AND NOW PUT YOURSELF IN MY PLACE FOR A MOMENT",
+ "raw": {
+ "normalized_reference": "and now put yourself in my place for a moment.",
+ "normalized_hypothesis": "and now put yourself in my place for a moment",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "and now put yourself in my place for a moment",
+ "normalized_hypothesis": "and now put yourself in my place for a moment",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0331",
+ "category": "phonetic_balance",
+ "text": "Fast, but endure.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0331.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "FAST BUT INDOHER",
+ "raw": {
+ "normalized_reference": "fast, but endure.",
+ "normalized_hypothesis": "fast but indoher",
+ "errors": 1,
+ "words": 3,
+ "wer": 0.3333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "fast but endure",
+ "normalized_hypothesis": "fast but indoher",
+ "errors": 1,
+ "words": 3,
+ "wer": 0.3333333333333333
+ }
+ },
+ {
+ "id": "release500_0332",
+ "category": "phonetic_balance",
+ "text": "You were looking squeamish this afternoon, he began.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0332.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "YOU WERE LOOKING SQUEAMISH THIS AFTERNOON HE BEGAN",
+ "raw": {
+ "normalized_reference": "you were looking squeamish this afternoon, he began.",
+ "normalized_hypothesis": "you were looking squeamish this afternoon he began",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "you were looking squeamish this afternoon he began",
+ "normalized_hypothesis": "you were looking squeamish this afternoon he began",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0333",
+ "category": "phonetic_balance",
+ "text": "It is the nearest refuge.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0333.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT IS THE NEAREST REFUGE",
+ "raw": {
+ "normalized_reference": "it is the nearest refuge.",
+ "normalized_hypothesis": "it is the nearest refuge",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it is the nearest refuge",
+ "normalized_hypothesis": "it is the nearest refuge",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0334",
+ "category": "phonetic_balance",
+ "text": "I want to die in it.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0334.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I WANT TO DIE IN IT",
+ "raw": {
+ "normalized_reference": "i want to die in it.",
+ "normalized_hypothesis": "i want to die in it",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i want to die in it",
+ "normalized_hypothesis": "i want to die in it",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0335",
+ "category": "phonetic_balance",
+ "text": "He stopped, and Philip nodded at the horrified question in his eyes.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0335.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE STOPPED AND ELEBOTTED AT THE HORRIFIED QUESTION AN HIS EYES",
+ "raw": {
+ "normalized_reference": "he stopped, and philip nodded at the horrified question in his eyes.",
+ "normalized_hypothesis": "he stopped and elebotted at the horrified question an his eyes",
+ "errors": 3,
+ "words": 12,
+ "wer": 0.25
+ },
+ "semantic": {
+ "normalized_reference": "he stopped and philip nodded at the horrified question in his eyes",
+ "normalized_hypothesis": "he stopped and elebotted at the horrified question an his eyes",
+ "errors": 3,
+ "words": 12,
+ "wer": 0.25
+ }
+ },
+ {
+ "id": "release500_0336",
+ "category": "phonetic_balance",
+ "text": "She turned, fearing that Jacques might see what was in her face.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0336.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SHE TURNED FEARING THAT JACK MIGHT SEE WHAT WAS IN HER FACE",
+ "raw": {
+ "normalized_reference": "she turned, fearing that jacques might see what was in her face.",
+ "normalized_hypothesis": "she turned fearing that jack might see what was in her face",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "she turned fearing that jacques might see what was in her face",
+ "normalized_hypothesis": "she turned fearing that jack might see what was in her face",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ }
+ },
+ {
+ "id": "release500_0337",
+ "category": "phonetic_balance",
+ "text": "It was a curious coincidence.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0337.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT WAS A CURIOUS COINCIDENCE",
+ "raw": {
+ "normalized_reference": "it was a curious coincidence.",
+ "normalized_hypothesis": "it was a curious coincidence",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it was a curious coincidence",
+ "normalized_hypothesis": "it was a curious coincidence",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0338",
+ "category": "phonetic_balance",
+ "text": "Daylight made no answer, and the door closed behind him.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0338.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "DAYLIGHT MADE NO ANSWER AND THE DOOR CLOSED BEHIND HIM",
+ "raw": {
+ "normalized_reference": "daylight made no answer, and the door closed behind him.",
+ "normalized_hypothesis": "daylight made no answer and the door closed behind him",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "daylight made no answer and the door closed behind him",
+ "normalized_hypothesis": "daylight made no answer and the door closed behind him",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0339",
+ "category": "phonetic_balance",
+ "text": "And he did hurt my arm.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0339.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "AND HE DID HURT MY ARM",
+ "raw": {
+ "normalized_reference": "and he did hurt my arm.",
+ "normalized_hypothesis": "and he did hurt my arm",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "and he did hurt my arm",
+ "normalized_hypothesis": "and he did hurt my arm",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0340",
+ "category": "phonetic_balance",
+ "text": "Dunham, can your boy go along with Jesse.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0340.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "DUNAM KAN YOUR BOY GO ALONG WITH JESSE",
+ "raw": {
+ "normalized_reference": "dunham, can your boy go along with jesse.",
+ "normalized_hypothesis": "dunam kan your boy go along with jesse",
+ "errors": 2,
+ "words": 8,
+ "wer": 0.25
+ },
+ "semantic": {
+ "normalized_reference": "dunham can your boy go along with jesse",
+ "normalized_hypothesis": "dunam kan your boy go along with jesse",
+ "errors": 2,
+ "words": 8,
+ "wer": 0.25
+ }
+ },
+ {
+ "id": "release500_0341",
+ "category": "phonetic_balance",
+ "text": "The creative joy, I murmured.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0341.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE CREATIVE JOIT I MURMURED",
+ "raw": {
+ "normalized_reference": "the creative joy, i murmured.",
+ "normalized_hypothesis": "the creative joit i murmured",
+ "errors": 1,
+ "words": 5,
+ "wer": 0.2
+ },
+ "semantic": {
+ "normalized_reference": "the creative joy i murmured",
+ "normalized_hypothesis": "the creative joit i murmured",
+ "errors": 1,
+ "words": 5,
+ "wer": 0.2
+ }
+ },
+ {
+ "id": "release500_0342",
+ "category": "phonetic_balance",
+ "text": "He pulled, and the log crashed down to break his back.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0342.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE PULLED AND THE LOG CRASHED DOWN TO BREAK HIS BACK",
+ "raw": {
+ "normalized_reference": "he pulled, and the log crashed down to break his back.",
+ "normalized_hypothesis": "he pulled and the log crashed down to break his back",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he pulled and the log crashed down to break his back",
+ "normalized_hypothesis": "he pulled and the log crashed down to break his back",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0343",
+ "category": "phonetic_balance",
+ "text": "By golly, the boy wins.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0343.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I GOLLY THE BOYWINS",
+ "raw": {
+ "normalized_reference": "by golly, the boy wins.",
+ "normalized_hypothesis": "i golly the boywins",
+ "errors": 3,
+ "words": 5,
+ "wer": 0.6
+ },
+ "semantic": {
+ "normalized_reference": "by golly the boy wins",
+ "normalized_hypothesis": "i golly the boywins",
+ "errors": 3,
+ "words": 5,
+ "wer": 0.6
+ }
+ },
+ {
+ "id": "release500_0344",
+ "category": "phonetic_balance",
+ "text": "The boy threw back his head with pride.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0344.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE BOY THREW BACK HIS HEAD WITH PRIDE",
+ "raw": {
+ "normalized_reference": "the boy threw back his head with pride.",
+ "normalized_hypothesis": "the boy threw back his head with pride",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the boy threw back his head with pride",
+ "normalized_hypothesis": "the boy threw back his head with pride",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0345",
+ "category": "phonetic_balance",
+ "text": "The last one I knew was an overseer.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0345.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE LAST ONE I KNEW WAS AN OVERSEER",
+ "raw": {
+ "normalized_reference": "the last one i knew was an overseer.",
+ "normalized_hypothesis": "the last one i knew was an overseer",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the last one i knew was an overseer",
+ "normalized_hypothesis": "the last one i knew was an overseer",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0346",
+ "category": "phonetic_balance",
+ "text": "Two years ago I gave up civilization for this.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0346.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "TWO YEARS AGO I GAVE UP CIVILIZATION FOR THIS",
+ "raw": {
+ "normalized_reference": "two years ago i gave up civilization for this.",
+ "normalized_hypothesis": "two years ago i gave up civilization for this",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "2 years ago i gave up civilization for this",
+ "normalized_hypothesis": "2 years ago i gave up civilization for this",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0347",
+ "category": "phonetic_balance",
+ "text": "The river bared its bosom, and snorting steamboats challenged the wilderness.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0347.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE RIVER BARED ITS BOSOM AND SNORTING STEAMBOATS CHALLENGED THE WILDERNESS",
+ "raw": {
+ "normalized_reference": "the river bared its bosom, and snorting steamboats challenged the wilderness.",
+ "normalized_hypothesis": "the river bared its bosom and snorting steamboats challenged the wilderness",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the river bared its bosom and snorting steamboats challenged the wilderness",
+ "normalized_hypothesis": "the river bared its bosom and snorting steamboats challenged the wilderness",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0348",
+ "category": "phonetic_balance",
+ "text": "Mab, she said.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0348.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "MABUT SHE SAID",
+ "raw": {
+ "normalized_reference": "mab, she said.",
+ "normalized_hypothesis": "mabut she said",
+ "errors": 1,
+ "words": 3,
+ "wer": 0.3333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "mab she said",
+ "normalized_hypothesis": "mabut she said",
+ "errors": 1,
+ "words": 3,
+ "wer": 0.3333333333333333
+ }
+ },
+ {
+ "id": "release500_0349",
+ "category": "phonetic_balance",
+ "text": "The regret in his voice was provocative of a second burst of laughter.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0349.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE REGRET IN HIS VOICE WAS PROVOCATIVE OF A SECOND BURST OF LAUGHTER",
+ "raw": {
+ "normalized_reference": "the regret in his voice was provocative of a second burst of laughter.",
+ "normalized_hypothesis": "the regret in his voice was provocative of a second burst of laughter",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the regret in his voice was provocative of a 2nd burst of laughter",
+ "normalized_hypothesis": "the regret in his voice was provocative of a 2nd burst of laughter",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0350",
+ "category": "phonetic_balance",
+ "text": "And each year something happened, and I did not go.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0350.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "AND EACH YEAR SOMETHING HAPPENED AND I DID NOT GO",
+ "raw": {
+ "normalized_reference": "and each year something happened, and i did not go.",
+ "normalized_hypothesis": "and each year something happened and i did not go",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "and each year something happened and i did not go",
+ "normalized_hypothesis": "and each year something happened and i did not go",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0351",
+ "category": "phonetic_balance",
+ "text": "Gad, your letter came just in time.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0351.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "GAD YOUR LETTER CAME JUST IN TIME",
+ "raw": {
+ "normalized_reference": "gad, your letter came just in time.",
+ "normalized_hypothesis": "gad your letter came just in time",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "gad your letter came just in time",
+ "normalized_hypothesis": "gad your letter came just in time",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0352",
+ "category": "phonetic_balance",
+ "text": "But if Pierre did not return, until tomorrow.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0352.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "BUT IF THARE DID NOT RETURN UNTIL TO MORROW",
+ "raw": {
+ "normalized_reference": "but if pierre did not return, until tomorrow.",
+ "normalized_hypothesis": "but if thare did not return until to morrow",
+ "errors": 3,
+ "words": 8,
+ "wer": 0.375
+ },
+ "semantic": {
+ "normalized_reference": "but if pierre did not return until tomorrow",
+ "normalized_hypothesis": "but if thare did not return until to morrow",
+ "errors": 3,
+ "words": 8,
+ "wer": 0.375
+ }
+ },
+ {
+ "id": "release500_0353",
+ "category": "phonetic_balance",
+ "text": "Billinger may arrive in time.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0353.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "BILLINGDER MAY ARRIVE IN TIME",
+ "raw": {
+ "normalized_reference": "billinger may arrive in time.",
+ "normalized_hypothesis": "billingder may arrive in time",
+ "errors": 1,
+ "words": 5,
+ "wer": 0.2
+ },
+ "semantic": {
+ "normalized_reference": "billinger may arrive in time",
+ "normalized_hypothesis": "billingder may arrive in time",
+ "errors": 1,
+ "words": 5,
+ "wer": 0.2
+ }
+ },
+ {
+ "id": "release500_0354",
+ "category": "phonetic_balance",
+ "text": "In such a tumbling of values was no time to sell.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0354.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IN SUCH A TUMBLING OF VALUES WAS NO TIME TO SELL",
+ "raw": {
+ "normalized_reference": "in such a tumbling of values was no time to sell.",
+ "normalized_hypothesis": "in such a tumbling of values was no time to sell",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "in such a tumbling of values was no time to sell",
+ "normalized_hypothesis": "in such a tumbling of values was no time to sell",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0355",
+ "category": "phonetic_balance",
+ "text": "There's not an iota of truth in it.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0355.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THERE'S NOT AN IOTA OF TRUTH IN IT",
+ "raw": {
+ "normalized_reference": "there's not an iota of truth in it.",
+ "normalized_hypothesis": "there's not an iota of truth in it",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "there is not an iota of truth in it",
+ "normalized_hypothesis": "there is not an iota of truth in it",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0356",
+ "category": "phonetic_balance",
+ "text": "It is dog eat dog, and you ate them up.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0356.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT IS DOG EAT DOD AND YOU EAT THEM UP",
+ "raw": {
+ "normalized_reference": "it is dog eat dog, and you ate them up.",
+ "normalized_hypothesis": "it is dog eat dod and you eat them up",
+ "errors": 2,
+ "words": 10,
+ "wer": 0.2
+ },
+ "semantic": {
+ "normalized_reference": "it is dog eat dog and you ate them up",
+ "normalized_hypothesis": "it is dog eat dod and you eat them up",
+ "errors": 2,
+ "words": 10,
+ "wer": 0.2
+ }
+ },
+ {
+ "id": "release500_0357",
+ "category": "phonetic_balance",
+ "text": "You can take a vacation on pay.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0357.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "YOU CAN TAKE A VACATION ON PAY",
+ "raw": {
+ "normalized_reference": "you can take a vacation on pay.",
+ "normalized_hypothesis": "you can take a vacation on pay",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "you can take a vacation on pay",
+ "normalized_hypothesis": "you can take a vacation on pay",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0358",
+ "category": "phonetic_balance",
+ "text": "I have seen myself that one man contemplated by Pascal's philosophic eye.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0358.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I HAVE SEEN MYSELF THAT ONE MAN CONTEMPLATED BY PASCALL'S PHILOSOPHIC EYE",
+ "raw": {
+ "normalized_reference": "i have seen myself that one man contemplated by pascal's philosophic eye.",
+ "normalized_hypothesis": "i have seen myself that one man contemplated by pascall's philosophic eye",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "i have seen myself that one man contemplated by pascal is philosophic eye",
+ "normalized_hypothesis": "i have seen myself that one man contemplated by pascall is philosophic eye",
+ "errors": 1,
+ "words": 13,
+ "wer": 0.07692307692307693
+ }
+ },
+ {
+ "id": "release500_0359",
+ "category": "phonetic_balance",
+ "text": "Ah, I had forgotten, he exclaimed.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0359.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "AH I HAD FORGOTTEN HE EXCLAIMED",
+ "raw": {
+ "normalized_reference": "ah, i had forgotten, he exclaimed.",
+ "normalized_hypothesis": "ah i had forgotten he exclaimed",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "ah i had forgotten he exclaimed",
+ "normalized_hypothesis": "ah i had forgotten he exclaimed",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0360",
+ "category": "phonetic_balance",
+ "text": "Providence had delivered him through the maelstrom.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0360.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "PROVIDENCE HAD DELIVERED HIM THROUGH THE MAILSTROME",
+ "raw": {
+ "normalized_reference": "providence had delivered him through the maelstrom.",
+ "normalized_hypothesis": "providence had delivered him through the mailstrome",
+ "errors": 1,
+ "words": 7,
+ "wer": 0.14285714285714285
+ },
+ "semantic": {
+ "normalized_reference": "providence had delivered him through the maelstrom",
+ "normalized_hypothesis": "providence had delivered him through the mailstrome",
+ "errors": 1,
+ "words": 7,
+ "wer": 0.14285714285714285
+ }
+ },
+ {
+ "id": "release500_0361",
+ "category": "phonetic_balance",
+ "text": "She was his now, forever.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0361.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SHE WAS HIS NOW FOREVER",
+ "raw": {
+ "normalized_reference": "she was his now, forever.",
+ "normalized_hypothesis": "she was his now forever",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "she was his now forever",
+ "normalized_hypothesis": "she was his now forever",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0362",
+ "category": "phonetic_balance",
+ "text": "It was unobtrusive, yet it was there.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0362.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT WAS UNOBTRUSIVE YET IT WAS THERE",
+ "raw": {
+ "normalized_reference": "it was unobtrusive, yet it was there.",
+ "normalized_hypothesis": "it was unobtrusive yet it was there",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it was unobtrusive yet it was there",
+ "normalized_hypothesis": "it was unobtrusive yet it was there",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0363",
+ "category": "phonetic_balance",
+ "text": "Men like Joe Goose dated existence from drunk to drunk.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0363.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "MEN LIKED JOGOU'S DATED EXISTENCE FROM DRUNK TO DRUNK",
+ "raw": {
+ "normalized_reference": "men like joe goose dated existence from drunk to drunk.",
+ "normalized_hypothesis": "men liked jogou's dated existence from drunk to drunk",
+ "errors": 3,
+ "words": 10,
+ "wer": 0.3
+ },
+ "semantic": {
+ "normalized_reference": "men like joe goose dated existence from drunk to drunk",
+ "normalized_hypothesis": "men liked jogou is dated existence from drunk to drunk",
+ "errors": 3,
+ "words": 10,
+ "wer": 0.3
+ }
+ },
+ {
+ "id": "release500_0364",
+ "category": "phonetic_balance",
+ "text": "I was completely lost in my work.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0364.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I WAS COMPLETELY LOST IN MY WORK",
+ "raw": {
+ "normalized_reference": "i was completely lost in my work.",
+ "normalized_hypothesis": "i was completely lost in my work",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i was completely lost in my work",
+ "normalized_hypothesis": "i was completely lost in my work",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0365",
+ "category": "phonetic_balance",
+ "text": "The very thing, Ernest agreed.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0365.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE VERY THING ERNEST AGREED",
+ "raw": {
+ "normalized_reference": "the very thing, ernest agreed.",
+ "normalized_hypothesis": "the very thing ernest agreed",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the very thing ernest agreed",
+ "normalized_hypothesis": "the very thing ernest agreed",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0366",
+ "category": "phonetic_balance",
+ "text": "He was the leader, and Tudor was his lieutenant.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0366.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE WAS THE LEADER AND TUDER WAS HIS LIEUTENANT",
+ "raw": {
+ "normalized_reference": "he was the leader, and tudor was his lieutenant.",
+ "normalized_hypothesis": "he was the leader and tuder was his lieutenant",
+ "errors": 1,
+ "words": 9,
+ "wer": 0.1111111111111111
+ },
+ "semantic": {
+ "normalized_reference": "he was the leader and tudor was his lieutenant",
+ "normalized_hypothesis": "he was the leader and tuder was his lieutenant",
+ "errors": 1,
+ "words": 9,
+ "wer": 0.1111111111111111
+ }
+ },
+ {
+ "id": "release500_0367",
+ "category": "phonetic_balance",
+ "text": "The Fire-Men wore animal skins around their waists and across their shoulders.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0367.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE FIREMEN WORE ANIMAL SKINS AROUND THEIR WAISTS AND ACROSS THEIR SHOULDERS",
+ "raw": {
+ "normalized_reference": "the fire-men wore animal skins around their waists and across their shoulders.",
+ "normalized_hypothesis": "the firemen wore animal skins around their waists and across their shoulders",
+ "errors": 2,
+ "words": 13,
+ "wer": 0.15384615384615385
+ },
+ "semantic": {
+ "normalized_reference": "the fire men wore animal skins around their waists and across their shoulders",
+ "normalized_hypothesis": "the firemen wore animal skins around their waists and across their shoulders",
+ "errors": 2,
+ "words": 13,
+ "wer": 0.15384615384615385
+ }
+ },
+ {
+ "id": "release500_0368",
+ "category": "phonetic_balance",
+ "text": "We could throw stones with our feet.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0368.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WE COULD THROW STONES WITH OUR FEET",
+ "raw": {
+ "normalized_reference": "we could throw stones with our feet.",
+ "normalized_hypothesis": "we could throw stones with our feet",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "we could throw stones with our feet",
+ "normalized_hypothesis": "we could throw stones with our feet",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0369",
+ "category": "phonetic_balance",
+ "text": "He turned sharply, and faced Gregson across the table.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0369.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE TURNED SHARPLY AND FACED GREGSON ACROSS THE TABLE",
+ "raw": {
+ "normalized_reference": "he turned sharply, and faced gregson across the table.",
+ "normalized_hypothesis": "he turned sharply and faced gregson across the table",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he turned sharply and faced gregson across the table",
+ "normalized_hypothesis": "he turned sharply and faced gregson across the table",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0370",
+ "category": "phonetic_balance",
+ "text": "They are to attack your camp tomorrow night.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0370.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THEY ARE TO ATTACK YOUR CAMP TO MORROW NIGHT",
+ "raw": {
+ "normalized_reference": "they are to attack your camp tomorrow night.",
+ "normalized_hypothesis": "they are to attack your camp to morrow night",
+ "errors": 2,
+ "words": 8,
+ "wer": 0.25
+ },
+ "semantic": {
+ "normalized_reference": "they are to attack your camp tomorrow night",
+ "normalized_hypothesis": "they are to attack your camp to morrow night",
+ "errors": 2,
+ "words": 8,
+ "wer": 0.25
+ }
+ },
+ {
+ "id": "release500_0371",
+ "category": "phonetic_balance",
+ "text": "And watch out for wet feet, was his parting advice.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0371.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "AND WATCH OUT FOR WHAT FEET WAS HIS PARTING ADVICE",
+ "raw": {
+ "normalized_reference": "and watch out for wet feet, was his parting advice.",
+ "normalized_hypothesis": "and watch out for what feet was his parting advice",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ },
+ "semantic": {
+ "normalized_reference": "and watch out for wet feet was his parting advice",
+ "normalized_hypothesis": "and watch out for what feet was his parting advice",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ }
+ },
+ {
+ "id": "release500_0372",
+ "category": "phonetic_balance",
+ "text": "Two weeks passed, and in that time Thorpe left camp three times.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0372.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "TWO WEEKS PASSED AND IN THAT TIME THORPE LEFT CAMP THREE TIMES",
+ "raw": {
+ "normalized_reference": "two weeks passed, and in that time thorpe left camp three times.",
+ "normalized_hypothesis": "two weeks passed and in that time thorpe left camp three times",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "2 weeks passed and in that time thorpe left camp 3 times",
+ "normalized_hypothesis": "2 weeks passed and in that time thorpe left camp 3 times",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0373",
+ "category": "phonetic_balance",
+ "text": "But ever his gaze returned to that Crouched Venus on the piano.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0373.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "BUT EVER HIS GAZE RETURNED TO THAT CROUCHED VENUS ON THE PIANO",
+ "raw": {
+ "normalized_reference": "but ever his gaze returned to that crouched venus on the piano.",
+ "normalized_hypothesis": "but ever his gaze returned to that crouched venus on the piano",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "but ever his gaze returned to that crouched venus on the piano",
+ "normalized_hypothesis": "but ever his gaze returned to that crouched venus on the piano",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0374",
+ "category": "phonetic_balance",
+ "text": "But she had become an automaton.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0374.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "BUT SHE HAD BECOME AN AUTOMATON",
+ "raw": {
+ "normalized_reference": "but she had become an automaton.",
+ "normalized_hypothesis": "but she had become an automaton",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "but she had become an automaton",
+ "normalized_hypothesis": "but she had become an automaton",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0375",
+ "category": "phonetic_balance",
+ "text": "I do not blame you for anything; remember that.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0375.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I DO NOT BLAME YOU FOR ANYTHING REMEMBER THAT",
+ "raw": {
+ "normalized_reference": "i do not blame you for anything; remember that.",
+ "normalized_hypothesis": "i do not blame you for anything remember that",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i do not blame you for anything remember that",
+ "normalized_hypothesis": "i do not blame you for anything remember that",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0376",
+ "category": "phonetic_balance",
+ "text": "The next thing to watch out for is bed sores.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0376.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE NEXT THING TO WATCH OUT FURDA'S BEDSORES",
+ "raw": {
+ "normalized_reference": "the next thing to watch out for is bed sores.",
+ "normalized_hypothesis": "the next thing to watch out furda's bedsores",
+ "errors": 4,
+ "words": 10,
+ "wer": 0.4
+ },
+ "semantic": {
+ "normalized_reference": "the next thing to watch out for is bed sores",
+ "normalized_hypothesis": "the next thing to watch out furda is bedsores",
+ "errors": 3,
+ "words": 10,
+ "wer": 0.3
+ }
+ },
+ {
+ "id": "release500_0377",
+ "category": "phonetic_balance",
+ "text": "He will follow us soon.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0377.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE WILL FOLLOW US SOON",
+ "raw": {
+ "normalized_reference": "he will follow us soon.",
+ "normalized_hypothesis": "he will follow us soon",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he will follow us soon",
+ "normalized_hypothesis": "he will follow us soon",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0378",
+ "category": "phonetic_balance",
+ "text": "His beady black eyes saw bargains where other men saw bankruptcy.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0378.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HIS BEADY BLACKEYES SAW BARGAINS WHERE OTHER MEN SAW BANKRUPTCY",
+ "raw": {
+ "normalized_reference": "his beady black eyes saw bargains where other men saw bankruptcy.",
+ "normalized_hypothesis": "his beady blackeyes saw bargains where other men saw bankruptcy",
+ "errors": 2,
+ "words": 11,
+ "wer": 0.18181818181818182
+ },
+ "semantic": {
+ "normalized_reference": "his beady black eyes saw bargains where other men saw bankruptcy",
+ "normalized_hypothesis": "his beady blackeyes saw bargains where other men saw bankruptcy",
+ "errors": 2,
+ "words": 11,
+ "wer": 0.18181818181818182
+ }
+ },
+ {
+ "id": "release500_0379",
+ "category": "phonetic_balance",
+ "text": "They robbed me a few years later.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0379.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THEY ROBBED ME A FEW YEARS LATER",
+ "raw": {
+ "normalized_reference": "they robbed me a few years later.",
+ "normalized_hypothesis": "they robbed me a few years later",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "they robbed me a few years later",
+ "normalized_hypothesis": "they robbed me a few years later",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0380",
+ "category": "phonetic_balance",
+ "text": "Gregson held a lighted match until it burnt his fingertips.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0380.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "GREGSON HELD A LIGHTED MATCH UNTIL IT BURNT HIS FINGER TIPS",
+ "raw": {
+ "normalized_reference": "gregson held a lighted match until it burnt his fingertips.",
+ "normalized_hypothesis": "gregson held a lighted match until it burnt his finger tips",
+ "errors": 2,
+ "words": 10,
+ "wer": 0.2
+ },
+ "semantic": {
+ "normalized_reference": "gregson held a lighted match until it burnt his fingertips",
+ "normalized_hypothesis": "gregson held a lighted match until it burnt his finger tips",
+ "errors": 2,
+ "words": 10,
+ "wer": 0.2
+ }
+ },
+ {
+ "id": "release500_0381",
+ "category": "phonetic_balance",
+ "text": "A burst of laughter was his reward.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0381.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "A BURST OF LAUGHTER WAS HIS REWARD",
+ "raw": {
+ "normalized_reference": "a burst of laughter was his reward.",
+ "normalized_hypothesis": "a burst of laughter was his reward",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "a burst of laughter was his reward",
+ "normalized_hypothesis": "a burst of laughter was his reward",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0382",
+ "category": "phonetic_balance",
+ "text": "He cried, and swung the club wildly.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0382.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE CRIED AND SWUNG THE CLUB WILDLY",
+ "raw": {
+ "normalized_reference": "he cried, and swung the club wildly.",
+ "normalized_hypothesis": "he cried and swung the club wildly",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he cried and swung the club wildly",
+ "normalized_hypothesis": "he cried and swung the club wildly",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0383",
+ "category": "phonetic_balance",
+ "text": "Sandel would never become a world champion.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0383.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SANDAL WOULD NEVER BECOME A WORLD CHAMPION",
+ "raw": {
+ "normalized_reference": "sandel would never become a world champion.",
+ "normalized_hypothesis": "sandal would never become a world champion",
+ "errors": 1,
+ "words": 7,
+ "wer": 0.14285714285714285
+ },
+ "semantic": {
+ "normalized_reference": "sandel would never become a world champion",
+ "normalized_hypothesis": "sandal would never become a world champion",
+ "errors": 1,
+ "words": 7,
+ "wer": 0.14285714285714285
+ }
+ },
+ {
+ "id": "release500_0384",
+ "category": "phonetic_balance",
+ "text": "And after the bath a shave would not be bad.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0384.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "AND AFTER THE BATH A SHAVE WOULD NOT BE BAD",
+ "raw": {
+ "normalized_reference": "and after the bath a shave would not be bad.",
+ "normalized_hypothesis": "and after the bath a shave would not be bad",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "and after the bath a shave would not be bad",
+ "normalized_hypothesis": "and after the bath a shave would not be bad",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0385",
+ "category": "phonetic_balance",
+ "text": "They saw each other for the first time in Boston.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0385.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THEY SOUGHT EACH OTHER FOR THE FIRST TIME IN BOSTON",
+ "raw": {
+ "normalized_reference": "they saw each other for the first time in boston.",
+ "normalized_hypothesis": "they sought each other for the first time in boston",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ },
+ "semantic": {
+ "normalized_reference": "they saw each other for the 1st time in boston",
+ "normalized_hypothesis": "they sought each other for the 1st time in boston",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ }
+ },
+ {
+ "id": "release500_0386",
+ "category": "phonetic_balance",
+ "text": "I had been born with no organic, chemical predisposition toward alcohol.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0386.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I HAD BEEN BORN WITH NO ORGANIC CHEMICAL PREDISPOSITION TOWARD ALCOHOL",
+ "raw": {
+ "normalized_reference": "i had been born with no organic, chemical predisposition toward alcohol.",
+ "normalized_hypothesis": "i had been born with no organic chemical predisposition toward alcohol",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i had been born with no organic chemical predisposition toward alcohol",
+ "normalized_hypothesis": "i had been born with no organic chemical predisposition toward alcohol",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0387",
+ "category": "phonetic_balance",
+ "text": "They had no fixed values, to be altered by adjectives and adverbs.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0387.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THEY HAD NO FIXED VALUES TO BE ALTERED BY ADJECTIVES AND ADVERBS",
+ "raw": {
+ "normalized_reference": "they had no fixed values, to be altered by adjectives and adverbs.",
+ "normalized_hypothesis": "they had no fixed values to be altered by adjectives and adverbs",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "they had no fixed values to be altered by adjectives and adverbs",
+ "normalized_hypothesis": "they had no fixed values to be altered by adjectives and adverbs",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0388",
+ "category": "phonetic_balance",
+ "text": "Nor was Elam Harnish an exception.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0388.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "NOR WAS THE LAMB HARNISH AN EXCEPTION",
+ "raw": {
+ "normalized_reference": "nor was elam harnish an exception.",
+ "normalized_hypothesis": "nor was the lamb harnish an exception",
+ "errors": 2,
+ "words": 6,
+ "wer": 0.3333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "nor was elam harnish an exception",
+ "normalized_hypothesis": "nor was the lamb harnish an exception",
+ "errors": 2,
+ "words": 6,
+ "wer": 0.3333333333333333
+ }
+ },
+ {
+ "id": "release500_0389",
+ "category": "phonetic_balance",
+ "text": "He was a wise hyena.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0389.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE WAS A WISE HYENA",
+ "raw": {
+ "normalized_reference": "he was a wise hyena.",
+ "normalized_hypothesis": "he was a wise hyena",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he was a wise hyena",
+ "normalized_hypothesis": "he was a wise hyena",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0390",
+ "category": "phonetic_balance",
+ "text": "And now behold the perversity of things.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0390.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "AND NOW BEHOLD THE PERVERSITY OF THINGS",
+ "raw": {
+ "normalized_reference": "and now behold the perversity of things.",
+ "normalized_hypothesis": "and now behold the perversity of things",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "and now behold the perversity of things",
+ "normalized_hypothesis": "and now behold the perversity of things",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0391",
+ "category": "phonetic_balance",
+ "text": "The Fire People, like ourselves, lived in caves.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0391.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE FIRE PEOPLE LIKE OURSELVES LIVED IN CAVES",
+ "raw": {
+ "normalized_reference": "the fire people, like ourselves, lived in caves.",
+ "normalized_hypothesis": "the fire people like ourselves lived in caves",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the fire people like ourselves lived in caves",
+ "normalized_hypothesis": "the fire people like ourselves lived in caves",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0392",
+ "category": "phonetic_balance",
+ "text": "To her the bridge was tambo, which is the native for taboo.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0392.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "TO HER THE BRIDGE WAS TAMBO WHICH IS THE NATIVE FOR TABOO",
+ "raw": {
+ "normalized_reference": "to her the bridge was tambo, which is the native for taboo.",
+ "normalized_hypothesis": "to her the bridge was tambo which is the native for taboo",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "to her the bridge was tambo which is the native for taboo",
+ "normalized_hypothesis": "to her the bridge was tambo which is the native for taboo",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0393",
+ "category": "phonetic_balance",
+ "text": "The voyage was our idea of a good time.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0393.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE VOYAGE WAS OUR IDEA OF A GOOD TIME",
+ "raw": {
+ "normalized_reference": "the voyage was our idea of a good time.",
+ "normalized_hypothesis": "the voyage was our idea of a good time",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the voyage was our idea of a good time",
+ "normalized_hypothesis": "the voyage was our idea of a good time",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0394",
+ "category": "phonetic_balance",
+ "text": "Of course much grumbling went on, and little outbursts were continually occurring.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0394.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "OF COURSE MUCH GRUMBLING WENT ON AND LITTLE OUTBURSTS WERE CONTINUALLY OCCURRING",
+ "raw": {
+ "normalized_reference": "of course much grumbling went on, and little outbursts were continually occurring.",
+ "normalized_hypothesis": "of course much grumbling went on and little outbursts were continually occurring",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "of course much grumbling went on and little outbursts were continually occurring",
+ "normalized_hypothesis": "of course much grumbling went on and little outbursts were continually occurring",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0395",
+ "category": "phonetic_balance",
+ "text": "Get down and dig in.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0395.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "GET DOWN AND DIG IN",
+ "raw": {
+ "normalized_reference": "get down and dig in.",
+ "normalized_hypothesis": "get down and dig in",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "get down and dig in",
+ "normalized_hypothesis": "get down and dig in",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0396",
+ "category": "phonetic_balance",
+ "text": "It was put together in a casual, helter-skelter sort of way.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0396.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT WAS PUT TOGETHER IN A CASUAL HELTER SKELTER SORT OF WAY",
+ "raw": {
+ "normalized_reference": "it was put together in a casual, helter-skelter sort of way.",
+ "normalized_hypothesis": "it was put together in a casual helter skelter sort of way",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it was put together in a casual helter skelter sort of way",
+ "normalized_hypothesis": "it was put together in a casual helter skelter sort of way",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0397",
+ "category": "phonetic_balance",
+ "text": "He deluged me, overwhelmed me with argument.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0397.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE DELUGED ME OVERWHELMED ME WITH ARGUMENT",
+ "raw": {
+ "normalized_reference": "he deluged me, overwhelmed me with argument.",
+ "normalized_hypothesis": "he deluged me overwhelmed me with argument",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he deluged me overwhelmed me with argument",
+ "normalized_hypothesis": "he deluged me overwhelmed me with argument",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0398",
+ "category": "phonetic_balance",
+ "text": "The girl faced him, her eyes shining with sudden fear.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0398.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE GIRL FACED HIM HER EYES SHINING WITH SUDDEN FEAR",
+ "raw": {
+ "normalized_reference": "the girl faced him, her eyes shining with sudden fear.",
+ "normalized_hypothesis": "the girl faced him her eyes shining with sudden fear",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the girl faced him her eyes shining with sudden fear",
+ "normalized_hypothesis": "the girl faced him her eyes shining with sudden fear",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0399",
+ "category": "phonetic_balance",
+ "text": "It will break our hearts and our backs to hoist anchor by hand.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0399.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT WILL BREAK OUR HEARTS AND OUR BACKS TO HOIST ANCHOR BY HAND",
+ "raw": {
+ "normalized_reference": "it will break our hearts and our backs to hoist anchor by hand.",
+ "normalized_hypothesis": "it will break our hearts and our backs to hoist anchor by hand",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it will break our hearts and our backs to hoist anchor by hand",
+ "normalized_hypothesis": "it will break our hearts and our backs to hoist anchor by hand",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0400",
+ "category": "phonetic_balance",
+ "text": "He made no reply as he waited for Whittemore to continue.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0400.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE MADE NO REPLY AS HE WAITED FOR WIDAMORE TO CONTINUE",
+ "raw": {
+ "normalized_reference": "he made no reply as he waited for whittemore to continue.",
+ "normalized_hypothesis": "he made no reply as he waited for widamore to continue",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ },
+ "semantic": {
+ "normalized_reference": "he made no reply as he waited for whittemore to continue",
+ "normalized_hypothesis": "he made no reply as he waited for widamore to continue",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ }
+ },
+ {
+ "id": "release500_0401",
+ "category": "phonetic_balance",
+ "text": "Who the devil gave it to you to be judge and jury.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0401.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WHO THE DEVIL GAVE IT TO YOU TO BE JUDGE AND JURY",
+ "raw": {
+ "normalized_reference": "who the devil gave it to you to be judge and jury.",
+ "normalized_hypothesis": "who the devil gave it to you to be judge and jury",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "who the devil gave it to you to be judge and jury",
+ "normalized_hypothesis": "who the devil gave it to you to be judge and jury",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0402",
+ "category": "phonetic_balance",
+ "text": "Daylight found himself charmed and made curious by the little man.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0402.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "DAYLIGHT FOUND HIMSELF CHARMED AND MADE CURIOUS BY THE LITTLE MAN",
+ "raw": {
+ "normalized_reference": "daylight found himself charmed and made curious by the little man.",
+ "normalized_hypothesis": "daylight found himself charmed and made curious by the little man",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "daylight found himself charmed and made curious by the little man",
+ "normalized_hypothesis": "daylight found himself charmed and made curious by the little man",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0403",
+ "category": "phonetic_balance",
+ "text": "It's a Yankee, Joan cried.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0403.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT'S A YANKEE JONE CRIED",
+ "raw": {
+ "normalized_reference": "it's a yankee, joan cried.",
+ "normalized_hypothesis": "it's a yankee jone cried",
+ "errors": 1,
+ "words": 5,
+ "wer": 0.2
+ },
+ "semantic": {
+ "normalized_reference": "it is a yankee joan cried",
+ "normalized_hypothesis": "it is a yankee jone cried",
+ "errors": 1,
+ "words": 6,
+ "wer": 0.16666666666666666
+ }
+ },
+ {
+ "id": "release500_0404",
+ "category": "phonetic_balance",
+ "text": "But what they want with your toothbrush is more than I can imagine.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0404.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "BUT WHAT THEY WANT WITH YOUR TOOTH BRUSH IS MORE THAN I CAN IMAGINE",
+ "raw": {
+ "normalized_reference": "but what they want with your toothbrush is more than i can imagine.",
+ "normalized_hypothesis": "but what they want with your tooth brush is more than i can imagine",
+ "errors": 2,
+ "words": 13,
+ "wer": 0.15384615384615385
+ },
+ "semantic": {
+ "normalized_reference": "but what they want with your toothbrush is more than i can imagine",
+ "normalized_hypothesis": "but what they want with your tooth brush is more than i can imagine",
+ "errors": 2,
+ "words": 13,
+ "wer": 0.15384615384615385
+ }
+ },
+ {
+ "id": "release500_0405",
+ "category": "phonetic_balance",
+ "text": "They were babbling and chattering all together.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0405.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THEY WERE BABBLING AND CHATTERING ALTOGETHER",
+ "raw": {
+ "normalized_reference": "they were babbling and chattering all together.",
+ "normalized_hypothesis": "they were babbling and chattering altogether",
+ "errors": 2,
+ "words": 7,
+ "wer": 0.2857142857142857
+ },
+ "semantic": {
+ "normalized_reference": "they were babbling and chattering all together",
+ "normalized_hypothesis": "they were babbling and chattering altogether",
+ "errors": 2,
+ "words": 7,
+ "wer": 0.2857142857142857
+ }
+ },
+ {
+ "id": "release500_0406",
+ "category": "phonetic_balance",
+ "text": "Men of Selden's stamp don't stop at women and children.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0406.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "MEN OF SELDOM'S STAMP DON'T STOP AT WOMEN AND CHILDREN",
+ "raw": {
+ "normalized_reference": "men of selden's stamp don't stop at women and children.",
+ "normalized_hypothesis": "men of seldom's stamp don't stop at women and children",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ },
+ "semantic": {
+ "normalized_reference": "men of selden is stamp do not stop at women and children",
+ "normalized_hypothesis": "men of seldom is stamp do not stop at women and children",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ }
+ },
+ {
+ "id": "release500_0407",
+ "category": "phonetic_balance",
+ "text": "Cameron looked at his hands with their long, sinewy fingers.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0407.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "CAMERON LOOKED AT HIS HANDS WITH THEIR LONG SINEWY FINGERS",
+ "raw": {
+ "normalized_reference": "cameron looked at his hands with their long, sinewy fingers.",
+ "normalized_hypothesis": "cameron looked at his hands with their long sinewy fingers",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "cameron looked at his hands with their long sinewy fingers",
+ "normalized_hypothesis": "cameron looked at his hands with their long sinewy fingers",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0408",
+ "category": "phonetic_balance",
+ "text": "Saxon waited, for she knew a fresh idea had struck Billy.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0408.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SAXON WAITED FOR SHE KNEW A FRESH IDEA HAD STRUCK BILLY",
+ "raw": {
+ "normalized_reference": "saxon waited, for she knew a fresh idea had struck billy.",
+ "normalized_hypothesis": "saxon waited for she knew a fresh idea had struck billy",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "saxon waited for she knew a fresh idea had struck billy",
+ "normalized_hypothesis": "saxon waited for she knew a fresh idea had struck billy",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0409",
+ "category": "phonetic_balance",
+ "text": "He saw only the effect in a general, sketchy way.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0409.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE SAW ONLY THE EFFECT IN A GENERAL SKETCHY WAY",
+ "raw": {
+ "normalized_reference": "he saw only the effect in a general, sketchy way.",
+ "normalized_hypothesis": "he saw only the effect in a general sketchy way",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he saw only the effect in a general sketchy way",
+ "normalized_hypothesis": "he saw only the effect in a general sketchy way",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0410",
+ "category": "phonetic_balance",
+ "text": "He began to follow the footprints of the dog.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0410.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE BEGAN TO FOLLOW THE FOOTPRINTS OF THE DAD",
+ "raw": {
+ "normalized_reference": "he began to follow the footprints of the dog.",
+ "normalized_hypothesis": "he began to follow the footprints of the dad",
+ "errors": 1,
+ "words": 9,
+ "wer": 0.1111111111111111
+ },
+ "semantic": {
+ "normalized_reference": "he began to follow the footprints of the dog",
+ "normalized_hypothesis": "he began to follow the footprints of the dad",
+ "errors": 1,
+ "words": 9,
+ "wer": 0.1111111111111111
+ }
+ },
+ {
+ "id": "release500_0411",
+ "category": "phonetic_balance",
+ "text": "Do you value your hide.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0411.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "DO YOU VALUE YOUR HIDE",
+ "raw": {
+ "normalized_reference": "do you value your hide.",
+ "normalized_hypothesis": "do you value your hide",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "do you value your hide",
+ "normalized_hypothesis": "do you value your hide",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0412",
+ "category": "phonetic_balance",
+ "text": "A rising tide of fat had submerged them.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0412.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "A RISING TIDE OF FAT HANT SUBMERGED THEM",
+ "raw": {
+ "normalized_reference": "a rising tide of fat had submerged them.",
+ "normalized_hypothesis": "a rising tide of fat hant submerged them",
+ "errors": 1,
+ "words": 8,
+ "wer": 0.125
+ },
+ "semantic": {
+ "normalized_reference": "a rising tide of fat had submerged them",
+ "normalized_hypothesis": "a rising tide of fat hant submerged them",
+ "errors": 1,
+ "words": 8,
+ "wer": 0.125
+ }
+ },
+ {
+ "id": "release500_0413",
+ "category": "phonetic_balance",
+ "text": "There was nothing on the rock.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0413.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THERE WAS NOTHING ON THE ROCK",
+ "raw": {
+ "normalized_reference": "there was nothing on the rock.",
+ "normalized_hypothesis": "there was nothing on the rock",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "there was nothing on the rock",
+ "normalized_hypothesis": "there was nothing on the rock",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0414",
+ "category": "phonetic_balance",
+ "text": "Whatever he guessed he locked away in the taboo room of Naomi.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0414.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WHATEVER HE GUESSED HE LOCKED AWAY IN THE TABOO ROOM OF NAOMI",
+ "raw": {
+ "normalized_reference": "whatever he guessed he locked away in the taboo room of naomi.",
+ "normalized_hypothesis": "whatever he guessed he locked away in the taboo room of naomi",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "whatever he guessed he locked away in the taboo room of naomi",
+ "normalized_hypothesis": "whatever he guessed he locked away in the taboo room of naomi",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0415",
+ "category": "phonetic_balance",
+ "text": "Will we ever forget it.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0415.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WILL WE EVER FORGET IT",
+ "raw": {
+ "normalized_reference": "will we ever forget it.",
+ "normalized_hypothesis": "will we ever forget it",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "will we ever forget it",
+ "normalized_hypothesis": "will we ever forget it",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0416",
+ "category": "phonetic_balance",
+ "text": "Jeanne and Pierre both gazed toward the great rock.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0416.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "JEAN AND PIERRE BOTH GAZED TOWARD THE GREAT ROCK",
+ "raw": {
+ "normalized_reference": "jeanne and pierre both gazed toward the great rock.",
+ "normalized_hypothesis": "jean and pierre both gazed toward the great rock",
+ "errors": 1,
+ "words": 9,
+ "wer": 0.1111111111111111
+ },
+ "semantic": {
+ "normalized_reference": "jeanne and pierre both gazed toward the great rock",
+ "normalized_hypothesis": "jean and pierre both gazed toward the great rock",
+ "errors": 1,
+ "words": 9,
+ "wer": 0.1111111111111111
+ }
+ },
+ {
+ "id": "release500_0417",
+ "category": "phonetic_balance",
+ "text": "Does the old boy often go off at half-cock that way.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0417.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "DOES THE OLD BOY OFTEN GO OFF AT HALF COCK THAT WAY",
+ "raw": {
+ "normalized_reference": "does the old boy often go off at half-cock that way.",
+ "normalized_hypothesis": "does the old boy often go off at half cock that way",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "does the old boy often go off at half cock that way",
+ "normalized_hypothesis": "does the old boy often go off at half cock that way",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0418",
+ "category": "phonetic_balance",
+ "text": "The temperature dropped to fifty below zero and remained there the whole trip.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0418.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE TEMPERATURE DROPPED TO FIFTY BELOW ZERO AND REMAINED THERE THE WHOLE TRIP",
+ "raw": {
+ "normalized_reference": "the temperature dropped to fifty below zero and remained there the whole trip.",
+ "normalized_hypothesis": "the temperature dropped to fifty below zero and remained there the whole trip",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the temperature dropped to 50 below 0 and remained there the whole trip",
+ "normalized_hypothesis": "the temperature dropped to 50 below 0 and remained there the whole trip",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0419",
+ "category": "phonetic_balance",
+ "text": "It's worth eight dollars.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0419.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT'S WORTH EIGHT DOLLARS",
+ "raw": {
+ "normalized_reference": "it's worth eight dollars.",
+ "normalized_hypothesis": "it's worth eight dollars",
+ "errors": 0,
+ "words": 4,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it is worth $8",
+ "normalized_hypothesis": "it is worth $8",
+ "errors": 0,
+ "words": 4,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0420",
+ "category": "phonetic_balance",
+ "text": "But I did not enjoy it long.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0420.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "BUT I DID NOT ENJOY IT LONG",
+ "raw": {
+ "normalized_reference": "but i did not enjoy it long.",
+ "normalized_hypothesis": "but i did not enjoy it long",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "but i did not enjoy it long",
+ "normalized_hypothesis": "but i did not enjoy it long",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0421",
+ "category": "phonetic_balance",
+ "text": "Shorty turned to their employers.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0421.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SHORTY TURNED TO THEIR EMPLOYERS",
+ "raw": {
+ "normalized_reference": "shorty turned to their employers.",
+ "normalized_hypothesis": "shorty turned to their employers",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "shorty turned to their employers",
+ "normalized_hypothesis": "shorty turned to their employers",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0422",
+ "category": "phonetic_balance",
+ "text": "It was the third or fourth time that Philip had heard MacDougall swear.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0422.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT WAS THE THIRD OR FOURTH TIME THAT PHILIP HAD HEARD MACDUBLE SWEAR",
+ "raw": {
+ "normalized_reference": "it was the third or fourth time that philip had heard macdougall swear.",
+ "normalized_hypothesis": "it was the third or fourth time that philip had heard macduble swear",
+ "errors": 1,
+ "words": 13,
+ "wer": 0.07692307692307693
+ },
+ "semantic": {
+ "normalized_reference": "it was the 3rd or 4th time that philip had heard macdougall swear",
+ "normalized_hypothesis": "it was the 3rd or 4th time that philip had heard macduble swear",
+ "errors": 1,
+ "words": 13,
+ "wer": 0.07692307692307693
+ }
+ },
+ {
+ "id": "release500_0423",
+ "category": "phonetic_balance",
+ "text": "Ernest saw in the affair the most sinister import.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0423.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "EARNEST SAW IN THE AFFAIR THE MOST SINISTER IMPORT",
+ "raw": {
+ "normalized_reference": "ernest saw in the affair the most sinister import.",
+ "normalized_hypothesis": "earnest saw in the affair the most sinister import",
+ "errors": 1,
+ "words": 9,
+ "wer": 0.1111111111111111
+ },
+ "semantic": {
+ "normalized_reference": "ernest saw in the affair the most sinister import",
+ "normalized_hypothesis": "earnest saw in the affair the most sinister import",
+ "errors": 1,
+ "words": 9,
+ "wer": 0.1111111111111111
+ }
+ },
+ {
+ "id": "release500_0424",
+ "category": "phonetic_balance",
+ "text": "Also, at regular intervals, he would mutter.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0424.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "ALSO AT REGULAR INTERVALS HE WOULD MUTTER",
+ "raw": {
+ "normalized_reference": "also, at regular intervals, he would mutter.",
+ "normalized_hypothesis": "also at regular intervals he would mutter",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "also at regular intervals he would mutter",
+ "normalized_hypothesis": "also at regular intervals he would mutter",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0425",
+ "category": "phonetic_balance",
+ "text": "He may anticipate the day of his death.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0425.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE MIGHT ANTICIPATE THE DAY OF HIS DEATH",
+ "raw": {
+ "normalized_reference": "he may anticipate the day of his death.",
+ "normalized_hypothesis": "he might anticipate the day of his death",
+ "errors": 1,
+ "words": 8,
+ "wer": 0.125
+ },
+ "semantic": {
+ "normalized_reference": "he may anticipate the day of his death",
+ "normalized_hypothesis": "he might anticipate the day of his death",
+ "errors": 1,
+ "words": 8,
+ "wer": 0.125
+ }
+ },
+ {
+ "id": "release500_0426",
+ "category": "phonetic_balance",
+ "text": "I defy any man to get a Solomon Island sore in California.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0426.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I DEFY ANY MAN TO GET A SOLOMON ISLAND SOR IN CALIFORNIA",
+ "raw": {
+ "normalized_reference": "i defy any man to get a solomon island sore in california.",
+ "normalized_hypothesis": "i defy any man to get a solomon island sor in california",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "i defy any man to get a solomon island sore in california",
+ "normalized_hypothesis": "i defy any man to get a solomon island sor in california",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ }
+ },
+ {
+ "id": "release500_0427",
+ "category": "phonetic_balance",
+ "text": "Now our figuring was all right.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0427.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "NOW OUR FIGURING WAS ALL RIGHT",
+ "raw": {
+ "normalized_reference": "now our figuring was all right.",
+ "normalized_hypothesis": "now our figuring was all right",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "now our figuring was all right",
+ "normalized_hypothesis": "now our figuring was all right",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0428",
+ "category": "phonetic_balance",
+ "text": "MacDougall tapped his forehead suspiciously with a stubby forefinger.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0428.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "MACDOUGLE TAPPED HIS FOREHEAD SUSPICIOUSLY WITH A STUBBY FOREFINGER",
+ "raw": {
+ "normalized_reference": "macdougall tapped his forehead suspiciously with a stubby forefinger.",
+ "normalized_hypothesis": "macdougle tapped his forehead suspiciously with a stubby forefinger",
+ "errors": 1,
+ "words": 9,
+ "wer": 0.1111111111111111
+ },
+ "semantic": {
+ "normalized_reference": "macdougall tapped his forehead suspiciously with a stubby forefinger",
+ "normalized_hypothesis": "macdougle tapped his forehead suspiciously with a stubby forefinger",
+ "errors": 1,
+ "words": 9,
+ "wer": 0.1111111111111111
+ }
+ },
+ {
+ "id": "release500_0429",
+ "category": "phonetic_balance",
+ "text": "O'Brien took off his coat and bared his right arm.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0429.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "OBRIEN TOOK OFF HIS COAT AND BARED HIS RIGHT ARM",
+ "raw": {
+ "normalized_reference": "o'brien took off his coat and bared his right arm.",
+ "normalized_hypothesis": "obrien took off his coat and bared his right arm",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ },
+ "semantic": {
+ "normalized_reference": "0 brien took off his coat and bared his right arm",
+ "normalized_hypothesis": "obrien took off his coat and bared his right arm",
+ "errors": 2,
+ "words": 11,
+ "wer": 0.18181818181818182
+ }
+ },
+ {
+ "id": "release500_0430",
+ "category": "phonetic_balance",
+ "text": "But it won't continue, she said with easy confidence.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0430.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "BUT IT WON'T CONTINUE SHE SAID WITH EASY CONFIDENCE",
+ "raw": {
+ "normalized_reference": "but it won't continue, she said with easy confidence.",
+ "normalized_hypothesis": "but it won't continue she said with easy confidence",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "but it will not continue she said with easy confidence",
+ "normalized_hypothesis": "but it will not continue she said with easy confidence",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0431",
+ "category": "phonetic_balance",
+ "text": "I was still weak from my prolonged immersion.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0431.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I WAS STILL WEAK FROM MY PROLONGED IMMERSION",
+ "raw": {
+ "normalized_reference": "i was still weak from my prolonged immersion.",
+ "normalized_hypothesis": "i was still weak from my prolonged immersion",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i was still weak from my prolonged immersion",
+ "normalized_hypothesis": "i was still weak from my prolonged immersion",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0432",
+ "category": "phonetic_balance",
+ "text": "Oolong was two hundred and fifty miles from the nearest land.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0432.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "BULLEN WAS TWO HUNDEDFAN FIFTY MILES FROM THE NEAREST LAND",
+ "raw": {
+ "normalized_reference": "oolong was two hundred and fifty miles from the nearest land.",
+ "normalized_hypothesis": "bullen was two hundedfan fifty miles from the nearest land",
+ "errors": 3,
+ "words": 11,
+ "wer": 0.2727272727272727
+ },
+ "semantic": {
+ "normalized_reference": "oolong was 250 miles from the nearest land",
+ "normalized_hypothesis": "bullen was 2 hundedfan 50 miles from the nearest land",
+ "errors": 4,
+ "words": 8,
+ "wer": 0.5
+ }
+ },
+ {
+ "id": "release500_0433",
+ "category": "phonetic_balance",
+ "text": "I just do appreciate it without being able to express my feelings.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0433.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I JUST DO APPRECIATE IT WITHOUT BEING ABLE TO EXPRESS MY FEELINGS",
+ "raw": {
+ "normalized_reference": "i just do appreciate it without being able to express my feelings.",
+ "normalized_hypothesis": "i just do appreciate it without being able to express my feelings",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i just do appreciate it without being able to express my feelings",
+ "normalized_hypothesis": "i just do appreciate it without being able to express my feelings",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0434",
+ "category": "phonetic_balance",
+ "text": "Now just what do you want to know.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0434.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "NOW JUST WHAT DO YOU WANT TO KNOW",
+ "raw": {
+ "normalized_reference": "now just what do you want to know.",
+ "normalized_hypothesis": "now just what do you want to know",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "now just what do you want to know",
+ "normalized_hypothesis": "now just what do you want to know",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0435",
+ "category": "phonetic_balance",
+ "text": "But she swung obediently on her heel into the wind.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0435.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "BUT SHE SWUNG OBEDIENTLY ON HER HEEL INTO THE WIND",
+ "raw": {
+ "normalized_reference": "but she swung obediently on her heel into the wind.",
+ "normalized_hypothesis": "but she swung obediently on her heel into the wind",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "but she swung obediently on her heel into the wind",
+ "normalized_hypothesis": "but she swung obediently on her heel into the wind",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0436",
+ "category": "phonetic_balance",
+ "text": "Ask him, I laughed, then turned to Pasquini.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0436.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "ASK HIM I LAUGHED THEN TURNED TO PASQUINI",
+ "raw": {
+ "normalized_reference": "ask him, i laughed, then turned to pasquini.",
+ "normalized_hypothesis": "ask him i laughed then turned to pasquini",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "ask him i laughed then turned to pasquini",
+ "normalized_hypothesis": "ask him i laughed then turned to pasquini",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0437",
+ "category": "phonetic_balance",
+ "text": "He did not rush in.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0437.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE DID NOT RUSH IN",
+ "raw": {
+ "normalized_reference": "he did not rush in.",
+ "normalized_hypothesis": "he did not rush in",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he did not rush in",
+ "normalized_hypothesis": "he did not rush in",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0438",
+ "category": "phonetic_balance",
+ "text": "They obeyed him, and went here and there at his commands.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0438.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THEY OBEYED HIM AND WENT HERE AND THERE AT HIS COMMANDS",
+ "raw": {
+ "normalized_reference": "they obeyed him, and went here and there at his commands.",
+ "normalized_hypothesis": "they obeyed him and went here and there at his commands",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "they obeyed him and went here and there at his commands",
+ "normalized_hypothesis": "they obeyed him and went here and there at his commands",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0439",
+ "category": "phonetic_balance",
+ "text": "Tom Spink has a harpoon.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0439.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "TOM SPINK HAS A HARPOON",
+ "raw": {
+ "normalized_reference": "tom spink has a harpoon.",
+ "normalized_hypothesis": "tom spink has a harpoon",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "tom spink has a harpoon",
+ "normalized_hypothesis": "tom spink has a harpoon",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0440",
+ "category": "phonetic_balance",
+ "text": "It was my reports from the north which chiefly induced people to buy.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0440.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT WAS MY REPORTS FROM THE NORTH WHICH CHIEFLY INDUCED PEOPLE TO BUY",
+ "raw": {
+ "normalized_reference": "it was my reports from the north which chiefly induced people to buy.",
+ "normalized_hypothesis": "it was my reports from the north which chiefly induced people to buy",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it was my reports from the north which chiefly induced people to buy",
+ "normalized_hypothesis": "it was my reports from the north which chiefly induced people to buy",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0441",
+ "category": "phonetic_balance",
+ "text": "Cherokee identified himself with his instinct.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0441.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "CHEROKEE IDENTIFIED HIMSELF WITH HIS INSTINCT",
+ "raw": {
+ "normalized_reference": "cherokee identified himself with his instinct.",
+ "normalized_hypothesis": "cherokee identified himself with his instinct",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "cherokee identified himself with his instinct",
+ "normalized_hypothesis": "cherokee identified himself with his instinct",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0442",
+ "category": "phonetic_balance",
+ "text": "I graduated last of my class.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0442.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I GRADUATED LAST OF MY CLASS",
+ "raw": {
+ "normalized_reference": "i graduated last of my class.",
+ "normalized_hypothesis": "i graduated last of my class",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i graduated last of my class",
+ "normalized_hypothesis": "i graduated last of my class",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0443",
+ "category": "phonetic_balance",
+ "text": "It was a gigantic inadequacy.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0443.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT WAS A GIGANTIC INADEQUACY",
+ "raw": {
+ "normalized_reference": "it was a gigantic inadequacy.",
+ "normalized_hypothesis": "it was a gigantic inadequacy",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it was a gigantic inadequacy",
+ "normalized_hypothesis": "it was a gigantic inadequacy",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0444",
+ "category": "phonetic_balance",
+ "text": "Did I possess too much vitality.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0444.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "DID I POSSESS TOO MUCH FATALITY",
+ "raw": {
+ "normalized_reference": "did i possess too much vitality.",
+ "normalized_hypothesis": "did i possess too much fatality",
+ "errors": 1,
+ "words": 6,
+ "wer": 0.16666666666666666
+ },
+ "semantic": {
+ "normalized_reference": "did i possess too much vitality",
+ "normalized_hypothesis": "did i possess too much fatality",
+ "errors": 1,
+ "words": 6,
+ "wer": 0.16666666666666666
+ }
+ },
+ {
+ "id": "release500_0445",
+ "category": "phonetic_balance",
+ "text": "Therefore, hurrah for the game.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0445.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THEREFORE HURRAH FOR THE GAME",
+ "raw": {
+ "normalized_reference": "therefore, hurrah for the game.",
+ "normalized_hypothesis": "therefore hurrah for the game",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "therefore hurrah for the game",
+ "normalized_hypothesis": "therefore hurrah for the game",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0446",
+ "category": "phonetic_balance",
+ "text": "He had fulfilled his duty and paid properly.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0446.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "E HAN FULFILLED HIS DUTY AND PAID PROPERLY",
+ "raw": {
+ "normalized_reference": "he had fulfilled his duty and paid properly.",
+ "normalized_hypothesis": "e han fulfilled his duty and paid properly",
+ "errors": 2,
+ "words": 8,
+ "wer": 0.25
+ },
+ "semantic": {
+ "normalized_reference": "he had fulfilled his duty and paid properly",
+ "normalized_hypothesis": "e han fulfilled his duty and paid properly",
+ "errors": 2,
+ "words": 8,
+ "wer": 0.25
+ }
+ },
+ {
+ "id": "release500_0447",
+ "category": "phonetic_balance",
+ "text": "How can you manage all alone, Mr Young.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0447.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HOW CAN YOU MANAGE ALL ALONE MISTER YON",
+ "raw": {
+ "normalized_reference": "how can you manage all alone, mr young.",
+ "normalized_hypothesis": "how can you manage all alone mister yon",
+ "errors": 2,
+ "words": 8,
+ "wer": 0.25
+ },
+ "semantic": {
+ "normalized_reference": "how can you manage all alone mister young",
+ "normalized_hypothesis": "how can you manage all alone mister yon",
+ "errors": 1,
+ "words": 8,
+ "wer": 0.125
+ }
+ },
+ {
+ "id": "release500_0448",
+ "category": "phonetic_balance",
+ "text": "Red-Eye never committed a more outrageous deed.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0448.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "RED EYE NEVER COMMITTED A MORE OUTRAGEOUS DEED",
+ "raw": {
+ "normalized_reference": "red-eye never committed a more outrageous deed.",
+ "normalized_hypothesis": "red eye never committed a more outrageous deed",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "red eye never committed a more outrageous deed",
+ "normalized_hypothesis": "red eye never committed a more outrageous deed",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0449",
+ "category": "phonetic_balance",
+ "text": "He would keep his faith with Gregson for the promised day or two.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0449.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE WOULD KEEP HIS FAITH WITH GREGSON FOR THE PROMISED DAY OR TWO",
+ "raw": {
+ "normalized_reference": "he would keep his faith with gregson for the promised day or two.",
+ "normalized_hypothesis": "he would keep his faith with gregson for the promised day or two",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he would keep his faith with gregson for the promised day or 2",
+ "normalized_hypothesis": "he would keep his faith with gregson for the promised day or 2",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0450",
+ "category": "phonetic_balance",
+ "text": "There was one chance, and only one, of saving Jeanne.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0450.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THERE WAS ONE CHANCE AND ONLY ONE OF SAVING JEAN",
+ "raw": {
+ "normalized_reference": "there was one chance, and only one, of saving jeanne.",
+ "normalized_hypothesis": "there was one chance and only one of saving jean",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ },
+ "semantic": {
+ "normalized_reference": "there was one chance and only one of saving jeanne",
+ "normalized_hypothesis": "there was one chance and only one of saving jean",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ }
+ },
+ {
+ "id": "release500_0451",
+ "category": "phonetic_balance",
+ "text": "I was beneath the water, suffocating and drowning.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0451.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I WAS BENEATH THE WATER SUFFOCATING AND DROWNING",
+ "raw": {
+ "normalized_reference": "i was beneath the water, suffocating and drowning.",
+ "normalized_hypothesis": "i was beneath the water suffocating and drowning",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i was beneath the water suffocating and drowning",
+ "normalized_hypothesis": "i was beneath the water suffocating and drowning",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0452",
+ "category": "phonetic_balance",
+ "text": "Give them their choice between a fine or an official whipping.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0452.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "GIVE THEM THEIR CHOICE BETWEEN A FINE OR AND OFFICIAL WHIPPING",
+ "raw": {
+ "normalized_reference": "give them their choice between a fine or an official whipping.",
+ "normalized_hypothesis": "give them their choice between a fine or and official whipping",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ },
+ "semantic": {
+ "normalized_reference": "give them their choice between a fine or an official whipping",
+ "normalized_hypothesis": "give them their choice between a fine or and official whipping",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ }
+ },
+ {
+ "id": "release500_0453",
+ "category": "phonetic_balance",
+ "text": "They must have been swept away by the chaotic currents.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0453.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THEY MUST HAVE BEEN SWEPT AWAY BY THE CHAOTIC CURRENTS",
+ "raw": {
+ "normalized_reference": "they must have been swept away by the chaotic currents.",
+ "normalized_hypothesis": "they must have been swept away by the chaotic currents",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "they must have been swept away by the chaotic currents",
+ "normalized_hypothesis": "they must have been swept away by the chaotic currents",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0454",
+ "category": "phonetic_balance",
+ "text": "You yellow giant thing of the frost.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0454.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "YOU YELLOW GIANT THING OF THE FROST",
+ "raw": {
+ "normalized_reference": "you yellow giant thing of the frost.",
+ "normalized_hypothesis": "you yellow giant thing of the frost",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "you yellow giant thing of the frost",
+ "normalized_hypothesis": "you yellow giant thing of the frost",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0455",
+ "category": "phonetic_balance",
+ "text": "He had peeled off his shirt and was wildly waving it.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0455.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE HAD PEELED OFF HIS SHIRT AND WAS WILDLY WAVING IT",
+ "raw": {
+ "normalized_reference": "he had peeled off his shirt and was wildly waving it.",
+ "normalized_hypothesis": "he had peeled off his shirt and was wildly waving it",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he had peeled off his shirt and was wildly waving it",
+ "normalized_hypothesis": "he had peeled off his shirt and was wildly waving it",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0456",
+ "category": "phonetic_balance",
+ "text": "To these he gave castor oil.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0456.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "TO THESE HE GAVE CASTOR OIL",
+ "raw": {
+ "normalized_reference": "to these he gave castor oil.",
+ "normalized_hypothesis": "to these he gave castor oil",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "to these he gave castor oil",
+ "normalized_hypothesis": "to these he gave castor oil",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0457",
+ "category": "phonetic_balance",
+ "text": "Tudor surveyed him with withering disgust.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0457.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "TUDOR SURVEYED HIM WITH WITHERING DISGUST",
+ "raw": {
+ "normalized_reference": "tudor surveyed him with withering disgust.",
+ "normalized_hypothesis": "tudor surveyed him with withering disgust",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "tudor surveyed him with withering disgust",
+ "normalized_hypothesis": "tudor surveyed him with withering disgust",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0458",
+ "category": "phonetic_balance",
+ "text": "After all, it was simply a mistake in judgment.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0458.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "AFTER ALL IT WAS SIMPLY A MISTAKE IN JUDGMENT",
+ "raw": {
+ "normalized_reference": "after all, it was simply a mistake in judgment.",
+ "normalized_hypothesis": "after all it was simply a mistake in judgment",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "after all it was simply a mistake in judgment",
+ "normalized_hypothesis": "after all it was simply a mistake in judgment",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0459",
+ "category": "phonetic_balance",
+ "text": "That is the strange part of it.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0459.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THAT IS THE STRANGE PART OF IT",
+ "raw": {
+ "normalized_reference": "that is the strange part of it.",
+ "normalized_hypothesis": "that is the strange part of it",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "that is the strange part of it",
+ "normalized_hypothesis": "that is the strange part of it",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0460",
+ "category": "phonetic_balance",
+ "text": "Not a wheel moved in his empire.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0460.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "NOT A WHEEL MOVED IN HIS EMPIRE",
+ "raw": {
+ "normalized_reference": "not a wheel moved in his empire.",
+ "normalized_hypothesis": "not a wheel moved in his empire",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "not a wheel moved in his empire",
+ "normalized_hypothesis": "not a wheel moved in his empire",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0461",
+ "category": "phonetic_balance",
+ "text": "Its diameter was not more than two hundred yards.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0461.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "ITS DIAMETER WAS NOT MORE THAN TWO HUNDRED YARDS",
+ "raw": {
+ "normalized_reference": "its diameter was not more than two hundred yards.",
+ "normalized_hypothesis": "its diameter was not more than two hundred yards",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "its diameter was not more than 200 yards",
+ "normalized_hypothesis": "its diameter was not more than 200 yards",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0462",
+ "category": "phonetic_balance",
+ "text": "Nimrod replied, with a slight manifestation of sensitiveness.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0462.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "NIMROD REPLIED WITH A SLIGHT MANIFESTATION OF SENSITIVENESS",
+ "raw": {
+ "normalized_reference": "nimrod replied, with a slight manifestation of sensitiveness.",
+ "normalized_hypothesis": "nimrod replied with a slight manifestation of sensitiveness",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "nimrod replied with a slight manifestation of sensitiveness",
+ "normalized_hypothesis": "nimrod replied with a slight manifestation of sensitiveness",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0463",
+ "category": "phonetic_balance",
+ "text": "I tried to read George Moore last night, and was dreadfully bored.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0463.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I TRIED TO READ GEORGE MOORE LAST NIGHT AND WAS DREADFULLY BORED",
+ "raw": {
+ "normalized_reference": "i tried to read george moore last night, and was dreadfully bored.",
+ "normalized_hypothesis": "i tried to read george moore last night and was dreadfully bored",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i tried to read george moore last night and was dreadfully bored",
+ "normalized_hypothesis": "i tried to read george moore last night and was dreadfully bored",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0464",
+ "category": "phonetic_balance",
+ "text": "Yes, it was a man who asked, a stranger.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0464.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "YES IT WAS A MAN WHO ASKED A STRANGER",
+ "raw": {
+ "normalized_reference": "yes, it was a man who asked, a stranger.",
+ "normalized_hypothesis": "yes it was a man who asked a stranger",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "yes it was a man who asked a stranger",
+ "normalized_hypothesis": "yes it was a man who asked a stranger",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0465",
+ "category": "phonetic_balance",
+ "text": "We were now good friends.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0465.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WE WERE NOW GOOD FRIENDS",
+ "raw": {
+ "normalized_reference": "we were now good friends.",
+ "normalized_hypothesis": "we were now good friends",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "we were now good friends",
+ "normalized_hypothesis": "we were now good friends",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0466",
+ "category": "phonetic_balance",
+ "text": "Yet, in accordance with Ernest's test of truth, it worked.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0466.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "YET IN ACCORDANCE WITH EARNEST TEST OF TRUTH IT WORKED",
+ "raw": {
+ "normalized_reference": "yet, in accordance with ernest's test of truth, it worked.",
+ "normalized_hypothesis": "yet in accordance with earnest test of truth it worked",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ },
+ "semantic": {
+ "normalized_reference": "yet in accordance with ernest is test of truth it worked",
+ "normalized_hypothesis": "yet in accordance with earnest test of truth it worked",
+ "errors": 2,
+ "words": 11,
+ "wer": 0.18181818181818182
+ }
+ },
+ {
+ "id": "release500_0467",
+ "category": "phonetic_balance",
+ "text": "It's the strap hangers that'll keep us from going under.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0467.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT'S THE STRAP HANGERS THAT'LL KEEP US FROM GOING UNDER",
+ "raw": {
+ "normalized_reference": "it's the strap hangers that'll keep us from going under.",
+ "normalized_hypothesis": "it's the strap hangers that'll keep us from going under",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it is the strap hangers that will keep us from going under",
+ "normalized_hypothesis": "it is the strap hangers that will keep us from going under",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0468",
+ "category": "phonetic_balance",
+ "text": "I never allow what can't be changed to annoy me.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0468.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I NEVER ALLOW WHAT CAN'T BE CHANGED TO ANNOY ME",
+ "raw": {
+ "normalized_reference": "i never allow what can't be changed to annoy me.",
+ "normalized_hypothesis": "i never allow what can't be changed to annoy me",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i never allow what can not be changed to annoy me",
+ "normalized_hypothesis": "i never allow what can not be changed to annoy me",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0469",
+ "category": "phonetic_balance",
+ "text": "After all superfluous flesh is gone what is left is stringy and resistant.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0469.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "AFTER ALL SUPERFLUOUS FLESH IS COME WHAT IS LEFT IS STRINGY AND RESISTENT",
+ "raw": {
+ "normalized_reference": "after all superfluous flesh is gone what is left is stringy and resistant.",
+ "normalized_hypothesis": "after all superfluous flesh is come what is left is stringy and resistent",
+ "errors": 2,
+ "words": 13,
+ "wer": 0.15384615384615385
+ },
+ "semantic": {
+ "normalized_reference": "after all superfluous flesh is gone what is left is stringy and resistant",
+ "normalized_hypothesis": "after all superfluous flesh is come what is left is stringy and resistent",
+ "errors": 2,
+ "words": 13,
+ "wer": 0.15384615384615385
+ }
+ },
+ {
+ "id": "release500_0470",
+ "category": "phonetic_balance",
+ "text": "Then he shouted, Shut up.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0470.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THEN HE SHOUTED SHUT UP",
+ "raw": {
+ "normalized_reference": "then he shouted, shut up.",
+ "normalized_hypothesis": "then he shouted shut up",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "then he shouted shut up",
+ "normalized_hypothesis": "then he shouted shut up",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0471",
+ "category": "phonetic_balance",
+ "text": "He had worshiped her, as Dante might have worshiped Beatrice.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0471.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE HAD WORSHIPPED HER AS DANTE MIGHT HAVE WORSHIPPED BEATRICE",
+ "raw": {
+ "normalized_reference": "he had worshiped her, as dante might have worshiped beatrice.",
+ "normalized_hypothesis": "he had worshipped her as dante might have worshipped beatrice",
+ "errors": 2,
+ "words": 10,
+ "wer": 0.2
+ },
+ "semantic": {
+ "normalized_reference": "he had worshiped her as dante might have worshiped beatrice",
+ "normalized_hypothesis": "he had worshipped her as dante might have worshipped beatrice",
+ "errors": 2,
+ "words": 10,
+ "wer": 0.2
+ }
+ },
+ {
+ "id": "release500_0472",
+ "category": "phonetic_balance",
+ "text": "He looked like one who had passed through an uncomfortable hour or two.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0472.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE LOOKED LIKE ONE WHO HAD PASSED THROUGH AN UNCOMFORTABLE HOUR OR TWO",
+ "raw": {
+ "normalized_reference": "he looked like one who had passed through an uncomfortable hour or two.",
+ "normalized_hypothesis": "he looked like one who had passed through an uncomfortable hour or two",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he looked like one who had passed through an uncomfortable hour or 2",
+ "normalized_hypothesis": "he looked like one who had passed through an uncomfortable hour or 2",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0473",
+ "category": "phonetic_balance",
+ "text": "The history of our westward-faring race is written in it.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0473.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE HISTORY OF OUR WESTWARD FARING RACE IS WRITTEN IN IT",
+ "raw": {
+ "normalized_reference": "the history of our westward-faring race is written in it.",
+ "normalized_hypothesis": "the history of our westward faring race is written in it",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the history of our westward faring race is written in it",
+ "normalized_hypothesis": "the history of our westward faring race is written in it",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0474",
+ "category": "phonetic_balance",
+ "text": "He may desire to escape pain, or to enjoy pleasure.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0474.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE MAY DESIRE TO ESCAPE PAIN OR TO ENJOY PLEASURE",
+ "raw": {
+ "normalized_reference": "he may desire to escape pain, or to enjoy pleasure.",
+ "normalized_hypothesis": "he may desire to escape pain or to enjoy pleasure",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he may desire to escape pain or to enjoy pleasure",
+ "normalized_hypothesis": "he may desire to escape pain or to enjoy pleasure",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0475",
+ "category": "phonetic_balance",
+ "text": "And the air was growing chilly.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0475.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "AND THE AIR WAS GROWING CHILLY",
+ "raw": {
+ "normalized_reference": "and the air was growing chilly.",
+ "normalized_hypothesis": "and the air was growing chilly",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "and the air was growing chilly",
+ "normalized_hypothesis": "and the air was growing chilly",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0476",
+ "category": "phonetic_balance",
+ "text": "Dig in; you're sure good, was Daylight's answer.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0476.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "DIG IN YOU'RE SURE GOOD WAS DAYLIGHT'S ANSWER",
+ "raw": {
+ "normalized_reference": "dig in; you're sure good, was daylight's answer.",
+ "normalized_hypothesis": "dig in you're sure good was daylight's answer",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "dig in you are sure good was daylight is answer",
+ "normalized_hypothesis": "dig in you are sure good was daylight is answer",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0477",
+ "category": "phonetic_balance",
+ "text": "My idea was that he would have more influence over the natives.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0477.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "MY IDEA WAS THAT HE WOULD HAVE MORE INFLUENCE OVER THE NATIVES",
+ "raw": {
+ "normalized_reference": "my idea was that he would have more influence over the natives.",
+ "normalized_hypothesis": "my idea was that he would have more influence over the natives",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "my idea was that he would have more influence over the natives",
+ "normalized_hypothesis": "my idea was that he would have more influence over the natives",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0478",
+ "category": "phonetic_balance",
+ "text": "Now run along, and tell them to hurry.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0478.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "NOW RUN ALONG AND TELL THEM TO HURRY",
+ "raw": {
+ "normalized_reference": "now run along, and tell them to hurry.",
+ "normalized_hypothesis": "now run along and tell them to hurry",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "now run along and tell them to hurry",
+ "normalized_hypothesis": "now run along and tell them to hurry",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0479",
+ "category": "phonetic_balance",
+ "text": "I have no idea, replied Philip.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0479.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I HAD NO IDEA REPLIED PHILETT",
+ "raw": {
+ "normalized_reference": "i have no idea, replied philip.",
+ "normalized_hypothesis": "i had no idea replied philett",
+ "errors": 2,
+ "words": 6,
+ "wer": 0.3333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "i have no idea replied philip",
+ "normalized_hypothesis": "i had no idea replied philett",
+ "errors": 2,
+ "words": 6,
+ "wer": 0.3333333333333333
+ }
+ },
+ {
+ "id": "release500_0480",
+ "category": "phonetic_balance",
+ "text": "For the rest, he was a mere automaton.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0480.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "FOR THE REST HE WAS A MERE AUTOMATON",
+ "raw": {
+ "normalized_reference": "for the rest, he was a mere automaton.",
+ "normalized_hypothesis": "for the rest he was a mere automaton",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "for the rest he was a mere automaton",
+ "normalized_hypothesis": "for the rest he was a mere automaton",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0481",
+ "category": "phonetic_balance",
+ "text": "Why, the average review is more nauseating than cod liver oil.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0481.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WHY THE AVERAGE REVIEW IS MORE NAUSEATING THAN COD LIVER OIL",
+ "raw": {
+ "normalized_reference": "why, the average review is more nauseating than cod liver oil.",
+ "normalized_hypothesis": "why the average review is more nauseating than cod liver oil",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "why the average review is more nauseating than cod liver oil",
+ "normalized_hypothesis": "why the average review is more nauseating than cod liver oil",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0482",
+ "category": "phonetic_balance",
+ "text": "I have been robbed, sir, I amended.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0482.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I HAVE BEEN ROBBED SIR I AMMENDED",
+ "raw": {
+ "normalized_reference": "i have been robbed, sir, i amended.",
+ "normalized_hypothesis": "i have been robbed sir i ammended",
+ "errors": 1,
+ "words": 7,
+ "wer": 0.14285714285714285
+ },
+ "semantic": {
+ "normalized_reference": "i have been robbed sir i amended",
+ "normalized_hypothesis": "i have been robbed sir i ammended",
+ "errors": 1,
+ "words": 7,
+ "wer": 0.14285714285714285
+ }
+ },
+ {
+ "id": "release500_0483",
+ "category": "phonetic_balance",
+ "text": "He moved away as quietly as he had come.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0483.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE MOVED AWAY AS QUIETLY AS HE HAD COME",
+ "raw": {
+ "normalized_reference": "he moved away as quietly as he had come.",
+ "normalized_hypothesis": "he moved away as quietly as he had come",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he moved away as quietly as he had come",
+ "normalized_hypothesis": "he moved away as quietly as he had come",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0484",
+ "category": "phonetic_balance",
+ "text": "A trickle of fresh blood ran over his face.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0484.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "A TRICKLE OF FRESH BLOOD RAN OVER HIS FACE",
+ "raw": {
+ "normalized_reference": "a trickle of fresh blood ran over his face.",
+ "normalized_hypothesis": "a trickle of fresh blood ran over his face",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "a trickle of fresh blood ran over his face",
+ "normalized_hypothesis": "a trickle of fresh blood ran over his face",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0485",
+ "category": "phonetic_balance",
+ "text": "Suppose you saw me at work through the window.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0485.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SUPPOSE YOU SAW ME AT WORK THROUGH THE WINDOW",
+ "raw": {
+ "normalized_reference": "suppose you saw me at work through the window.",
+ "normalized_hypothesis": "suppose you saw me at work through the window",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "suppose you saw me at work through the window",
+ "normalized_hypothesis": "suppose you saw me at work through the window",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0486",
+ "category": "phonetic_balance",
+ "text": "But who was Eileen's double.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0486.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "BUT WHO WAS ILEEAN'S DOUBLE",
+ "raw": {
+ "normalized_reference": "but who was eileen's double.",
+ "normalized_hypothesis": "but who was ileean's double",
+ "errors": 1,
+ "words": 5,
+ "wer": 0.2
+ },
+ "semantic": {
+ "normalized_reference": "but who was eileen is double",
+ "normalized_hypothesis": "but who was ileean is double",
+ "errors": 1,
+ "words": 6,
+ "wer": 0.16666666666666666
+ }
+ },
+ {
+ "id": "release500_0487",
+ "category": "phonetic_balance",
+ "text": "Darkness hid him from Jeanne.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0487.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "DARKNESS HID HIM FROM JEAN",
+ "raw": {
+ "normalized_reference": "darkness hid him from jeanne.",
+ "normalized_hypothesis": "darkness hid him from jean",
+ "errors": 1,
+ "words": 5,
+ "wer": 0.2
+ },
+ "semantic": {
+ "normalized_reference": "darkness hid him from jeanne",
+ "normalized_hypothesis": "darkness hid him from jean",
+ "errors": 1,
+ "words": 5,
+ "wer": 0.2
+ }
+ },
+ {
+ "id": "release500_0488",
+ "category": "phonetic_balance",
+ "text": "By this time Charley was as enraged as the Greek.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0488.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "BY THIS TIME CHARLEY WAS AS ENRAGED AS THE GREET",
+ "raw": {
+ "normalized_reference": "by this time charley was as enraged as the greek.",
+ "normalized_hypothesis": "by this time charley was as enraged as the greet",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ },
+ "semantic": {
+ "normalized_reference": "by this time charley was as enraged as the greek",
+ "normalized_hypothesis": "by this time charley was as enraged as the greet",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ }
+ },
+ {
+ "id": "release500_0489",
+ "category": "phonetic_balance",
+ "text": "I am writing these lines in Honolulu, Hawaii.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0489.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I'M WRITING THESE LINES IN HONOLULU AWAII",
+ "raw": {
+ "normalized_reference": "i am writing these lines in honolulu, hawaii.",
+ "normalized_hypothesis": "i'm writing these lines in honolulu awaii",
+ "errors": 3,
+ "words": 8,
+ "wer": 0.375
+ },
+ "semantic": {
+ "normalized_reference": "i am writing these lines in honolulu hawaii",
+ "normalized_hypothesis": "i am writing these lines in honolulu awaii",
+ "errors": 1,
+ "words": 8,
+ "wer": 0.125
+ }
+ },
+ {
+ "id": "release500_0490",
+ "category": "phonetic_balance",
+ "text": "Nobody knows how the natives got them.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0490.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "NOBODY KNOWS HOW THE NATIVES GOT THEM",
+ "raw": {
+ "normalized_reference": "nobody knows how the natives got them.",
+ "normalized_hypothesis": "nobody knows how the natives got them",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "nobody knows how the natives got them",
+ "normalized_hypothesis": "nobody knows how the natives got them",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0491",
+ "category": "phonetic_balance",
+ "text": "And how would we ever find ourselves.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0491.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "AND HOW WOULD WE EVER FIND OURSELVES",
+ "raw": {
+ "normalized_reference": "and how would we ever find ourselves.",
+ "normalized_hypothesis": "and how would we ever find ourselves",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "and how would we ever find ourselves",
+ "normalized_hypothesis": "and how would we ever find ourselves",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0492",
+ "category": "phonetic_balance",
+ "text": "They are coming ashore, whoever they are.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0492.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THEY ARE COMING ASHORE WHOEVER THEY ARE",
+ "raw": {
+ "normalized_reference": "they are coming ashore, whoever they are.",
+ "normalized_hypothesis": "they are coming ashore whoever they are",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "they are coming ashore whoever they are",
+ "normalized_hypothesis": "they are coming ashore whoever they are",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0493",
+ "category": "phonetic_balance",
+ "text": "This is my fifth voyage.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0493.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THIS IS MY FIFTH VOYAGE",
+ "raw": {
+ "normalized_reference": "this is my fifth voyage.",
+ "normalized_hypothesis": "this is my fifth voyage",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "this is my 5th voyage",
+ "normalized_hypothesis": "this is my 5th voyage",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0494",
+ "category": "phonetic_balance",
+ "text": "And yet if she came he had no words to say.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0494.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "AND YET IF SHE CAME HE HAD NO WORDS TO SAY",
+ "raw": {
+ "normalized_reference": "and yet if she came he had no words to say.",
+ "normalized_hypothesis": "and yet if she came he had no words to say",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "and yet if she came he had no words to say",
+ "normalized_hypothesis": "and yet if she came he had no words to say",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0495",
+ "category": "phonetic_balance",
+ "text": "You must sleep, he urged.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0495.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "YOU MUST SLEEP HE URGED",
+ "raw": {
+ "normalized_reference": "you must sleep, he urged.",
+ "normalized_hypothesis": "you must sleep he urged",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "you must sleep he urged",
+ "normalized_hypothesis": "you must sleep he urged",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0496",
+ "category": "phonetic_balance",
+ "text": "Could the incident have anything to do with Jeanne and Pierre.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0496.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "COULD THE INCIDENT HAVE ANYTHING TO DO WITH JEAN AND PIERRE",
+ "raw": {
+ "normalized_reference": "could the incident have anything to do with jeanne and pierre.",
+ "normalized_hypothesis": "could the incident have anything to do with jean and pierre",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ },
+ "semantic": {
+ "normalized_reference": "could the incident have anything to do with jeanne and pierre",
+ "normalized_hypothesis": "could the incident have anything to do with jean and pierre",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ }
+ },
+ {
+ "id": "release500_0497",
+ "category": "phonetic_balance",
+ "text": "Each day she became a more vital part of him.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0497.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "EACH DAY SHE BECAME A MORE VITAL PART OF HIM",
+ "raw": {
+ "normalized_reference": "each day she became a more vital part of him.",
+ "normalized_hypothesis": "each day she became a more vital part of him",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "each day she became a more vital part of him",
+ "normalized_hypothesis": "each day she became a more vital part of him",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0498",
+ "category": "phonetic_balance",
+ "text": "But why continue the tirade, for tirade it was.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0498.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "BUT WHY CONTINUE THE TARRADE FOR TARADE IT WAS",
+ "raw": {
+ "normalized_reference": "but why continue the tirade, for tirade it was.",
+ "normalized_hypothesis": "but why continue the tarrade for tarade it was",
+ "errors": 2,
+ "words": 9,
+ "wer": 0.2222222222222222
+ },
+ "semantic": {
+ "normalized_reference": "but why continue the tirade for tirade it was",
+ "normalized_hypothesis": "but why continue the tarrade for tarade it was",
+ "errors": 2,
+ "words": 9,
+ "wer": 0.2222222222222222
+ }
+ },
+ {
+ "id": "release500_0499",
+ "category": "phonetic_balance",
+ "text": "That is what distinguishes all of us from the lower animals.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0499.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THAT IS WHAT DISTINGUISHES ALL OF US FROM THE LOWER ANIMALS",
+ "raw": {
+ "normalized_reference": "that is what distinguishes all of us from the lower animals.",
+ "normalized_hypothesis": "that is what distinguishes all of us from the lower animals",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "that is what distinguishes all of us from the lower animals",
+ "normalized_hypothesis": "that is what distinguishes all of us from the lower animals",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0500",
+ "category": "phonetic_balance",
+ "text": "There was something pathetic in the girl's attitude now.",
+ "model": "inflect-micro-v2",
+ "audio": "benchmark/audio/inflect-micro-v2/release500_0500.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THERE WAS SOMETHING PATHETIC IN THE GIRL'S ATTITUDE NOW",
+ "raw": {
+ "normalized_reference": "there was something pathetic in the girl's attitude now.",
+ "normalized_hypothesis": "there was something pathetic in the girl's attitude now",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "there was something pathetic in the girl is attitude now",
+ "normalized_hypothesis": "there was something pathetic in the girl is attitude now",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0001",
+ "category": "conversational",
+ "text": "It wasn't until later that I realized what had actually happened.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0001.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT WASN'T UNTIL LATER THAT I REALIZED WHAT HAD ACTUALLY HAPPENED",
+ "raw": {
+ "normalized_reference": "it wasn't until later that i realized what had actually happened.",
+ "normalized_hypothesis": "it wasn't until later that i realized what had actually happened",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it was not until later that i realized what had actually happened",
+ "normalized_hypothesis": "it was not until later that i realized what had actually happened",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0002",
+ "category": "descriptive",
+ "text": "The light at that time of afternoon is perfect for about twenty minutes.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0002.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE LIGHT AT THAT TIME OF AFTERNOON IS PERFECT FOR ABOUT TWENTY MINUTES",
+ "raw": {
+ "normalized_reference": "the light at that time of afternoon is perfect for about twenty minutes.",
+ "normalized_hypothesis": "the light at that time of afternoon is perfect for about twenty minutes",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the light at that time of afternoon is perfect for about 20 minutes",
+ "normalized_hypothesis": "the light at that time of afternoon is perfect for about 20 minutes",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0003",
+ "category": "emotional",
+ "text": "That was the scariest thing I've ever done, and also maybe the best.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0003.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THAT WAS THE SCARRIEST THING I'VE EVER DONE AND ALSO MAY BE THE BEST",
+ "raw": {
+ "normalized_reference": "that was the scariest thing i've ever done, and also maybe the best.",
+ "normalized_hypothesis": "that was the scarriest thing i've ever done and also may be the best",
+ "errors": 3,
+ "words": 13,
+ "wer": 0.23076923076923078
+ },
+ "semantic": {
+ "normalized_reference": "that was the scariest thing i have ever done and also maybe the best",
+ "normalized_hypothesis": "that was the scarriest thing i have ever done and also may be the best",
+ "errors": 3,
+ "words": 14,
+ "wer": 0.21428571428571427
+ }
+ },
+ {
+ "id": "release500_0004",
+ "category": "homographs",
+ "text": "After I record the record, please present the present to the bass player.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0004.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "AFTER I RECORD THE RECORD PLEASE PRESENT THE PRESENT TO THE BASE PLAYER",
+ "raw": {
+ "normalized_reference": "after i record the record, please present the present to the bass player.",
+ "normalized_hypothesis": "after i record the record please present the present to the base player",
+ "errors": 1,
+ "words": 13,
+ "wer": 0.07692307692307693
+ },
+ "semantic": {
+ "normalized_reference": "after i record the record please present the present to the bass player",
+ "normalized_hypothesis": "after i record the record please present the present to the base player",
+ "errors": 1,
+ "words": 13,
+ "wer": 0.07692307692307693
+ }
+ },
+ {
+ "id": "release500_0005",
+ "category": "long",
+ "text": "If I'm being completely honest, I had no idea what I was doing for most of it, and I think that might be the most normal thing about the whole experience.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0005.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IF I'M BEING COMPLETELY HONEST I HAD NO IDEA WHAT I WAS DOING FOR MOST OF IT AND I THINK THAT MIGHT BE THE MOST NORMAL THING ABOUT THE WHOLE EXPERIENCE",
+ "raw": {
+ "normalized_reference": "if i'm being completely honest, i had no idea what i was doing for most of it, and i think that might be the most normal thing about the whole experience.",
+ "normalized_hypothesis": "if i'm being completely honest i had no idea what i was doing for most of it and i think that might be the most normal thing about the whole experience",
+ "errors": 0,
+ "words": 31,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "if i am being completely honest i had no idea what i was doing for most of it and i think that might be the most normal thing about the whole experience",
+ "normalized_hypothesis": "if i am being completely honest i had no idea what i was doing for most of it and i think that might be the most normal thing about the whole experience",
+ "errors": 0,
+ "words": 32,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0006",
+ "category": "long_range",
+ "text": "The committee postponed its decision after discovering that three pages of the report had been printed in the wrong order.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0006.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE COMMITTEE POSTPONED ITS DECISION AFTER DISCOVERING THAT THREE PAGES OF THE REPORT HAD BEEN PRINTED IN THE WRONG ORDER",
+ "raw": {
+ "normalized_reference": "the committee postponed its decision after discovering that three pages of the report had been printed in the wrong order.",
+ "normalized_hypothesis": "the committee postponed its decision after discovering that three pages of the report had been printed in the wrong order",
+ "errors": 0,
+ "words": 20,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the committee postponed its decision after discovering that 3 pages of the report had been printed in the wrong order",
+ "normalized_hypothesis": "the committee postponed its decision after discovering that 3 pages of the report had been printed in the wrong order",
+ "errors": 0,
+ "words": 20,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0007",
+ "category": "names_places",
+ "text": "Gwendolyn photographed the eucalyptus trees outside Ljubljana.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0007.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "GWENDOLEN PHOTOGRAPHED THE EUCALYPTIS TREES OUTSIDE LIUBLIANA",
+ "raw": {
+ "normalized_reference": "gwendolyn photographed the eucalyptus trees outside ljubljana.",
+ "normalized_hypothesis": "gwendolen photographed the eucalyptis trees outside liubliana",
+ "errors": 3,
+ "words": 7,
+ "wer": 0.42857142857142855
+ },
+ "semantic": {
+ "normalized_reference": "gwendolyn photographed the eucalyptus trees outside ljubljana",
+ "normalized_hypothesis": "gwendolen photographed the eucalyptis trees outside liubliana",
+ "errors": 3,
+ "words": 7,
+ "wer": 0.42857142857142855
+ }
+ },
+ {
+ "id": "release500_0008",
+ "category": "narrative",
+ "text": "He had this theory, and the more time passes, the more I think he was right.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0008.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE HAD THIS THEORY AND THE MORE TIME PASSES THE MORE I THINK HE WAS RIGHT",
+ "raw": {
+ "normalized_reference": "he had this theory, and the more time passes, the more i think he was right.",
+ "normalized_hypothesis": "he had this theory and the more time passes the more i think he was right",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he had this theory and the more time passes the more i think he was right",
+ "normalized_hypothesis": "he had this theory and the more time passes the more i think he was right",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0009",
+ "category": "numbers",
+ "text": "The package weighs twelve point six kilograms and arrived on July twenty-first.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0009.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE PACKAGE WEIGHS TWELVE POINT SIX KILOGRAMS AND ARRIVED ON JULY TWENTY FIRST",
+ "raw": {
+ "normalized_reference": "the package weighs twelve point six kilograms and arrived on july twenty-first.",
+ "normalized_hypothesis": "the package weighs twelve point six kilograms and arrived on july twenty first",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the package weighs 12.6 kilograms and arrived on july 21st",
+ "normalized_hypothesis": "the package weighs 12.6 kilograms and arrived on july 21st",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0010",
+ "category": "punctuation",
+ "text": "First, close the window; second, turn off the lamp; finally, lock the door.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0010.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "FIRST CLOSE THE WINDOW SECOND TURN OFF THE LAMP FINALLY LOCK THE DOOR",
+ "raw": {
+ "normalized_reference": "first, close the window; second, turn off the lamp; finally, lock the door.",
+ "normalized_hypothesis": "first close the window second turn off the lamp finally lock the door",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "1st close the window 2nd turn off the lamp finally lock the door",
+ "normalized_hypothesis": "1st close the window 2nd turn off the lamp finally lock the door",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0011",
+ "category": "question",
+ "text": "When did you first realize something was wrong?",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0011.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WHEN DID YOU FIRST REALIZE SOMETHING WAS WRONG",
+ "raw": {
+ "normalized_reference": "when did you first realize something was wrong?",
+ "normalized_hypothesis": "when did you first realize something was wrong",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "when did you 1st realize something was wrong",
+ "normalized_hypothesis": "when did you 1st realize something was wrong",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0012",
+ "category": "rare_words",
+ "text": "The archaeologist catalogued a Byzantine reliquary and several anachronistic fragments.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0012.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE ARCHAEOLOGIST CATALOGUED A BIZANTINE RELIQUARY AND SEVERAL ANICRONISTIC FRAGMENTS",
+ "raw": {
+ "normalized_reference": "the archaeologist catalogued a byzantine reliquary and several anachronistic fragments.",
+ "normalized_hypothesis": "the archaeologist catalogued a bizantine reliquary and several anicronistic fragments",
+ "errors": 2,
+ "words": 10,
+ "wer": 0.2
+ },
+ "semantic": {
+ "normalized_reference": "the archaeologist catalogued a byzantine reliquary and several anachronistic fragments",
+ "normalized_hypothesis": "the archaeologist catalogued a bizantine reliquary and several anicronistic fragments",
+ "errors": 2,
+ "words": 10,
+ "wer": 0.2
+ }
+ },
+ {
+ "id": "release500_0013",
+ "category": "short",
+ "text": "I have no idea.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0013.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I HAVE NO IDEA",
+ "raw": {
+ "normalized_reference": "i have no idea.",
+ "normalized_hypothesis": "i have no idea",
+ "errors": 0,
+ "words": 4,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i have no idea",
+ "normalized_hypothesis": "i have no idea",
+ "errors": 0,
+ "words": 4,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0014",
+ "category": "technical",
+ "text": "The system runs on three core components that all have to stay in sync.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0014.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE SYSTEM RUNS ON THREE CORPS COMPONENTS THAT ALL HAVE TO STAY AND SINK",
+ "raw": {
+ "normalized_reference": "the system runs on three core components that all have to stay in sync.",
+ "normalized_hypothesis": "the system runs on three corps components that all have to stay and sink",
+ "errors": 3,
+ "words": 14,
+ "wer": 0.21428571428571427
+ },
+ "semantic": {
+ "normalized_reference": "the system runs on 3 core components that all have to stay in sync",
+ "normalized_hypothesis": "the system runs on 3 corps components that all have to stay and sink",
+ "errors": 3,
+ "words": 14,
+ "wer": 0.21428571428571427
+ }
+ },
+ {
+ "id": "release500_0015",
+ "category": "conversational",
+ "text": "We improvised the entire second half and somehow nobody noticed.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0015.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WE IMPROVISED THE ENTIRE SECOND HALF AND SOMEHOW NOBODY NOTICED",
+ "raw": {
+ "normalized_reference": "we improvised the entire second half and somehow nobody noticed.",
+ "normalized_hypothesis": "we improvised the entire second half and somehow nobody noticed",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "we improvised the entire 2nd half and somehow nobody noticed",
+ "normalized_hypothesis": "we improvised the entire 2nd half and somehow nobody noticed",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0016",
+ "category": "descriptive",
+ "text": "There's a smell in old buildings that's impossible to describe but immediately familiar.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0016.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THERE'S A SMELL IN OLD BUILDINGS THAT'S IMPOSSIBLE TO DESCRIBE BUT IMMEDIATELY FAMILIAR",
+ "raw": {
+ "normalized_reference": "there's a smell in old buildings that's impossible to describe but immediately familiar.",
+ "normalized_hypothesis": "there's a smell in old buildings that's impossible to describe but immediately familiar",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "there is a smell in old buildings that is impossible to describe but immediately familiar",
+ "normalized_hypothesis": "there is a smell in old buildings that is impossible to describe but immediately familiar",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0017",
+ "category": "emotional",
+ "text": "I specifically asked for one thing. One thing. And it still didn't happen.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0017.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I SPECIFICALLY ASKED FOR ONE THING ONE THING AND IT STILL DIDN'T HAPPEN",
+ "raw": {
+ "normalized_reference": "i specifically asked for one thing. one thing. and it still didn't happen.",
+ "normalized_hypothesis": "i specifically asked for one thing one thing and it still didn't happen",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i specifically asked for one thing one thing and it still did not happen",
+ "normalized_hypothesis": "i specifically asked for one thing one thing and it still did not happen",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0018",
+ "category": "homographs",
+ "text": "They refuse to place the refuse beside the produce aisle.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0018.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THEY REFUSED TO PLACE THE REFUSE BESIDE THE PRODUCE ISLE",
+ "raw": {
+ "normalized_reference": "they refuse to place the refuse beside the produce aisle.",
+ "normalized_hypothesis": "they refused to place the refuse beside the produce isle",
+ "errors": 2,
+ "words": 10,
+ "wer": 0.2
+ },
+ "semantic": {
+ "normalized_reference": "they refuse to place the refuse beside the produce aisle",
+ "normalized_hypothesis": "they refused to place the refuse beside the produce isle",
+ "errors": 2,
+ "words": 10,
+ "wer": 0.2
+ }
+ },
+ {
+ "id": "release500_0019",
+ "category": "long",
+ "text": "There were days last year when I genuinely didn't know how I was going to get through it, and somehow I did, and I still don't fully understand how.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0019.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THERE WERE DAYS LAST YEAR WHEN I GENUINELY DIDN'T KNOW HOW I WAS GOING TO GET THROUGH IT AND SOMEHOW I DID AND I STILL DON'T FULLY UNDERSTAND HOW",
+ "raw": {
+ "normalized_reference": "there were days last year when i genuinely didn't know how i was going to get through it, and somehow i did, and i still don't fully understand how.",
+ "normalized_hypothesis": "there were days last year when i genuinely didn't know how i was going to get through it and somehow i did and i still don't fully understand how",
+ "errors": 0,
+ "words": 29,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "there were days last year when i genuinely did not know how i was going to get through it and somehow i did and i still do not fully understand how",
+ "normalized_hypothesis": "there were days last year when i genuinely did not know how i was going to get through it and somehow i did and i still do not fully understand how",
+ "errors": 0,
+ "words": 31,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0020",
+ "category": "long_range",
+ "text": "Before announcing the winner, the judge thanked the volunteers who had worked through the storm to repair the damaged stage.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0020.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "BEFORE ANNOUNCING THE WINNER THE JUDGE THANKED THE VOLUNTEERS WHO HAD WORKED THROUGH THIS STORM TO REPAIR THE DAMAGED STAGE",
+ "raw": {
+ "normalized_reference": "before announcing the winner, the judge thanked the volunteers who had worked through the storm to repair the damaged stage.",
+ "normalized_hypothesis": "before announcing the winner the judge thanked the volunteers who had worked through this storm to repair the damaged stage",
+ "errors": 1,
+ "words": 20,
+ "wer": 0.05
+ },
+ "semantic": {
+ "normalized_reference": "before announcing the winner the judge thanked the volunteers who had worked through the storm to repair the damaged stage",
+ "normalized_hypothesis": "before announcing the winner the judge thanked the volunteers who had worked through this storm to repair the damaged stage",
+ "errors": 1,
+ "words": 20,
+ "wer": 0.05
+ }
+ },
+ {
+ "id": "release500_0021",
+ "category": "names_places",
+ "text": "Professor Nguyen mailed the fluorescent poster to Reykjavik on Thursday.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0021.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "PROFESSOR NUMYAN MAILED THE FLORISCENT POSTER TO RAKE JAVIC ON THURSDAY",
+ "raw": {
+ "normalized_reference": "professor nguyen mailed the fluorescent poster to reykjavik on thursday.",
+ "normalized_hypothesis": "professor numyan mailed the floriscent poster to rake javic on thursday",
+ "errors": 4,
+ "words": 10,
+ "wer": 0.4
+ },
+ "semantic": {
+ "normalized_reference": "professor nguyen mailed the fluorescent poster to reykjavik on thursday",
+ "normalized_hypothesis": "professor numyan mailed the floriscent poster to rake javic on thursday",
+ "errors": 4,
+ "words": 10,
+ "wer": 0.4
+ }
+ },
+ {
+ "id": "release500_0022",
+ "category": "narrative",
+ "text": "The plan was simple. The execution was anything but.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0022.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE PLAN WAS SIMPLE THE EXECUTION WAS ANYTHING BUT",
+ "raw": {
+ "normalized_reference": "the plan was simple. the execution was anything but.",
+ "normalized_hypothesis": "the plan was simple the execution was anything but",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the plan was simple the execution was anything but",
+ "normalized_hypothesis": "the plan was simple the execution was anything but",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0023",
+ "category": "numbers",
+ "text": "Her score improved from eighty-two percent to ninety-seven percent in two weeks.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0023.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "BERSCORE IMPROVED FROM EIGHTY TWO PER CENT TO NINETY SEVEN PER CENT IN TWO WEEKS",
+ "raw": {
+ "normalized_reference": "her score improved from eighty-two percent to ninety-seven percent in two weeks.",
+ "normalized_hypothesis": "berscore improved from eighty two per cent to ninety seven per cent in two weeks",
+ "errors": 6,
+ "words": 14,
+ "wer": 0.42857142857142855
+ },
+ "semantic": {
+ "normalized_reference": "her score improved from 82% to 97% in 2 weeks",
+ "normalized_hypothesis": "berscore improved from 82% to 97% in 2 weeks",
+ "errors": 2,
+ "words": 10,
+ "wer": 0.2
+ }
+ },
+ {
+ "id": "release500_0024",
+ "category": "punctuation",
+ "text": "Why would anyone pack spoons, batteries, and a violin for one short trip?",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0024.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WHY WOULD ANY ONE PACK SPOONS BATTERIES AND A VIOLIN FOR ONE SHORT TRIP",
+ "raw": {
+ "normalized_reference": "why would anyone pack spoons, batteries, and a violin for one short trip?",
+ "normalized_hypothesis": "why would any one pack spoons batteries and a violin for one short trip",
+ "errors": 2,
+ "words": 13,
+ "wer": 0.15384615384615385
+ },
+ "semantic": {
+ "normalized_reference": "why would anyone pack spoons batteries and a violin for one short trip",
+ "normalized_hypothesis": "why would any one pack spoons batteries and a violin for one short trip",
+ "errors": 2,
+ "words": 13,
+ "wer": 0.15384615384615385
+ }
+ },
+ {
+ "id": "release500_0025",
+ "category": "question",
+ "text": "How long have you been sitting on this?",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0025.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HOW LONG HAVE YOU BEEN SITTING ON THIS",
+ "raw": {
+ "normalized_reference": "how long have you been sitting on this?",
+ "normalized_hypothesis": "how long have you been sitting on this",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "how long have you been sitting on this",
+ "normalized_hypothesis": "how long have you been sitting on this",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0026",
+ "category": "rare_words",
+ "text": "A mischievous chameleon rested beside the chrysanthemum and the bougainvillea.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0026.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "A MISCHIEVOUS CAMELION RESTED BESIDE THE CHRYSANTHEMUM AND THE BAUDIINGVILIA",
+ "raw": {
+ "normalized_reference": "a mischievous chameleon rested beside the chrysanthemum and the bougainvillea.",
+ "normalized_hypothesis": "a mischievous camelion rested beside the chrysanthemum and the baudiingvilia",
+ "errors": 2,
+ "words": 10,
+ "wer": 0.2
+ },
+ "semantic": {
+ "normalized_reference": "a mischievous chameleon rested beside the chrysanthemum and the bougainvillea",
+ "normalized_hypothesis": "a mischievous camelion rested beside the chrysanthemum and the baudiingvilia",
+ "errors": 2,
+ "words": 10,
+ "wer": 0.2
+ }
+ },
+ {
+ "id": "release500_0027",
+ "category": "short",
+ "text": "We don't talk about that.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0027.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WE DON'T TALK ABOUT THAT",
+ "raw": {
+ "normalized_reference": "we don't talk about that.",
+ "normalized_hypothesis": "we don't talk about that",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "we do not talk about that",
+ "normalized_hypothesis": "we do not talk about that",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0028",
+ "category": "technical",
+ "text": "Hash the UTF eight payload with S H A two fifty-six before opening the checkpoint.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0028.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "ASH THE U T F IF EGHT PALOD WITH S H A TWO FIFTY SIX BEFORE OPENING THE CHECK POINT",
+ "raw": {
+ "normalized_reference": "hash the utf eight payload with s h a two fifty-six before opening the checkpoint.",
+ "normalized_hypothesis": "ash the u t f if eght palod with s h a two fifty six before opening the check point",
+ "errors": 9,
+ "words": 16,
+ "wer": 0.5625
+ },
+ "semantic": {
+ "normalized_reference": "hash the utf 8 payload with s h a 256 before opening the checkpoint",
+ "normalized_hypothesis": "ash the u t f if eght palod with s h a 256 before opening the check point",
+ "errors": 9,
+ "words": 14,
+ "wer": 0.6428571428571429
+ }
+ },
+ {
+ "id": "release500_0029",
+ "category": "conversational",
+ "text": "I thought I had it all figured out, and then this happened.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0029.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I THOUGHT I HAD IT ALL FIGURED OUT AND THEN THIS HAPPENED",
+ "raw": {
+ "normalized_reference": "i thought i had it all figured out, and then this happened.",
+ "normalized_hypothesis": "i thought i had it all figured out and then this happened",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i thought i had it all figured out and then this happened",
+ "normalized_hypothesis": "i thought i had it all figured out and then this happened",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0030",
+ "category": "descriptive",
+ "text": "The kind of quiet you only get at three in the morning in a city.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0030.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE KIND OF QUIET YOU ONLY GET AT THREE IN THE MORNING IN A CITY",
+ "raw": {
+ "normalized_reference": "the kind of quiet you only get at three in the morning in a city.",
+ "normalized_hypothesis": "the kind of quiet you only get at three in the morning in a city",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the kind of quiet you only get at 3 in the morning in a city",
+ "normalized_hypothesis": "the kind of quiet you only get at 3 in the morning in a city",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0031",
+ "category": "emotional",
+ "text": "The joy of it was almost overwhelming, like too much good all at once.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0031.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE JOY OF IT WAS ALMOST OVERWHELMING LIKE TOO MUCH GOOD ALL AT ONCE",
+ "raw": {
+ "normalized_reference": "the joy of it was almost overwhelming, like too much good all at once.",
+ "normalized_hypothesis": "the joy of it was almost overwhelming like too much good all at once",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the joy of it was almost overwhelming like too much good all at once",
+ "normalized_hypothesis": "the joy of it was almost overwhelming like too much good all at once",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0032",
+ "category": "homographs",
+ "text": "Please close the door while I stand close to the microphone.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0032.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "PLEASE CLOSE THE DOOR WHILE I STAND CLOSE TO THE MICROPHONE",
+ "raw": {
+ "normalized_reference": "please close the door while i stand close to the microphone.",
+ "normalized_hypothesis": "please close the door while i stand close to the microphone",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "please close the door while i stand close to the microphone",
+ "normalized_hypothesis": "please close the door while i stand close to the microphone",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0033",
+ "category": "long",
+ "text": "I think the most underrated skill in the world is knowing when to stop explaining yourself and just let people draw their own conclusions.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0033.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I THINK THE MOST UNDERRATED SKILL IN THE WORLD IS KNOWING WHEN TO STOP EXPLAINING YOURSELF AND JUST LET PEOPLE DRAW THEIR OWN CONCLUSIONS",
+ "raw": {
+ "normalized_reference": "i think the most underrated skill in the world is knowing when to stop explaining yourself and just let people draw their own conclusions.",
+ "normalized_hypothesis": "i think the most underrated skill in the world is knowing when to stop explaining yourself and just let people draw their own conclusions",
+ "errors": 0,
+ "words": 24,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i think the most underrated skill in the world is knowing when to stop explaining yourself and just let people draw their own conclusions",
+ "normalized_hypothesis": "i think the most underrated skill in the world is knowing when to stop explaining yourself and just let people draw their own conclusions",
+ "errors": 0,
+ "words": 24,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0034",
+ "category": "long_range",
+ "text": "When the final bus disappeared around the corner, Nadia remembered that her umbrella, notebook, and lunch were still inside the quiet office.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0034.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WHEN THE FINAL BUST DISAPPEARED AROUND THE CORNER NODYA REMEMBERED THAT HER UMBRELLA NOTE BOOK AND LUNCH WERE STILL INSIDE THE QUIET OFFICE",
+ "raw": {
+ "normalized_reference": "when the final bus disappeared around the corner, nadia remembered that her umbrella, notebook, and lunch were still inside the quiet office.",
+ "normalized_hypothesis": "when the final bust disappeared around the corner nodya remembered that her umbrella note book and lunch were still inside the quiet office",
+ "errors": 4,
+ "words": 22,
+ "wer": 0.18181818181818182
+ },
+ "semantic": {
+ "normalized_reference": "when the final bus disappeared around the corner nadia remembered that her umbrella notebook and lunch were still inside the quiet office",
+ "normalized_hypothesis": "when the final bust disappeared around the corner nodya remembered that her umbrella note book and lunch were still inside the quiet office",
+ "errors": 4,
+ "words": 22,
+ "wer": 0.18181818181818182
+ }
+ },
+ {
+ "id": "release500_0035",
+ "category": "names_places",
+ "text": "Aurelia and Joaquin crossed the Schuylkill River before noon.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0035.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "ORLYA AND JOQUING CROSSED THE SHILKO RIVER BEFORE NOON",
+ "raw": {
+ "normalized_reference": "aurelia and joaquin crossed the schuylkill river before noon.",
+ "normalized_hypothesis": "orlya and joquing crossed the shilko river before noon",
+ "errors": 3,
+ "words": 9,
+ "wer": 0.3333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "aurelia and joaquin crossed the schuylkill river before noon",
+ "normalized_hypothesis": "orlya and joquing crossed the shilko river before noon",
+ "errors": 3,
+ "words": 9,
+ "wer": 0.3333333333333333
+ }
+ },
+ {
+ "id": "release500_0036",
+ "category": "narrative",
+ "text": "The village had a name, but nobody used it. Everyone just called it home.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0036.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE VILLAGE HAD A NAME BUT NOBODY USED IT EVERYONE JUST CALLED AT HOME",
+ "raw": {
+ "normalized_reference": "the village had a name, but nobody used it. everyone just called it home.",
+ "normalized_hypothesis": "the village had a name but nobody used it everyone just called at home",
+ "errors": 1,
+ "words": 14,
+ "wer": 0.07142857142857142
+ },
+ "semantic": {
+ "normalized_reference": "the village had a name but nobody used it everyone just called it home",
+ "normalized_hypothesis": "the village had a name but nobody used it everyone just called at home",
+ "errors": 1,
+ "words": 14,
+ "wer": 0.07142857142857142
+ }
+ },
+ {
+ "id": "release500_0037",
+ "category": "numbers",
+ "text": "Increase the gain by three decibels, wait five seconds, then reduce it by one.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0037.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "INCREASE THE GAIN BY THREE DESIBALS WEIGHT FIVE SECONDS THEN REDUCE IT BY ONE",
+ "raw": {
+ "normalized_reference": "increase the gain by three decibels, wait five seconds, then reduce it by one.",
+ "normalized_hypothesis": "increase the gain by three desibals weight five seconds then reduce it by one",
+ "errors": 2,
+ "words": 14,
+ "wer": 0.14285714285714285
+ },
+ "semantic": {
+ "normalized_reference": "increase the gain by 3 decibels wait 5 seconds then reduce it by one",
+ "normalized_hypothesis": "increase the gain by 3 desibals weight 5 seconds then reduce it by one",
+ "errors": 2,
+ "words": 14,
+ "wer": 0.14285714285714285
+ }
+ },
+ {
+ "id": "release500_0038",
+ "category": "punctuation",
+ "text": "The result was unexpected: clear, stable, and surprisingly natural.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0038.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE RESULT WAS UNEXPECTED CLEAR STABLE AND SURPRISINGLY NATURAL",
+ "raw": {
+ "normalized_reference": "the result was unexpected: clear, stable, and surprisingly natural.",
+ "normalized_hypothesis": "the result was unexpected clear stable and surprisingly natural",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the result was unexpected clear stable and surprisingly natural",
+ "normalized_hypothesis": "the result was unexpected clear stable and surprisingly natural",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0039",
+ "category": "question",
+ "text": "Who told you that was the right way to do it?",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0039.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WHO TOLD YOU THAT WAS THE RIGHT WAY TO DO IT",
+ "raw": {
+ "normalized_reference": "who told you that was the right way to do it?",
+ "normalized_hypothesis": "who told you that was the right way to do it",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "who told you that was the right way to do it",
+ "normalized_hypothesis": "who told you that was the right way to do it",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0040",
+ "category": "rare_words",
+ "text": "The meticulous cartographer examined an iridescent archipelago through binoculars.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0040.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE METICULOUS CARTOGRAPHER EXAMINED AN IRIDESCENT ARCHIPELAGO THROUGH BONOCULARS",
+ "raw": {
+ "normalized_reference": "the meticulous cartographer examined an iridescent archipelago through binoculars.",
+ "normalized_hypothesis": "the meticulous cartographer examined an iridescent archipelago through bonoculars",
+ "errors": 1,
+ "words": 9,
+ "wer": 0.1111111111111111
+ },
+ "semantic": {
+ "normalized_reference": "the meticulous cartographer examined an iridescent archipelago through binoculars",
+ "normalized_hypothesis": "the meticulous cartographer examined an iridescent archipelago through bonoculars",
+ "errors": 1,
+ "words": 9,
+ "wer": 0.1111111111111111
+ }
+ },
+ {
+ "id": "release500_0041",
+ "category": "short",
+ "text": "I mean. Yeah. Kind of.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0041.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I MEAN THAT YET TAT KIND OF",
+ "raw": {
+ "normalized_reference": "i mean. yeah. kind of.",
+ "normalized_hypothesis": "i mean that yet tat kind of",
+ "errors": 3,
+ "words": 5,
+ "wer": 0.6
+ },
+ "semantic": {
+ "normalized_reference": "i mean yeah kind of",
+ "normalized_hypothesis": "i mean that yet tat kind of",
+ "errors": 3,
+ "words": 5,
+ "wer": 0.6
+ }
+ },
+ {
+ "id": "release500_0042",
+ "category": "technical",
+ "text": "The logs are your best friend when something behaves unexpectedly.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0042.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE LOGS ARE YOUR BEST FRIEND WHEN SOMETHING BEHAVES UNEXPECTEDLY",
+ "raw": {
+ "normalized_reference": "the logs are your best friend when something behaves unexpectedly.",
+ "normalized_hypothesis": "the logs are your best friend when something behaves unexpectedly",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the logs are your best friend when something behaves unexpectedly",
+ "normalized_hypothesis": "the logs are your best friend when something behaves unexpectedly",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0043",
+ "category": "conversational",
+ "text": "He kept talking and I kept nodding and neither of us was listening.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0043.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE KEPT TALKING AND I KEPT NODDING AND NEITHER OF US WAS LISTENING",
+ "raw": {
+ "normalized_reference": "he kept talking and i kept nodding and neither of us was listening.",
+ "normalized_hypothesis": "he kept talking and i kept nodding and neither of us was listening",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he kept talking and i kept nodding and neither of us was listening",
+ "normalized_hypothesis": "he kept talking and i kept nodding and neither of us was listening",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0044",
+ "category": "descriptive",
+ "text": "It's one of those places where everyone seems to know each other already.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0044.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT'S ONE OF THOSE PLACES WHERE EVERYONE SEEMS TO KNOW EACH OTHER ALREADY",
+ "raw": {
+ "normalized_reference": "it's one of those places where everyone seems to know each other already.",
+ "normalized_hypothesis": "it's one of those places where everyone seems to know each other already",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it is one of those places where everyone seems to know each other already",
+ "normalized_hypothesis": "it is one of those places where everyone seems to know each other already",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0045",
+ "category": "emotional",
+ "text": "He said he was proud of me and nobody had ever said that to me before.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0045.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE SAID HE WAS PROUD OF ME AND NOBODY HAD EVER SAID THAT TO ME BEFORE",
+ "raw": {
+ "normalized_reference": "he said he was proud of me and nobody had ever said that to me before.",
+ "normalized_hypothesis": "he said he was proud of me and nobody had ever said that to me before",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he said he was proud of me and nobody had ever said that to me before",
+ "normalized_hypothesis": "he said he was proud of me and nobody had ever said that to me before",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0046",
+ "category": "homographs",
+ "text": "The minute detail took a minute to explain clearly.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0046.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE MINUTE DETAIL TOOK A MINUTE TO EXPLAIN CLEARLY",
+ "raw": {
+ "normalized_reference": "the minute detail took a minute to explain clearly.",
+ "normalized_hypothesis": "the minute detail took a minute to explain clearly",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the minute detail took a minute to explain clearly",
+ "normalized_hypothesis": "the minute detail took a minute to explain clearly",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0047",
+ "category": "long",
+ "text": "The strange thing about really wanting something is that getting it never quite feels the way you imagined it would, and not getting it never quite feels as bad.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0047.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE STRANGE THING ABOUT REALLY WANTING SOMETHING IS THAT GETTING IT NEVER QUITE FEELS THE WAY YOU IMAGINED IT WOULD AND NOT GETTING IT NEVER QUITE FEELS AS BAD",
+ "raw": {
+ "normalized_reference": "the strange thing about really wanting something is that getting it never quite feels the way you imagined it would, and not getting it never quite feels as bad.",
+ "normalized_hypothesis": "the strange thing about really wanting something is that getting it never quite feels the way you imagined it would and not getting it never quite feels as bad",
+ "errors": 0,
+ "words": 29,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the strange thing about really wanting something is that getting it never quite feels the way you imagined it would and not getting it never quite feels as bad",
+ "normalized_hypothesis": "the strange thing about really wanting something is that getting it never quite feels the way you imagined it would and not getting it never quite feels as bad",
+ "errors": 0,
+ "words": 29,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0048",
+ "category": "long_range",
+ "text": "Even though the recipe looked simple at first, preparing the sauce required patience, careful timing, and a much larger pan.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0048.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "EVEN THOUGH THE RECIPE LOOKED SIMPLE AT FIRST PREPARING THE SAUCE REQUIRED PATIENT'S CAREFUL TIMING AND A MUCH LARGER PAN",
+ "raw": {
+ "normalized_reference": "even though the recipe looked simple at first, preparing the sauce required patience, careful timing, and a much larger pan.",
+ "normalized_hypothesis": "even though the recipe looked simple at first preparing the sauce required patient's careful timing and a much larger pan",
+ "errors": 1,
+ "words": 20,
+ "wer": 0.05
+ },
+ "semantic": {
+ "normalized_reference": "even though the recipe looked simple at 1st preparing the sauce required patience careful timing and a much larger pan",
+ "normalized_hypothesis": "even though the recipe looked simple at 1st preparing the sauce required patient is careful timing and a much larger pan",
+ "errors": 2,
+ "words": 20,
+ "wer": 0.1
+ }
+ },
+ {
+ "id": "release500_0049",
+ "category": "names_places",
+ "text": "Xiomara met Siobhan near the old library in Worcestershire.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0049.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "DEMARRA MET CHAVONNTE NEAR THE OLD LIBRARY IN WORESTERSHIRE",
+ "raw": {
+ "normalized_reference": "xiomara met siobhan near the old library in worcestershire.",
+ "normalized_hypothesis": "demarra met chavonnte near the old library in worestershire",
+ "errors": 3,
+ "words": 9,
+ "wer": 0.3333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "xiomara met siobhan near the old library in worcestershire",
+ "normalized_hypothesis": "demarra met chavonnte near the old library in worestershire",
+ "errors": 3,
+ "words": 9,
+ "wer": 0.3333333333333333
+ }
+ },
+ {
+ "id": "release500_0050",
+ "category": "narrative",
+ "text": "She walked into the room and everyone went quiet, not because she was loud, but because she wasn't.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0050.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SHE WALKED INTO THE ROOM AND EVERYONE WENT QUIET NOT BECAUSE SHE WAS LOUD BUT BECAUSE SHE WASN'T",
+ "raw": {
+ "normalized_reference": "she walked into the room and everyone went quiet, not because she was loud, but because she wasn't.",
+ "normalized_hypothesis": "she walked into the room and everyone went quiet not because she was loud but because she wasn't",
+ "errors": 0,
+ "words": 18,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "she walked into the room and everyone went quiet not because she was loud but because she was not",
+ "normalized_hypothesis": "she walked into the room and everyone went quiet not because she was loud but because she was not",
+ "errors": 0,
+ "words": 19,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0051",
+ "category": "numbers",
+ "text": "At exactly seven forty-five, Morgan placed eighteen dollars and fifty cents beside the notebook.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0051.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "EXACTLY SEVEN FORTY FIVE MORGAN PLACED EIGHTEEN DOLLARS AND FIFTY CENTS BESIDE THE NOTEBOOK",
+ "raw": {
+ "normalized_reference": "at exactly seven forty-five, morgan placed eighteen dollars and fifty cents beside the notebook.",
+ "normalized_hypothesis": "exactly seven forty five morgan placed eighteen dollars and fifty cents beside the notebook",
+ "errors": 1,
+ "words": 15,
+ "wer": 0.06666666666666667
+ },
+ "semantic": {
+ "normalized_reference": "at exactly 745 morgan placed $18.50 beside the notebook",
+ "normalized_hypothesis": "exactly 745 morgan placed $18.50 beside the notebook",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ }
+ },
+ {
+ "id": "release500_0052",
+ "category": "punctuation",
+ "text": "Does punctuation work? Yes: commas, semicolons, and periods should create sensible boundaries.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0052.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "DOES PUNCTUATION WORK YES COMMAS SEMICOLANS AND PERIODS SHOULD CREATE SENSIBLE BOUNDARIES",
+ "raw": {
+ "normalized_reference": "does punctuation work? yes: commas, semicolons, and periods should create sensible boundaries.",
+ "normalized_hypothesis": "does punctuation work yes commas semicolans and periods should create sensible boundaries",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "does punctuation work yes commas semicolons and periods should create sensible boundaries",
+ "normalized_hypothesis": "does punctuation work yes commas semicolans and periods should create sensible boundaries",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ }
+ },
+ {
+ "id": "release500_0053",
+ "category": "question",
+ "text": "Have you ever just stopped and thought about how strange all of this is?",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0053.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HAVE YOU EVER JUST STOPPED AND THOUGHT ABOUT HOW STRANGE ALL OF THIS IS",
+ "raw": {
+ "normalized_reference": "have you ever just stopped and thought about how strange all of this is?",
+ "normalized_hypothesis": "have you ever just stopped and thought about how strange all of this is",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "have you ever just stopped and thought about how strange all of this is",
+ "normalized_hypothesis": "have you ever just stopped and thought about how strange all of this is",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0054",
+ "category": "rare_words",
+ "text": "Her idiosyncratic explanation sounded plausible but remained fundamentally ambiguous.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0054.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HER IDIOSOCRATIC EXPLANATION SOUNDED PLAUSIBLE BUT REMAINED FUNDAMENTALLY AMBIGUOUS",
+ "raw": {
+ "normalized_reference": "her idiosyncratic explanation sounded plausible but remained fundamentally ambiguous.",
+ "normalized_hypothesis": "her idiosocratic explanation sounded plausible but remained fundamentally ambiguous",
+ "errors": 1,
+ "words": 9,
+ "wer": 0.1111111111111111
+ },
+ "semantic": {
+ "normalized_reference": "her idiosyncratic explanation sounded plausible but remained fundamentally ambiguous",
+ "normalized_hypothesis": "her idiosocratic explanation sounded plausible but remained fundamentally ambiguous",
+ "errors": 1,
+ "words": 9,
+ "wer": 0.1111111111111111
+ }
+ },
+ {
+ "id": "release500_0055",
+ "category": "short",
+ "text": "That tracks.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0055.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THAT TRACKS",
+ "raw": {
+ "normalized_reference": "that tracks.",
+ "normalized_hypothesis": "that tracks",
+ "errors": 0,
+ "words": 2,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "that tracks",
+ "normalized_hypothesis": "that tracks",
+ "errors": 0,
+ "words": 2,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0056",
+ "category": "technical",
+ "text": "The key thing to remember is that order matters here more than you'd think.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0056.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE KEY THING TO REMEMBER IS THAT ORDER MATTERS HEAR MORE THAN YOU'D THINK",
+ "raw": {
+ "normalized_reference": "the key thing to remember is that order matters here more than you'd think.",
+ "normalized_hypothesis": "the key thing to remember is that order matters hear more than you'd think",
+ "errors": 1,
+ "words": 14,
+ "wer": 0.07142857142857142
+ },
+ "semantic": {
+ "normalized_reference": "the key thing to remember is that order matters here more than you would think",
+ "normalized_hypothesis": "the key thing to remember is that order matters hear more than you would think",
+ "errors": 1,
+ "words": 15,
+ "wer": 0.06666666666666667
+ }
+ },
+ {
+ "id": "release500_0057",
+ "category": "conversational",
+ "text": "I would not say that again in this room if I were you.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0057.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I WOULD NOT SAY THAT AGAIN IN THIS ROOM IF I WERE YOU",
+ "raw": {
+ "normalized_reference": "i would not say that again in this room if i were you.",
+ "normalized_hypothesis": "i would not say that again in this room if i were you",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i would not say that again in this room if i were you",
+ "normalized_hypothesis": "i would not say that again in this room if i were you",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0058",
+ "category": "descriptive",
+ "text": "The room was filled with the quiet that means something is about to change.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0058.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE ROOM WAS FILLED WITH THE QUIET THAT MEANS SOMETHING IS ABOUT TO CHANGE",
+ "raw": {
+ "normalized_reference": "the room was filled with the quiet that means something is about to change.",
+ "normalized_hypothesis": "the room was filled with the quiet that means something is about to change",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the room was filled with the quiet that means something is about to change",
+ "normalized_hypothesis": "the room was filled with the quiet that means something is about to change",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0059",
+ "category": "emotional",
+ "text": "I didn't expect to feel that way about it, but I did.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0059.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I DIDN'T EXPECT TO FEEL THAT WAY ABOUT IT BUT I DID",
+ "raw": {
+ "normalized_reference": "i didn't expect to feel that way about it, but i did.",
+ "normalized_hypothesis": "i didn't expect to feel that way about it but i did",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i did not expect to feel that way about it but i did",
+ "normalized_hypothesis": "i did not expect to feel that way about it but i did",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0060",
+ "category": "homographs",
+ "text": "I object to the strange object that was left near the entrance.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0060.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I OBJECT TO THE STRANGE OBJECT THAT WAS LEFT NEAR THE ENTRANCE",
+ "raw": {
+ "normalized_reference": "i object to the strange object that was left near the entrance.",
+ "normalized_hypothesis": "i object to the strange object that was left near the entrance",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i object to the strange object that was left near the entrance",
+ "normalized_hypothesis": "i object to the strange object that was left near the entrance",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0061",
+ "category": "long",
+ "text": "I've learned more from the things that went wrong than from the things that worked, which is a frustrating way to have to learn things but apparently the only way that stuck.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0061.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I'VE LEARNED MORE FROM THE THINGS THAT WENT WRONG THAN FROM THE THINGS THAT WORKED WHICH IS A FRUSTRATING WAY TO HAVE TO LEARN THINGS BUT APPARENTLY THE ONLY WAY THAT STOD",
+ "raw": {
+ "normalized_reference": "i've learned more from the things that went wrong than from the things that worked, which is a frustrating way to have to learn things but apparently the only way that stuck.",
+ "normalized_hypothesis": "i've learned more from the things that went wrong than from the things that worked which is a frustrating way to have to learn things but apparently the only way that stod",
+ "errors": 1,
+ "words": 32,
+ "wer": 0.03125
+ },
+ "semantic": {
+ "normalized_reference": "i have learned more from the things that went wrong than from the things that worked which is a frustrating way to have to learn things but apparently the only way that stuck",
+ "normalized_hypothesis": "i have learned more from the things that went wrong than from the things that worked which is a frustrating way to have to learn things but apparently the only way that stod",
+ "errors": 1,
+ "words": 33,
+ "wer": 0.030303030303030304
+ }
+ },
+ {
+ "id": "release500_0062",
+ "category": "long_range",
+ "text": "Because every nearby hotel was full, the exhausted hikers followed the narrow road until they found a farmhouse with one empty room.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0062.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "BECAUSE EVERY NEARBY HOTEL WAS FULL THE EXHAUSTED HIKERS FOLLOWED THE NARROW ROAD UNTIL THEY FOUND A FARMHOUSE WITH ONE EMPTY ROOM",
+ "raw": {
+ "normalized_reference": "because every nearby hotel was full, the exhausted hikers followed the narrow road until they found a farmhouse with one empty room.",
+ "normalized_hypothesis": "because every nearby hotel was full the exhausted hikers followed the narrow road until they found a farmhouse with one empty room",
+ "errors": 0,
+ "words": 22,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "because every nearby hotel was full the exhausted hikers followed the narrow road until they found a farmhouse with one empty room",
+ "normalized_hypothesis": "because every nearby hotel was full the exhausted hikers followed the narrow road until they found a farmhouse with one empty room",
+ "errors": 0,
+ "words": 22,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0063",
+ "category": "names_places",
+ "text": "The itinerary sends Beatrice from Albuquerque to Des Moines and then to Boise.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0063.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE ITINERARY SENDS BEATRICE FROM ALBAQIRQUY TO DUOIGN AND THEN TO BOYS",
+ "raw": {
+ "normalized_reference": "the itinerary sends beatrice from albuquerque to des moines and then to boise.",
+ "normalized_hypothesis": "the itinerary sends beatrice from albaqirquy to duoign and then to boys",
+ "errors": 4,
+ "words": 13,
+ "wer": 0.3076923076923077
+ },
+ "semantic": {
+ "normalized_reference": "the itinerary sends beatrice from albuquerque to des moines and then to boise",
+ "normalized_hypothesis": "the itinerary sends beatrice from albaqirquy to duoign and then to boys",
+ "errors": 4,
+ "words": 13,
+ "wer": 0.3076923076923077
+ }
+ },
+ {
+ "id": "release500_0064",
+ "category": "narrative",
+ "text": "We had been driving for six hours when we finally saw the sign for the town.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0064.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WE HAD BEEN DRIVING FOR SIX HOURS WHEN WE FINALLY SAW THE SIGN FOR THE TOWN",
+ "raw": {
+ "normalized_reference": "we had been driving for six hours when we finally saw the sign for the town.",
+ "normalized_hypothesis": "we had been driving for six hours when we finally saw the sign for the town",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "we had been driving for 6 hours when we finally saw the sign for the town",
+ "normalized_hypothesis": "we had been driving for 6 hours when we finally saw the sign for the town",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0065",
+ "category": "numbers",
+ "text": "Call extension zero four nine after eight thirty in the morning.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0065.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "CALL EXTENSION ZERO FOUR NINE AFTER EIGHT THIRTY IN THE MORNING",
+ "raw": {
+ "normalized_reference": "call extension zero four nine after eight thirty in the morning.",
+ "normalized_hypothesis": "call extension zero four nine after eight thirty in the morning",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "call extension 049 after 830 in the morning",
+ "normalized_hypothesis": "call extension 049 after 830 in the morning",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0066",
+ "category": "punctuation",
+ "text": "Stop - listen carefully - then answer the question without guessing.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0066.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "STOP LISTEN CAREFULLY THEN ANSWER THE QUESTION WITHOUT GUESSING",
+ "raw": {
+ "normalized_reference": "stop - listen carefully - then answer the question without guessing.",
+ "normalized_hypothesis": "stop listen carefully then answer the question without guessing",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "stop listen carefully then answer the question without guessing",
+ "normalized_hypothesis": "stop listen carefully then answer the question without guessing",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0067",
+ "category": "question",
+ "text": "Does it matter at this point, honestly?",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0067.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "DOES IT MATTER AT THIS POINT HONESTLY",
+ "raw": {
+ "normalized_reference": "does it matter at this point, honestly?",
+ "normalized_hypothesis": "does it matter at this point honestly",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "does it matter at this point honestly",
+ "normalized_hypothesis": "does it matter at this point honestly",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0068",
+ "category": "rare_words",
+ "text": "An entrepreneurial meteorologist described the unprecedented atmospheric phenomenon.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0068.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "AN ONTRIPERNURIAL METEOROLOGIST DESCRIBED THE UNPRECEDENTED ATMOSPHERIC PHENOMENON",
+ "raw": {
+ "normalized_reference": "an entrepreneurial meteorologist described the unprecedented atmospheric phenomenon.",
+ "normalized_hypothesis": "an ontripernurial meteorologist described the unprecedented atmospheric phenomenon",
+ "errors": 1,
+ "words": 8,
+ "wer": 0.125
+ },
+ "semantic": {
+ "normalized_reference": "an entrepreneurial meteorologist described the unprecedented atmospheric phenomenon",
+ "normalized_hypothesis": "an ontripernurial meteorologist described the unprecedented atmospheric phenomenon",
+ "errors": 1,
+ "words": 8,
+ "wer": 0.125
+ }
+ },
+ {
+ "id": "release500_0069",
+ "category": "short",
+ "text": "That's not what I expected.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0069.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THAT'S NOT WHAT I EXPECTED",
+ "raw": {
+ "normalized_reference": "that's not what i expected.",
+ "normalized_hypothesis": "that's not what i expected",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "that is not what i expected",
+ "normalized_hypothesis": "that is not what i expected",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0070",
+ "category": "technical",
+ "text": "There's a known issue with that version that causes failures under load.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0070.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THERE'S A KNOWN ISSUE WITH THAT VERSION THAT CAUSES FAILURES UNDER LOAD",
+ "raw": {
+ "normalized_reference": "there's a known issue with that version that causes failures under load.",
+ "normalized_hypothesis": "there's a known issue with that version that causes failures under load",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "there is a known issue with that version that causes failures under load",
+ "normalized_hypothesis": "there is a known issue with that version that causes failures under load",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0071",
+ "category": "conversational",
+ "text": "Don't look now but I think that's him over there.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0071.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "DON'T LOOK NOW BUT I THINK THAT'S HIM OVER THERE",
+ "raw": {
+ "normalized_reference": "don't look now but i think that's him over there.",
+ "normalized_hypothesis": "don't look now but i think that's him over there",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "do not look now but i think that is him over there",
+ "normalized_hypothesis": "do not look now but i think that is him over there",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0072",
+ "category": "descriptive",
+ "text": "The city at night looks completely different from how it feels during the day.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0072.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE CITY AT NIGHT LOOKS COMPLETELY DIFFERENT FROM HOW IT FEELS DURING THE DAY",
+ "raw": {
+ "normalized_reference": "the city at night looks completely different from how it feels during the day.",
+ "normalized_hypothesis": "the city at night looks completely different from how it feels during the day",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the city at night looks completely different from how it feels during the day",
+ "normalized_hypothesis": "the city at night looks completely different from how it feels during the day",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0073",
+ "category": "emotional",
+ "text": "It's the kind of happy that makes you nervous because you know it can't last.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0073.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT'S THE KIND OF HAPPY THAT MAKES YOU NERVOUS BECAUSE YOU KNOW IT CAN'T LAST",
+ "raw": {
+ "normalized_reference": "it's the kind of happy that makes you nervous because you know it can't last.",
+ "normalized_hypothesis": "it's the kind of happy that makes you nervous because you know it can't last",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it is the kind of happy that makes you nervous because you know it can not last",
+ "normalized_hypothesis": "it is the kind of happy that makes you nervous because you know it can not last",
+ "errors": 0,
+ "words": 17,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0074",
+ "category": "homographs",
+ "text": "The dove dove beneath the branch while the wind began to wind the loose cable.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0074.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE DUB DOVE BENEATH THE BRANCH WHILE THE WIND BEGAN TO WIND THE LOOSE CABLE",
+ "raw": {
+ "normalized_reference": "the dove dove beneath the branch while the wind began to wind the loose cable.",
+ "normalized_hypothesis": "the dub dove beneath the branch while the wind began to wind the loose cable",
+ "errors": 1,
+ "words": 15,
+ "wer": 0.06666666666666667
+ },
+ "semantic": {
+ "normalized_reference": "the dove dove beneath the branch while the wind began to wind the loose cable",
+ "normalized_hypothesis": "the dub dove beneath the branch while the wind began to wind the loose cable",
+ "errors": 1,
+ "words": 15,
+ "wer": 0.06666666666666667
+ }
+ },
+ {
+ "id": "release500_0075",
+ "category": "long",
+ "text": "I've thought a lot about what I would do differently if I could go back, and the honest answer is probably not as much as I'd like to think.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0075.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I'VE THOUGHT A LOT ABOUT WHAT I WOULD DO DIFFERENTLY IF I COULD GO BACK AND THE HONEST ANSWER IS PROBABLY NOT AS MUCH AS I'D LIKE TO THINK",
+ "raw": {
+ "normalized_reference": "i've thought a lot about what i would do differently if i could go back, and the honest answer is probably not as much as i'd like to think.",
+ "normalized_hypothesis": "i've thought a lot about what i would do differently if i could go back and the honest answer is probably not as much as i'd like to think",
+ "errors": 0,
+ "words": 29,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i have thought a lot about what i would do differently if i could go back and the honest answer is probably not as much as i would like to think",
+ "normalized_hypothesis": "i have thought a lot about what i would do differently if i could go back and the honest answer is probably not as much as i would like to think",
+ "errors": 0,
+ "words": 31,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0076",
+ "category": "long_range",
+ "text": "Although the weather changed without warning, the musicians finished packing before anyone opened the enormous glass door.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0076.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "ALTHOUGH THE WEATHER CHANGED WITHOUT WARNING THE MUSICIANS FINISHED PACKING BEFORE ANYONE OPENED THE ENORMOUS GLASS DOOR",
+ "raw": {
+ "normalized_reference": "although the weather changed without warning, the musicians finished packing before anyone opened the enormous glass door.",
+ "normalized_hypothesis": "although the weather changed without warning the musicians finished packing before anyone opened the enormous glass door",
+ "errors": 0,
+ "words": 17,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "although the weather changed without warning the musicians finished packing before anyone opened the enormous glass door",
+ "normalized_hypothesis": "although the weather changed without warning the musicians finished packing before anyone opened the enormous glass door",
+ "errors": 0,
+ "words": 17,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0077",
+ "category": "names_places",
+ "text": "Saskatchewan looked unusually bright beneath the pale February sunrise.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0077.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SUSCACHEUIN LOOKED UNUSUALLY BRIGHT BENEATH THE PALE FEBRUARY SUNRISE",
+ "raw": {
+ "normalized_reference": "saskatchewan looked unusually bright beneath the pale february sunrise.",
+ "normalized_hypothesis": "suscacheuin looked unusually bright beneath the pale february sunrise",
+ "errors": 1,
+ "words": 9,
+ "wer": 0.1111111111111111
+ },
+ "semantic": {
+ "normalized_reference": "saskatchewan looked unusually bright beneath the pale february sunrise",
+ "normalized_hypothesis": "suscacheuin looked unusually bright beneath the pale february sunrise",
+ "errors": 1,
+ "words": 9,
+ "wer": 0.1111111111111111
+ }
+ },
+ {
+ "id": "release500_0078",
+ "category": "narrative",
+ "text": "She handled it with a grace I genuinely did not expect.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0078.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SHE HANDLED IT WITH A GRACE I GENUINELY DID NOT EXPECT",
+ "raw": {
+ "normalized_reference": "she handled it with a grace i genuinely did not expect.",
+ "normalized_hypothesis": "she handled it with a grace i genuinely did not expect",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "she handled it with a grace i genuinely did not expect",
+ "normalized_hypothesis": "she handled it with a grace i genuinely did not expect",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0079",
+ "category": "numbers",
+ "text": "Apartment four hundred seven is across from room twenty-nine on the third floor.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0079.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "PARTMENT FOUR HUNDRED SEVEN IS ACROSS FROM ROOM TWENTY NINE ON THE THIRD FLOOR",
+ "raw": {
+ "normalized_reference": "apartment four hundred seven is across from room twenty-nine on the third floor.",
+ "normalized_hypothesis": "partment four hundred seven is across from room twenty nine on the third floor",
+ "errors": 1,
+ "words": 14,
+ "wer": 0.07142857142857142
+ },
+ "semantic": {
+ "normalized_reference": "apartment 407 is across from room 29 on the 3rd floor",
+ "normalized_hypothesis": "partment 407 is across from room 29 on the 3rd floor",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ }
+ },
+ {
+ "id": "release500_0080",
+ "category": "punctuation",
+ "text": "\"Not yet,\" Rowan whispered. \"Give me another minute... please.\"",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0080.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "NOT YET ROWAN WHISPERED GIVE ME ANOTHER MINUTE PLEASE",
+ "raw": {
+ "normalized_reference": "\"not yet,\" rowan whispered. \"give me another minute... please.\"",
+ "normalized_hypothesis": "not yet rowan whispered give me another minute please",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "not yet rowan whispered give me another minute please",
+ "normalized_hypothesis": "not yet rowan whispered give me another minute please",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0081",
+ "category": "question",
+ "text": "What are we actually trying to accomplish here?",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0081.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WHAT ARE WE ACTUALLY TRYING TO ACCOMPLISH HERE",
+ "raw": {
+ "normalized_reference": "what are we actually trying to accomplish here?",
+ "normalized_hypothesis": "what are we actually trying to accomplish here",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "what are we actually trying to accomplish here",
+ "normalized_hypothesis": "what are we actually trying to accomplish here",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0082",
+ "category": "rare_words",
+ "text": "The physician recommended a noninvasive procedure for the recurring neurological symptoms.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0082.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE PHYSICIAN RECOMMENDED A NON EVASIVE PROCEDURE FOR THE RECURRING NEUROLOGICAL SYMPTOMS",
+ "raw": {
+ "normalized_reference": "the physician recommended a noninvasive procedure for the recurring neurological symptoms.",
+ "normalized_hypothesis": "the physician recommended a non evasive procedure for the recurring neurological symptoms",
+ "errors": 2,
+ "words": 11,
+ "wer": 0.18181818181818182
+ },
+ "semantic": {
+ "normalized_reference": "the physician recommended a noninvasive procedure for the recurring neurological symptoms",
+ "normalized_hypothesis": "the physician recommended a non evasive procedure for the recurring neurological symptoms",
+ "errors": 2,
+ "words": 11,
+ "wer": 0.18181818181818182
+ }
+ },
+ {
+ "id": "release500_0083",
+ "category": "short",
+ "text": "Are you serious right now?",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0083.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "ARE YOU SERIOUS RIGHT NOW",
+ "raw": {
+ "normalized_reference": "are you serious right now?",
+ "normalized_hypothesis": "are you serious right now",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "are you serious right now",
+ "normalized_hypothesis": "are you serious right now",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0084",
+ "category": "technical",
+ "text": "The performance improvement is real, but it comes with tradeoffs.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0084.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE PERFORMANCE IMPROVEMENT IS REAL BUT IT COMES WITH TRADOS",
+ "raw": {
+ "normalized_reference": "the performance improvement is real, but it comes with tradeoffs.",
+ "normalized_hypothesis": "the performance improvement is real but it comes with trados",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ },
+ "semantic": {
+ "normalized_reference": "the performance improvement is real but it comes with tradeoffs",
+ "normalized_hypothesis": "the performance improvement is real but it comes with trados",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ }
+ },
+ {
+ "id": "release500_0085",
+ "category": "conversational",
+ "text": "I don't know. It's just been one of those weeks, you know?",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0085.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I DON'T KNOW IT'S JUST BEEN ONE OF THOSE WEEKS YOU KNOW",
+ "raw": {
+ "normalized_reference": "i don't know. it's just been one of those weeks, you know?",
+ "normalized_hypothesis": "i don't know it's just been one of those weeks you know",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i do not know it is just been one of those weeks you know",
+ "normalized_hypothesis": "i do not know it is just been one of those weeks you know",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0086",
+ "category": "descriptive",
+ "text": "The moment you stop expecting perfection, things get a lot more interesting.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0086.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE MOMENT YOU STOP EXPECTING PERFECTION THINGS GET A LOT MORE INTERESTING",
+ "raw": {
+ "normalized_reference": "the moment you stop expecting perfection, things get a lot more interesting.",
+ "normalized_hypothesis": "the moment you stop expecting perfection things get a lot more interesting",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the moment you stop expecting perfection things get a lot more interesting",
+ "normalized_hypothesis": "the moment you stop expecting perfection things get a lot more interesting",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0087",
+ "category": "emotional",
+ "text": "It's the kind of thing that stays with you even years later.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0087.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT'S THE KIND OF THING THAT STAYS WITH YOU EVEN YEARS LATER",
+ "raw": {
+ "normalized_reference": "it's the kind of thing that stays with you even years later.",
+ "normalized_hypothesis": "it's the kind of thing that stays with you even years later",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it is the kind of thing that stays with you even years later",
+ "normalized_hypothesis": "it is the kind of thing that stays with you even years later",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0088",
+ "category": "long",
+ "text": "She told me once that the most important skill she'd developed was learning to disagree with someone without making them feel they were wrong.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0088.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SHE TOLD ME ONCE THAT THE MOST IMPORTANT SKILL SHE DEVELOPED WAS LEARNING TO DISAGREE WITH SOMEONE WITHOUT MAKING THEM FEEL THEY WERE WRONG",
+ "raw": {
+ "normalized_reference": "she told me once that the most important skill she'd developed was learning to disagree with someone without making them feel they were wrong.",
+ "normalized_hypothesis": "she told me once that the most important skill she developed was learning to disagree with someone without making them feel they were wrong",
+ "errors": 1,
+ "words": 24,
+ "wer": 0.041666666666666664
+ },
+ "semantic": {
+ "normalized_reference": "she told me once that the most important skill she would developed was learning to disagree with someone without making them feel they were wrong",
+ "normalized_hypothesis": "she told me once that the most important skill she developed was learning to disagree with someone without making them feel they were wrong",
+ "errors": 1,
+ "words": 25,
+ "wer": 0.04
+ }
+ },
+ {
+ "id": "release500_0089",
+ "category": "narrative",
+ "text": "He stood at the edge of the cliff and wondered, not for the first time, how he'd gotten here.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0089.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "KEY STOOD AT THE EDGE OF THE CLIFF AND WONDERED NOT FOR THE FIRST TIME HOW HE'D GOTTEN HERE",
+ "raw": {
+ "normalized_reference": "he stood at the edge of the cliff and wondered, not for the first time, how he'd gotten here.",
+ "normalized_hypothesis": "key stood at the edge of the cliff and wondered not for the first time how he'd gotten here",
+ "errors": 1,
+ "words": 19,
+ "wer": 0.05263157894736842
+ },
+ "semantic": {
+ "normalized_reference": "he stood at the edge of the cliff and wondered not for the 1st time how he would gotten here",
+ "normalized_hypothesis": "key stood at the edge of the cliff and wondered not for the 1st time how he would gotten here",
+ "errors": 1,
+ "words": 20,
+ "wer": 0.05
+ }
+ },
+ {
+ "id": "release500_0090",
+ "category": "question",
+ "text": "Why does asking for help feel so hard when you actually need it?",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0090.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WHY DOES ASKING FOR A HELP FEEL SO HARD WHEN YOU ACTUALLY NEED IT",
+ "raw": {
+ "normalized_reference": "why does asking for help feel so hard when you actually need it?",
+ "normalized_hypothesis": "why does asking for a help feel so hard when you actually need it",
+ "errors": 1,
+ "words": 13,
+ "wer": 0.07692307692307693
+ },
+ "semantic": {
+ "normalized_reference": "why does asking for help feel so hard when you actually need it",
+ "normalized_hypothesis": "why does asking for a help feel so hard when you actually need it",
+ "errors": 1,
+ "words": 13,
+ "wer": 0.07692307692307693
+ }
+ },
+ {
+ "id": "release500_0091",
+ "category": "short",
+ "text": "That's fair.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0091.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THAT'S FAIR",
+ "raw": {
+ "normalized_reference": "that's fair.",
+ "normalized_hypothesis": "that's fair",
+ "errors": 0,
+ "words": 2,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "that is fair",
+ "normalized_hypothesis": "that is fair",
+ "errors": 0,
+ "words": 3,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0092",
+ "category": "technical",
+ "text": "If you run into that error, the fix is usually just restarting the service.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0092.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IF YOU RUN INTO THAT ERROR THE FIX IS USUALLY JUST RESTARTING THE SERVICE",
+ "raw": {
+ "normalized_reference": "if you run into that error, the fix is usually just restarting the service.",
+ "normalized_hypothesis": "if you run into that error the fix is usually just restarting the service",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "if you run into that error the fix is usually just restarting the service",
+ "normalized_hypothesis": "if you run into that error the fix is usually just restarting the service",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0093",
+ "category": "conversational",
+ "text": "It was a disaster in the best possible way.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0093.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT WAS A DISASTER IN THE BEST POSSIBLE WAY",
+ "raw": {
+ "normalized_reference": "it was a disaster in the best possible way.",
+ "normalized_hypothesis": "it was a disaster in the best possible way",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it was a disaster in the best possible way",
+ "normalized_hypothesis": "it was a disaster in the best possible way",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0094",
+ "category": "descriptive",
+ "text": "It had the look of a place that used to mean something to someone.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0094.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT HAD THE LOOK OF A PLACE THAT USED TO MEAN SOMETHING TO SOME ONE",
+ "raw": {
+ "normalized_reference": "it had the look of a place that used to mean something to someone.",
+ "normalized_hypothesis": "it had the look of a place that used to mean something to some one",
+ "errors": 2,
+ "words": 14,
+ "wer": 0.14285714285714285
+ },
+ "semantic": {
+ "normalized_reference": "it had the look of a place that used to mean something to someone",
+ "normalized_hypothesis": "it had the look of a place that used to mean something to some one",
+ "errors": 2,
+ "words": 14,
+ "wer": 0.14285714285714285
+ }
+ },
+ {
+ "id": "release500_0095",
+ "category": "emotional",
+ "text": "There's this feeling you get when something finally works after weeks of failure.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0095.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THERE'S THIS FEELING YOU GET WHEN SOMETHING FINALLY WORKS AFTER WEEKS OF FAILURE",
+ "raw": {
+ "normalized_reference": "there's this feeling you get when something finally works after weeks of failure.",
+ "normalized_hypothesis": "there's this feeling you get when something finally works after weeks of failure",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "there is this feeling you get when something finally works after weeks of failure",
+ "normalized_hypothesis": "there is this feeling you get when something finally works after weeks of failure",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0096",
+ "category": "long",
+ "text": "He had a theory that most arguments aren't actually about what they seem to be about, and the longer I live the more I think he was onto something.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0096.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE HAD A THEORY THAT MOST ARDUMENTS AREN'T ACTUALLY ABOUT WHAT THEY SEEM TO BE ABOUT AND THE LONGER I LIVE THE MORE I THINK HE WAS ON TO SOMETHING",
+ "raw": {
+ "normalized_reference": "he had a theory that most arguments aren't actually about what they seem to be about, and the longer i live the more i think he was onto something.",
+ "normalized_hypothesis": "he had a theory that most arduments aren't actually about what they seem to be about and the longer i live the more i think he was on to something",
+ "errors": 3,
+ "words": 29,
+ "wer": 0.10344827586206896
+ },
+ "semantic": {
+ "normalized_reference": "he had a theory that most arguments are not actually about what they seem to be about and the longer i live the more i think he was onto something",
+ "normalized_hypothesis": "he had a theory that most arduments are not actually about what they seem to be about and the longer i live the more i think he was on to something",
+ "errors": 3,
+ "words": 30,
+ "wer": 0.1
+ }
+ },
+ {
+ "id": "release500_0097",
+ "category": "narrative",
+ "text": "The thing nobody tells you about starting over is how ordinary most of it feels.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0097.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE THING NOBODY TELLS YOU ABOUT STARTING OVER IS HOW ORDINARY MOST OF IT FEELS",
+ "raw": {
+ "normalized_reference": "the thing nobody tells you about starting over is how ordinary most of it feels.",
+ "normalized_hypothesis": "the thing nobody tells you about starting over is how ordinary most of it feels",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the thing nobody tells you about starting over is how ordinary most of it feels",
+ "normalized_hypothesis": "the thing nobody tells you about starting over is how ordinary most of it feels",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0098",
+ "category": "question",
+ "text": "What exactly are we waiting for at this point?",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0098.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WHAT EXACTLY ARE WE WAITING FOR AT THIS POINT",
+ "raw": {
+ "normalized_reference": "what exactly are we waiting for at this point?",
+ "normalized_hypothesis": "what exactly are we waiting for at this point",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "what exactly are we waiting for at this point",
+ "normalized_hypothesis": "what exactly are we waiting for at this point",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0099",
+ "category": "short",
+ "text": "I stand by what I said.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0099.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I STAND BY WHAT I SAID",
+ "raw": {
+ "normalized_reference": "i stand by what i said.",
+ "normalized_hypothesis": "i stand by what i said",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i stand by what i said",
+ "normalized_hypothesis": "i stand by what i said",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0100",
+ "category": "technical",
+ "text": "Double-check that value before you push anything to production.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0100.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "DOUBLE CHECK THAT VALUE BEFORE YOU PUSH ANYTHING TO PRODUCTION",
+ "raw": {
+ "normalized_reference": "double-check that value before you push anything to production.",
+ "normalized_hypothesis": "double check that value before you push anything to production",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "double check that value before you push anything to production",
+ "normalized_hypothesis": "double check that value before you push anything to production",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0101",
+ "category": "conversational",
+ "text": "I understand your point; I just do not agree with the conclusion.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0101.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I UNDERSTAND YOUR POINT I JUST DO NOT AGREE WITH THE CONCLUSION",
+ "raw": {
+ "normalized_reference": "i understand your point; i just do not agree with the conclusion.",
+ "normalized_hypothesis": "i understand your point i just do not agree with the conclusion",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i understand your point i just do not agree with the conclusion",
+ "normalized_hypothesis": "i understand your point i just do not agree with the conclusion",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0102",
+ "category": "descriptive",
+ "text": "The sky that day was a strange shade of orange I'd never seen before.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0102.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE SKY THAT DAY WAS A STRANGE SHADE OF ORANGE I'D NEVER SEEN BEFORE",
+ "raw": {
+ "normalized_reference": "the sky that day was a strange shade of orange i'd never seen before.",
+ "normalized_hypothesis": "the sky that day was a strange shade of orange i'd never seen before",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the sky that day was a strange shade of orange i would never seen before",
+ "normalized_hypothesis": "the sky that day was a strange shade of orange i would never seen before",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0103",
+ "category": "emotional",
+ "text": "It hurt more than I expected, and I thought I was prepared.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0103.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT HURT MORE THAN I EXPECTED AND I THOUGHT I WAS PREPARED",
+ "raw": {
+ "normalized_reference": "it hurt more than i expected, and i thought i was prepared.",
+ "normalized_hypothesis": "it hurt more than i expected and i thought i was prepared",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it hurt more than i expected and i thought i was prepared",
+ "normalized_hypothesis": "it hurt more than i expected and i thought i was prepared",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0104",
+ "category": "long",
+ "text": "The hardest part wasn't making the decision, it was sitting with the uncertainty of not knowing whether it was the right one for months afterward.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0104.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE HARDEST PART WASN'T MAKING THE DECISION IT WAS SITTING WITH THE UNCERTAINTY OF NOT KNOWING WHETHER IT WAS THE RIGHT ONE FOR MONTHS AFTERWARD",
+ "raw": {
+ "normalized_reference": "the hardest part wasn't making the decision, it was sitting with the uncertainty of not knowing whether it was the right one for months afterward.",
+ "normalized_hypothesis": "the hardest part wasn't making the decision it was sitting with the uncertainty of not knowing whether it was the right one for months afterward",
+ "errors": 0,
+ "words": 25,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the hardest part was not making the decision it was sitting with the uncertainty of not knowing whether it was the right one for months afterward",
+ "normalized_hypothesis": "the hardest part was not making the decision it was sitting with the uncertainty of not knowing whether it was the right one for months afterward",
+ "errors": 0,
+ "words": 26,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0105",
+ "category": "narrative",
+ "text": "There was this one day last summer when everything just clicked into place.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0105.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THERE WAS THIS ONE DAY LAST SUMMER WHEN EVERYTHING JUST CLICKED INTO PLACE",
+ "raw": {
+ "normalized_reference": "there was this one day last summer when everything just clicked into place.",
+ "normalized_hypothesis": "there was this one day last summer when everything just clicked into place",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "there was this one day last summer when everything just clicked into place",
+ "normalized_hypothesis": "there was this one day last summer when everything just clicked into place",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0106",
+ "category": "question",
+ "text": "Have you ever thought about why we keep doing things the same way when they stop working?",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0106.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HAVE YOU EVER THOUGHT ABOUT WHY WE KEEP DOING THINGS THE SAME WAY WHEN THEY STOP WORKING",
+ "raw": {
+ "normalized_reference": "have you ever thought about why we keep doing things the same way when they stop working?",
+ "normalized_hypothesis": "have you ever thought about why we keep doing things the same way when they stop working",
+ "errors": 0,
+ "words": 17,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "have you ever thought about why we keep doing things the same way when they stop working",
+ "normalized_hypothesis": "have you ever thought about why we keep doing things the same way when they stop working",
+ "errors": 0,
+ "words": 17,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0107",
+ "category": "short",
+ "text": "I can't believe you just said that.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0107.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I CAN'T BELIEVE YOU JUST SAID THAT",
+ "raw": {
+ "normalized_reference": "i can't believe you just said that.",
+ "normalized_hypothesis": "i can't believe you just said that",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i can not believe you just said that",
+ "normalized_hypothesis": "i can not believe you just said that",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0108",
+ "category": "technical",
+ "text": "You'll want to isolate that variable before drawing any conclusions.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0108.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "YOU'LL WANT TO ISOLATE THAT VARIABLE BEFORE DRAWING ANY CONCLUSIONS",
+ "raw": {
+ "normalized_reference": "you'll want to isolate that variable before drawing any conclusions.",
+ "normalized_hypothesis": "you'll want to isolate that variable before drawing any conclusions",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "you will want to isolate that variable before drawing any conclusions",
+ "normalized_hypothesis": "you will want to isolate that variable before drawing any conclusions",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0109",
+ "category": "conversational",
+ "text": "The meeting that could have been an email lasted two hours.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0109.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE MEETING THAT COULD HAVE BEEN IN EMELE LASTED TWO HOURS",
+ "raw": {
+ "normalized_reference": "the meeting that could have been an email lasted two hours.",
+ "normalized_hypothesis": "the meeting that could have been in emele lasted two hours",
+ "errors": 2,
+ "words": 11,
+ "wer": 0.18181818181818182
+ },
+ "semantic": {
+ "normalized_reference": "the meeting that could have been an email lasted 2 hours",
+ "normalized_hypothesis": "the meeting that could have been in emele lasted 2 hours",
+ "errors": 2,
+ "words": 11,
+ "wer": 0.18181818181818182
+ }
+ },
+ {
+ "id": "release500_0110",
+ "category": "descriptive",
+ "text": "There's a specific kind of tired that sleep doesn't fix.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0110.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THERE'S A SPECIFIC KIND OF TIRE THAT SLEEP DOESN'T FIX",
+ "raw": {
+ "normalized_reference": "there's a specific kind of tired that sleep doesn't fix.",
+ "normalized_hypothesis": "there's a specific kind of tire that sleep doesn't fix",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ },
+ "semantic": {
+ "normalized_reference": "there is a specific kind of tired that sleep does not fix",
+ "normalized_hypothesis": "there is a specific kind of tire that sleep does not fix",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ }
+ },
+ {
+ "id": "release500_0111",
+ "category": "emotional",
+ "text": "Something about that moment broke something open in me that I hadn't realized was closed.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0111.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SOMETHING ABOUT THAT MOMENT BROKE SOMETHING OPEN IN ME THAT I HADN'T REALIZED WAS CLOSED",
+ "raw": {
+ "normalized_reference": "something about that moment broke something open in me that i hadn't realized was closed.",
+ "normalized_hypothesis": "something about that moment broke something open in me that i hadn't realized was closed",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "something about that moment broke something open in me that i had not realized was closed",
+ "normalized_hypothesis": "something about that moment broke something open in me that i had not realized was closed",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0112",
+ "category": "long",
+ "text": "The thing about making something good is that you don't usually know it's good while you're making it. You just know you're not done yet.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0112.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE THING ABOUT MAKING SOMETHING GOOD IS THAT YOU DON'T USUALLY KNOW IT'S GOOD WHILE YOU'RE MAKING IT YOU JUST KNOW YOU'RE NOT DONE YET",
+ "raw": {
+ "normalized_reference": "the thing about making something good is that you don't usually know it's good while you're making it. you just know you're not done yet.",
+ "normalized_hypothesis": "the thing about making something good is that you don't usually know it's good while you're making it you just know you're not done yet",
+ "errors": 0,
+ "words": 25,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the thing about making something good is that you do not usually know it is good while you are making it you just know you are not done yet",
+ "normalized_hypothesis": "the thing about making something good is that you do not usually know it is good while you are making it you just know you are not done yet",
+ "errors": 0,
+ "words": 29,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0113",
+ "category": "narrative",
+ "text": "She opened the door slowly, as if she expected to find something she wasn't ready for.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0113.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SHE OPENED THE DOOR SLOWLY AS IF SHE EXPECTED TO FIND SOMETHING SHE WASN'T READY FOR",
+ "raw": {
+ "normalized_reference": "she opened the door slowly, as if she expected to find something she wasn't ready for.",
+ "normalized_hypothesis": "she opened the door slowly as if she expected to find something she wasn't ready for",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "she opened the door slowly as if she expected to find something she was not ready for",
+ "normalized_hypothesis": "she opened the door slowly as if she expected to find something she was not ready for",
+ "errors": 0,
+ "words": 17,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0114",
+ "category": "question",
+ "text": "What would you do differently if you could go back?",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0114.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WHAT WOULD YOU DO DIFFERENTLY IF YOU COULD GO BACK",
+ "raw": {
+ "normalized_reference": "what would you do differently if you could go back?",
+ "normalized_hypothesis": "what would you do differently if you could go back",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "what would you do differently if you could go back",
+ "normalized_hypothesis": "what would you do differently if you could go back",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0115",
+ "category": "short",
+ "text": "Okay. Okay okay okay.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0115.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "BOQAD I OK O K OK",
+ "raw": {
+ "normalized_reference": "okay. okay okay okay.",
+ "normalized_hypothesis": "boqad i ok o k ok",
+ "errors": 6,
+ "words": 4,
+ "wer": 1.5
+ },
+ "semantic": {
+ "normalized_reference": "okay okay okay okay",
+ "normalized_hypothesis": "boqad i ok 0 k ok",
+ "errors": 6,
+ "words": 4,
+ "wer": 1.5
+ }
+ },
+ {
+ "id": "release500_0116",
+ "category": "technical",
+ "text": "The latency goes up significantly at scale, which is expected.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0116.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE LATENCY GOES UP SIGNIFICANTLY AT SCALE WHICH IS EXPECTED",
+ "raw": {
+ "normalized_reference": "the latency goes up significantly at scale, which is expected.",
+ "normalized_hypothesis": "the latency goes up significantly at scale which is expected",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the latency goes up significantly at scale which is expected",
+ "normalized_hypothesis": "the latency goes up significantly at scale which is expected",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0117",
+ "category": "conversational",
+ "text": "We ended up staying until two in the morning talking about nothing.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0117.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WE ENDED UP STAYING UNTIL TWO IN THE MORNING TALKING ABOUT NOTHING",
+ "raw": {
+ "normalized_reference": "we ended up staying until two in the morning talking about nothing.",
+ "normalized_hypothesis": "we ended up staying until two in the morning talking about nothing",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "we ended up staying until 2 in the morning talking about nothing",
+ "normalized_hypothesis": "we ended up staying until 2 in the morning talking about nothing",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0118",
+ "category": "descriptive",
+ "text": "The rain started slowly, and then all at once, the way it usually does.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0118.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE RAIN STARTED SLOWLY AND THEN ALL AT ONCE THE WAY IT USUALLY DOES",
+ "raw": {
+ "normalized_reference": "the rain started slowly, and then all at once, the way it usually does.",
+ "normalized_hypothesis": "the rain started slowly and then all at once the way it usually does",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the rain started slowly and then all at once the way it usually does",
+ "normalized_hypothesis": "the rain started slowly and then all at once the way it usually does",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0119",
+ "category": "emotional",
+ "text": "I honestly did not think I was going to make it through that year.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0119.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I HONESTLY DID NOT THINK I WAS GOING TO MAKE IT THROUGH THAT YEAR",
+ "raw": {
+ "normalized_reference": "i honestly did not think i was going to make it through that year.",
+ "normalized_hypothesis": "i honestly did not think i was going to make it through that year",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i honestly did not think i was going to make it through that year",
+ "normalized_hypothesis": "i honestly did not think i was going to make it through that year",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0120",
+ "category": "long",
+ "text": "There's an assumption that people who seem confident all the time actually feel confident, and in my experience that's almost never true.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0120.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THERE'S AN ASSUMPTION THAT PEOPLE WHO SEEM CONFIDENT ALL THE TIME ACTUALLY FEEL CONFIDENT AND IN MY EXPERIENCE THAT'S ALMOST NEVER TRUE",
+ "raw": {
+ "normalized_reference": "there's an assumption that people who seem confident all the time actually feel confident, and in my experience that's almost never true.",
+ "normalized_hypothesis": "there's an assumption that people who seem confident all the time actually feel confident and in my experience that's almost never true",
+ "errors": 0,
+ "words": 22,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "there is an assumption that people who seem confident all the time actually feel confident and in my experience that is almost never true",
+ "normalized_hypothesis": "there is an assumption that people who seem confident all the time actually feel confident and in my experience that is almost never true",
+ "errors": 0,
+ "words": 24,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0121",
+ "category": "narrative",
+ "text": "The moment I opened the door I knew something was different, even if I couldn't say what.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0121.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE MOMENT I OPENED THE DOOR I KNEW SOMETHING WAS DIFFERENT EVEN IF I COULDN'T SAY WHAT",
+ "raw": {
+ "normalized_reference": "the moment i opened the door i knew something was different, even if i couldn't say what.",
+ "normalized_hypothesis": "the moment i opened the door i knew something was different even if i couldn't say what",
+ "errors": 0,
+ "words": 17,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the moment i opened the door i knew something was different even if i could not say what",
+ "normalized_hypothesis": "the moment i opened the door i knew something was different even if i could not say what",
+ "errors": 0,
+ "words": 18,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0122",
+ "category": "question",
+ "text": "How do you explain something that doesn't make sense even to you?",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0122.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "COW DO YOU EXPLAIN SOMETHING THAT DOESN'T MAKE SENSE EVEN TO YOU",
+ "raw": {
+ "normalized_reference": "how do you explain something that doesn't make sense even to you?",
+ "normalized_hypothesis": "cow do you explain something that doesn't make sense even to you",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "how do you explain something that does not make sense even to you",
+ "normalized_hypothesis": "cow do you explain something that does not make sense even to you",
+ "errors": 1,
+ "words": 13,
+ "wer": 0.07692307692307693
+ }
+ },
+ {
+ "id": "release500_0123",
+ "category": "short",
+ "text": "Hold on a second.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0123.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HOLD ON A SECOND",
+ "raw": {
+ "normalized_reference": "hold on a second.",
+ "normalized_hypothesis": "hold on a second",
+ "errors": 0,
+ "words": 4,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "hold on a 2nd",
+ "normalized_hypothesis": "hold on a 2nd",
+ "errors": 0,
+ "words": 4,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0124",
+ "category": "technical",
+ "text": "A deterministic split prevents duplicate normalized text from leaking into evaluation.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0124.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "A DETERMINISTIC SPLIT PREVENTS DUPLICATE NORMALIZED TEXT FROM LEAKING INTO EVALUATION",
+ "raw": {
+ "normalized_reference": "a deterministic split prevents duplicate normalized text from leaking into evaluation.",
+ "normalized_hypothesis": "a deterministic split prevents duplicate normalized text from leaking into evaluation",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "a deterministic split prevents duplicate normalized text from leaking into evaluation",
+ "normalized_hypothesis": "a deterministic split prevents duplicate normalized text from leaking into evaluation",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0125",
+ "category": "conversational",
+ "text": "Honestly, that was much better than I expected, but the ending still felt rushed.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0125.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HONESTLY THAT WAS MUCH BETTER THAN I EXPECTED BUT THE ENDING STILL FELT RUSHED",
+ "raw": {
+ "normalized_reference": "honestly, that was much better than i expected, but the ending still felt rushed.",
+ "normalized_hypothesis": "honestly that was much better than i expected but the ending still felt rushed",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "honestly that was much better than i expected but the ending still felt rushed",
+ "normalized_hypothesis": "honestly that was much better than i expected but the ending still felt rushed",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0126",
+ "category": "descriptive",
+ "text": "She had a habit of tilting her head slightly when thinking hard about something.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0126.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SHE HAD A HABIT OF TILTING HER HEAD SLIGHTLY WHEN THINKING HARD ABOUT SOMETHING",
+ "raw": {
+ "normalized_reference": "she had a habit of tilting her head slightly when thinking hard about something.",
+ "normalized_hypothesis": "she had a habit of tilting her head slightly when thinking hard about something",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "she had a habit of tilting her head slightly when thinking hard about something",
+ "normalized_hypothesis": "she had a habit of tilting her head slightly when thinking hard about something",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0127",
+ "category": "emotional",
+ "text": "I cannot stress enough how much I did not want to be right about this.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0127.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I CANNOT STRESS ENOUGH HOW MUCH I DID NOT WANT TO BE RIGHT ABOUT THIS",
+ "raw": {
+ "normalized_reference": "i cannot stress enough how much i did not want to be right about this.",
+ "normalized_hypothesis": "i cannot stress enough how much i did not want to be right about this",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i cannot stress enough how much i did not want to be right about this",
+ "normalized_hypothesis": "i cannot stress enough how much i did not want to be right about this",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0128",
+ "category": "long",
+ "text": "I used to think the goal was to get to a point where nothing was hard anymore, and now I think that point doesn't exist and I'm not sure I'd want it to.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0128.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I USED TO THINK THE GOAL WAS TO GET TO A POINT WHERE NOTHING WAS HARD ANIMORE AND NOW I THINK THAT POINT DOESN'T EXIST AND I'M NOT SURE I'D WANT IT TO",
+ "raw": {
+ "normalized_reference": "i used to think the goal was to get to a point where nothing was hard anymore, and now i think that point doesn't exist and i'm not sure i'd want it to.",
+ "normalized_hypothesis": "i used to think the goal was to get to a point where nothing was hard animore and now i think that point doesn't exist and i'm not sure i'd want it to",
+ "errors": 1,
+ "words": 33,
+ "wer": 0.030303030303030304
+ },
+ "semantic": {
+ "normalized_reference": "i used to think the goal was to get to a point where nothing was hard anymore and now i think that point does not exist and i am not sure i would want it to",
+ "normalized_hypothesis": "i used to think the goal was to get to a point where nothing was hard animore and now i think that point does not exist and i am not sure i would want it to",
+ "errors": 1,
+ "words": 36,
+ "wer": 0.027777777777777776
+ }
+ },
+ {
+ "id": "release500_0129",
+ "category": "narrative",
+ "text": "So this is the part of the story where everything goes completely sideways.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0129.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SO THIS IS THE PART OF THE STORY WHERE EVERYTHING GOES COMPLETELY SIDEWAYS",
+ "raw": {
+ "normalized_reference": "so this is the part of the story where everything goes completely sideways.",
+ "normalized_hypothesis": "so this is the part of the story where everything goes completely sideways",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "so this is the part of the story where everything goes completely sideways",
+ "normalized_hypothesis": "so this is the part of the story where everything goes completely sideways",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0130",
+ "category": "question",
+ "text": "Is there a version of this where it actually works out okay?",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0130.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IS THERE A VERSION OF THIS WHERE IT ACTUALLY WORKS OUT O Q",
+ "raw": {
+ "normalized_reference": "is there a version of this where it actually works out okay?",
+ "normalized_hypothesis": "is there a version of this where it actually works out o q",
+ "errors": 2,
+ "words": 12,
+ "wer": 0.16666666666666666
+ },
+ "semantic": {
+ "normalized_reference": "is there a version of this where it actually works out okay",
+ "normalized_hypothesis": "is there a version of this where it actually works out 0 q",
+ "errors": 2,
+ "words": 12,
+ "wer": 0.16666666666666666
+ }
+ },
+ {
+ "id": "release500_0131",
+ "category": "short",
+ "text": "Oh no. Oh no no no.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0131.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "OH NO BOT OH NO NO NO",
+ "raw": {
+ "normalized_reference": "oh no. oh no no no.",
+ "normalized_hypothesis": "oh no bot oh no no no",
+ "errors": 1,
+ "words": 6,
+ "wer": 0.16666666666666666
+ },
+ "semantic": {
+ "normalized_reference": "0 no 0 no no no",
+ "normalized_hypothesis": "0 no bot 0 no no no",
+ "errors": 1,
+ "words": 6,
+ "wer": 0.16666666666666666
+ }
+ },
+ {
+ "id": "release500_0132",
+ "category": "technical",
+ "text": "Always test on a copy before modifying the original.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0132.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "ALWAYS TEST ON A COPY BEFORE MODIFYING THE ORIGINAL",
+ "raw": {
+ "normalized_reference": "always test on a copy before modifying the original.",
+ "normalized_hypothesis": "always test on a copy before modifying the original",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "always test on a copy before modifying the original",
+ "normalized_hypothesis": "always test on a copy before modifying the original",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0133",
+ "category": "conversational",
+ "text": "I wasn't even supposed to be there. That's the wild part.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0133.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I WASN'T EVEN SUPPOSED TO BE THERE THAT'S THE WILD PART",
+ "raw": {
+ "normalized_reference": "i wasn't even supposed to be there. that's the wild part.",
+ "normalized_hypothesis": "i wasn't even supposed to be there that's the wild part",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i was not even supposed to be there that is the wild part",
+ "normalized_hypothesis": "i was not even supposed to be there that is the wild part",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0134",
+ "category": "descriptive",
+ "text": "It's a small town. Not the kind you leave. The kind you leave and come back to.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0134.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT'S A SMALL TOWN NOT THE KIND YOU LEAVE THE KIND YOU LEAVE AND COME BACK TO",
+ "raw": {
+ "normalized_reference": "it's a small town. not the kind you leave. the kind you leave and come back to.",
+ "normalized_hypothesis": "it's a small town not the kind you leave the kind you leave and come back to",
+ "errors": 0,
+ "words": 17,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it is a small town not the kind you leave the kind you leave and come back to",
+ "normalized_hypothesis": "it is a small town not the kind you leave the kind you leave and come back to",
+ "errors": 0,
+ "words": 18,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0135",
+ "category": "emotional",
+ "text": "She cried. I cried. We all just sort of stood there crying.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0135.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SHE CRIED I CRYE NUT WE ALL JUST SORT OF STOOD THERE CRYING",
+ "raw": {
+ "normalized_reference": "she cried. i cried. we all just sort of stood there crying.",
+ "normalized_hypothesis": "she cried i crye nut we all just sort of stood there crying",
+ "errors": 2,
+ "words": 12,
+ "wer": 0.16666666666666666
+ },
+ "semantic": {
+ "normalized_reference": "she cried i cried we all just sort of stood there crying",
+ "normalized_hypothesis": "she cried i crye nut we all just sort of stood there crying",
+ "errors": 2,
+ "words": 12,
+ "wer": 0.16666666666666666
+ }
+ },
+ {
+ "id": "release500_0136",
+ "category": "long",
+ "text": "There's a version of being busy that's actually just a very elaborate way of avoiding the one thing you know you should be doing.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0136.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THERE'S A VERSION OF BEING BUSY THAT'S ACTUALLY JUST A VERY ELABORATE WAY OF AVOIDING THE ONE THING YOU KNOW YOU SHOULD BE DOING",
+ "raw": {
+ "normalized_reference": "there's a version of being busy that's actually just a very elaborate way of avoiding the one thing you know you should be doing.",
+ "normalized_hypothesis": "there's a version of being busy that's actually just a very elaborate way of avoiding the one thing you know you should be doing",
+ "errors": 0,
+ "words": 24,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "there is a version of being busy that is actually just a very elaborate way of avoiding the one thing you know you should be doing",
+ "normalized_hypothesis": "there is a version of being busy that is actually just a very elaborate way of avoiding the one thing you know you should be doing",
+ "errors": 0,
+ "words": 26,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0137",
+ "category": "narrative",
+ "text": "The letter arrived on a Tuesday, which felt appropriate somehow.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0137.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE LETTER ARRIVED ON A TUESDAY WHICH FELT APPROPRIATE SOMEHOW",
+ "raw": {
+ "normalized_reference": "the letter arrived on a tuesday, which felt appropriate somehow.",
+ "normalized_hypothesis": "the letter arrived on a tuesday which felt appropriate somehow",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the letter arrived on a tuesday which felt appropriate somehow",
+ "normalized_hypothesis": "the letter arrived on a tuesday which felt appropriate somehow",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0138",
+ "category": "question",
+ "text": "What exactly did she say when you told her?",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0138.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WHAT EXACTLY DID SHE SAY WHEN YOU TOLD HER",
+ "raw": {
+ "normalized_reference": "what exactly did she say when you told her?",
+ "normalized_hypothesis": "what exactly did she say when you told her",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "what exactly did she say when you told her",
+ "normalized_hypothesis": "what exactly did she say when you told her",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0139",
+ "category": "short",
+ "text": "No way. Absolutely not.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0139.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "NO WETA ABSOLUTELY NOT",
+ "raw": {
+ "normalized_reference": "no way. absolutely not.",
+ "normalized_hypothesis": "no weta absolutely not",
+ "errors": 1,
+ "words": 4,
+ "wer": 0.25
+ },
+ "semantic": {
+ "normalized_reference": "no way absolutely not",
+ "normalized_hypothesis": "no weta absolutely not",
+ "errors": 1,
+ "words": 4,
+ "wer": 0.25
+ }
+ },
+ {
+ "id": "release500_0140",
+ "category": "technical",
+ "text": "The first thing you want to do is make sure everything is backed up.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0140.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE FIRST THING YOU WANT TO DO IS MAKE SURE EVERYTHING IS BACKED UP",
+ "raw": {
+ "normalized_reference": "the first thing you want to do is make sure everything is backed up.",
+ "normalized_hypothesis": "the first thing you want to do is make sure everything is backed up",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the 1st thing you want to do is make sure everything is backed up",
+ "normalized_hypothesis": "the 1st thing you want to do is make sure everything is backed up",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0141",
+ "category": "conversational",
+ "text": "I have the most insane idea and you need to just hear me out.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0141.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I HAVE THE MOST INSANE IDEA AND YOU NEED TO JUST HEAR ME OUT",
+ "raw": {
+ "normalized_reference": "i have the most insane idea and you need to just hear me out.",
+ "normalized_hypothesis": "i have the most insane idea and you need to just hear me out",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i have the most insane idea and you need to just hear me out",
+ "normalized_hypothesis": "i have the most insane idea and you need to just hear me out",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0142",
+ "category": "descriptive",
+ "text": "There's a coffee shop on the corner that's been there longer than anyone can remember.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0142.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THERE'S A COFFEE SHOP ON THE CORNER THAT'S BEEN THERE LONGER THAN ANYONE CAN REMEMBER",
+ "raw": {
+ "normalized_reference": "there's a coffee shop on the corner that's been there longer than anyone can remember.",
+ "normalized_hypothesis": "there's a coffee shop on the corner that's been there longer than anyone can remember",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "there is a coffee shop on the corner that has been there longer than anyone can remember",
+ "normalized_hypothesis": "there is a coffee shop on the corner that has been there longer than anyone can remember",
+ "errors": 0,
+ "words": 17,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0143",
+ "category": "emotional",
+ "text": "I kept it together until I got to the car, and then I completely fell apart.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0143.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I KEPT IT TOGETHER UNTIL I GOT TO THE CAR AND THEN I COMPLETELY FELL APART",
+ "raw": {
+ "normalized_reference": "i kept it together until i got to the car, and then i completely fell apart.",
+ "normalized_hypothesis": "i kept it together until i got to the car and then i completely fell apart",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i kept it together until i got to the car and then i completely fell apart",
+ "normalized_hypothesis": "i kept it together until i got to the car and then i completely fell apart",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0144",
+ "category": "long",
+ "text": "The older I get, the more I think that most of what we call wisdom is just pattern recognition built from having been wrong enough times to start noticing.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0144.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE OLDER I GET THE MORE I THINK THAT MOST OF WHAT WE CALL WISDOM IS JUST PATERN RECOGNITION BUILT FROM HAVING BEEN RUN ENOUGH TIMES TO START NOTICING",
+ "raw": {
+ "normalized_reference": "the older i get, the more i think that most of what we call wisdom is just pattern recognition built from having been wrong enough times to start noticing.",
+ "normalized_hypothesis": "the older i get the more i think that most of what we call wisdom is just patern recognition built from having been run enough times to start noticing",
+ "errors": 2,
+ "words": 29,
+ "wer": 0.06896551724137931
+ },
+ "semantic": {
+ "normalized_reference": "the older i get the more i think that most of what we call wisdom is just pattern recognition built from having been wrong enough times to start noticing",
+ "normalized_hypothesis": "the older i get the more i think that most of what we call wisdom is just patern recognition built from having been run enough times to start noticing",
+ "errors": 2,
+ "words": 29,
+ "wer": 0.06896551724137931
+ }
+ },
+ {
+ "id": "release500_0145",
+ "category": "narrative",
+ "text": "I remember the exact moment I realized I had made a huge mistake.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0145.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I REMEMBER THE EXACT MOMENT I REALIZED I HAD MADE A HUGE MISTAKE",
+ "raw": {
+ "normalized_reference": "i remember the exact moment i realized i had made a huge mistake.",
+ "normalized_hypothesis": "i remember the exact moment i realized i had made a huge mistake",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i remember the exact moment i realized i had made a huge mistake",
+ "normalized_hypothesis": "i remember the exact moment i realized i had made a huge mistake",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0146",
+ "category": "question",
+ "text": "Do you think it would have gone differently if we'd left earlier?",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0146.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "DO YOU THINK IT WOULD HAVE GOME DIFFERENTLY IF WE'D LEFT EARLIER",
+ "raw": {
+ "normalized_reference": "do you think it would have gone differently if we'd left earlier?",
+ "normalized_hypothesis": "do you think it would have gome differently if we'd left earlier",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "do you think it would have gone differently if we would left earlier",
+ "normalized_hypothesis": "do you think it would have gome differently if we would left earlier",
+ "errors": 1,
+ "words": 13,
+ "wer": 0.07692307692307693
+ }
+ },
+ {
+ "id": "release500_0147",
+ "category": "short",
+ "text": "You have to be kidding me.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0147.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "YOU HAVE TO BE KITTING ME",
+ "raw": {
+ "normalized_reference": "you have to be kidding me.",
+ "normalized_hypothesis": "you have to be kitting me",
+ "errors": 1,
+ "words": 6,
+ "wer": 0.16666666666666666
+ },
+ "semantic": {
+ "normalized_reference": "you have to be kidding me",
+ "normalized_hypothesis": "you have to be kitting me",
+ "errors": 1,
+ "words": 6,
+ "wer": 0.16666666666666666
+ }
+ },
+ {
+ "id": "release500_0148",
+ "category": "technical",
+ "text": "If the checksum fails, restart the worker process without deleting the cached tensors.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0148.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IF THE CHEKSUM FAILS RESTART THE WORKER PROCESS WITHOUT DELEADING THE CASH TENSERS",
+ "raw": {
+ "normalized_reference": "if the checksum fails, restart the worker process without deleting the cached tensors.",
+ "normalized_hypothesis": "if the cheksum fails restart the worker process without deleading the cash tensers",
+ "errors": 4,
+ "words": 13,
+ "wer": 0.3076923076923077
+ },
+ "semantic": {
+ "normalized_reference": "if the checksum fails restart the worker process without deleting the cached tensors",
+ "normalized_hypothesis": "if the cheksum fails restart the worker process without deleading the cash tensers",
+ "errors": 4,
+ "words": 13,
+ "wer": 0.3076923076923077
+ }
+ },
+ {
+ "id": "release500_0149",
+ "category": "conversational",
+ "text": "I need you to understand how close that was to not working.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0149.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I NEED YOU TO UNDERSTAND HOW CLOSE THAT WAS TO NOT WORKING",
+ "raw": {
+ "normalized_reference": "i need you to understand how close that was to not working.",
+ "normalized_hypothesis": "i need you to understand how close that was to not working",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i need you to understand how close that was to not working",
+ "normalized_hypothesis": "i need you to understand how close that was to not working",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0150",
+ "category": "descriptive",
+ "text": "It was the sort of morning where the whole world looks rinsed clean.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0150.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT WAS THE SORT OF MORNING WHERE THE WHOLE WORLD LOOKS RINCED CLEAN",
+ "raw": {
+ "normalized_reference": "it was the sort of morning where the whole world looks rinsed clean.",
+ "normalized_hypothesis": "it was the sort of morning where the whole world looks rinced clean",
+ "errors": 1,
+ "words": 13,
+ "wer": 0.07692307692307693
+ },
+ "semantic": {
+ "normalized_reference": "it was the sort of morning where the whole world looks rinsed clean",
+ "normalized_hypothesis": "it was the sort of morning where the whole world looks rinced clean",
+ "errors": 1,
+ "words": 13,
+ "wer": 0.07692307692307693
+ }
+ },
+ {
+ "id": "release500_0151",
+ "category": "emotional",
+ "text": "She started laughing and I started laughing and we couldn't stop for a full minute.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0151.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SHE STARTED LAUGHING AND I STARTED LAUGHING AND WE COULDN'T STOP FOR A FULL MINUTE",
+ "raw": {
+ "normalized_reference": "she started laughing and i started laughing and we couldn't stop for a full minute.",
+ "normalized_hypothesis": "she started laughing and i started laughing and we couldn't stop for a full minute",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "she started laughing and i started laughing and we could not stop for a full minute",
+ "normalized_hypothesis": "she started laughing and i started laughing and we could not stop for a full minute",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0152",
+ "category": "long",
+ "text": "We spent three years building something we were genuinely proud of, and then the market shifted and none of it mattered the way we thought it would.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0152.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WE SPENT THREE YEARS BUILDING SOMETHING WE WERE GENUINELY PROUD OF AND THEN THE MARKET SHIFTED AND NONE OF IT MATTERED THE WAY WE THOUGHT IT WOULD",
+ "raw": {
+ "normalized_reference": "we spent three years building something we were genuinely proud of, and then the market shifted and none of it mattered the way we thought it would.",
+ "normalized_hypothesis": "we spent three years building something we were genuinely proud of and then the market shifted and none of it mattered the way we thought it would",
+ "errors": 0,
+ "words": 27,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "we spent 3 years building something we were genuinely proud of and then the market shifted and none of it mattered the way we thought it would",
+ "normalized_hypothesis": "we spent 3 years building something we were genuinely proud of and then the market shifted and none of it mattered the way we thought it would",
+ "errors": 0,
+ "words": 27,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0153",
+ "category": "narrative",
+ "text": "And somehow, despite all of that, it actually turned out okay.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0153.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "AND SOMEHOW DESPITE ALL OF THAT IT ACTUALLY TURNED OUT OQUT",
+ "raw": {
+ "normalized_reference": "and somehow, despite all of that, it actually turned out okay.",
+ "normalized_hypothesis": "and somehow despite all of that it actually turned out oqut",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ },
+ "semantic": {
+ "normalized_reference": "and somehow despite all of that it actually turned out okay",
+ "normalized_hypothesis": "and somehow despite all of that it actually turned out oqut",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ }
+ },
+ {
+ "id": "release500_0154",
+ "category": "question",
+ "text": "Is that something you actually want, or something you think you should want?",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0154.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IS THAT SOMETHING YOU ACTUALLY WANT OR SOMETHING YOU THINK YOU SHOULD WANT",
+ "raw": {
+ "normalized_reference": "is that something you actually want, or something you think you should want?",
+ "normalized_hypothesis": "is that something you actually want or something you think you should want",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "is that something you actually want or something you think you should want",
+ "normalized_hypothesis": "is that something you actually want or something you think you should want",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0155",
+ "category": "short",
+ "text": "This is fine.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0155.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THIS IS FINE",
+ "raw": {
+ "normalized_reference": "this is fine.",
+ "normalized_hypothesis": "this is fine",
+ "errors": 0,
+ "words": 3,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "this is fine",
+ "normalized_hypothesis": "this is fine",
+ "errors": 0,
+ "words": 3,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0156",
+ "category": "technical",
+ "text": "The tricky part isn't the setup, it's making sure all dependencies line up.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0156.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE TRICKY PART ISN'T THE SET UP IT'S MAKING SURE ALL DEPENDENCIES LIE NOW",
+ "raw": {
+ "normalized_reference": "the tricky part isn't the setup, it's making sure all dependencies line up.",
+ "normalized_hypothesis": "the tricky part isn't the set up it's making sure all dependencies lie now",
+ "errors": 4,
+ "words": 13,
+ "wer": 0.3076923076923077
+ },
+ "semantic": {
+ "normalized_reference": "the tricky part is not the setup it is making sure all dependencies line up",
+ "normalized_hypothesis": "the tricky part is not the set up it is making sure all dependencies lie now",
+ "errors": 4,
+ "words": 15,
+ "wer": 0.26666666666666666
+ }
+ },
+ {
+ "id": "release500_0157",
+ "category": "conversational",
+ "text": "He called it a minor setback. I called it a disaster.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0157.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "KEY CALLED IT A MINOR SETBACK THAT I CALLED IT AI DISASTER",
+ "raw": {
+ "normalized_reference": "he called it a minor setback. i called it a disaster.",
+ "normalized_hypothesis": "key called it a minor setback that i called it ai disaster",
+ "errors": 3,
+ "words": 11,
+ "wer": 0.2727272727272727
+ },
+ "semantic": {
+ "normalized_reference": "he called it a minor setback i called it a disaster",
+ "normalized_hypothesis": "key called it a minor setback that i called it ai disaster",
+ "errors": 3,
+ "words": 11,
+ "wer": 0.2727272727272727
+ }
+ },
+ {
+ "id": "release500_0158",
+ "category": "descriptive",
+ "text": "The whole place smelled like pine and old wood and cold air.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0158.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE WHOLE PLACE SMELLED LIKE PINE IN OLD WOOD AND COLD AIR",
+ "raw": {
+ "normalized_reference": "the whole place smelled like pine and old wood and cold air.",
+ "normalized_hypothesis": "the whole place smelled like pine in old wood and cold air",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "the whole place smelled like pine and old wood and cold air",
+ "normalized_hypothesis": "the whole place smelled like pine in old wood and cold air",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ }
+ },
+ {
+ "id": "release500_0159",
+ "category": "emotional",
+ "text": "I've never been so relieved in my entire life.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0159.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I'VE NEVER BEEN SO RELIEVED IN MY ENTIRE LIFE",
+ "raw": {
+ "normalized_reference": "i've never been so relieved in my entire life.",
+ "normalized_hypothesis": "i've never been so relieved in my entire life",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i have never been so relieved in my entire life",
+ "normalized_hypothesis": "i have never been so relieved in my entire life",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0160",
+ "category": "narrative",
+ "text": "I wrote it all down that night so I wouldn't forget, and I still haven't looked at it.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0160.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I WROTE IT ALL DOWN THAT NIGHT SO I WOULDN'T FORGET AND I STILL HAVEN'T LOOKED AT IT",
+ "raw": {
+ "normalized_reference": "i wrote it all down that night so i wouldn't forget, and i still haven't looked at it.",
+ "normalized_hypothesis": "i wrote it all down that night so i wouldn't forget and i still haven't looked at it",
+ "errors": 0,
+ "words": 18,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i wrote it all down that night so i would not forget and i still have not looked at it",
+ "normalized_hypothesis": "i wrote it all down that night so i would not forget and i still have not looked at it",
+ "errors": 0,
+ "words": 20,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0161",
+ "category": "short",
+ "text": "That's kind of wild actually.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0161.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THAT'S KIND OF WILD ACTUALLY",
+ "raw": {
+ "normalized_reference": "that's kind of wild actually.",
+ "normalized_hypothesis": "that's kind of wild actually",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "that is kind of wild actually",
+ "normalized_hypothesis": "that is kind of wild actually",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0162",
+ "category": "technical",
+ "text": "Version pinning will save you a lot of pain later down the line.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0162.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "VERSION PINNING WILL SAVE YOU A LOT OF PAIN LATER DOWN THE LINE",
+ "raw": {
+ "normalized_reference": "version pinning will save you a lot of pain later down the line.",
+ "normalized_hypothesis": "version pinning will save you a lot of pain later down the line",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "version pinning will save you a lot of pain later down the line",
+ "normalized_hypothesis": "version pinning will save you a lot of pain later down the line",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0163",
+ "category": "conversational",
+ "text": "At some point you just have to commit and see what happens.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0163.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "AT SOME POINT YOU JUST HAVE TO COMMIT SEE WHAT HAPPENS",
+ "raw": {
+ "normalized_reference": "at some point you just have to commit and see what happens.",
+ "normalized_hypothesis": "at some point you just have to commit see what happens",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "at some point you just have to commit and see what happens",
+ "normalized_hypothesis": "at some point you just have to commit see what happens",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ }
+ },
+ {
+ "id": "release500_0164",
+ "category": "emotional",
+ "text": "Nobody told me it was going to be this hard, and honestly, I'm grateful.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0164.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "NOBODY TOLD ME IT WAS GOING TO BE THIS HARD AND HONESTLY I'M GRATEFUL",
+ "raw": {
+ "normalized_reference": "nobody told me it was going to be this hard, and honestly, i'm grateful.",
+ "normalized_hypothesis": "nobody told me it was going to be this hard and honestly i'm grateful",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "nobody told me it was going to be this hard and honestly i am grateful",
+ "normalized_hypothesis": "nobody told me it was going to be this hard and honestly i am grateful",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0165",
+ "category": "narrative",
+ "text": "Chapter seven. In which everything that could go wrong did, and a few things that couldn't.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0165.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "CHAPTER SEVEN IN WHICH EVERYTHING THAT COULD GO WRONG DID AND A FEW THINGS THAT COULDN'T",
+ "raw": {
+ "normalized_reference": "chapter seven. in which everything that could go wrong did, and a few things that couldn't.",
+ "normalized_hypothesis": "chapter seven in which everything that could go wrong did and a few things that couldn't",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "chapter 7 in which everything that could go wrong did and a few things that could not",
+ "normalized_hypothesis": "chapter 7 in which everything that could go wrong did and a few things that could not",
+ "errors": 0,
+ "words": 17,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0166",
+ "category": "short",
+ "text": "Wait, what?",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0166.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WAIT WHAT",
+ "raw": {
+ "normalized_reference": "wait, what?",
+ "normalized_hypothesis": "wait what",
+ "errors": 0,
+ "words": 2,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "wait what",
+ "normalized_hypothesis": "wait what",
+ "errors": 0,
+ "words": 2,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0167",
+ "category": "technical",
+ "text": "The graphics processor remained idle because the data loader exhausted every central processing core.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0167.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE GRAPHICS PROCESSER REMAINED IDLE BECAUSE THE DATA LOADER EXHAUSTED EVERY CENTRAL PROCESSING CORE",
+ "raw": {
+ "normalized_reference": "the graphics processor remained idle because the data loader exhausted every central processing core.",
+ "normalized_hypothesis": "the graphics processer remained idle because the data loader exhausted every central processing core",
+ "errors": 1,
+ "words": 14,
+ "wer": 0.07142857142857142
+ },
+ "semantic": {
+ "normalized_reference": "the graphics processor remained idle because the data loader exhausted every central processing core",
+ "normalized_hypothesis": "the graphics processer remained idle because the data loader exhausted every central processing core",
+ "errors": 1,
+ "words": 14,
+ "wer": 0.07142857142857142
+ }
+ },
+ {
+ "id": "release500_0168",
+ "category": "conversational",
+ "text": "Could you speak a little more slowly? I am trying to write every detail down.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0168.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "COULD YOU SPEAK A LITTLE MORE SLOWLY IAM TRYING TO WRITE EVERY DETAIL DOWN",
+ "raw": {
+ "normalized_reference": "could you speak a little more slowly? i am trying to write every detail down.",
+ "normalized_hypothesis": "could you speak a little more slowly iam trying to write every detail down",
+ "errors": 2,
+ "words": 15,
+ "wer": 0.13333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "could you speak a little more slowly i am trying to write every detail down",
+ "normalized_hypothesis": "could you speak a little more slowly iam trying to write every detail down",
+ "errors": 2,
+ "words": 15,
+ "wer": 0.13333333333333333
+ }
+ },
+ {
+ "id": "release500_0169",
+ "category": "emotional",
+ "text": "I was so excited I could barely form a complete sentence.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0169.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I WAS SO EXCITED I COULD BARELY FORM A COMPLETE SENTENCE",
+ "raw": {
+ "normalized_reference": "i was so excited i could barely form a complete sentence.",
+ "normalized_hypothesis": "i was so excited i could barely form a complete sentence",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i was so excited i could barely form a complete sentence",
+ "normalized_hypothesis": "i was so excited i could barely form a complete sentence",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0170",
+ "category": "narrative",
+ "text": "He looked at me like I had said something in a language he didn't speak.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0170.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE LOOKED AT ME LIKE I AD SAID SOMETHING IN A LANGUAGE HE DIDN'T SPEAK",
+ "raw": {
+ "normalized_reference": "he looked at me like i had said something in a language he didn't speak.",
+ "normalized_hypothesis": "he looked at me like i ad said something in a language he didn't speak",
+ "errors": 1,
+ "words": 15,
+ "wer": 0.06666666666666667
+ },
+ "semantic": {
+ "normalized_reference": "he looked at me like i had said something in a language he did not speak",
+ "normalized_hypothesis": "he looked at me like i ad said something in a language he did not speak",
+ "errors": 1,
+ "words": 16,
+ "wer": 0.0625
+ }
+ },
+ {
+ "id": "release500_0171",
+ "category": "short",
+ "text": "That's actually hilarious.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0171.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THAT'S ACTUALLY HILLARIOUS",
+ "raw": {
+ "normalized_reference": "that's actually hilarious.",
+ "normalized_hypothesis": "that's actually hillarious",
+ "errors": 1,
+ "words": 3,
+ "wer": 0.3333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "that is actually hilarious",
+ "normalized_hypothesis": "that is actually hillarious",
+ "errors": 1,
+ "words": 4,
+ "wer": 0.25
+ }
+ },
+ {
+ "id": "release500_0172",
+ "category": "technical",
+ "text": "The encoder maps phonemes, stress, and phrase boundaries into a shared hidden representation.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0172.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE INCODER MAPS PHONIUMS STRESS AND PRAYS BOUNDARIES INTO A SHARED HIDDEN REPRESENTATION",
+ "raw": {
+ "normalized_reference": "the encoder maps phonemes, stress, and phrase boundaries into a shared hidden representation.",
+ "normalized_hypothesis": "the incoder maps phoniums stress and prays boundaries into a shared hidden representation",
+ "errors": 3,
+ "words": 13,
+ "wer": 0.23076923076923078
+ },
+ "semantic": {
+ "normalized_reference": "the encoder maps phonemes stress and phrase boundaries into a shared hidden representation",
+ "normalized_hypothesis": "the incoder maps phoniums stress and prays boundaries into a shared hidden representation",
+ "errors": 3,
+ "words": 13,
+ "wer": 0.23076923076923078
+ }
+ },
+ {
+ "id": "release500_0173",
+ "category": "conversational",
+ "text": "She showed up three hours late and acted like nothing had happened.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0173.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SHE SHOWED UP THREE HOURS LATE AND ACTED LIKE NOTHING HAD HAPPENED",
+ "raw": {
+ "normalized_reference": "she showed up three hours late and acted like nothing had happened.",
+ "normalized_hypothesis": "she showed up three hours late and acted like nothing had happened",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "she showed up 3 hours late and acted like nothing had happened",
+ "normalized_hypothesis": "she showed up 3 hours late and acted like nothing had happened",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0174",
+ "category": "emotional",
+ "text": "Oh my god. This is the best thing I've ever tasted.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0174.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "O MY GOD THIS IS THE BEST THING I'VE EVER TASTED",
+ "raw": {
+ "normalized_reference": "oh my god. this is the best thing i've ever tasted.",
+ "normalized_hypothesis": "o my god this is the best thing i've ever tasted",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ },
+ "semantic": {
+ "normalized_reference": "0 my god this is the best thing i have ever tasted",
+ "normalized_hypothesis": "0 my god this is the best thing i have ever tasted",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0175",
+ "category": "narrative",
+ "text": "The note was short, which was somehow more frightening than a long one would have been.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0175.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE NOTE WAS SHORT WHICH WAS SOMEHOW MORE FRIGHTENING THAN A LONG ONE WOULD HAVE BEEN",
+ "raw": {
+ "normalized_reference": "the note was short, which was somehow more frightening than a long one would have been.",
+ "normalized_hypothesis": "the note was short which was somehow more frightening than a long one would have been",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the note was short which was somehow more frightening than a long one would have been",
+ "normalized_hypothesis": "the note was short which was somehow more frightening than a long one would have been",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0176",
+ "category": "short",
+ "text": "Noted.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0176.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "NOTED",
+ "raw": {
+ "normalized_reference": "noted.",
+ "normalized_hypothesis": "noted",
+ "errors": 0,
+ "words": 1,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "noted",
+ "normalized_hypothesis": "noted",
+ "errors": 0,
+ "words": 1,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0177",
+ "category": "technical",
+ "text": "It's a simple process once you understand the underlying structure.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0177.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT'S A SIMPLE PROCESS ONCE YOU UNDERSTAND THE UNDERLYING STRUCTURE",
+ "raw": {
+ "normalized_reference": "it's a simple process once you understand the underlying structure.",
+ "normalized_hypothesis": "it's a simple process once you understand the underlying structure",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it is a simple process once you understand the underlying structure",
+ "normalized_hypothesis": "it is a simple process once you understand the underlying structure",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0178",
+ "category": "conversational",
+ "text": "You have to hear this story because it's almost too good to be true.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0178.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "YOU HAVE TO HEAR THIS STORY BECAUSE IT'S ALMOST TOO GOOD TO BE TRUE",
+ "raw": {
+ "normalized_reference": "you have to hear this story because it's almost too good to be true.",
+ "normalized_hypothesis": "you have to hear this story because it's almost too good to be true",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "you have to hear this story because it is almost too good to be true",
+ "normalized_hypothesis": "you have to hear this story because it is almost too good to be true",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0179",
+ "category": "emotional",
+ "text": "I have never felt more seen in my entire life than in that moment.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0179.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I HAVE NEVER FELT MORE SEEN IN MY ENTIRE LIFE THAN IN THAT MOMENT",
+ "raw": {
+ "normalized_reference": "i have never felt more seen in my entire life than in that moment.",
+ "normalized_hypothesis": "i have never felt more seen in my entire life than in that moment",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i have never felt more seen in my entire life than in that moment",
+ "normalized_hypothesis": "i have never felt more seen in my entire life than in that moment",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0180",
+ "category": "narrative",
+ "text": "She knew. She had always known. She just waited for the right moment.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0180.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SHE KNEW THAT SHE HAD ALWAYS KNOWN SHE JUST WAITED FOR THE RIGHT MOMENT",
+ "raw": {
+ "normalized_reference": "she knew. she had always known. she just waited for the right moment.",
+ "normalized_hypothesis": "she knew that she had always known she just waited for the right moment",
+ "errors": 1,
+ "words": 13,
+ "wer": 0.07692307692307693
+ },
+ "semantic": {
+ "normalized_reference": "she knew she had always known she just waited for the right moment",
+ "normalized_hypothesis": "she knew that she had always known she just waited for the right moment",
+ "errors": 1,
+ "words": 13,
+ "wer": 0.07692307692307693
+ }
+ },
+ {
+ "id": "release500_0181",
+ "category": "short",
+ "text": "Yeah, no. Hard pass.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0181.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "YEH NO DOC HARD PASS",
+ "raw": {
+ "normalized_reference": "yeah, no. hard pass.",
+ "normalized_hypothesis": "yeh no doc hard pass",
+ "errors": 2,
+ "words": 4,
+ "wer": 0.5
+ },
+ "semantic": {
+ "normalized_reference": "yeah no hard pass",
+ "normalized_hypothesis": "yeh no doc hard pass",
+ "errors": 2,
+ "words": 4,
+ "wer": 0.5
+ }
+ },
+ {
+ "id": "release500_0182",
+ "category": "technical",
+ "text": "Please verify the sample rate, hop length, and mel filter configuration before inference.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0182.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "PLEASE VERIFY THE SAMPLE RATE POPLENGTH AND MELL FILTER CONFIGURATION BEFORE INFERENCE",
+ "raw": {
+ "normalized_reference": "please verify the sample rate, hop length, and mel filter configuration before inference.",
+ "normalized_hypothesis": "please verify the sample rate poplength and mell filter configuration before inference",
+ "errors": 3,
+ "words": 13,
+ "wer": 0.23076923076923078
+ },
+ "semantic": {
+ "normalized_reference": "please verify the sample rate hop length and mel filter configuration before inference",
+ "normalized_hypothesis": "please verify the sample rate poplength and mell filter configuration before inference",
+ "errors": 3,
+ "words": 13,
+ "wer": 0.23076923076923078
+ }
+ },
+ {
+ "id": "release500_0183",
+ "category": "conversational",
+ "text": "That's the worst idea I've ever heard and I'm completely in.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0183.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THAT'S THE WORST IDEA I'VE EVER HEARD AND I'M COMPLETELY IN",
+ "raw": {
+ "normalized_reference": "that's the worst idea i've ever heard and i'm completely in.",
+ "normalized_hypothesis": "that's the worst idea i've ever heard and i'm completely in",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "that is the worst idea i have ever heard and i am completely in",
+ "normalized_hypothesis": "that is the worst idea i have ever heard and i am completely in",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0184",
+ "category": "emotional",
+ "text": "I remember thinking: this is it. This is the thing I've been waiting for.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0184.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I REMEMBER THINKING THIS IS IT THIS IS THE THING I'VE BEEN WAITING FOR",
+ "raw": {
+ "normalized_reference": "i remember thinking: this is it. this is the thing i've been waiting for.",
+ "normalized_hypothesis": "i remember thinking this is it this is the thing i've been waiting for",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i remember thinking this is it this is the thing i have been waiting for",
+ "normalized_hypothesis": "i remember thinking this is it this is the thing i have been waiting for",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0185",
+ "category": "narrative",
+ "text": "We never talked about it again, which I think was the right call.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0185.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WE NEVER TALKED ABOUT IT AGAIN WHICH I THINK WAS THE RIGHT CALL",
+ "raw": {
+ "normalized_reference": "we never talked about it again, which i think was the right call.",
+ "normalized_hypothesis": "we never talked about it again which i think was the right call",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "we never talked about it again which i think was the right call",
+ "normalized_hypothesis": "we never talked about it again which i think was the right call",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0186",
+ "category": "short",
+ "text": "In hindsight, obvious.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0186.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IN HINDSIGHT OBVIOUS",
+ "raw": {
+ "normalized_reference": "in hindsight, obvious.",
+ "normalized_hypothesis": "in hindsight obvious",
+ "errors": 0,
+ "words": 3,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "in hindsight obvious",
+ "normalized_hypothesis": "in hindsight obvious",
+ "errors": 0,
+ "words": 3,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0187",
+ "category": "technical",
+ "text": "Once it's configured correctly, it basically runs itself.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0187.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "ONCE IT'S CONFIGURED CORRECTLY IT BASICALLY RUNS ITSELF",
+ "raw": {
+ "normalized_reference": "once it's configured correctly, it basically runs itself.",
+ "normalized_hypothesis": "once it's configured correctly it basically runs itself",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "once it is configured correctly it basically runs itself",
+ "normalized_hypothesis": "once it is configured correctly it basically runs itself",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0188",
+ "category": "conversational",
+ "text": "You got engaged? When? How? Tell me everything right now.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0188.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "YOU GOT ENGAGED WEN CAU TELL ME EVERYTHING RIGHT NOW",
+ "raw": {
+ "normalized_reference": "you got engaged? when? how? tell me everything right now.",
+ "normalized_hypothesis": "you got engaged wen cau tell me everything right now",
+ "errors": 2,
+ "words": 10,
+ "wer": 0.2
+ },
+ "semantic": {
+ "normalized_reference": "you got engaged when how tell me everything right now",
+ "normalized_hypothesis": "you got engaged wen cau tell me everything right now",
+ "errors": 2,
+ "words": 10,
+ "wer": 0.2
+ }
+ },
+ {
+ "id": "release500_0189",
+ "category": "emotional",
+ "text": "It's hard to explain how much that moment meant to me.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0189.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT'S HARD TO EXPLAIN HOW MUCH THAT MOMENT MEANT TO ME",
+ "raw": {
+ "normalized_reference": "it's hard to explain how much that moment meant to me.",
+ "normalized_hypothesis": "it's hard to explain how much that moment meant to me",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it is hard to explain how much that moment meant to me",
+ "normalized_hypothesis": "it is hard to explain how much that moment meant to me",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0190",
+ "category": "narrative",
+ "text": "He told the story three different times that night and somehow it got better each time.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0190.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "KEI TOLD THIS STORY THREE DIFFERENT TIMES THAT NIGHT AND SOMEHOW IT GOT BETTER EACH TIME",
+ "raw": {
+ "normalized_reference": "he told the story three different times that night and somehow it got better each time.",
+ "normalized_hypothesis": "kei told this story three different times that night and somehow it got better each time",
+ "errors": 2,
+ "words": 16,
+ "wer": 0.125
+ },
+ "semantic": {
+ "normalized_reference": "he told the story 3 different times that night and somehow it got better each time",
+ "normalized_hypothesis": "kei told this story 3 different times that night and somehow it got better each time",
+ "errors": 2,
+ "words": 16,
+ "wer": 0.125
+ }
+ },
+ {
+ "id": "release500_0191",
+ "category": "short",
+ "text": "Right. Moving on.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0191.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "RIGHT AT MOVING ON",
+ "raw": {
+ "normalized_reference": "right. moving on.",
+ "normalized_hypothesis": "right at moving on",
+ "errors": 1,
+ "words": 3,
+ "wer": 0.3333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "right moving on",
+ "normalized_hypothesis": "right at moving on",
+ "errors": 1,
+ "words": 3,
+ "wer": 0.3333333333333333
+ }
+ },
+ {
+ "id": "release500_0192",
+ "category": "conversational",
+ "text": "No, I did not move your keys; check beside the small ceramic bowl.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0192.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "NO I DID NOT MOVE ER KEYS CHECK BESIDE THE SMALL SERAMIC BOWL",
+ "raw": {
+ "normalized_reference": "no, i did not move your keys; check beside the small ceramic bowl.",
+ "normalized_hypothesis": "no i did not move er keys check beside the small seramic bowl",
+ "errors": 2,
+ "words": 13,
+ "wer": 0.15384615384615385
+ },
+ "semantic": {
+ "normalized_reference": "no i did not move your keys check beside the small ceramic bowl",
+ "normalized_hypothesis": "no i did not move er keys check beside the small seramic bowl",
+ "errors": 2,
+ "words": 13,
+ "wer": 0.15384615384615385
+ }
+ },
+ {
+ "id": "release500_0193",
+ "category": "emotional",
+ "text": "I was devastated. I'm not going to pretend otherwise.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0193.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I WAS DEVASTATED IAM NOT GOING TO PRETEND OTHERWISE",
+ "raw": {
+ "normalized_reference": "i was devastated. i'm not going to pretend otherwise.",
+ "normalized_hypothesis": "i was devastated iam not going to pretend otherwise",
+ "errors": 1,
+ "words": 9,
+ "wer": 0.1111111111111111
+ },
+ "semantic": {
+ "normalized_reference": "i was devastated i am not going to pretend otherwise",
+ "normalized_hypothesis": "i was devastated iam not going to pretend otherwise",
+ "errors": 2,
+ "words": 10,
+ "wer": 0.2
+ }
+ },
+ {
+ "id": "release500_0194",
+ "category": "narrative",
+ "text": "That was the last summer before everything changed, though we didn't know it yet.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0194.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THAT WAS THE LAST SUMMER BEFORE EVERYTHING CHANGED THOUGH WE DIDN'T KNOW IT YET",
+ "raw": {
+ "normalized_reference": "that was the last summer before everything changed, though we didn't know it yet.",
+ "normalized_hypothesis": "that was the last summer before everything changed though we didn't know it yet",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "that was the last summer before everything changed though we did not know it yet",
+ "normalized_hypothesis": "that was the last summer before everything changed though we did not know it yet",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0195",
+ "category": "short",
+ "text": "Bold of you to assume.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0195.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "BOLD OF YOU TO ASSUME",
+ "raw": {
+ "normalized_reference": "bold of you to assume.",
+ "normalized_hypothesis": "bold of you to assume",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "bold of you to assume",
+ "normalized_hypothesis": "bold of you to assume",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0196",
+ "category": "conversational",
+ "text": "It's one of those decisions that makes total sense in the moment.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0196.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT'S ONE OF THOSE DECISIONS THAT MAKES TOTAL SENSE IN THE MOMENT",
+ "raw": {
+ "normalized_reference": "it's one of those decisions that makes total sense in the moment.",
+ "normalized_hypothesis": "it's one of those decisions that makes total sense in the moment",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it is one of those decisions that makes total sense in the moment",
+ "normalized_hypothesis": "it is one of those decisions that makes total sense in the moment",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0197",
+ "category": "emotional",
+ "text": "I was furious. Not the loud kind. The quiet kind that lasts for days.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0197.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I WAS FURIOUS NOT THE LOUD KIND THE QUIET KIND THAT LASTS FOR DAYS",
+ "raw": {
+ "normalized_reference": "i was furious. not the loud kind. the quiet kind that lasts for days.",
+ "normalized_hypothesis": "i was furious not the loud kind the quiet kind that lasts for days",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i was furious not the loud kind the quiet kind that lasts for days",
+ "normalized_hypothesis": "i was furious not the loud kind the quiet kind that lasts for days",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0198",
+ "category": "narrative",
+ "text": "That's the part of the story I usually leave out.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0198.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THAT'S THE PART OF THE STORY I USUALLY LEAVE OUT",
+ "raw": {
+ "normalized_reference": "that's the part of the story i usually leave out.",
+ "normalized_hypothesis": "that's the part of the story i usually leave out",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "that is the part of the story i usually leave out",
+ "normalized_hypothesis": "that is the part of the story i usually leave out",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0199",
+ "category": "short",
+ "text": "I've said too much.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0199.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I'VE SAID TOO MUCH",
+ "raw": {
+ "normalized_reference": "i've said too much.",
+ "normalized_hypothesis": "i've said too much",
+ "errors": 0,
+ "words": 4,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i have said too much",
+ "normalized_hypothesis": "i have said too much",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0200",
+ "category": "conversational",
+ "text": "I've been faking that I understood this the entire time.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0200.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I'VE BEEN FAKING THAT I UNDERSTOOD THIS THE ENTIRE TIME",
+ "raw": {
+ "normalized_reference": "i've been faking that i understood this the entire time.",
+ "normalized_hypothesis": "i've been faking that i understood this the entire time",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i have been faking that i understood this the entire time",
+ "normalized_hypothesis": "i have been faking that i understood this the entire time",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0201",
+ "category": "phonetic_balance",
+ "text": "Peace be unto you and grace before the Lord.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0201.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "PEACE BE UNTO YOU AND GRACE BEFORE THE LORD",
+ "raw": {
+ "normalized_reference": "peace be unto you and grace before the lord.",
+ "normalized_hypothesis": "peace be unto you and grace before the lord",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "peace be unto you and grace before the lord",
+ "normalized_hypothesis": "peace be unto you and grace before the lord",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0202",
+ "category": "phonetic_balance",
+ "text": "From now on we're pals.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0202.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "FROM NOW ON WER HALSE",
+ "raw": {
+ "normalized_reference": "from now on we're pals.",
+ "normalized_hypothesis": "from now on wer halse",
+ "errors": 2,
+ "words": 5,
+ "wer": 0.4
+ },
+ "semantic": {
+ "normalized_reference": "from now on we are pals",
+ "normalized_hypothesis": "from now on wer halse",
+ "errors": 3,
+ "words": 6,
+ "wer": 0.5
+ }
+ },
+ {
+ "id": "release500_0203",
+ "category": "phonetic_balance",
+ "text": "Here we allow our solicitors to look after our legal work.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0203.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HERE WE ALLOW OUR SOLICITORS TO LOOK AFTER OUR LEGAL WORK",
+ "raw": {
+ "normalized_reference": "here we allow our solicitors to look after our legal work.",
+ "normalized_hypothesis": "here we allow our solicitors to look after our legal work",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "here we allow our solicitors to look after our legal work",
+ "normalized_hypothesis": "here we allow our solicitors to look after our legal work",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0204",
+ "category": "phonetic_balance",
+ "text": "Her words sent a strange chill through Philip.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0204.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HER WORDS SENT A STRANGE CHILL THROUGH PHILOTT",
+ "raw": {
+ "normalized_reference": "her words sent a strange chill through philip.",
+ "normalized_hypothesis": "her words sent a strange chill through philott",
+ "errors": 1,
+ "words": 8,
+ "wer": 0.125
+ },
+ "semantic": {
+ "normalized_reference": "her words sent a strange chill through philip",
+ "normalized_hypothesis": "her words sent a strange chill through philott",
+ "errors": 1,
+ "words": 8,
+ "wer": 0.125
+ }
+ },
+ {
+ "id": "release500_0205",
+ "category": "phonetic_balance",
+ "text": "It was introduced by Representative Dick of Ohio.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0205.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT WAS INTRODUCED BY REPRESENTATIVE DICK OF OHIO",
+ "raw": {
+ "normalized_reference": "it was introduced by representative dick of ohio.",
+ "normalized_hypothesis": "it was introduced by representative dick of ohio",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it was introduced by representative dick of ohio",
+ "normalized_hypothesis": "it was introduced by representative dick of ohio",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0206",
+ "category": "phonetic_balance",
+ "text": "It's the aurora borealis.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0206.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT'S THE AURORA BORIALIS",
+ "raw": {
+ "normalized_reference": "it's the aurora borealis.",
+ "normalized_hypothesis": "it's the aurora borialis",
+ "errors": 1,
+ "words": 4,
+ "wer": 0.25
+ },
+ "semantic": {
+ "normalized_reference": "it is the aurora borealis",
+ "normalized_hypothesis": "it is the aurora borialis",
+ "errors": 1,
+ "words": 5,
+ "wer": 0.2
+ }
+ },
+ {
+ "id": "release500_0207",
+ "category": "phonetic_balance",
+ "text": "The boy grew and prospered.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0207.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE BOY GREW AND PROSPERED",
+ "raw": {
+ "normalized_reference": "the boy grew and prospered.",
+ "normalized_hypothesis": "the boy grew and prospered",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the boy grew and prospered",
+ "normalized_hypothesis": "the boy grew and prospered",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0208",
+ "category": "phonetic_balance",
+ "text": "His voice was passionately rebellious.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0208.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HIS VOICE WAS PASSIONATELY REBELLIOUS",
+ "raw": {
+ "normalized_reference": "his voice was passionately rebellious.",
+ "normalized_hypothesis": "his voice was passionately rebellious",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "his voice was passionately rebellious",
+ "normalized_hypothesis": "his voice was passionately rebellious",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0209",
+ "category": "phonetic_balance",
+ "text": "That's where they cut off the Scottish Chiefs and killed all hands.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0209.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THAT'S WHERE THEY CUT OFF THE SCOTTISH CHIEFS AND KILLED ALL HANDS",
+ "raw": {
+ "normalized_reference": "that's where they cut off the scottish chiefs and killed all hands.",
+ "normalized_hypothesis": "that's where they cut off the scottish chiefs and killed all hands",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "that is where they cut off the scottish chiefs and killed all hands",
+ "normalized_hypothesis": "that is where they cut off the scottish chiefs and killed all hands",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0210",
+ "category": "phonetic_balance",
+ "text": "And Tom King patiently endured.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0210.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "AND TOM KEEN PATIENTLY ENDURED",
+ "raw": {
+ "normalized_reference": "and tom king patiently endured.",
+ "normalized_hypothesis": "and tom keen patiently endured",
+ "errors": 1,
+ "words": 5,
+ "wer": 0.2
+ },
+ "semantic": {
+ "normalized_reference": "and tom king patiently endured",
+ "normalized_hypothesis": "and tom keen patiently endured",
+ "errors": 1,
+ "words": 5,
+ "wer": 0.2
+ }
+ },
+ {
+ "id": "release500_0211",
+ "category": "phonetic_balance",
+ "text": "From that moment his friendship for Belize turns to hatred and jealousy.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0211.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "FROM THAT MOMENT HIS FRIENDSHIP FOR BELLIS TURNS TO HATRED AND JEALOUSY",
+ "raw": {
+ "normalized_reference": "from that moment his friendship for belize turns to hatred and jealousy.",
+ "normalized_hypothesis": "from that moment his friendship for bellis turns to hatred and jealousy",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "from that moment his friendship for belize turns to hatred and jealousy",
+ "normalized_hypothesis": "from that moment his friendship for bellis turns to hatred and jealousy",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ }
+ },
+ {
+ "id": "release500_0212",
+ "category": "phonetic_balance",
+ "text": "Then you don't believe in altruism.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0212.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THEN YOU DON'T BELIEVE IN ALTRUISM",
+ "raw": {
+ "normalized_reference": "then you don't believe in altruism.",
+ "normalized_hypothesis": "then you don't believe in altruism",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "then you do not believe in altruism",
+ "normalized_hypothesis": "then you do not believe in altruism",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0213",
+ "category": "phonetic_balance",
+ "text": "What Pascal glimpsed with the vision of a seer, I have lived.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0213.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WHAT PASCAL GLIMPSED WITH THE VISION OF A SEER I HAVE LIVED",
+ "raw": {
+ "normalized_reference": "what pascal glimpsed with the vision of a seer, i have lived.",
+ "normalized_hypothesis": "what pascal glimpsed with the vision of a seer i have lived",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "what pascal glimpsed with the vision of a seer i have lived",
+ "normalized_hypothesis": "what pascal glimpsed with the vision of a seer i have lived",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0214",
+ "category": "phonetic_balance",
+ "text": "You were making them talk shop, Ruth charged him.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0214.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "YOU WERE MAKING THEM TALK SHOP RUFE CHARGED HIM",
+ "raw": {
+ "normalized_reference": "you were making them talk shop, ruth charged him.",
+ "normalized_hypothesis": "you were making them talk shop rufe charged him",
+ "errors": 1,
+ "words": 9,
+ "wer": 0.1111111111111111
+ },
+ "semantic": {
+ "normalized_reference": "you were making them talk shop ruth charged him",
+ "normalized_hypothesis": "you were making them talk shop rufe charged him",
+ "errors": 1,
+ "words": 9,
+ "wer": 0.1111111111111111
+ }
+ },
+ {
+ "id": "release500_0215",
+ "category": "phonetic_balance",
+ "text": "Philip snatched at the letter which Gregson held out to him.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0215.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "PHILIP SNATCHED AT THE LETTER WHICH GREGGSON HELD OUT TO HIM",
+ "raw": {
+ "normalized_reference": "philip snatched at the letter which gregson held out to him.",
+ "normalized_hypothesis": "philip snatched at the letter which greggson held out to him",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ },
+ "semantic": {
+ "normalized_reference": "philip snatched at the letter which gregson held out to him",
+ "normalized_hypothesis": "philip snatched at the letter which greggson held out to him",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ }
+ },
+ {
+ "id": "release500_0216",
+ "category": "phonetic_balance",
+ "text": "For a time the exciting thrill of his adventure was gone.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0216.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "FOR A TIME THE EXCITING THRILL OF HIS ADVENTURE WAS COME",
+ "raw": {
+ "normalized_reference": "for a time the exciting thrill of his adventure was gone.",
+ "normalized_hypothesis": "for a time the exciting thrill of his adventure was come",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ },
+ "semantic": {
+ "normalized_reference": "for a time the exciting thrill of his adventure was gone",
+ "normalized_hypothesis": "for a time the exciting thrill of his adventure was come",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ }
+ },
+ {
+ "id": "release500_0217",
+ "category": "phonetic_balance",
+ "text": "Obviously, it was a disease that could be contracted by contact.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0217.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "OBVIOUSLY IT WAS A DISEASE THAT COULD BE CONTRACTED BY CONTACT",
+ "raw": {
+ "normalized_reference": "obviously, it was a disease that could be contracted by contact.",
+ "normalized_hypothesis": "obviously it was a disease that could be contracted by contact",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "obviously it was a disease that could be contracted by contact",
+ "normalized_hypothesis": "obviously it was a disease that could be contracted by contact",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0218",
+ "category": "phonetic_balance",
+ "text": "Blind with rage, he darted in.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0218.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "BLIND WITH RAGE HE DARTED IN",
+ "raw": {
+ "normalized_reference": "blind with rage, he darted in.",
+ "normalized_hypothesis": "blind with rage he darted in",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "blind with rage he darted in",
+ "normalized_hypothesis": "blind with rage he darted in",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0219",
+ "category": "phonetic_balance",
+ "text": "You live on an income which your father earned.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0219.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "YOU LIVE ON AN INCOME WHICH YOUR FATHER EARNED",
+ "raw": {
+ "normalized_reference": "you live on an income which your father earned.",
+ "normalized_hypothesis": "you live on an income which your father earned",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "you live on an income which your father earned",
+ "normalized_hypothesis": "you live on an income which your father earned",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0220",
+ "category": "phonetic_balance",
+ "text": "Something about it seemed to fascinate him, to challenge his presence.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0220.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SOMETHING ABOUT IT SEEMED TO FASCINATE HIM TO CHALLENGE HIS PRESENCE",
+ "raw": {
+ "normalized_reference": "something about it seemed to fascinate him, to challenge his presence.",
+ "normalized_hypothesis": "something about it seemed to fascinate him to challenge his presence",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "something about it seemed to fascinate him to challenge his presence",
+ "normalized_hypothesis": "something about it seemed to fascinate him to challenge his presence",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0221",
+ "category": "phonetic_balance",
+ "text": "Her beautiful hair was done up in shining coils.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0221.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HER BEAUTIFUL HAIR WAS DONE UP IN SHINING COILS",
+ "raw": {
+ "normalized_reference": "her beautiful hair was done up in shining coils.",
+ "normalized_hypothesis": "her beautiful hair was done up in shining coils",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "her beautiful hair was done up in shining coils",
+ "normalized_hypothesis": "her beautiful hair was done up in shining coils",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0222",
+ "category": "phonetic_balance",
+ "text": "It does, was her audacious answer.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0222.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT DOES WAS HER AUDACIOUS ANSWER",
+ "raw": {
+ "normalized_reference": "it does, was her audacious answer.",
+ "normalized_hypothesis": "it does was her audacious answer",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it does was her audacious answer",
+ "normalized_hypothesis": "it does was her audacious answer",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0223",
+ "category": "phonetic_balance",
+ "text": "He caught himself with a jerk.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0223.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE CAUGHT HIMSELF WITH A JERK",
+ "raw": {
+ "normalized_reference": "he caught himself with a jerk.",
+ "normalized_hypothesis": "he caught himself with a jerk",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he caught himself with a jerk",
+ "normalized_hypothesis": "he caught himself with a jerk",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0224",
+ "category": "phonetic_balance",
+ "text": "He did not believe in the burning of daylight for such a luxury.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0224.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE DID NOT BELIEVE IN THE BURNING OF DAYLIGHT FOR SUCH A LUXURY",
+ "raw": {
+ "normalized_reference": "he did not believe in the burning of daylight for such a luxury.",
+ "normalized_hypothesis": "he did not believe in the burning of daylight for such a luxury",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he did not believe in the burning of daylight for such a luxury",
+ "normalized_hypothesis": "he did not believe in the burning of daylight for such a luxury",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0225",
+ "category": "phonetic_balance",
+ "text": "He had comparatively no advantages at first.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0225.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE HAD COMPARATIVELY NO ADVANTAGES AT FIRST",
+ "raw": {
+ "normalized_reference": "he had comparatively no advantages at first.",
+ "normalized_hypothesis": "he had comparatively no advantages at first",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he had comparatively no advantages at 1st",
+ "normalized_hypothesis": "he had comparatively no advantages at 1st",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0226",
+ "category": "phonetic_balance",
+ "text": "Such men believe, when they come together.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0226.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SUCH MEN BELIEVE WHEN THEY COME TOGETHER",
+ "raw": {
+ "normalized_reference": "such men believe, when they come together.",
+ "normalized_hypothesis": "such men believe when they come together",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "such men believe when they come together",
+ "normalized_hypothesis": "such men believe when they come together",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0227",
+ "category": "phonetic_balance",
+ "text": "The ship should be in within a week or ten days.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0227.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE SHIP SHOULD BE IN WITHIN A WEEK OR TEN DAYS",
+ "raw": {
+ "normalized_reference": "the ship should be in within a week or ten days.",
+ "normalized_hypothesis": "the ship should be in within a week or ten days",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the ship should be in within a week or 10 days",
+ "normalized_hypothesis": "the ship should be in within a week or 10 days",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0228",
+ "category": "phonetic_balance",
+ "text": "Such a dog the wise driver kills, or turns loose.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0228.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SUCH A DOT THE WISE DRIVER KILLS OR TURNS LOOSE",
+ "raw": {
+ "normalized_reference": "such a dog the wise driver kills, or turns loose.",
+ "normalized_hypothesis": "such a dot the wise driver kills or turns loose",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ },
+ "semantic": {
+ "normalized_reference": "such a dog the wise driver kills or turns loose",
+ "normalized_hypothesis": "such a dot the wise driver kills or turns loose",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ }
+ },
+ {
+ "id": "release500_0229",
+ "category": "phonetic_balance",
+ "text": "You have heard always how he was the lover of the Princess Naomi.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0229.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "YOU HAVE HEARD ALWAYS HOW HE WAS THE LOVER OF THE PRINCESS NAOMI",
+ "raw": {
+ "normalized_reference": "you have heard always how he was the lover of the princess naomi.",
+ "normalized_hypothesis": "you have heard always how he was the lover of the princess naomi",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "you have heard always how he was the lover of the princess naomi",
+ "normalized_hypothesis": "you have heard always how he was the lover of the princess naomi",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0230",
+ "category": "phonetic_balance",
+ "text": "I want to know how all this is possible.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0230.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I WANT TO KNOW HOW ALL THIS IS POSSIBLE",
+ "raw": {
+ "normalized_reference": "i want to know how all this is possible.",
+ "normalized_hypothesis": "i want to know how all this is possible",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i want to know how all this is possible",
+ "normalized_hypothesis": "i want to know how all this is possible",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0231",
+ "category": "phonetic_balance",
+ "text": "It was my idea to a tee.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0231.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT WAS MY IDEA TURETI",
+ "raw": {
+ "normalized_reference": "it was my idea to a tee.",
+ "normalized_hypothesis": "it was my idea tureti",
+ "errors": 3,
+ "words": 7,
+ "wer": 0.42857142857142855
+ },
+ "semantic": {
+ "normalized_reference": "it was my idea to a tee",
+ "normalized_hypothesis": "it was my idea tureti",
+ "errors": 3,
+ "words": 7,
+ "wer": 0.42857142857142855
+ }
+ },
+ {
+ "id": "release500_0232",
+ "category": "phonetic_balance",
+ "text": "They likewise are disinclined to being eaten.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0232.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THEY LIKEWISE ARE DISINCLINED TO BEING EATEN",
+ "raw": {
+ "normalized_reference": "they likewise are disinclined to being eaten.",
+ "normalized_hypothesis": "they likewise are disinclined to being eaten",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "they likewise are disinclined to being eaten",
+ "normalized_hypothesis": "they likewise are disinclined to being eaten",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0233",
+ "category": "phonetic_balance",
+ "text": "This state of mind comes of an undue prominence of the ego.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0233.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THIS STATE OF MIND COMES OF AN UNDUE PROMINENCE ON THE EGO",
+ "raw": {
+ "normalized_reference": "this state of mind comes of an undue prominence of the ego.",
+ "normalized_hypothesis": "this state of mind comes of an undue prominence on the ego",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "this state of mind comes of an undue prominence of the ego",
+ "normalized_hypothesis": "this state of mind comes of an undue prominence on the ego",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ }
+ },
+ {
+ "id": "release500_0234",
+ "category": "phonetic_balance",
+ "text": "Then it is as I said, Womble announced with finality.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0234.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THEN IT IS AS I SAID WAMBLE ANNOUNCED WITH FINALITY",
+ "raw": {
+ "normalized_reference": "then it is as i said, womble announced with finality.",
+ "normalized_hypothesis": "then it is as i said wamble announced with finality",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ },
+ "semantic": {
+ "normalized_reference": "then it is as i said womble announced with finality",
+ "normalized_hypothesis": "then it is as i said wamble announced with finality",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ }
+ },
+ {
+ "id": "release500_0235",
+ "category": "phonetic_balance",
+ "text": "I can see that knife now.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0235.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I CAN SEE THAT KNIFE NOW",
+ "raw": {
+ "normalized_reference": "i can see that knife now.",
+ "normalized_hypothesis": "i can see that knife now",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i can see that knife now",
+ "normalized_hypothesis": "i can see that knife now",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0236",
+ "category": "phonetic_balance",
+ "text": "If I meet her again I shall apologize, said Eileen.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0236.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IF I MEET HER AGAIN I SHALL APOLOGIZE SAID ILEEN",
+ "raw": {
+ "normalized_reference": "if i meet her again i shall apologize, said eileen.",
+ "normalized_hypothesis": "if i meet her again i shall apologize said ileen",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ },
+ "semantic": {
+ "normalized_reference": "if i meet her again i shall apologize said eileen",
+ "normalized_hypothesis": "if i meet her again i shall apologize said ileen",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ }
+ },
+ {
+ "id": "release500_0237",
+ "category": "phonetic_balance",
+ "text": "A wildly exciting time was his during the week preceding Thursday the eighteenth.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0237.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE WILDLY EXCITING TIME WAS HIS DURING THE WEEK PRECEDING THURSDAY THE EIGHTEENTH",
+ "raw": {
+ "normalized_reference": "a wildly exciting time was his during the week preceding thursday the eighteenth.",
+ "normalized_hypothesis": "the wildly exciting time was his during the week preceding thursday the eighteenth",
+ "errors": 1,
+ "words": 13,
+ "wer": 0.07692307692307693
+ },
+ "semantic": {
+ "normalized_reference": "a wildly exciting time was his during the week preceding thursday the 18th",
+ "normalized_hypothesis": "the wildly exciting time was his during the week preceding thursday the 18th",
+ "errors": 1,
+ "words": 13,
+ "wer": 0.07692307692307693
+ }
+ },
+ {
+ "id": "release500_0238",
+ "category": "phonetic_balance",
+ "text": "In what bucolic school of fence he had been taught was beyond imagining.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0238.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IN WHAT BUCALIC SCHOOL OF FENCE HE HAD BEEN TAUGHT WAS BEYOND IMAGINING",
+ "raw": {
+ "normalized_reference": "in what bucolic school of fence he had been taught was beyond imagining.",
+ "normalized_hypothesis": "in what bucalic school of fence he had been taught was beyond imagining",
+ "errors": 1,
+ "words": 13,
+ "wer": 0.07692307692307693
+ },
+ "semantic": {
+ "normalized_reference": "in what bucolic school of fence he had been taught was beyond imagining",
+ "normalized_hypothesis": "in what bucalic school of fence he had been taught was beyond imagining",
+ "errors": 1,
+ "words": 13,
+ "wer": 0.07692307692307693
+ }
+ },
+ {
+ "id": "release500_0239",
+ "category": "phonetic_balance",
+ "text": "So she said, the irate skipper dashed on.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0239.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SO SHE SAID THE IRATE SKIPPER DASHED ON",
+ "raw": {
+ "normalized_reference": "so she said, the irate skipper dashed on.",
+ "normalized_hypothesis": "so she said the irate skipper dashed on",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "so she said the irate skipper dashed on",
+ "normalized_hypothesis": "so she said the irate skipper dashed on",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0240",
+ "category": "phonetic_balance",
+ "text": "It is not an attempt to smash the market.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0240.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IS NOT AN ATTEMPT TO SMASH THE MARKET",
+ "raw": {
+ "normalized_reference": "it is not an attempt to smash the market.",
+ "normalized_hypothesis": "is not an attempt to smash the market",
+ "errors": 1,
+ "words": 9,
+ "wer": 0.1111111111111111
+ },
+ "semantic": {
+ "normalized_reference": "it is not an attempt to smash the market",
+ "normalized_hypothesis": "is not an attempt to smash the market",
+ "errors": 1,
+ "words": 9,
+ "wer": 0.1111111111111111
+ }
+ },
+ {
+ "id": "release500_0241",
+ "category": "phonetic_balance",
+ "text": "He destroyed everything that had belonged to the woman.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0241.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE DESTROYED EVERYTHING THAT HAD BELONGED TO THE WOMAN",
+ "raw": {
+ "normalized_reference": "he destroyed everything that had belonged to the woman.",
+ "normalized_hypothesis": "he destroyed everything that had belonged to the woman",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he destroyed everything that had belonged to the woman",
+ "normalized_hypothesis": "he destroyed everything that had belonged to the woman",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0242",
+ "category": "phonetic_balance",
+ "text": "Did Thorpe go to see any one in Churchill.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0242.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "DID THORPE GO TO SEE ANYONE IN CHURCHILL",
+ "raw": {
+ "normalized_reference": "did thorpe go to see any one in churchill.",
+ "normalized_hypothesis": "did thorpe go to see anyone in churchill",
+ "errors": 2,
+ "words": 9,
+ "wer": 0.2222222222222222
+ },
+ "semantic": {
+ "normalized_reference": "did thorpe go to see any one in churchill",
+ "normalized_hypothesis": "did thorpe go to see anyone in churchill",
+ "errors": 2,
+ "words": 9,
+ "wer": 0.2222222222222222
+ }
+ },
+ {
+ "id": "release500_0243",
+ "category": "phonetic_balance",
+ "text": "Many other similar disconcerting omissions will be noticed in the Manuscript.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0243.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "MANY OTHER SIMILAR DISCONCERTING OMISSIONS WILL BE NOTICED IN THE MANUSCRIPT",
+ "raw": {
+ "normalized_reference": "many other similar disconcerting omissions will be noticed in the manuscript.",
+ "normalized_hypothesis": "many other similar disconcerting omissions will be noticed in the manuscript",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "many other similar disconcerting omissions will be noticed in the manuscript",
+ "normalized_hypothesis": "many other similar disconcerting omissions will be noticed in the manuscript",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0244",
+ "category": "phonetic_balance",
+ "text": "The ringing of the big bell aroused him.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0244.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE RINGING OF THE BIG BELL AROUSED HIM",
+ "raw": {
+ "normalized_reference": "the ringing of the big bell aroused him.",
+ "normalized_hypothesis": "the ringing of the big bell aroused him",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the ringing of the big bell aroused him",
+ "normalized_hypothesis": "the ringing of the big bell aroused him",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0245",
+ "category": "phonetic_balance",
+ "text": "A sinewy hand, dripping with water, was clutching the rail.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0245.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "A SINEWY HAND DRIPPING WITH WATER WAS CLUTCHING THE RAIL",
+ "raw": {
+ "normalized_reference": "a sinewy hand, dripping with water, was clutching the rail.",
+ "normalized_hypothesis": "a sinewy hand dripping with water was clutching the rail",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "a sinewy hand dripping with water was clutching the rail",
+ "normalized_hypothesis": "a sinewy hand dripping with water was clutching the rail",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0246",
+ "category": "phonetic_balance",
+ "text": "Enters now the psychology of the situation.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0246.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "ENTERS NOW THE PSYCHOLOGY OF THE SITUATION",
+ "raw": {
+ "normalized_reference": "enters now the psychology of the situation.",
+ "normalized_hypothesis": "enters now the psychology of the situation",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "enters now the psychology of the situation",
+ "normalized_hypothesis": "enters now the psychology of the situation",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0247",
+ "category": "phonetic_balance",
+ "text": "Wash your hands of me.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0247.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WASH YOUR HANDS OF ME",
+ "raw": {
+ "normalized_reference": "wash your hands of me.",
+ "normalized_hypothesis": "wash your hands of me",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "wash your hands of me",
+ "normalized_hypothesis": "wash your hands of me",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0248",
+ "category": "phonetic_balance",
+ "text": "He confessed that the sketch had startled him.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0248.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE CONFESSED THAT THE SKETCH HAD STARTLED HIM",
+ "raw": {
+ "normalized_reference": "he confessed that the sketch had startled him.",
+ "normalized_hypothesis": "he confessed that the sketch had startled him",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he confessed that the sketch had startled him",
+ "normalized_hypothesis": "he confessed that the sketch had startled him",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0249",
+ "category": "phonetic_balance",
+ "text": "Her face was against his breast.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0249.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HER FACE WAS AGAINST HIS BREAST",
+ "raw": {
+ "normalized_reference": "her face was against his breast.",
+ "normalized_hypothesis": "her face was against his breast",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "her face was against his breast",
+ "normalized_hypothesis": "her face was against his breast",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0250",
+ "category": "phonetic_balance",
+ "text": "He was a pariah; a wanderer without a friend or a home.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0250.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE WAS A PARIAH A WANDERER WITHOUT A FRIEND OR A HOME",
+ "raw": {
+ "normalized_reference": "he was a pariah; a wanderer without a friend or a home.",
+ "normalized_hypothesis": "he was a pariah a wanderer without a friend or a home",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he was a pariah a wanderer without a friend or a home",
+ "normalized_hypothesis": "he was a pariah a wanderer without a friend or a home",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0251",
+ "category": "phonetic_balance",
+ "text": "A cry of joy burst from Philip's lips.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0251.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "A CRY OF JOY BURST FROM PHILIP'S LIPS",
+ "raw": {
+ "normalized_reference": "a cry of joy burst from philip's lips.",
+ "normalized_hypothesis": "a cry of joy burst from philip's lips",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "a cry of joy burst from philip is lips",
+ "normalized_hypothesis": "a cry of joy burst from philip is lips",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0252",
+ "category": "phonetic_balance",
+ "text": "And the big Persian knew of his existence before he did of hers.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0252.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "AND THE BIG PERSIAN KNEW OF HIS EXISTENCE BEFORE HE DID OF HERS",
+ "raw": {
+ "normalized_reference": "and the big persian knew of his existence before he did of hers.",
+ "normalized_hypothesis": "and the big persian knew of his existence before he did of hers",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "and the big persian knew of his existence before he did of hers",
+ "normalized_hypothesis": "and the big persian knew of his existence before he did of hers",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0253",
+ "category": "phonetic_balance",
+ "text": "The men stared into each other's face.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0253.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE MEN STARED INTO EACH OTHER'S FACE",
+ "raw": {
+ "normalized_reference": "the men stared into each other's face.",
+ "normalized_hypothesis": "the men stared into each other's face",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the men stared into each other is face",
+ "normalized_hypothesis": "the men stared into each other is face",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0254",
+ "category": "phonetic_balance",
+ "text": "It is the fire, partly, she said.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0254.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "TIS THE FIRE PARTLY CHE SAID",
+ "raw": {
+ "normalized_reference": "it is the fire, partly, she said.",
+ "normalized_hypothesis": "tis the fire partly che said",
+ "errors": 3,
+ "words": 7,
+ "wer": 0.42857142857142855
+ },
+ "semantic": {
+ "normalized_reference": "it is the fire partly she said",
+ "normalized_hypothesis": "tis the fire partly che said",
+ "errors": 3,
+ "words": 7,
+ "wer": 0.42857142857142855
+ }
+ },
+ {
+ "id": "release500_0255",
+ "category": "phonetic_balance",
+ "text": "It seemed nearer to him since he had seen and talked with Gregson.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0255.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT SEEMED NEARER TO HIM SINCE HE HAD SEEN AND TALKED WITH GREGSON",
+ "raw": {
+ "normalized_reference": "it seemed nearer to him since he had seen and talked with gregson.",
+ "normalized_hypothesis": "it seemed nearer to him since he had seen and talked with gregson",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it seemed nearer to him since he had seen and talked with gregson",
+ "normalized_hypothesis": "it seemed nearer to him since he had seen and talked with gregson",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0256",
+ "category": "phonetic_balance",
+ "text": "Among my minor afflictions, I may mention a new and mysterious one.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0256.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "AMONG MY MINOR AFFLICTIONS I MAY MENTION A NEW AND MYSTERIOUS ONE",
+ "raw": {
+ "normalized_reference": "among my minor afflictions, i may mention a new and mysterious one.",
+ "normalized_hypothesis": "among my minor afflictions i may mention a new and mysterious one",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "among my minor afflictions i may mention a new and mysterious one",
+ "normalized_hypothesis": "among my minor afflictions i may mention a new and mysterious one",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0257",
+ "category": "phonetic_balance",
+ "text": "Between him and all domestic animals there must be no hostilities.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0257.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "BETWEEN THEM AND ALL DOMESTIC ANIMALS THERE MUST BE NO HOSTILITIES",
+ "raw": {
+ "normalized_reference": "between him and all domestic animals there must be no hostilities.",
+ "normalized_hypothesis": "between them and all domestic animals there must be no hostilities",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ },
+ "semantic": {
+ "normalized_reference": "between him and all domestic animals there must be no hostilities",
+ "normalized_hypothesis": "between them and all domestic animals there must be no hostilities",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ }
+ },
+ {
+ "id": "release500_0258",
+ "category": "phonetic_balance",
+ "text": "With these arguments he convinced himself that he should go on alone.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0258.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WITH THESE ARGUMENTS HE CONVINCED HIMSELF THAT HE SHOULD GO ON ALONE",
+ "raw": {
+ "normalized_reference": "with these arguments he convinced himself that he should go on alone.",
+ "normalized_hypothesis": "with these arguments he convinced himself that he should go on alone",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "with these arguments he convinced himself that he should go on alone",
+ "normalized_hypothesis": "with these arguments he convinced himself that he should go on alone",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0259",
+ "category": "phonetic_balance",
+ "text": "We are consumed in our own flesh-pots.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0259.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WE ARE CONSUMED IN OUR OWN FLESH POTS",
+ "raw": {
+ "normalized_reference": "we are consumed in our own flesh-pots.",
+ "normalized_hypothesis": "we are consumed in our own flesh pots",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "we are consumed in our own flesh pots",
+ "normalized_hypothesis": "we are consumed in our own flesh pots",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0260",
+ "category": "phonetic_balance",
+ "text": "Mercedes screamed, cried, laughed, and manifested the chaotic abandonment of hysteria.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0260.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "MERCEDES SCREAMED CRIED LAUGHED AND MANIFESTED THE CHAOTIC ABANDONMENT OF HYSTERIA",
+ "raw": {
+ "normalized_reference": "mercedes screamed, cried, laughed, and manifested the chaotic abandonment of hysteria.",
+ "normalized_hypothesis": "mercedes screamed cried laughed and manifested the chaotic abandonment of hysteria",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "mercedes screamed cried laughed and manifested the chaotic abandonment of hysteria",
+ "normalized_hypothesis": "mercedes screamed cried laughed and manifested the chaotic abandonment of hysteria",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0261",
+ "category": "phonetic_balance",
+ "text": "Philip made no effort to follow.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0261.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "PHILIP MADE NO EFFORT TO FOLLOW",
+ "raw": {
+ "normalized_reference": "philip made no effort to follow.",
+ "normalized_hypothesis": "philip made no effort to follow",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "philip made no effort to follow",
+ "normalized_hypothesis": "philip made no effort to follow",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0262",
+ "category": "phonetic_balance",
+ "text": "She said with chattering teeth.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0262.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SHE SAID WITH CHATTERING TEETH",
+ "raw": {
+ "normalized_reference": "she said with chattering teeth.",
+ "normalized_hypothesis": "she said with chattering teeth",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "she said with chattering teeth",
+ "normalized_hypothesis": "she said with chattering teeth",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0263",
+ "category": "phonetic_balance",
+ "text": "Halfway around the track one donkey got into an argument with its rider.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0263.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HALF WAY AROUND THE TRACK ONE DONKEY GOT INTO AN ARGUMENT WITH ITS RIDER",
+ "raw": {
+ "normalized_reference": "halfway around the track one donkey got into an argument with its rider.",
+ "normalized_hypothesis": "half way around the track one donkey got into an argument with its rider",
+ "errors": 2,
+ "words": 13,
+ "wer": 0.15384615384615385
+ },
+ "semantic": {
+ "normalized_reference": "halfway around the track one donkey got into an argument with its rider",
+ "normalized_hypothesis": "half way around the track one donkey got into an argument with its rider",
+ "errors": 2,
+ "words": 13,
+ "wer": 0.15384615384615385
+ }
+ },
+ {
+ "id": "release500_0264",
+ "category": "phonetic_balance",
+ "text": "Earth and gravel seemed to fill the pan.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0264.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "EARTH AND GRAVEL SEEMED TO FILL THE PEN",
+ "raw": {
+ "normalized_reference": "earth and gravel seemed to fill the pan.",
+ "normalized_hypothesis": "earth and gravel seemed to fill the pen",
+ "errors": 1,
+ "words": 8,
+ "wer": 0.125
+ },
+ "semantic": {
+ "normalized_reference": "earth and gravel seemed to fill the pan",
+ "normalized_hypothesis": "earth and gravel seemed to fill the pen",
+ "errors": 1,
+ "words": 8,
+ "wer": 0.125
+ }
+ },
+ {
+ "id": "release500_0265",
+ "category": "phonetic_balance",
+ "text": "Her achievements with cocoanuts were a revelation.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0265.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HER ACHIEVEMENTS WITH COCOA NUTS WERE A REVELATION",
+ "raw": {
+ "normalized_reference": "her achievements with cocoanuts were a revelation.",
+ "normalized_hypothesis": "her achievements with cocoa nuts were a revelation",
+ "errors": 2,
+ "words": 7,
+ "wer": 0.2857142857142857
+ },
+ "semantic": {
+ "normalized_reference": "her achievements with cocoanuts were a revelation",
+ "normalized_hypothesis": "her achievements with cocoa nuts were a revelation",
+ "errors": 2,
+ "words": 7,
+ "wer": 0.2857142857142857
+ }
+ },
+ {
+ "id": "release500_0266",
+ "category": "phonetic_balance",
+ "text": "For a full minute he crouched and listened.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0266.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "FOR A FULL MINUTE HE CROUCHED AND LISTENED",
+ "raw": {
+ "normalized_reference": "for a full minute he crouched and listened.",
+ "normalized_hypothesis": "for a full minute he crouched and listened",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "for a full minute he crouched and listened",
+ "normalized_hypothesis": "for a full minute he crouched and listened",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0267",
+ "category": "phonetic_balance",
+ "text": "It was edged with ice.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0267.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT WAS EDGED WITH ICE",
+ "raw": {
+ "normalized_reference": "it was edged with ice.",
+ "normalized_hypothesis": "it was edged with ice",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it was edged with ice",
+ "normalized_hypothesis": "it was edged with ice",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0268",
+ "category": "phonetic_balance",
+ "text": "The Portuguese boy passed the Hawaiian.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0268.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE PORTUGUESE BOY PASSED THE HAWAIAN",
+ "raw": {
+ "normalized_reference": "the portuguese boy passed the hawaiian.",
+ "normalized_hypothesis": "the portuguese boy passed the hawaian",
+ "errors": 1,
+ "words": 6,
+ "wer": 0.16666666666666666
+ },
+ "semantic": {
+ "normalized_reference": "the portuguese boy passed the hawaiian",
+ "normalized_hypothesis": "the portuguese boy passed the hawaian",
+ "errors": 1,
+ "words": 6,
+ "wer": 0.16666666666666666
+ }
+ },
+ {
+ "id": "release500_0269",
+ "category": "phonetic_balance",
+ "text": "But he did not broach it, preferring to mature it carefully.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0269.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "BUT HE DID NOT BROACH IT PREFERRING TO MATURE IT CAREFULLY",
+ "raw": {
+ "normalized_reference": "but he did not broach it, preferring to mature it carefully.",
+ "normalized_hypothesis": "but he did not broach it preferring to mature it carefully",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "but he did not broach it preferring to mature it carefully",
+ "normalized_hypothesis": "but he did not broach it preferring to mature it carefully",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0270",
+ "category": "phonetic_balance",
+ "text": "Come on, Del Mar challenged.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0270.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "COME ALONG DALMAR CHALLENGED",
+ "raw": {
+ "normalized_reference": "come on, del mar challenged.",
+ "normalized_hypothesis": "come along dalmar challenged",
+ "errors": 3,
+ "words": 5,
+ "wer": 0.6
+ },
+ "semantic": {
+ "normalized_reference": "come on del mar challenged",
+ "normalized_hypothesis": "come along dalmar challenged",
+ "errors": 3,
+ "words": 5,
+ "wer": 0.6
+ }
+ },
+ {
+ "id": "release500_0271",
+ "category": "phonetic_balance",
+ "text": "For a full half minute Jeanne looked at him without speaking.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0271.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "FOR A FULL HALF MINUTE JEAN LOOKED AT HIM WITHOUT SPEAKING",
+ "raw": {
+ "normalized_reference": "for a full half minute jeanne looked at him without speaking.",
+ "normalized_hypothesis": "for a full half minute jean looked at him without speaking",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ },
+ "semantic": {
+ "normalized_reference": "for a full half minute jeanne looked at him without speaking",
+ "normalized_hypothesis": "for a full half minute jean looked at him without speaking",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ }
+ },
+ {
+ "id": "release500_0272",
+ "category": "phonetic_balance",
+ "text": "He bore no grudges and had few enemies.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0272.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE BORE NO GRUDGES AND HAD FEW ENEMIES",
+ "raw": {
+ "normalized_reference": "he bore no grudges and had few enemies.",
+ "normalized_hypothesis": "he bore no grudges and had few enemies",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he bore no grudges and had few enemies",
+ "normalized_hypothesis": "he bore no grudges and had few enemies",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0273",
+ "category": "phonetic_balance",
+ "text": "I'll go over tomorrow afternoon.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0273.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I'LL GO OVER TO MORROW AFTERNOON",
+ "raw": {
+ "normalized_reference": "i'll go over tomorrow afternoon.",
+ "normalized_hypothesis": "i'll go over to morrow afternoon",
+ "errors": 2,
+ "words": 5,
+ "wer": 0.4
+ },
+ "semantic": {
+ "normalized_reference": "i will go over tomorrow afternoon",
+ "normalized_hypothesis": "i will go over to morrow afternoon",
+ "errors": 2,
+ "words": 6,
+ "wer": 0.3333333333333333
+ }
+ },
+ {
+ "id": "release500_0274",
+ "category": "phonetic_balance",
+ "text": "Well, there are better men in Hawaii, that's all.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0274.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WELL THERE ARE BETTER MEN IN HAWAI'I THAT'S ALL",
+ "raw": {
+ "normalized_reference": "well, there are better men in hawaii, that's all.",
+ "normalized_hypothesis": "well there are better men in hawai'i that's all",
+ "errors": 1,
+ "words": 9,
+ "wer": 0.1111111111111111
+ },
+ "semantic": {
+ "normalized_reference": "well there are better men in hawaii that is all",
+ "normalized_hypothesis": "well there are better men in hawai i that is all",
+ "errors": 2,
+ "words": 10,
+ "wer": 0.2
+ }
+ },
+ {
+ "id": "release500_0275",
+ "category": "phonetic_balance",
+ "text": "He was an athlete and a giant.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0275.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "KE WAS AN ATHLETE AND A GIANT",
+ "raw": {
+ "normalized_reference": "he was an athlete and a giant.",
+ "normalized_hypothesis": "ke was an athlete and a giant",
+ "errors": 1,
+ "words": 7,
+ "wer": 0.14285714285714285
+ },
+ "semantic": {
+ "normalized_reference": "he was an athlete and a giant",
+ "normalized_hypothesis": "ke was an athlete and a giant",
+ "errors": 1,
+ "words": 7,
+ "wer": 0.14285714285714285
+ }
+ },
+ {
+ "id": "release500_0276",
+ "category": "phonetic_balance",
+ "text": "McVeigh when he returns from a trip to Honolulu.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0276.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "MAC VAY WHEN HE RETURNS FROM A TRIP TO HONOLULU",
+ "raw": {
+ "normalized_reference": "mcveigh when he returns from a trip to honolulu.",
+ "normalized_hypothesis": "mac vay when he returns from a trip to honolulu",
+ "errors": 2,
+ "words": 9,
+ "wer": 0.2222222222222222
+ },
+ "semantic": {
+ "normalized_reference": "mcveigh when he returns from a trip to honolulu",
+ "normalized_hypothesis": "mac vay when he returns from a trip to honolulu",
+ "errors": 2,
+ "words": 9,
+ "wer": 0.2222222222222222
+ }
+ },
+ {
+ "id": "release500_0277",
+ "category": "phonetic_balance",
+ "text": "You're a devil for fighting, and will surely win.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0277.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "YOU'RE A DEVIL FOR FIGHTING AND WE'LL SURELY WIN",
+ "raw": {
+ "normalized_reference": "you're a devil for fighting, and will surely win.",
+ "normalized_hypothesis": "you're a devil for fighting and we'll surely win",
+ "errors": 1,
+ "words": 9,
+ "wer": 0.1111111111111111
+ },
+ "semantic": {
+ "normalized_reference": "you are a devil for fighting and will surely win",
+ "normalized_hypothesis": "you are a devil for fighting and we will surely win",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ }
+ },
+ {
+ "id": "release500_0278",
+ "category": "phonetic_balance",
+ "text": "All an appearance can know is mirage.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0278.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "ALL AN APPEARANCE CAN KNOW IS MORAGE",
+ "raw": {
+ "normalized_reference": "all an appearance can know is mirage.",
+ "normalized_hypothesis": "all an appearance can know is morage",
+ "errors": 1,
+ "words": 7,
+ "wer": 0.14285714285714285
+ },
+ "semantic": {
+ "normalized_reference": "all an appearance can know is mirage",
+ "normalized_hypothesis": "all an appearance can know is morage",
+ "errors": 1,
+ "words": 7,
+ "wer": 0.14285714285714285
+ }
+ },
+ {
+ "id": "release500_0279",
+ "category": "phonetic_balance",
+ "text": "You fired me out of your house, in short.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0279.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "YOU FIRED ME OUT OF YOUR HOUSE IN SHORT",
+ "raw": {
+ "normalized_reference": "you fired me out of your house, in short.",
+ "normalized_hypothesis": "you fired me out of your house in short",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "you fired me out of your house in short",
+ "normalized_hypothesis": "you fired me out of your house in short",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0280",
+ "category": "phonetic_balance",
+ "text": "He made sure that the magazine was loaded, and resumed his paddling.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0280.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE MADE SURE THAT THE MAGAZINE WAS LOADED AND RESUMED HIS PADDLING",
+ "raw": {
+ "normalized_reference": "he made sure that the magazine was loaded, and resumed his paddling.",
+ "normalized_hypothesis": "he made sure that the magazine was loaded and resumed his paddling",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he made sure that the magazine was loaded and resumed his paddling",
+ "normalized_hypothesis": "he made sure that the magazine was loaded and resumed his paddling",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0281",
+ "category": "phonetic_balance",
+ "text": "Change chairs, Daylight commanded.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0281.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "CHANGE CHAIRS DAYLIGHT COMMANDED",
+ "raw": {
+ "normalized_reference": "change chairs, daylight commanded.",
+ "normalized_hypothesis": "change chairs daylight commanded",
+ "errors": 0,
+ "words": 4,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "change chairs daylight commanded",
+ "normalized_hypothesis": "change chairs daylight commanded",
+ "errors": 0,
+ "words": 4,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0282",
+ "category": "phonetic_balance",
+ "text": "By that answer my professional medical prestige stood or fell.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0282.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "BY THAT ANSWER MY PROFESSIONAL MEDICAL PRESTIGE STODOR FELL",
+ "raw": {
+ "normalized_reference": "by that answer my professional medical prestige stood or fell.",
+ "normalized_hypothesis": "by that answer my professional medical prestige stodor fell",
+ "errors": 2,
+ "words": 10,
+ "wer": 0.2
+ },
+ "semantic": {
+ "normalized_reference": "by that answer my professional medical prestige stood or fell",
+ "normalized_hypothesis": "by that answer my professional medical prestige stodor fell",
+ "errors": 2,
+ "words": 10,
+ "wer": 0.2
+ }
+ },
+ {
+ "id": "release500_0283",
+ "category": "phonetic_balance",
+ "text": "The moon had already begun its westward decline.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0283.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE MOON HAD ALREADY BEGUN ITS WESTWARD DECLINE",
+ "raw": {
+ "normalized_reference": "the moon had already begun its westward decline.",
+ "normalized_hypothesis": "the moon had already begun its westward decline",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the moon had already begun its westward decline",
+ "normalized_hypothesis": "the moon had already begun its westward decline",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0284",
+ "category": "phonetic_balance",
+ "text": "At sea, Monday, March 16, 1908.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0284.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "AT SE MONDAY MARCH SIXTEENTH ONE THOUSAND NINE HUNDRED EIGHT",
+ "raw": {
+ "normalized_reference": "at sea, monday, march 16, 1908.",
+ "normalized_hypothesis": "at se monday march sixteenth one thousand nine hundred eight",
+ "errors": 7,
+ "words": 6,
+ "wer": 1.1666666666666667
+ },
+ "semantic": {
+ "normalized_reference": "at sea monday march 16 1908",
+ "normalized_hypothesis": "at se monday march 16th 1908",
+ "errors": 2,
+ "words": 6,
+ "wer": 0.3333333333333333
+ }
+ },
+ {
+ "id": "release500_0285",
+ "category": "phonetic_balance",
+ "text": "And wherever I ranged, the way lay along alcohol-drenched roads.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0285.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "AND WHEREVER I RANGED THE WAY LAY ALONG ALCOHAL DRENCHED ROADS",
+ "raw": {
+ "normalized_reference": "and wherever i ranged, the way lay along alcohol-drenched roads.",
+ "normalized_hypothesis": "and wherever i ranged the way lay along alcohal drenched roads",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ },
+ "semantic": {
+ "normalized_reference": "and wherever i ranged the way lay along alcohol drenched roads",
+ "normalized_hypothesis": "and wherever i ranged the way lay along alcohal drenched roads",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ }
+ },
+ {
+ "id": "release500_0286",
+ "category": "phonetic_balance",
+ "text": "The date was nearly eighteen years old.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0286.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE DATE WAS NEARLY EIGHTEEN YEARS OLD",
+ "raw": {
+ "normalized_reference": "the date was nearly eighteen years old.",
+ "normalized_hypothesis": "the date was nearly eighteen years old",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the date was nearly 18 years old",
+ "normalized_hypothesis": "the date was nearly 18 years old",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0287",
+ "category": "phonetic_balance",
+ "text": "Very early in my life, I separated from my mother.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0287.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "VERY EARLY IN MY LIFE I SEPARATED FROM MY MOTHER",
+ "raw": {
+ "normalized_reference": "very early in my life, i separated from my mother.",
+ "normalized_hypothesis": "very early in my life i separated from my mother",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "very early in my life i separated from my mother",
+ "normalized_hypothesis": "very early in my life i separated from my mother",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0288",
+ "category": "phonetic_balance",
+ "text": "There was pride and strength, the ring of triumph in his voice.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0288.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THERE WAS PRIDE AND STRENGTH THE RING OF TRIUMPH IN HIS VOICE",
+ "raw": {
+ "normalized_reference": "there was pride and strength, the ring of triumph in his voice.",
+ "normalized_hypothesis": "there was pride and strength the ring of triumph in his voice",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "there was pride and strength the ring of triumph in his voice",
+ "normalized_hypothesis": "there was pride and strength the ring of triumph in his voice",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0289",
+ "category": "phonetic_balance",
+ "text": "From my earliest recollection my sleep was a period of terror.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0289.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "FROM MY EARLIEST RECOLLECTION MY SLEEP WAS A PERIOD OF TERROR",
+ "raw": {
+ "normalized_reference": "from my earliest recollection my sleep was a period of terror.",
+ "normalized_hypothesis": "from my earliest recollection my sleep was a period of terror",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "from my earliest recollection my sleep was a period of terror",
+ "normalized_hypothesis": "from my earliest recollection my sleep was a period of terror",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0290",
+ "category": "phonetic_balance",
+ "text": "We'll have to watch our chances.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0290.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WE'LL HAVE TO WATCH OUR CHANCES",
+ "raw": {
+ "normalized_reference": "we'll have to watch our chances.",
+ "normalized_hypothesis": "we'll have to watch our chances",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "we will have to watch our chances",
+ "normalized_hypothesis": "we will have to watch our chances",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0291",
+ "category": "phonetic_balance",
+ "text": "Close beside him gleamed the white fangs of the wolf-dog.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0291.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "CLOSE BESIDE HIM BLEANED THE WHITE FANGS OF THE WOLF DOG",
+ "raw": {
+ "normalized_reference": "close beside him gleamed the white fangs of the wolf-dog.",
+ "normalized_hypothesis": "close beside him bleaned the white fangs of the wolf dog",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ },
+ "semantic": {
+ "normalized_reference": "close beside him gleamed the white fangs of the wolf dog",
+ "normalized_hypothesis": "close beside him bleaned the white fangs of the wolf dog",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ }
+ },
+ {
+ "id": "release500_0292",
+ "category": "phonetic_balance",
+ "text": "Would you be satisfied with that one hundredth part of me.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0292.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WOULD YOU BE SATISFIED WITH THAT ONE HUNDREDTH PART OF ME",
+ "raw": {
+ "normalized_reference": "would you be satisfied with that one hundredth part of me.",
+ "normalized_hypothesis": "would you be satisfied with that one hundredth part of me",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "would you be satisfied with that 100th part of me",
+ "normalized_hypothesis": "would you be satisfied with that 100th part of me",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0293",
+ "category": "phonetic_balance",
+ "text": "Two gigantic owls were tearing at the carcass.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0293.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "TWO GIGANTIC OWLS WERE TEARING AT THE CARCASE",
+ "raw": {
+ "normalized_reference": "two gigantic owls were tearing at the carcass.",
+ "normalized_hypothesis": "two gigantic owls were tearing at the carcase",
+ "errors": 1,
+ "words": 8,
+ "wer": 0.125
+ },
+ "semantic": {
+ "normalized_reference": "2 gigantic owls were tearing at the carcass",
+ "normalized_hypothesis": "2 gigantic owls were tearing at the carcase",
+ "errors": 1,
+ "words": 8,
+ "wer": 0.125
+ }
+ },
+ {
+ "id": "release500_0294",
+ "category": "phonetic_balance",
+ "text": "The lush vegetation of that sheltered spot make a natural shield.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0294.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE LUSH VEGETATION OF THAT SHELTERED SPOT MAKE A NATURAL SHIELD",
+ "raw": {
+ "normalized_reference": "the lush vegetation of that sheltered spot make a natural shield.",
+ "normalized_hypothesis": "the lush vegetation of that sheltered spot make a natural shield",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the lush vegetation of that sheltered spot make a natural shield",
+ "normalized_hypothesis": "the lush vegetation of that sheltered spot make a natural shield",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0295",
+ "category": "phonetic_balance",
+ "text": "Below him the shadow was broken into a pool of rippling starlight.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0295.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "BELOW HIM THE SHADOW WAS BROKEN INTO A POOL OF RIPPLING STARLIGHT",
+ "raw": {
+ "normalized_reference": "below him the shadow was broken into a pool of rippling starlight.",
+ "normalized_hypothesis": "below him the shadow was broken into a pool of rippling starlight",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "below him the shadow was broken into a pool of rippling starlight",
+ "normalized_hypothesis": "below him the shadow was broken into a pool of rippling starlight",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0296",
+ "category": "phonetic_balance",
+ "text": "Lots of men take women buggy riding.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0296.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "LOTS OF MEN TAKE WOMEN BUGGY RIDING",
+ "raw": {
+ "normalized_reference": "lots of men take women buggy riding.",
+ "normalized_hypothesis": "lots of men take women buggy riding",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "lots of men take women buggy riding",
+ "normalized_hypothesis": "lots of men take women buggy riding",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0297",
+ "category": "phonetic_balance",
+ "text": "But it contributed to the smash.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0297.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "BUT IT CONTRIBUTED TO THI SMASH",
+ "raw": {
+ "normalized_reference": "but it contributed to the smash.",
+ "normalized_hypothesis": "but it contributed to thi smash",
+ "errors": 1,
+ "words": 6,
+ "wer": 0.16666666666666666
+ },
+ "semantic": {
+ "normalized_reference": "but it contributed to the smash",
+ "normalized_hypothesis": "but it contributed to thi smash",
+ "errors": 1,
+ "words": 6,
+ "wer": 0.16666666666666666
+ }
+ },
+ {
+ "id": "release500_0298",
+ "category": "phonetic_balance",
+ "text": "No man ate of the seal meat or the oil.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0298.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "NO MAN ATE OF THE SEAL MEAT OR THE OIL",
+ "raw": {
+ "normalized_reference": "no man ate of the seal meat or the oil.",
+ "normalized_hypothesis": "no man ate of the seal meat or the oil",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "no man ate of the seal meat or the oil",
+ "normalized_hypothesis": "no man ate of the seal meat or the oil",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0299",
+ "category": "phonetic_balance",
+ "text": "You're going in for grab sharing.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0299.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "YOU'RE GOING IN FOR A GRAB SHARING",
+ "raw": {
+ "normalized_reference": "you're going in for grab sharing.",
+ "normalized_hypothesis": "you're going in for a grab sharing",
+ "errors": 1,
+ "words": 6,
+ "wer": 0.16666666666666666
+ },
+ "semantic": {
+ "normalized_reference": "you are going in for grab sharing",
+ "normalized_hypothesis": "you are going in for a grab sharing",
+ "errors": 1,
+ "words": 7,
+ "wer": 0.14285714285714285
+ }
+ },
+ {
+ "id": "release500_0300",
+ "category": "phonetic_balance",
+ "text": "He it was that lived to found the family of the Patino.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0300.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE IT WAS THAT LIVED TO FOUND THE FAMILY OF THE PATINO",
+ "raw": {
+ "normalized_reference": "he it was that lived to found the family of the patino.",
+ "normalized_hypothesis": "he it was that lived to found the family of the patino",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he it was that lived to found the family of the patino",
+ "normalized_hypothesis": "he it was that lived to found the family of the patino",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0301",
+ "category": "phonetic_balance",
+ "text": "For such countries nothing remained but reorganization.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0301.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "FOR SUCH COUNTRIES NOTHING REMAINED BUT REORGANIZATION",
+ "raw": {
+ "normalized_reference": "for such countries nothing remained but reorganization.",
+ "normalized_hypothesis": "for such countries nothing remained but reorganization",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "for such countries nothing remained but reorganization",
+ "normalized_hypothesis": "for such countries nothing remained but reorganization",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0302",
+ "category": "phonetic_balance",
+ "text": "Stand off butcher and baker and all the rest.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0302.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "STAND OFF BUTCHER AND BAKER AND ALL THE REST",
+ "raw": {
+ "normalized_reference": "stand off butcher and baker and all the rest.",
+ "normalized_hypothesis": "stand off butcher and baker and all the rest",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "stand off butcher and baker and all the rest",
+ "normalized_hypothesis": "stand off butcher and baker and all the rest",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0303",
+ "category": "phonetic_balance",
+ "text": "All eyes, however, were staring at him in certitude of expectancy.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0303.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "ALL EYES HOWEVER WERE STARING AT HIM IN CERTITUDE OF EXPECTANCY",
+ "raw": {
+ "normalized_reference": "all eyes, however, were staring at him in certitude of expectancy.",
+ "normalized_hypothesis": "all eyes however were staring at him in certitude of expectancy",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "all eyes however were staring at him in certitude of expectancy",
+ "normalized_hypothesis": "all eyes however were staring at him in certitude of expectancy",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0304",
+ "category": "phonetic_balance",
+ "text": "Two days ago Jeanne learned where her father's men were hiding.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0304.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "TWO DAYS AGO JEAN LEARNED WHERE HER FATHER'S MEN WERE HIDING",
+ "raw": {
+ "normalized_reference": "two days ago jeanne learned where her father's men were hiding.",
+ "normalized_hypothesis": "two days ago jean learned where her father's men were hiding",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ },
+ "semantic": {
+ "normalized_reference": "2 days ago jeanne learned where her father is men were hiding",
+ "normalized_hypothesis": "2 days ago jean learned where her father is men were hiding",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ }
+ },
+ {
+ "id": "release500_0305",
+ "category": "phonetic_balance",
+ "text": "His previous wives had never lived long enough to bear him children.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0305.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HIS PREVIOUS WIVES HAD NEVER LIVED LONG ENOUGH TO BEAR HIM CHILDREN",
+ "raw": {
+ "normalized_reference": "his previous wives had never lived long enough to bear him children.",
+ "normalized_hypothesis": "his previous wives had never lived long enough to bear him children",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "his previous wives had never lived long enough to bear him children",
+ "normalized_hypothesis": "his previous wives had never lived long enough to bear him children",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0306",
+ "category": "phonetic_balance",
+ "text": "But this little defect did not worry him.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0306.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "BUT THIS LITTLE DEFICT DID NOT WORRY HIM",
+ "raw": {
+ "normalized_reference": "but this little defect did not worry him.",
+ "normalized_hypothesis": "but this little defict did not worry him",
+ "errors": 1,
+ "words": 8,
+ "wer": 0.125
+ },
+ "semantic": {
+ "normalized_reference": "but this little defect did not worry him",
+ "normalized_hypothesis": "but this little defict did not worry him",
+ "errors": 1,
+ "words": 8,
+ "wer": 0.125
+ }
+ },
+ {
+ "id": "release500_0307",
+ "category": "phonetic_balance",
+ "text": "Eli Harding asked, as Shunk started to follow.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0307.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "ELI HARDING ASKED AS SCHUNK STARTED TO FOLLOW",
+ "raw": {
+ "normalized_reference": "eli harding asked, as shunk started to follow.",
+ "normalized_hypothesis": "eli harding asked as schunk started to follow",
+ "errors": 1,
+ "words": 8,
+ "wer": 0.125
+ },
+ "semantic": {
+ "normalized_reference": "eli harding asked as shunk started to follow",
+ "normalized_hypothesis": "eli harding asked as schunk started to follow",
+ "errors": 1,
+ "words": 8,
+ "wer": 0.125
+ }
+ },
+ {
+ "id": "release500_0308",
+ "category": "phonetic_balance",
+ "text": "And this was their sole conversation throughout the meal.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0308.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "AND THIS WAS THEIR SOLE CONVERSATION THROUGHOUT THE MEAL",
+ "raw": {
+ "normalized_reference": "and this was their sole conversation throughout the meal.",
+ "normalized_hypothesis": "and this was their sole conversation throughout the meal",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "and this was their sole conversation throughout the meal",
+ "normalized_hypothesis": "and this was their sole conversation throughout the meal",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0309",
+ "category": "phonetic_balance",
+ "text": "Then, again, Tudor had such an irritating way about him.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0309.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THEN AGAIN TUDER HAD SUCH AN IRRITATING WAY ABOUT HIM",
+ "raw": {
+ "normalized_reference": "then, again, tudor had such an irritating way about him.",
+ "normalized_hypothesis": "then again tuder had such an irritating way about him",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ },
+ "semantic": {
+ "normalized_reference": "then again tudor had such an irritating way about him",
+ "normalized_hypothesis": "then again tuder had such an irritating way about him",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ }
+ },
+ {
+ "id": "release500_0310",
+ "category": "phonetic_balance",
+ "text": "In it there was something that was almost tragedy.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0310.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IN IT THERE WAS SOMETHING THAT WAS ALMOST TRAGEDY",
+ "raw": {
+ "normalized_reference": "in it there was something that was almost tragedy.",
+ "normalized_hypothesis": "in it there was something that was almost tragedy",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "in it there was something that was almost tragedy",
+ "normalized_hypothesis": "in it there was something that was almost tragedy",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0311",
+ "category": "phonetic_balance",
+ "text": "Thirty pounds, said the captain with finality.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0311.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THIRTY POUNDS SAID THE CAPTAIN WITH FINALITY",
+ "raw": {
+ "normalized_reference": "thirty pounds, said the captain with finality.",
+ "normalized_hypothesis": "thirty pounds said the captain with finality",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "\u00a330 said the captain with finality",
+ "normalized_hypothesis": "\u00a330 said the captain with finality",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0312",
+ "category": "phonetic_balance",
+ "text": "It was over when he made his way through the ring of spectators.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0312.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT WAS OVER WHEN HE MADE HIS WAY THROUGH THE REIGN OF SPETATORS",
+ "raw": {
+ "normalized_reference": "it was over when he made his way through the ring of spectators.",
+ "normalized_hypothesis": "it was over when he made his way through the reign of spetators",
+ "errors": 2,
+ "words": 13,
+ "wer": 0.15384615384615385
+ },
+ "semantic": {
+ "normalized_reference": "it was over when he made his way through the ring of spectators",
+ "normalized_hypothesis": "it was over when he made his way through the reign of spetators",
+ "errors": 2,
+ "words": 13,
+ "wer": 0.15384615384615385
+ }
+ },
+ {
+ "id": "release500_0313",
+ "category": "phonetic_balance",
+ "text": "She had died from cold and starvation.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0313.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SHE HAD DIED FROM COLD AT STARVATION",
+ "raw": {
+ "normalized_reference": "she had died from cold and starvation.",
+ "normalized_hypothesis": "she had died from cold at starvation",
+ "errors": 1,
+ "words": 7,
+ "wer": 0.14285714285714285
+ },
+ "semantic": {
+ "normalized_reference": "she had died from cold and starvation",
+ "normalized_hypothesis": "she had died from cold at starvation",
+ "errors": 1,
+ "words": 7,
+ "wer": 0.14285714285714285
+ }
+ },
+ {
+ "id": "release500_0314",
+ "category": "phonetic_balance",
+ "text": "We fished sharks on Niihau together.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0314.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WE FISHED SHARKS ON NIHOW TOGETHER",
+ "raw": {
+ "normalized_reference": "we fished sharks on niihau together.",
+ "normalized_hypothesis": "we fished sharks on nihow together",
+ "errors": 1,
+ "words": 6,
+ "wer": 0.16666666666666666
+ },
+ "semantic": {
+ "normalized_reference": "we fished sharks on niihau together",
+ "normalized_hypothesis": "we fished sharks on nihow together",
+ "errors": 1,
+ "words": 6,
+ "wer": 0.16666666666666666
+ }
+ },
+ {
+ "id": "release500_0315",
+ "category": "phonetic_balance",
+ "text": "His slim hands gripped the edges of the table.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0315.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HIS SLIM HANDS GRIPPED THE EDGES OF THE TABLE",
+ "raw": {
+ "normalized_reference": "his slim hands gripped the edges of the table.",
+ "normalized_hypothesis": "his slim hands gripped the edges of the table",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "his slim hands gripped the edges of the table",
+ "normalized_hypothesis": "his slim hands gripped the edges of the table",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0316",
+ "category": "phonetic_balance",
+ "text": "We had been chased by them ourselves, more than once.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0316.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WE HAD BEEN CHASED BY THEM OURSELVES MORE THAN ONCE",
+ "raw": {
+ "normalized_reference": "we had been chased by them ourselves, more than once.",
+ "normalized_hypothesis": "we had been chased by them ourselves more than once",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "we had been chased by them ourselves more than once",
+ "normalized_hypothesis": "we had been chased by them ourselves more than once",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0317",
+ "category": "phonetic_balance",
+ "text": "Or have they already devised one.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0317.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "OR HAVE THEY ALREADY DEVISED ONE",
+ "raw": {
+ "normalized_reference": "or have they already devised one.",
+ "normalized_hypothesis": "or have they already devised one",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "or have they already devised one",
+ "normalized_hypothesis": "or have they already devised one",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0318",
+ "category": "phonetic_balance",
+ "text": "I cannot follow you, she said.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0318.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I CANNOT FOLLOW YOU SHE SAID",
+ "raw": {
+ "normalized_reference": "i cannot follow you, she said.",
+ "normalized_hypothesis": "i cannot follow you she said",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i cannot follow you she said",
+ "normalized_hypothesis": "i cannot follow you she said",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0319",
+ "category": "phonetic_balance",
+ "text": "He went down in midstream, searching the shadows of both shores.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0319.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "KEY WENT DOWN IN MIDSTREAM SEARCHING THE SHADOWS OF BOTH SHORES",
+ "raw": {
+ "normalized_reference": "he went down in midstream, searching the shadows of both shores.",
+ "normalized_hypothesis": "key went down in midstream searching the shadows of both shores",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ },
+ "semantic": {
+ "normalized_reference": "he went down in midstream searching the shadows of both shores",
+ "normalized_hypothesis": "key went down in midstream searching the shadows of both shores",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ }
+ },
+ {
+ "id": "release500_0320",
+ "category": "phonetic_balance",
+ "text": "I was sick once -- typhoid.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0320.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I WAS SICK ONCE TYPHOID",
+ "raw": {
+ "normalized_reference": "i was sick once -- typhoid.",
+ "normalized_hypothesis": "i was sick once typhoid",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i was sick once typhoid",
+ "normalized_hypothesis": "i was sick once typhoid",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0321",
+ "category": "phonetic_balance",
+ "text": "Why not like any railroad station or ferry depot.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0321.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WHY NOT LIKE ANY RAILROAD STATION OR FURRY DEPOT",
+ "raw": {
+ "normalized_reference": "why not like any railroad station or ferry depot.",
+ "normalized_hypothesis": "why not like any railroad station or furry depot",
+ "errors": 1,
+ "words": 9,
+ "wer": 0.1111111111111111
+ },
+ "semantic": {
+ "normalized_reference": "why not like any railroad station or ferry depot",
+ "normalized_hypothesis": "why not like any railroad station or furry depot",
+ "errors": 1,
+ "words": 9,
+ "wer": 0.1111111111111111
+ }
+ },
+ {
+ "id": "release500_0322",
+ "category": "phonetic_balance",
+ "text": "Red-Eye swung back and forth on the branch farther down.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0322.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "RED EYE SWUNG BACK AND FORTH ON THE BRANCH FARTHER DOWN",
+ "raw": {
+ "normalized_reference": "red-eye swung back and forth on the branch farther down.",
+ "normalized_hypothesis": "red eye swung back and forth on the branch farther down",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "red eye swung back and forth on the branch farther down",
+ "normalized_hypothesis": "red eye swung back and forth on the branch farther down",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0323",
+ "category": "phonetic_balance",
+ "text": "Within himself he called it no longer his own.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0323.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WITHIN HIMSELF HE CALLED IT NO LONGER HIS OWN",
+ "raw": {
+ "normalized_reference": "within himself he called it no longer his own.",
+ "normalized_hypothesis": "within himself he called it no longer his own",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "within himself he called it no longer his own",
+ "normalized_hypothesis": "within himself he called it no longer his own",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0324",
+ "category": "phonetic_balance",
+ "text": "It lasted as a deterrent for two days.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0324.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT LASTED AS A DETERRENT FOR TWO DAYS",
+ "raw": {
+ "normalized_reference": "it lasted as a deterrent for two days.",
+ "normalized_hypothesis": "it lasted as a deterrent for two days",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it lasted as a deterrent for 2 days",
+ "normalized_hypothesis": "it lasted as a deterrent for 2 days",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0325",
+ "category": "phonetic_balance",
+ "text": "He loved to play Chinese lottery.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0325.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE LOVED TO PLAY CHINESE LOTTERY",
+ "raw": {
+ "normalized_reference": "he loved to play chinese lottery.",
+ "normalized_hypothesis": "he loved to play chinese lottery",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he loved to play chinese lottery",
+ "normalized_hypothesis": "he loved to play chinese lottery",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0326",
+ "category": "phonetic_balance",
+ "text": "Lord, but I'm glad to see you again, Phil.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0326.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "LORD BUT I'M GLAD TO SEE YOU AGAIN PHIL",
+ "raw": {
+ "normalized_reference": "lord, but i'm glad to see you again, phil.",
+ "normalized_hypothesis": "lord but i'm glad to see you again phil",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "lord but i am glad to see you again phil",
+ "normalized_hypothesis": "lord but i am glad to see you again phil",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0327",
+ "category": "phonetic_balance",
+ "text": "There is not an iota of truth in it, certainly not.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0327.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THERE IS NOT AN IOTA OF TRUTH IN IT CERTAINLY NOT",
+ "raw": {
+ "normalized_reference": "there is not an iota of truth in it, certainly not.",
+ "normalized_hypothesis": "there is not an iota of truth in it certainly not",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "there is not an iota of truth in it certainly not",
+ "normalized_hypothesis": "there is not an iota of truth in it certainly not",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0328",
+ "category": "phonetic_balance",
+ "text": "She added, with genuine sympathy in her face and voice.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0328.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SHE ADDED WITH GENUINE SYMPATHY IN HER FACE AND VOICE",
+ "raw": {
+ "normalized_reference": "she added, with genuine sympathy in her face and voice.",
+ "normalized_hypothesis": "she added with genuine sympathy in her face and voice",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "she added with genuine sympathy in her face and voice",
+ "normalized_hypothesis": "she added with genuine sympathy in her face and voice",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0329",
+ "category": "phonetic_balance",
+ "text": "We have plenty of capital ourselves, and yet we want more.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0329.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WE HAVE PLENTY OF CAPITAL OURSELVES AND YET WE WANT MORE",
+ "raw": {
+ "normalized_reference": "we have plenty of capital ourselves, and yet we want more.",
+ "normalized_hypothesis": "we have plenty of capital ourselves and yet we want more",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "we have plenty of capital ourselves and yet we want more",
+ "normalized_hypothesis": "we have plenty of capital ourselves and yet we want more",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0330",
+ "category": "phonetic_balance",
+ "text": "And now put yourself in my place for a moment.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0330.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "AND NOW PUT YOURSELF IN MY PLACE FOR A MOMENT",
+ "raw": {
+ "normalized_reference": "and now put yourself in my place for a moment.",
+ "normalized_hypothesis": "and now put yourself in my place for a moment",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "and now put yourself in my place for a moment",
+ "normalized_hypothesis": "and now put yourself in my place for a moment",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0331",
+ "category": "phonetic_balance",
+ "text": "Fast, but endure.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0331.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "FAST BUT ENDURE",
+ "raw": {
+ "normalized_reference": "fast, but endure.",
+ "normalized_hypothesis": "fast but endure",
+ "errors": 0,
+ "words": 3,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "fast but endure",
+ "normalized_hypothesis": "fast but endure",
+ "errors": 0,
+ "words": 3,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0332",
+ "category": "phonetic_balance",
+ "text": "You were looking squeamish this afternoon, he began.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0332.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "YOU WERE LOOKING SQUEAMISH THIS AFTERNOON HE BEGAN",
+ "raw": {
+ "normalized_reference": "you were looking squeamish this afternoon, he began.",
+ "normalized_hypothesis": "you were looking squeamish this afternoon he began",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "you were looking squeamish this afternoon he began",
+ "normalized_hypothesis": "you were looking squeamish this afternoon he began",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0333",
+ "category": "phonetic_balance",
+ "text": "It is the nearest refuge.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0333.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IS THE NEAREST REFUGE",
+ "raw": {
+ "normalized_reference": "it is the nearest refuge.",
+ "normalized_hypothesis": "is the nearest refuge",
+ "errors": 1,
+ "words": 5,
+ "wer": 0.2
+ },
+ "semantic": {
+ "normalized_reference": "it is the nearest refuge",
+ "normalized_hypothesis": "is the nearest refuge",
+ "errors": 1,
+ "words": 5,
+ "wer": 0.2
+ }
+ },
+ {
+ "id": "release500_0334",
+ "category": "phonetic_balance",
+ "text": "I want to die in it.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0334.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I WANT TO DIE IN IT",
+ "raw": {
+ "normalized_reference": "i want to die in it.",
+ "normalized_hypothesis": "i want to die in it",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i want to die in it",
+ "normalized_hypothesis": "i want to die in it",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0335",
+ "category": "phonetic_balance",
+ "text": "He stopped, and Philip nodded at the horrified question in his eyes.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0335.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "KEY STOPPED AND PHILIP NODDED AT THE HORRIFIED QUESTION IN HIS EYES",
+ "raw": {
+ "normalized_reference": "he stopped, and philip nodded at the horrified question in his eyes.",
+ "normalized_hypothesis": "key stopped and philip nodded at the horrified question in his eyes",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "he stopped and philip nodded at the horrified question in his eyes",
+ "normalized_hypothesis": "key stopped and philip nodded at the horrified question in his eyes",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ }
+ },
+ {
+ "id": "release500_0336",
+ "category": "phonetic_balance",
+ "text": "She turned, fearing that Jacques might see what was in her face.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0336.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SHE TURNED FEARING THAT JACK MIGHT SEE WHAT WAS IN HER FACE",
+ "raw": {
+ "normalized_reference": "she turned, fearing that jacques might see what was in her face.",
+ "normalized_hypothesis": "she turned fearing that jack might see what was in her face",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "she turned fearing that jacques might see what was in her face",
+ "normalized_hypothesis": "she turned fearing that jack might see what was in her face",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ }
+ },
+ {
+ "id": "release500_0337",
+ "category": "phonetic_balance",
+ "text": "It was a curious coincidence.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0337.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT WAS A CURIOUS COINCIDENCE",
+ "raw": {
+ "normalized_reference": "it was a curious coincidence.",
+ "normalized_hypothesis": "it was a curious coincidence",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it was a curious coincidence",
+ "normalized_hypothesis": "it was a curious coincidence",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0338",
+ "category": "phonetic_balance",
+ "text": "Daylight made no answer, and the door closed behind him.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0338.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "DAYLIGHT MADE NO ANSWER AND THE DOOR CLOSED BEHIND HIM",
+ "raw": {
+ "normalized_reference": "daylight made no answer, and the door closed behind him.",
+ "normalized_hypothesis": "daylight made no answer and the door closed behind him",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "daylight made no answer and the door closed behind him",
+ "normalized_hypothesis": "daylight made no answer and the door closed behind him",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0339",
+ "category": "phonetic_balance",
+ "text": "And he did hurt my arm.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0339.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "AND HE DID HURT MY ARM",
+ "raw": {
+ "normalized_reference": "and he did hurt my arm.",
+ "normalized_hypothesis": "and he did hurt my arm",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "and he did hurt my arm",
+ "normalized_hypothesis": "and he did hurt my arm",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0340",
+ "category": "phonetic_balance",
+ "text": "Dunham, can your boy go along with Jesse.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0340.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "DUNAM CAN YEUR BOY GO ALONG WITH JESSE",
+ "raw": {
+ "normalized_reference": "dunham, can your boy go along with jesse.",
+ "normalized_hypothesis": "dunam can yeur boy go along with jesse",
+ "errors": 2,
+ "words": 8,
+ "wer": 0.25
+ },
+ "semantic": {
+ "normalized_reference": "dunham can your boy go along with jesse",
+ "normalized_hypothesis": "dunam can yeur boy go along with jesse",
+ "errors": 2,
+ "words": 8,
+ "wer": 0.25
+ }
+ },
+ {
+ "id": "release500_0341",
+ "category": "phonetic_balance",
+ "text": "The creative joy, I murmured.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0341.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE CREATIVE JOY I MURMURED",
+ "raw": {
+ "normalized_reference": "the creative joy, i murmured.",
+ "normalized_hypothesis": "the creative joy i murmured",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the creative joy i murmured",
+ "normalized_hypothesis": "the creative joy i murmured",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0342",
+ "category": "phonetic_balance",
+ "text": "He pulled, and the log crashed down to break his back.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0342.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "KEY PULLED AND THE LOG CRASHED DOWN TO BREAK HIS BACK",
+ "raw": {
+ "normalized_reference": "he pulled, and the log crashed down to break his back.",
+ "normalized_hypothesis": "key pulled and the log crashed down to break his back",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ },
+ "semantic": {
+ "normalized_reference": "he pulled and the log crashed down to break his back",
+ "normalized_hypothesis": "key pulled and the log crashed down to break his back",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ }
+ },
+ {
+ "id": "release500_0343",
+ "category": "phonetic_balance",
+ "text": "By golly, the boy wins.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0343.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I CALLE THE BOY WINS",
+ "raw": {
+ "normalized_reference": "by golly, the boy wins.",
+ "normalized_hypothesis": "i calle the boy wins",
+ "errors": 2,
+ "words": 5,
+ "wer": 0.4
+ },
+ "semantic": {
+ "normalized_reference": "by golly the boy wins",
+ "normalized_hypothesis": "i calle the boy wins",
+ "errors": 2,
+ "words": 5,
+ "wer": 0.4
+ }
+ },
+ {
+ "id": "release500_0344",
+ "category": "phonetic_balance",
+ "text": "The boy threw back his head with pride.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0344.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE BOY THREW BACK HIS HEAD WITH PRIDE",
+ "raw": {
+ "normalized_reference": "the boy threw back his head with pride.",
+ "normalized_hypothesis": "the boy threw back his head with pride",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the boy threw back his head with pride",
+ "normalized_hypothesis": "the boy threw back his head with pride",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0345",
+ "category": "phonetic_balance",
+ "text": "The last one I knew was an overseer.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0345.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE LAST ONE I KNEW WAS AN OVERSEER",
+ "raw": {
+ "normalized_reference": "the last one i knew was an overseer.",
+ "normalized_hypothesis": "the last one i knew was an overseer",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the last one i knew was an overseer",
+ "normalized_hypothesis": "the last one i knew was an overseer",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0346",
+ "category": "phonetic_balance",
+ "text": "Two years ago I gave up civilization for this.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0346.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "TWO YEARS AGO I GAVE UP CIVILIZATION FOR THIS",
+ "raw": {
+ "normalized_reference": "two years ago i gave up civilization for this.",
+ "normalized_hypothesis": "two years ago i gave up civilization for this",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "2 years ago i gave up civilization for this",
+ "normalized_hypothesis": "2 years ago i gave up civilization for this",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0347",
+ "category": "phonetic_balance",
+ "text": "The river bared its bosom, and snorting steamboats challenged the wilderness.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0347.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE RIVER BARED ITS BOSOM AND SNORTING STEAMBOATS CHALLENGED THE WILDERNESS",
+ "raw": {
+ "normalized_reference": "the river bared its bosom, and snorting steamboats challenged the wilderness.",
+ "normalized_hypothesis": "the river bared its bosom and snorting steamboats challenged the wilderness",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the river bared its bosom and snorting steamboats challenged the wilderness",
+ "normalized_hypothesis": "the river bared its bosom and snorting steamboats challenged the wilderness",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0348",
+ "category": "phonetic_balance",
+ "text": "Mab, she said.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0348.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "MAB CHE SAID",
+ "raw": {
+ "normalized_reference": "mab, she said.",
+ "normalized_hypothesis": "mab che said",
+ "errors": 1,
+ "words": 3,
+ "wer": 0.3333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "mab she said",
+ "normalized_hypothesis": "mab che said",
+ "errors": 1,
+ "words": 3,
+ "wer": 0.3333333333333333
+ }
+ },
+ {
+ "id": "release500_0349",
+ "category": "phonetic_balance",
+ "text": "The regret in his voice was provocative of a second burst of laughter.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0349.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE REGRET IN HIS VOICE WAS PROVOCATIVE OF A SECOND BURST OF LAUGHTER",
+ "raw": {
+ "normalized_reference": "the regret in his voice was provocative of a second burst of laughter.",
+ "normalized_hypothesis": "the regret in his voice was provocative of a second burst of laughter",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the regret in his voice was provocative of a 2nd burst of laughter",
+ "normalized_hypothesis": "the regret in his voice was provocative of a 2nd burst of laughter",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0350",
+ "category": "phonetic_balance",
+ "text": "And each year something happened, and I did not go.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0350.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "AND EACH YEAR SOMETHING HAPPENED AND I DID NOT GO",
+ "raw": {
+ "normalized_reference": "and each year something happened, and i did not go.",
+ "normalized_hypothesis": "and each year something happened and i did not go",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "and each year something happened and i did not go",
+ "normalized_hypothesis": "and each year something happened and i did not go",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0351",
+ "category": "phonetic_balance",
+ "text": "Gad, your letter came just in time.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0351.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "GAD YOUR LETTER CAME JUST IN TIME",
+ "raw": {
+ "normalized_reference": "gad, your letter came just in time.",
+ "normalized_hypothesis": "gad your letter came just in time",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "gad your letter came just in time",
+ "normalized_hypothesis": "gad your letter came just in time",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0352",
+ "category": "phonetic_balance",
+ "text": "But if Pierre did not return, until tomorrow.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0352.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "BUT IF PIERRE DID NOT RETURN UNTIL TO MORROW",
+ "raw": {
+ "normalized_reference": "but if pierre did not return, until tomorrow.",
+ "normalized_hypothesis": "but if pierre did not return until to morrow",
+ "errors": 2,
+ "words": 8,
+ "wer": 0.25
+ },
+ "semantic": {
+ "normalized_reference": "but if pierre did not return until tomorrow",
+ "normalized_hypothesis": "but if pierre did not return until to morrow",
+ "errors": 2,
+ "words": 8,
+ "wer": 0.25
+ }
+ },
+ {
+ "id": "release500_0353",
+ "category": "phonetic_balance",
+ "text": "Billinger may arrive in time.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0353.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "DILINGER MAY ARRIVE IN TIME",
+ "raw": {
+ "normalized_reference": "billinger may arrive in time.",
+ "normalized_hypothesis": "dilinger may arrive in time",
+ "errors": 1,
+ "words": 5,
+ "wer": 0.2
+ },
+ "semantic": {
+ "normalized_reference": "billinger may arrive in time",
+ "normalized_hypothesis": "dilinger may arrive in time",
+ "errors": 1,
+ "words": 5,
+ "wer": 0.2
+ }
+ },
+ {
+ "id": "release500_0354",
+ "category": "phonetic_balance",
+ "text": "In such a tumbling of values was no time to sell.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0354.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IN SUCH A TUMBLING OF VALUES WAS NO TIME TO SELL",
+ "raw": {
+ "normalized_reference": "in such a tumbling of values was no time to sell.",
+ "normalized_hypothesis": "in such a tumbling of values was no time to sell",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "in such a tumbling of values was no time to sell",
+ "normalized_hypothesis": "in such a tumbling of values was no time to sell",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0355",
+ "category": "phonetic_balance",
+ "text": "There's not an iota of truth in it.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0355.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THERE'S NOT AN IODA OF TRUTH IN IT",
+ "raw": {
+ "normalized_reference": "there's not an iota of truth in it.",
+ "normalized_hypothesis": "there's not an ioda of truth in it",
+ "errors": 1,
+ "words": 8,
+ "wer": 0.125
+ },
+ "semantic": {
+ "normalized_reference": "there is not an iota of truth in it",
+ "normalized_hypothesis": "there is not an ioda of truth in it",
+ "errors": 1,
+ "words": 9,
+ "wer": 0.1111111111111111
+ }
+ },
+ {
+ "id": "release500_0356",
+ "category": "phonetic_balance",
+ "text": "It is dog eat dog, and you ate them up.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0356.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "DIS DOG EAT DOG AND YOU EAT THEM U",
+ "raw": {
+ "normalized_reference": "it is dog eat dog, and you ate them up.",
+ "normalized_hypothesis": "dis dog eat dog and you eat them u",
+ "errors": 4,
+ "words": 10,
+ "wer": 0.4
+ },
+ "semantic": {
+ "normalized_reference": "it is dog eat dog and you ate them up",
+ "normalized_hypothesis": "dis dog eat dog and you eat them u",
+ "errors": 4,
+ "words": 10,
+ "wer": 0.4
+ }
+ },
+ {
+ "id": "release500_0357",
+ "category": "phonetic_balance",
+ "text": "You can take a vacation on pay.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0357.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "YOU CAN TAKE A VACATION ON PAY",
+ "raw": {
+ "normalized_reference": "you can take a vacation on pay.",
+ "normalized_hypothesis": "you can take a vacation on pay",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "you can take a vacation on pay",
+ "normalized_hypothesis": "you can take a vacation on pay",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0358",
+ "category": "phonetic_balance",
+ "text": "I have seen myself that one man contemplated by Pascal's philosophic eye.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0358.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I HAVE SEEN MYSELF THAT ONE MAN CONTEMPLATED BY PASCOLL'S PHILOSOPHIC EYE",
+ "raw": {
+ "normalized_reference": "i have seen myself that one man contemplated by pascal's philosophic eye.",
+ "normalized_hypothesis": "i have seen myself that one man contemplated by pascoll's philosophic eye",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "i have seen myself that one man contemplated by pascal is philosophic eye",
+ "normalized_hypothesis": "i have seen myself that one man contemplated by pascoll is philosophic eye",
+ "errors": 1,
+ "words": 13,
+ "wer": 0.07692307692307693
+ }
+ },
+ {
+ "id": "release500_0359",
+ "category": "phonetic_balance",
+ "text": "Ah, I had forgotten, he exclaimed.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0359.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "A I HAD FORGOTTEN KEY EXCLAIMED",
+ "raw": {
+ "normalized_reference": "ah, i had forgotten, he exclaimed.",
+ "normalized_hypothesis": "a i had forgotten key exclaimed",
+ "errors": 2,
+ "words": 6,
+ "wer": 0.3333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "ah i had forgotten he exclaimed",
+ "normalized_hypothesis": "a i had forgotten key exclaimed",
+ "errors": 2,
+ "words": 6,
+ "wer": 0.3333333333333333
+ }
+ },
+ {
+ "id": "release500_0360",
+ "category": "phonetic_balance",
+ "text": "Providence had delivered him through the maelstrom.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0360.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "PROVIDENCE HAD DELIVERED HIM THROUGH THE MAILSTROM",
+ "raw": {
+ "normalized_reference": "providence had delivered him through the maelstrom.",
+ "normalized_hypothesis": "providence had delivered him through the mailstrom",
+ "errors": 1,
+ "words": 7,
+ "wer": 0.14285714285714285
+ },
+ "semantic": {
+ "normalized_reference": "providence had delivered him through the maelstrom",
+ "normalized_hypothesis": "providence had delivered him through the mailstrom",
+ "errors": 1,
+ "words": 7,
+ "wer": 0.14285714285714285
+ }
+ },
+ {
+ "id": "release500_0361",
+ "category": "phonetic_balance",
+ "text": "She was his now, forever.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0361.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SHE WAS HIS NOW FOR EVER",
+ "raw": {
+ "normalized_reference": "she was his now, forever.",
+ "normalized_hypothesis": "she was his now for ever",
+ "errors": 2,
+ "words": 5,
+ "wer": 0.4
+ },
+ "semantic": {
+ "normalized_reference": "she was his now forever",
+ "normalized_hypothesis": "she was his now for ever",
+ "errors": 2,
+ "words": 5,
+ "wer": 0.4
+ }
+ },
+ {
+ "id": "release500_0362",
+ "category": "phonetic_balance",
+ "text": "It was unobtrusive, yet it was there.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0362.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT WAS UNOBTRUSIVE YET IT WAS THERE",
+ "raw": {
+ "normalized_reference": "it was unobtrusive, yet it was there.",
+ "normalized_hypothesis": "it was unobtrusive yet it was there",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it was unobtrusive yet it was there",
+ "normalized_hypothesis": "it was unobtrusive yet it was there",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0363",
+ "category": "phonetic_balance",
+ "text": "Men like Joe Goose dated existence from drunk to drunk.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0363.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "MEN LIKE JOE GOOSE DATED EXISTENCE FROM DRUNK TO DRUNK",
+ "raw": {
+ "normalized_reference": "men like joe goose dated existence from drunk to drunk.",
+ "normalized_hypothesis": "men like joe goose dated existence from drunk to drunk",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "men like joe goose dated existence from drunk to drunk",
+ "normalized_hypothesis": "men like joe goose dated existence from drunk to drunk",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0364",
+ "category": "phonetic_balance",
+ "text": "I was completely lost in my work.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0364.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I WAS COMPLETELY LOST IN MY WORK",
+ "raw": {
+ "normalized_reference": "i was completely lost in my work.",
+ "normalized_hypothesis": "i was completely lost in my work",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i was completely lost in my work",
+ "normalized_hypothesis": "i was completely lost in my work",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0365",
+ "category": "phonetic_balance",
+ "text": "The very thing, Ernest agreed.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0365.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE VERY THING ERNEST AGREED",
+ "raw": {
+ "normalized_reference": "the very thing, ernest agreed.",
+ "normalized_hypothesis": "the very thing ernest agreed",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the very thing ernest agreed",
+ "normalized_hypothesis": "the very thing ernest agreed",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0366",
+ "category": "phonetic_balance",
+ "text": "He was the leader, and Tudor was his lieutenant.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0366.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE WAS THE LEADER AND TUDOR WAS HIS LIEUTENANT",
+ "raw": {
+ "normalized_reference": "he was the leader, and tudor was his lieutenant.",
+ "normalized_hypothesis": "he was the leader and tudor was his lieutenant",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he was the leader and tudor was his lieutenant",
+ "normalized_hypothesis": "he was the leader and tudor was his lieutenant",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0367",
+ "category": "phonetic_balance",
+ "text": "The Fire-Men wore animal skins around their waists and across their shoulders.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0367.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE FIREMEN WORE ANIMALS SKINS AROUND THEIR WAISTS AND ACROSS THEIR SHOULDERS",
+ "raw": {
+ "normalized_reference": "the fire-men wore animal skins around their waists and across their shoulders.",
+ "normalized_hypothesis": "the firemen wore animals skins around their waists and across their shoulders",
+ "errors": 3,
+ "words": 13,
+ "wer": 0.23076923076923078
+ },
+ "semantic": {
+ "normalized_reference": "the fire men wore animal skins around their waists and across their shoulders",
+ "normalized_hypothesis": "the firemen wore animals skins around their waists and across their shoulders",
+ "errors": 3,
+ "words": 13,
+ "wer": 0.23076923076923078
+ }
+ },
+ {
+ "id": "release500_0368",
+ "category": "phonetic_balance",
+ "text": "We could throw stones with our feet.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0368.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WE COULD THROW STONES WITH OUR FEET",
+ "raw": {
+ "normalized_reference": "we could throw stones with our feet.",
+ "normalized_hypothesis": "we could throw stones with our feet",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "we could throw stones with our feet",
+ "normalized_hypothesis": "we could throw stones with our feet",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0369",
+ "category": "phonetic_balance",
+ "text": "He turned sharply, and faced Gregson across the table.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0369.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE TURNED SHARPLY AND FACED GREGSON ACROSS THE TABLE",
+ "raw": {
+ "normalized_reference": "he turned sharply, and faced gregson across the table.",
+ "normalized_hypothesis": "he turned sharply and faced gregson across the table",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he turned sharply and faced gregson across the table",
+ "normalized_hypothesis": "he turned sharply and faced gregson across the table",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0370",
+ "category": "phonetic_balance",
+ "text": "They are to attack your camp tomorrow night.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0370.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THEY ARE TO ATTACK YOUR CAMP TO MORROW NIGHT",
+ "raw": {
+ "normalized_reference": "they are to attack your camp tomorrow night.",
+ "normalized_hypothesis": "they are to attack your camp to morrow night",
+ "errors": 2,
+ "words": 8,
+ "wer": 0.25
+ },
+ "semantic": {
+ "normalized_reference": "they are to attack your camp tomorrow night",
+ "normalized_hypothesis": "they are to attack your camp to morrow night",
+ "errors": 2,
+ "words": 8,
+ "wer": 0.25
+ }
+ },
+ {
+ "id": "release500_0371",
+ "category": "phonetic_balance",
+ "text": "And watch out for wet feet, was his parting advice.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0371.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "AND WATCH OUT FOR WET FEET WAS HIS PARTING ADVICE",
+ "raw": {
+ "normalized_reference": "and watch out for wet feet, was his parting advice.",
+ "normalized_hypothesis": "and watch out for wet feet was his parting advice",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "and watch out for wet feet was his parting advice",
+ "normalized_hypothesis": "and watch out for wet feet was his parting advice",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0372",
+ "category": "phonetic_balance",
+ "text": "Two weeks passed, and in that time Thorpe left camp three times.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0372.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "TWO WEEKS PASSED AND IN THAT TIME THORPE LEFT CAMP THREE TIMES",
+ "raw": {
+ "normalized_reference": "two weeks passed, and in that time thorpe left camp three times.",
+ "normalized_hypothesis": "two weeks passed and in that time thorpe left camp three times",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "2 weeks passed and in that time thorpe left camp 3 times",
+ "normalized_hypothesis": "2 weeks passed and in that time thorpe left camp 3 times",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0373",
+ "category": "phonetic_balance",
+ "text": "But ever his gaze returned to that Crouched Venus on the piano.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0373.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "BUT EVER HIS GAZE RETURNED TO THAT CROUCHED VENUS ON THE PIANO",
+ "raw": {
+ "normalized_reference": "but ever his gaze returned to that crouched venus on the piano.",
+ "normalized_hypothesis": "but ever his gaze returned to that crouched venus on the piano",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "but ever his gaze returned to that crouched venus on the piano",
+ "normalized_hypothesis": "but ever his gaze returned to that crouched venus on the piano",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0374",
+ "category": "phonetic_balance",
+ "text": "But she had become an automaton.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0374.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "BUT SHE HAD BECOME AN AUTOMATAN",
+ "raw": {
+ "normalized_reference": "but she had become an automaton.",
+ "normalized_hypothesis": "but she had become an automatan",
+ "errors": 1,
+ "words": 6,
+ "wer": 0.16666666666666666
+ },
+ "semantic": {
+ "normalized_reference": "but she had become an automaton",
+ "normalized_hypothesis": "but she had become an automatan",
+ "errors": 1,
+ "words": 6,
+ "wer": 0.16666666666666666
+ }
+ },
+ {
+ "id": "release500_0375",
+ "category": "phonetic_balance",
+ "text": "I do not blame you for anything; remember that.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0375.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I DO NOT BLAME YOU FOR ANYTHING REMEMBER THAT",
+ "raw": {
+ "normalized_reference": "i do not blame you for anything; remember that.",
+ "normalized_hypothesis": "i do not blame you for anything remember that",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i do not blame you for anything remember that",
+ "normalized_hypothesis": "i do not blame you for anything remember that",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0376",
+ "category": "phonetic_balance",
+ "text": "The next thing to watch out for is bed sores.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0376.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE NEXT THING TO WATCH OUT FOR HIS BEDSORES",
+ "raw": {
+ "normalized_reference": "the next thing to watch out for is bed sores.",
+ "normalized_hypothesis": "the next thing to watch out for his bedsores",
+ "errors": 3,
+ "words": 10,
+ "wer": 0.3
+ },
+ "semantic": {
+ "normalized_reference": "the next thing to watch out for is bed sores",
+ "normalized_hypothesis": "the next thing to watch out for his bedsores",
+ "errors": 3,
+ "words": 10,
+ "wer": 0.3
+ }
+ },
+ {
+ "id": "release500_0377",
+ "category": "phonetic_balance",
+ "text": "He will follow us soon.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0377.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE WILL FOLLOW US SOON",
+ "raw": {
+ "normalized_reference": "he will follow us soon.",
+ "normalized_hypothesis": "he will follow us soon",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he will follow us soon",
+ "normalized_hypothesis": "he will follow us soon",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0378",
+ "category": "phonetic_balance",
+ "text": "His beady black eyes saw bargains where other men saw bankruptcy.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0378.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HIS BEADY BLACK EYES SAW BARGAINS WHERE OTHER MEN SAW BANKRUPTCY",
+ "raw": {
+ "normalized_reference": "his beady black eyes saw bargains where other men saw bankruptcy.",
+ "normalized_hypothesis": "his beady black eyes saw bargains where other men saw bankruptcy",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "his beady black eyes saw bargains where other men saw bankruptcy",
+ "normalized_hypothesis": "his beady black eyes saw bargains where other men saw bankruptcy",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0379",
+ "category": "phonetic_balance",
+ "text": "They robbed me a few years later.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0379.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THEY ROBBED ME A FEW YEARS LATER",
+ "raw": {
+ "normalized_reference": "they robbed me a few years later.",
+ "normalized_hypothesis": "they robbed me a few years later",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "they robbed me a few years later",
+ "normalized_hypothesis": "they robbed me a few years later",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0380",
+ "category": "phonetic_balance",
+ "text": "Gregson held a lighted match until it burnt his fingertips.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0380.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "CRIGSON HELD TA LIGHTED MATCH UNTIL IT BURNED HIS FINGER TIPS",
+ "raw": {
+ "normalized_reference": "gregson held a lighted match until it burnt his fingertips.",
+ "normalized_hypothesis": "crigson held ta lighted match until it burned his finger tips",
+ "errors": 5,
+ "words": 10,
+ "wer": 0.5
+ },
+ "semantic": {
+ "normalized_reference": "gregson held a lighted match until it burnt his fingertips",
+ "normalized_hypothesis": "crigson held ta lighted match until it burned his finger tips",
+ "errors": 5,
+ "words": 10,
+ "wer": 0.5
+ }
+ },
+ {
+ "id": "release500_0381",
+ "category": "phonetic_balance",
+ "text": "A burst of laughter was his reward.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0381.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "A BURST OF LAUGHTER WAS HIS REWARD",
+ "raw": {
+ "normalized_reference": "a burst of laughter was his reward.",
+ "normalized_hypothesis": "a burst of laughter was his reward",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "a burst of laughter was his reward",
+ "normalized_hypothesis": "a burst of laughter was his reward",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0382",
+ "category": "phonetic_balance",
+ "text": "He cried, and swung the club wildly.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0382.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "KE CRIED AND SWUNG THE CLUB WILDLY",
+ "raw": {
+ "normalized_reference": "he cried, and swung the club wildly.",
+ "normalized_hypothesis": "ke cried and swung the club wildly",
+ "errors": 1,
+ "words": 7,
+ "wer": 0.14285714285714285
+ },
+ "semantic": {
+ "normalized_reference": "he cried and swung the club wildly",
+ "normalized_hypothesis": "ke cried and swung the club wildly",
+ "errors": 1,
+ "words": 7,
+ "wer": 0.14285714285714285
+ }
+ },
+ {
+ "id": "release500_0383",
+ "category": "phonetic_balance",
+ "text": "Sandel would never become a world champion.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0383.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SANDALL WOULD NEVER BECOME A WORLD CHAMPION",
+ "raw": {
+ "normalized_reference": "sandel would never become a world champion.",
+ "normalized_hypothesis": "sandall would never become a world champion",
+ "errors": 1,
+ "words": 7,
+ "wer": 0.14285714285714285
+ },
+ "semantic": {
+ "normalized_reference": "sandel would never become a world champion",
+ "normalized_hypothesis": "sandall would never become a world champion",
+ "errors": 1,
+ "words": 7,
+ "wer": 0.14285714285714285
+ }
+ },
+ {
+ "id": "release500_0384",
+ "category": "phonetic_balance",
+ "text": "And after the bath a shave would not be bad.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0384.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "AND AFTER THE BATHA SHAVE WOULD NOT BE BAD",
+ "raw": {
+ "normalized_reference": "and after the bath a shave would not be bad.",
+ "normalized_hypothesis": "and after the batha shave would not be bad",
+ "errors": 2,
+ "words": 10,
+ "wer": 0.2
+ },
+ "semantic": {
+ "normalized_reference": "and after the bath a shave would not be bad",
+ "normalized_hypothesis": "and after the batha shave would not be bad",
+ "errors": 2,
+ "words": 10,
+ "wer": 0.2
+ }
+ },
+ {
+ "id": "release500_0385",
+ "category": "phonetic_balance",
+ "text": "They saw each other for the first time in Boston.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0385.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THEY SAW EACH OTHER FOR THE FIRST TIME IN BOSTON",
+ "raw": {
+ "normalized_reference": "they saw each other for the first time in boston.",
+ "normalized_hypothesis": "they saw each other for the first time in boston",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "they saw each other for the 1st time in boston",
+ "normalized_hypothesis": "they saw each other for the 1st time in boston",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0386",
+ "category": "phonetic_balance",
+ "text": "I had been born with no organic, chemical predisposition toward alcohol.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0386.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I HAD BEEN BORN WITH NO ORGANIC CHEMICAL PREDISPOSITION TOWARD ALCOHOL",
+ "raw": {
+ "normalized_reference": "i had been born with no organic, chemical predisposition toward alcohol.",
+ "normalized_hypothesis": "i had been born with no organic chemical predisposition toward alcohol",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i had been born with no organic chemical predisposition toward alcohol",
+ "normalized_hypothesis": "i had been born with no organic chemical predisposition toward alcohol",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0387",
+ "category": "phonetic_balance",
+ "text": "They had no fixed values, to be altered by adjectives and adverbs.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0387.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THEY HAD NO FIXED VALUES TO BE ALTERED BY ADJECTIVES AND ADVERBS",
+ "raw": {
+ "normalized_reference": "they had no fixed values, to be altered by adjectives and adverbs.",
+ "normalized_hypothesis": "they had no fixed values to be altered by adjectives and adverbs",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "they had no fixed values to be altered by adjectives and adverbs",
+ "normalized_hypothesis": "they had no fixed values to be altered by adjectives and adverbs",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0388",
+ "category": "phonetic_balance",
+ "text": "Nor was Elam Harnish an exception.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0388.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "NOR WAS THE LAMB HARNISH AN EXCEPTION",
+ "raw": {
+ "normalized_reference": "nor was elam harnish an exception.",
+ "normalized_hypothesis": "nor was the lamb harnish an exception",
+ "errors": 2,
+ "words": 6,
+ "wer": 0.3333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "nor was elam harnish an exception",
+ "normalized_hypothesis": "nor was the lamb harnish an exception",
+ "errors": 2,
+ "words": 6,
+ "wer": 0.3333333333333333
+ }
+ },
+ {
+ "id": "release500_0389",
+ "category": "phonetic_balance",
+ "text": "He was a wise hyena.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0389.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "KE WAS A WISE HYENA",
+ "raw": {
+ "normalized_reference": "he was a wise hyena.",
+ "normalized_hypothesis": "ke was a wise hyena",
+ "errors": 1,
+ "words": 5,
+ "wer": 0.2
+ },
+ "semantic": {
+ "normalized_reference": "he was a wise hyena",
+ "normalized_hypothesis": "ke was a wise hyena",
+ "errors": 1,
+ "words": 5,
+ "wer": 0.2
+ }
+ },
+ {
+ "id": "release500_0390",
+ "category": "phonetic_balance",
+ "text": "And now behold the perversity of things.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0390.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "AND NOW BEHOLD THE PERVERSITY OF THINGS",
+ "raw": {
+ "normalized_reference": "and now behold the perversity of things.",
+ "normalized_hypothesis": "and now behold the perversity of things",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "and now behold the perversity of things",
+ "normalized_hypothesis": "and now behold the perversity of things",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0391",
+ "category": "phonetic_balance",
+ "text": "The Fire People, like ourselves, lived in caves.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0391.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE FIRE PEOPLE LIKE OURSELVES LIVED IN CAVES",
+ "raw": {
+ "normalized_reference": "the fire people, like ourselves, lived in caves.",
+ "normalized_hypothesis": "the fire people like ourselves lived in caves",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the fire people like ourselves lived in caves",
+ "normalized_hypothesis": "the fire people like ourselves lived in caves",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0392",
+ "category": "phonetic_balance",
+ "text": "To her the bridge was tambo, which is the native for taboo.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0392.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "TO HER THE BRIDGE WAS TAMBO WHICH IS THE NATIVE FOR TABOO",
+ "raw": {
+ "normalized_reference": "to her the bridge was tambo, which is the native for taboo.",
+ "normalized_hypothesis": "to her the bridge was tambo which is the native for taboo",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "to her the bridge was tambo which is the native for taboo",
+ "normalized_hypothesis": "to her the bridge was tambo which is the native for taboo",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0393",
+ "category": "phonetic_balance",
+ "text": "The voyage was our idea of a good time.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0393.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE VOYAGE WAS OUR IDEA OF A GOOD TIME",
+ "raw": {
+ "normalized_reference": "the voyage was our idea of a good time.",
+ "normalized_hypothesis": "the voyage was our idea of a good time",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the voyage was our idea of a good time",
+ "normalized_hypothesis": "the voyage was our idea of a good time",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0394",
+ "category": "phonetic_balance",
+ "text": "Of course much grumbling went on, and little outbursts were continually occurring.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0394.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "OF COURSE MUCH GRUMBLING WENT ON AND LITTLE OUTBURSTS WERE CONTINUALLY OCCURRING",
+ "raw": {
+ "normalized_reference": "of course much grumbling went on, and little outbursts were continually occurring.",
+ "normalized_hypothesis": "of course much grumbling went on and little outbursts were continually occurring",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "of course much grumbling went on and little outbursts were continually occurring",
+ "normalized_hypothesis": "of course much grumbling went on and little outbursts were continually occurring",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0395",
+ "category": "phonetic_balance",
+ "text": "Get down and dig in.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0395.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "ET DOWNIT AND DIG IN",
+ "raw": {
+ "normalized_reference": "get down and dig in.",
+ "normalized_hypothesis": "et downit and dig in",
+ "errors": 2,
+ "words": 5,
+ "wer": 0.4
+ },
+ "semantic": {
+ "normalized_reference": "get down and dig in",
+ "normalized_hypothesis": "et downit and dig in",
+ "errors": 2,
+ "words": 5,
+ "wer": 0.4
+ }
+ },
+ {
+ "id": "release500_0396",
+ "category": "phonetic_balance",
+ "text": "It was put together in a casual, helter-skelter sort of way.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0396.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT WAS PUT TOGETHER IN A CASUAL CALTESCELTER SORT OF WAY",
+ "raw": {
+ "normalized_reference": "it was put together in a casual, helter-skelter sort of way.",
+ "normalized_hypothesis": "it was put together in a casual caltescelter sort of way",
+ "errors": 2,
+ "words": 12,
+ "wer": 0.16666666666666666
+ },
+ "semantic": {
+ "normalized_reference": "it was put together in a casual helter skelter sort of way",
+ "normalized_hypothesis": "it was put together in a casual caltescelter sort of way",
+ "errors": 2,
+ "words": 12,
+ "wer": 0.16666666666666666
+ }
+ },
+ {
+ "id": "release500_0397",
+ "category": "phonetic_balance",
+ "text": "He deluged me, overwhelmed me with argument.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0397.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE DELUGED ME OVERWHELMED ME WITH ARGUMENT",
+ "raw": {
+ "normalized_reference": "he deluged me, overwhelmed me with argument.",
+ "normalized_hypothesis": "he deluged me overwhelmed me with argument",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he deluged me overwhelmed me with argument",
+ "normalized_hypothesis": "he deluged me overwhelmed me with argument",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0398",
+ "category": "phonetic_balance",
+ "text": "The girl faced him, her eyes shining with sudden fear.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0398.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE GIRL FACED HIM HER EYES SHINING WITH SUDDEN FEAR",
+ "raw": {
+ "normalized_reference": "the girl faced him, her eyes shining with sudden fear.",
+ "normalized_hypothesis": "the girl faced him her eyes shining with sudden fear",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the girl faced him her eyes shining with sudden fear",
+ "normalized_hypothesis": "the girl faced him her eyes shining with sudden fear",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0399",
+ "category": "phonetic_balance",
+ "text": "It will break our hearts and our backs to hoist anchor by hand.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0399.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT WILL BREAK OUR HEARTS AND OUR BACKS TO HOIST ANCHOR BY HAND",
+ "raw": {
+ "normalized_reference": "it will break our hearts and our backs to hoist anchor by hand.",
+ "normalized_hypothesis": "it will break our hearts and our backs to hoist anchor by hand",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it will break our hearts and our backs to hoist anchor by hand",
+ "normalized_hypothesis": "it will break our hearts and our backs to hoist anchor by hand",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0400",
+ "category": "phonetic_balance",
+ "text": "He made no reply as he waited for Whittemore to continue.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0400.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE MADE NO REPLY AS HE WAITED FOR WHITTAMORE TO CONTINUE",
+ "raw": {
+ "normalized_reference": "he made no reply as he waited for whittemore to continue.",
+ "normalized_hypothesis": "he made no reply as he waited for whittamore to continue",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ },
+ "semantic": {
+ "normalized_reference": "he made no reply as he waited for whittemore to continue",
+ "normalized_hypothesis": "he made no reply as he waited for whittamore to continue",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ }
+ },
+ {
+ "id": "release500_0401",
+ "category": "phonetic_balance",
+ "text": "Who the devil gave it to you to be judge and jury.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0401.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WHO THE DEVIL GAVE IT TO YOU TO BE JUDGE AND JURY",
+ "raw": {
+ "normalized_reference": "who the devil gave it to you to be judge and jury.",
+ "normalized_hypothesis": "who the devil gave it to you to be judge and jury",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "who the devil gave it to you to be judge and jury",
+ "normalized_hypothesis": "who the devil gave it to you to be judge and jury",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0402",
+ "category": "phonetic_balance",
+ "text": "Daylight found himself charmed and made curious by the little man.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0402.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "DAYLIGHT FOUND HIMSELF CHARMED AND MADE CURIOUS BY THE LITTLE MAN",
+ "raw": {
+ "normalized_reference": "daylight found himself charmed and made curious by the little man.",
+ "normalized_hypothesis": "daylight found himself charmed and made curious by the little man",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "daylight found himself charmed and made curious by the little man",
+ "normalized_hypothesis": "daylight found himself charmed and made curious by the little man",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0403",
+ "category": "phonetic_balance",
+ "text": "It's a Yankee, Joan cried.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0403.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT'S A YANKEE JONE CRIED",
+ "raw": {
+ "normalized_reference": "it's a yankee, joan cried.",
+ "normalized_hypothesis": "it's a yankee jone cried",
+ "errors": 1,
+ "words": 5,
+ "wer": 0.2
+ },
+ "semantic": {
+ "normalized_reference": "it is a yankee joan cried",
+ "normalized_hypothesis": "it is a yankee jone cried",
+ "errors": 1,
+ "words": 6,
+ "wer": 0.16666666666666666
+ }
+ },
+ {
+ "id": "release500_0404",
+ "category": "phonetic_balance",
+ "text": "But what they want with your toothbrush is more than I can imagine.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0404.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "BUT WHAT THEY WANT WITH YOUR TWO FRESHES MORE THAN I CAN IMAGINE",
+ "raw": {
+ "normalized_reference": "but what they want with your toothbrush is more than i can imagine.",
+ "normalized_hypothesis": "but what they want with your two freshes more than i can imagine",
+ "errors": 2,
+ "words": 13,
+ "wer": 0.15384615384615385
+ },
+ "semantic": {
+ "normalized_reference": "but what they want with your toothbrush is more than i can imagine",
+ "normalized_hypothesis": "but what they want with your 2 freshes more than i can imagine",
+ "errors": 2,
+ "words": 13,
+ "wer": 0.15384615384615385
+ }
+ },
+ {
+ "id": "release500_0405",
+ "category": "phonetic_balance",
+ "text": "They were babbling and chattering all together.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0405.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THEY WERE BABBLING AND CHATTERING ALTOGETHER",
+ "raw": {
+ "normalized_reference": "they were babbling and chattering all together.",
+ "normalized_hypothesis": "they were babbling and chattering altogether",
+ "errors": 2,
+ "words": 7,
+ "wer": 0.2857142857142857
+ },
+ "semantic": {
+ "normalized_reference": "they were babbling and chattering all together",
+ "normalized_hypothesis": "they were babbling and chattering altogether",
+ "errors": 2,
+ "words": 7,
+ "wer": 0.2857142857142857
+ }
+ },
+ {
+ "id": "release500_0406",
+ "category": "phonetic_balance",
+ "text": "Men of Selden's stamp don't stop at women and children.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0406.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "MEN OF SELDOM'S STAMP BONE STOP AT WOMEN IN CHILDREN",
+ "raw": {
+ "normalized_reference": "men of selden's stamp don't stop at women and children.",
+ "normalized_hypothesis": "men of seldom's stamp bone stop at women in children",
+ "errors": 3,
+ "words": 10,
+ "wer": 0.3
+ },
+ "semantic": {
+ "normalized_reference": "men of selden is stamp do not stop at women and children",
+ "normalized_hypothesis": "men of seldom is stamp bone stop at women in children",
+ "errors": 4,
+ "words": 12,
+ "wer": 0.3333333333333333
+ }
+ },
+ {
+ "id": "release500_0407",
+ "category": "phonetic_balance",
+ "text": "Cameron looked at his hands with their long, sinewy fingers.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0407.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "CAMERON LOOKED AT HIS HANDS WITH THEIR LONG SINEWY FINGERS",
+ "raw": {
+ "normalized_reference": "cameron looked at his hands with their long, sinewy fingers.",
+ "normalized_hypothesis": "cameron looked at his hands with their long sinewy fingers",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "cameron looked at his hands with their long sinewy fingers",
+ "normalized_hypothesis": "cameron looked at his hands with their long sinewy fingers",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0408",
+ "category": "phonetic_balance",
+ "text": "Saxon waited, for she knew a fresh idea had struck Billy.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0408.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SAXON WAITED FOR SHE KNEW A FRESH IDEA HAD STRUCK BILLY",
+ "raw": {
+ "normalized_reference": "saxon waited, for she knew a fresh idea had struck billy.",
+ "normalized_hypothesis": "saxon waited for she knew a fresh idea had struck billy",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "saxon waited for she knew a fresh idea had struck billy",
+ "normalized_hypothesis": "saxon waited for she knew a fresh idea had struck billy",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0409",
+ "category": "phonetic_balance",
+ "text": "He saw only the effect in a general, sketchy way.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0409.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "KEIS SAW ONLY THE EFFECT IN A GENERAL SKETCHY WAY",
+ "raw": {
+ "normalized_reference": "he saw only the effect in a general, sketchy way.",
+ "normalized_hypothesis": "keis saw only the effect in a general sketchy way",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ },
+ "semantic": {
+ "normalized_reference": "he saw only the effect in a general sketchy way",
+ "normalized_hypothesis": "keis saw only the effect in a general sketchy way",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ }
+ },
+ {
+ "id": "release500_0410",
+ "category": "phonetic_balance",
+ "text": "He began to follow the footprints of the dog.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0410.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE BEGAN TO FOLLOW THE FOOTPRINTS OF THE DOCK",
+ "raw": {
+ "normalized_reference": "he began to follow the footprints of the dog.",
+ "normalized_hypothesis": "he began to follow the footprints of the dock",
+ "errors": 1,
+ "words": 9,
+ "wer": 0.1111111111111111
+ },
+ "semantic": {
+ "normalized_reference": "he began to follow the footprints of the dog",
+ "normalized_hypothesis": "he began to follow the footprints of the dock",
+ "errors": 1,
+ "words": 9,
+ "wer": 0.1111111111111111
+ }
+ },
+ {
+ "id": "release500_0411",
+ "category": "phonetic_balance",
+ "text": "Do you value your hide.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0411.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "DO YOU VALUE YOUR HIDE",
+ "raw": {
+ "normalized_reference": "do you value your hide.",
+ "normalized_hypothesis": "do you value your hide",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "do you value your hide",
+ "normalized_hypothesis": "do you value your hide",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0412",
+ "category": "phonetic_balance",
+ "text": "A rising tide of fat had submerged them.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0412.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE RISING TIDE OF FAT HAD SUBMERGED THEM",
+ "raw": {
+ "normalized_reference": "a rising tide of fat had submerged them.",
+ "normalized_hypothesis": "the rising tide of fat had submerged them",
+ "errors": 1,
+ "words": 8,
+ "wer": 0.125
+ },
+ "semantic": {
+ "normalized_reference": "a rising tide of fat had submerged them",
+ "normalized_hypothesis": "the rising tide of fat had submerged them",
+ "errors": 1,
+ "words": 8,
+ "wer": 0.125
+ }
+ },
+ {
+ "id": "release500_0413",
+ "category": "phonetic_balance",
+ "text": "There was nothing on the rock.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0413.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THERE WAS NOTHING ON THE ROCK",
+ "raw": {
+ "normalized_reference": "there was nothing on the rock.",
+ "normalized_hypothesis": "there was nothing on the rock",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "there was nothing on the rock",
+ "normalized_hypothesis": "there was nothing on the rock",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0414",
+ "category": "phonetic_balance",
+ "text": "Whatever he guessed he locked away in the taboo room of Naomi.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0414.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WHATEVER HE GUESSED HE LOCKED AWAY IN THE TABOO ROOM OF NAOMI",
+ "raw": {
+ "normalized_reference": "whatever he guessed he locked away in the taboo room of naomi.",
+ "normalized_hypothesis": "whatever he guessed he locked away in the taboo room of naomi",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "whatever he guessed he locked away in the taboo room of naomi",
+ "normalized_hypothesis": "whatever he guessed he locked away in the taboo room of naomi",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0415",
+ "category": "phonetic_balance",
+ "text": "Will we ever forget it.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0415.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WON'L WE EVER FORGET IT",
+ "raw": {
+ "normalized_reference": "will we ever forget it.",
+ "normalized_hypothesis": "won'l we ever forget it",
+ "errors": 1,
+ "words": 5,
+ "wer": 0.2
+ },
+ "semantic": {
+ "normalized_reference": "will we ever forget it",
+ "normalized_hypothesis": "won l we ever forget it",
+ "errors": 2,
+ "words": 5,
+ "wer": 0.4
+ }
+ },
+ {
+ "id": "release500_0416",
+ "category": "phonetic_balance",
+ "text": "Jeanne and Pierre both gazed toward the great rock.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0416.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "JEAN AND PIERRE BOTH GAZED TOWARD THE GREAT ROCK",
+ "raw": {
+ "normalized_reference": "jeanne and pierre both gazed toward the great rock.",
+ "normalized_hypothesis": "jean and pierre both gazed toward the great rock",
+ "errors": 1,
+ "words": 9,
+ "wer": 0.1111111111111111
+ },
+ "semantic": {
+ "normalized_reference": "jeanne and pierre both gazed toward the great rock",
+ "normalized_hypothesis": "jean and pierre both gazed toward the great rock",
+ "errors": 1,
+ "words": 9,
+ "wer": 0.1111111111111111
+ }
+ },
+ {
+ "id": "release500_0417",
+ "category": "phonetic_balance",
+ "text": "Does the old boy often go off at half-cock that way.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0417.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "DOES THE OLD BOY OFTEN GO OFF AT HALFCOCK THAT WAY",
+ "raw": {
+ "normalized_reference": "does the old boy often go off at half-cock that way.",
+ "normalized_hypothesis": "does the old boy often go off at halfcock that way",
+ "errors": 2,
+ "words": 12,
+ "wer": 0.16666666666666666
+ },
+ "semantic": {
+ "normalized_reference": "does the old boy often go off at half cock that way",
+ "normalized_hypothesis": "does the old boy often go off at halfcock that way",
+ "errors": 2,
+ "words": 12,
+ "wer": 0.16666666666666666
+ }
+ },
+ {
+ "id": "release500_0418",
+ "category": "phonetic_balance",
+ "text": "The temperature dropped to fifty below zero and remained there the whole trip.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0418.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE TEMPERATURE DROPPED TO FIFTY BELOW ZERO AND REMAINED THERE THE WHOLE TRIP",
+ "raw": {
+ "normalized_reference": "the temperature dropped to fifty below zero and remained there the whole trip.",
+ "normalized_hypothesis": "the temperature dropped to fifty below zero and remained there the whole trip",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the temperature dropped to 50 below 0 and remained there the whole trip",
+ "normalized_hypothesis": "the temperature dropped to 50 below 0 and remained there the whole trip",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0419",
+ "category": "phonetic_balance",
+ "text": "It's worth eight dollars.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0419.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT'S WORTH EIGHT DOLLARS",
+ "raw": {
+ "normalized_reference": "it's worth eight dollars.",
+ "normalized_hypothesis": "it's worth eight dollars",
+ "errors": 0,
+ "words": 4,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it is worth $8",
+ "normalized_hypothesis": "it is worth $8",
+ "errors": 0,
+ "words": 4,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0420",
+ "category": "phonetic_balance",
+ "text": "But I did not enjoy it long.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0420.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "BUT I DID NOT ENJOY IT LONG",
+ "raw": {
+ "normalized_reference": "but i did not enjoy it long.",
+ "normalized_hypothesis": "but i did not enjoy it long",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "but i did not enjoy it long",
+ "normalized_hypothesis": "but i did not enjoy it long",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0421",
+ "category": "phonetic_balance",
+ "text": "Shorty turned to their employers.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0421.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SHORTY TURNED TO THEIR EMPLOYERS",
+ "raw": {
+ "normalized_reference": "shorty turned to their employers.",
+ "normalized_hypothesis": "shorty turned to their employers",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "shorty turned to their employers",
+ "normalized_hypothesis": "shorty turned to their employers",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0422",
+ "category": "phonetic_balance",
+ "text": "It was the third or fourth time that Philip had heard MacDougall swear.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0422.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT WAS THE THIRD OR FOURTH TIME THAT PHILIP HAD HEARD MAC DUBELL SWEAR",
+ "raw": {
+ "normalized_reference": "it was the third or fourth time that philip had heard macdougall swear.",
+ "normalized_hypothesis": "it was the third or fourth time that philip had heard mac dubell swear",
+ "errors": 2,
+ "words": 13,
+ "wer": 0.15384615384615385
+ },
+ "semantic": {
+ "normalized_reference": "it was the 3rd or 4th time that philip had heard macdougall swear",
+ "normalized_hypothesis": "it was the 3rd or 4th time that philip had heard mac dubell swear",
+ "errors": 2,
+ "words": 13,
+ "wer": 0.15384615384615385
+ }
+ },
+ {
+ "id": "release500_0423",
+ "category": "phonetic_balance",
+ "text": "Ernest saw in the affair the most sinister import.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0423.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "ERNEST SAW IN THE AFFAIR THE MOST SINISTER IMPORT",
+ "raw": {
+ "normalized_reference": "ernest saw in the affair the most sinister import.",
+ "normalized_hypothesis": "ernest saw in the affair the most sinister import",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "ernest saw in the affair the most sinister import",
+ "normalized_hypothesis": "ernest saw in the affair the most sinister import",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0424",
+ "category": "phonetic_balance",
+ "text": "Also, at regular intervals, he would mutter.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0424.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "ALSO AT REGULAR INTERVALS HE WOULD MUTTER",
+ "raw": {
+ "normalized_reference": "also, at regular intervals, he would mutter.",
+ "normalized_hypothesis": "also at regular intervals he would mutter",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "also at regular intervals he would mutter",
+ "normalized_hypothesis": "also at regular intervals he would mutter",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0425",
+ "category": "phonetic_balance",
+ "text": "He may anticipate the day of his death.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0425.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE MAY ANTICIPATE THE DAY OF HIS DEATH",
+ "raw": {
+ "normalized_reference": "he may anticipate the day of his death.",
+ "normalized_hypothesis": "he may anticipate the day of his death",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he may anticipate the day of his death",
+ "normalized_hypothesis": "he may anticipate the day of his death",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0426",
+ "category": "phonetic_balance",
+ "text": "I defy any man to get a Solomon Island sore in California.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0426.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I DEFY ANY MAN TO GET AS SOLOMON ISLANDS OR IN CALIFORNIA",
+ "raw": {
+ "normalized_reference": "i defy any man to get a solomon island sore in california.",
+ "normalized_hypothesis": "i defy any man to get as solomon islands or in california",
+ "errors": 3,
+ "words": 12,
+ "wer": 0.25
+ },
+ "semantic": {
+ "normalized_reference": "i defy any man to get a solomon island sore in california",
+ "normalized_hypothesis": "i defy any man to get as solomon islands or in california",
+ "errors": 3,
+ "words": 12,
+ "wer": 0.25
+ }
+ },
+ {
+ "id": "release500_0427",
+ "category": "phonetic_balance",
+ "text": "Now our figuring was all right.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0427.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "NOW OUR FIGURING WAS ALL RIGHT",
+ "raw": {
+ "normalized_reference": "now our figuring was all right.",
+ "normalized_hypothesis": "now our figuring was all right",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "now our figuring was all right",
+ "normalized_hypothesis": "now our figuring was all right",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0428",
+ "category": "phonetic_balance",
+ "text": "MacDougall tapped his forehead suspiciously with a stubby forefinger.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0428.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "MACDUGALE TAPPED HIS FOREHEAD SUSPICIOUSLY WITH A STUBBY FOREFINGER",
+ "raw": {
+ "normalized_reference": "macdougall tapped his forehead suspiciously with a stubby forefinger.",
+ "normalized_hypothesis": "macdugale tapped his forehead suspiciously with a stubby forefinger",
+ "errors": 1,
+ "words": 9,
+ "wer": 0.1111111111111111
+ },
+ "semantic": {
+ "normalized_reference": "macdougall tapped his forehead suspiciously with a stubby forefinger",
+ "normalized_hypothesis": "macdugale tapped his forehead suspiciously with a stubby forefinger",
+ "errors": 1,
+ "words": 9,
+ "wer": 0.1111111111111111
+ }
+ },
+ {
+ "id": "release500_0429",
+ "category": "phonetic_balance",
+ "text": "O'Brien took off his coat and bared his right arm.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0429.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "O'BRIEN TOOK OFF HIS COAT AND BARED HIS RIGHT ARM",
+ "raw": {
+ "normalized_reference": "o'brien took off his coat and bared his right arm.",
+ "normalized_hypothesis": "o'brien took off his coat and bared his right arm",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "0 brien took off his coat and bared his right arm",
+ "normalized_hypothesis": "0 brien took off his coat and bared his right arm",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0430",
+ "category": "phonetic_balance",
+ "text": "But it won't continue, she said with easy confidence.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0430.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "BUT IT WON'T CONTINUE SHE SAID WITH EASY CONFIDENCE",
+ "raw": {
+ "normalized_reference": "but it won't continue, she said with easy confidence.",
+ "normalized_hypothesis": "but it won't continue she said with easy confidence",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "but it will not continue she said with easy confidence",
+ "normalized_hypothesis": "but it will not continue she said with easy confidence",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0431",
+ "category": "phonetic_balance",
+ "text": "I was still weak from my prolonged immersion.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0431.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I WAS STILL WEAK FROM MY PROLONGED IMMERSION",
+ "raw": {
+ "normalized_reference": "i was still weak from my prolonged immersion.",
+ "normalized_hypothesis": "i was still weak from my prolonged immersion",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i was still weak from my prolonged immersion",
+ "normalized_hypothesis": "i was still weak from my prolonged immersion",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0432",
+ "category": "phonetic_balance",
+ "text": "Oolong was two hundred and fifty miles from the nearest land.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0432.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "ULON WAS TWO HUNDRED AND FIFTY MILES FROM THE NEAREST LAND",
+ "raw": {
+ "normalized_reference": "oolong was two hundred and fifty miles from the nearest land.",
+ "normalized_hypothesis": "ulon was two hundred and fifty miles from the nearest land",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ },
+ "semantic": {
+ "normalized_reference": "oolong was 250 miles from the nearest land",
+ "normalized_hypothesis": "ulon was 250 miles from the nearest land",
+ "errors": 1,
+ "words": 8,
+ "wer": 0.125
+ }
+ },
+ {
+ "id": "release500_0433",
+ "category": "phonetic_balance",
+ "text": "I just do appreciate it without being able to express my feelings.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0433.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I JUST DO APPRECIATE IT WITHOUT BEING ABLE TO EXPRESS MY FEELINGS",
+ "raw": {
+ "normalized_reference": "i just do appreciate it without being able to express my feelings.",
+ "normalized_hypothesis": "i just do appreciate it without being able to express my feelings",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i just do appreciate it without being able to express my feelings",
+ "normalized_hypothesis": "i just do appreciate it without being able to express my feelings",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0434",
+ "category": "phonetic_balance",
+ "text": "Now just what do you want to know.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0434.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "NOW JUST WHAT DO YOU WANT TO KNOW",
+ "raw": {
+ "normalized_reference": "now just what do you want to know.",
+ "normalized_hypothesis": "now just what do you want to know",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "now just what do you want to know",
+ "normalized_hypothesis": "now just what do you want to know",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0435",
+ "category": "phonetic_balance",
+ "text": "But she swung obediently on her heel into the wind.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0435.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "BUT SHE SWUNG OBEDIENTLY ON HER HEEL INTO THE WIND",
+ "raw": {
+ "normalized_reference": "but she swung obediently on her heel into the wind.",
+ "normalized_hypothesis": "but she swung obediently on her heel into the wind",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "but she swung obediently on her heel into the wind",
+ "normalized_hypothesis": "but she swung obediently on her heel into the wind",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0436",
+ "category": "phonetic_balance",
+ "text": "Ask him, I laughed, then turned to Pasquini.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0436.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "ASK HIM I LAUGHED THEN TURNED TO BASQUINI",
+ "raw": {
+ "normalized_reference": "ask him, i laughed, then turned to pasquini.",
+ "normalized_hypothesis": "ask him i laughed then turned to basquini",
+ "errors": 1,
+ "words": 8,
+ "wer": 0.125
+ },
+ "semantic": {
+ "normalized_reference": "ask him i laughed then turned to pasquini",
+ "normalized_hypothesis": "ask him i laughed then turned to basquini",
+ "errors": 1,
+ "words": 8,
+ "wer": 0.125
+ }
+ },
+ {
+ "id": "release500_0437",
+ "category": "phonetic_balance",
+ "text": "He did not rush in.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0437.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE DID NOT RUSH IN",
+ "raw": {
+ "normalized_reference": "he did not rush in.",
+ "normalized_hypothesis": "he did not rush in",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he did not rush in",
+ "normalized_hypothesis": "he did not rush in",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0438",
+ "category": "phonetic_balance",
+ "text": "They obeyed him, and went here and there at his commands.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0438.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THEY OBEYED HIM AND WENT HERE AND THERE AT HIS COMMANDS",
+ "raw": {
+ "normalized_reference": "they obeyed him, and went here and there at his commands.",
+ "normalized_hypothesis": "they obeyed him and went here and there at his commands",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "they obeyed him and went here and there at his commands",
+ "normalized_hypothesis": "they obeyed him and went here and there at his commands",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0439",
+ "category": "phonetic_balance",
+ "text": "Tom Spink has a harpoon.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0439.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "TOM SPINQUESSA HARPOON",
+ "raw": {
+ "normalized_reference": "tom spink has a harpoon.",
+ "normalized_hypothesis": "tom spinquessa harpoon",
+ "errors": 3,
+ "words": 5,
+ "wer": 0.6
+ },
+ "semantic": {
+ "normalized_reference": "tom spink has a harpoon",
+ "normalized_hypothesis": "tom spinquessa harpoon",
+ "errors": 3,
+ "words": 5,
+ "wer": 0.6
+ }
+ },
+ {
+ "id": "release500_0440",
+ "category": "phonetic_balance",
+ "text": "It was my reports from the north which chiefly induced people to buy.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0440.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT WAS MY REPORTS FROM THE NORTH WHICH CHIEFLY INDUCED PEOPLE TO BUY",
+ "raw": {
+ "normalized_reference": "it was my reports from the north which chiefly induced people to buy.",
+ "normalized_hypothesis": "it was my reports from the north which chiefly induced people to buy",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it was my reports from the north which chiefly induced people to buy",
+ "normalized_hypothesis": "it was my reports from the north which chiefly induced people to buy",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0441",
+ "category": "phonetic_balance",
+ "text": "Cherokee identified himself with his instinct.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0441.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "CHEROKEE IDENTIFIED HIMSELF WITH HIS INSTINCT",
+ "raw": {
+ "normalized_reference": "cherokee identified himself with his instinct.",
+ "normalized_hypothesis": "cherokee identified himself with his instinct",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "cherokee identified himself with his instinct",
+ "normalized_hypothesis": "cherokee identified himself with his instinct",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0442",
+ "category": "phonetic_balance",
+ "text": "I graduated last of my class.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0442.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I GRADUATED LAST OF MY CLASS",
+ "raw": {
+ "normalized_reference": "i graduated last of my class.",
+ "normalized_hypothesis": "i graduated last of my class",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i graduated last of my class",
+ "normalized_hypothesis": "i graduated last of my class",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0443",
+ "category": "phonetic_balance",
+ "text": "It was a gigantic inadequacy.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0443.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT WAS A GIGANTIC INADEQUACY",
+ "raw": {
+ "normalized_reference": "it was a gigantic inadequacy.",
+ "normalized_hypothesis": "it was a gigantic inadequacy",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it was a gigantic inadequacy",
+ "normalized_hypothesis": "it was a gigantic inadequacy",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0444",
+ "category": "phonetic_balance",
+ "text": "Did I possess too much vitality.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0444.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "DID I POSSESS TOO MUCH VITALITY",
+ "raw": {
+ "normalized_reference": "did i possess too much vitality.",
+ "normalized_hypothesis": "did i possess too much vitality",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "did i possess too much vitality",
+ "normalized_hypothesis": "did i possess too much vitality",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0445",
+ "category": "phonetic_balance",
+ "text": "Therefore, hurrah for the game.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0445.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THEREFORE HURRAH FOR THE GAME",
+ "raw": {
+ "normalized_reference": "therefore, hurrah for the game.",
+ "normalized_hypothesis": "therefore hurrah for the game",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "therefore hurrah for the game",
+ "normalized_hypothesis": "therefore hurrah for the game",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0446",
+ "category": "phonetic_balance",
+ "text": "He had fulfilled his duty and paid properly.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0446.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE HAD FULFILLED HIS DUTY AND PAID PROPERLY",
+ "raw": {
+ "normalized_reference": "he had fulfilled his duty and paid properly.",
+ "normalized_hypothesis": "he had fulfilled his duty and paid properly",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he had fulfilled his duty and paid properly",
+ "normalized_hypothesis": "he had fulfilled his duty and paid properly",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0447",
+ "category": "phonetic_balance",
+ "text": "How can you manage all alone, Mr Young.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0447.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HOW CAN YOU MANAGE ALL ALONE MISTER YOUNG",
+ "raw": {
+ "normalized_reference": "how can you manage all alone, mr young.",
+ "normalized_hypothesis": "how can you manage all alone mister young",
+ "errors": 1,
+ "words": 8,
+ "wer": 0.125
+ },
+ "semantic": {
+ "normalized_reference": "how can you manage all alone mister young",
+ "normalized_hypothesis": "how can you manage all alone mister young",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0448",
+ "category": "phonetic_balance",
+ "text": "Red-Eye never committed a more outrageous deed.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0448.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "RED EYE NEVER COMMITTED A MORE OUTRAGEOUS DEED",
+ "raw": {
+ "normalized_reference": "red-eye never committed a more outrageous deed.",
+ "normalized_hypothesis": "red eye never committed a more outrageous deed",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "red eye never committed a more outrageous deed",
+ "normalized_hypothesis": "red eye never committed a more outrageous deed",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0449",
+ "category": "phonetic_balance",
+ "text": "He would keep his faith with Gregson for the promised day or two.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0449.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE WOULD KEEP HIS FAITH WITH GREGSON FOR THE PROMISED DAY OR TWO",
+ "raw": {
+ "normalized_reference": "he would keep his faith with gregson for the promised day or two.",
+ "normalized_hypothesis": "he would keep his faith with gregson for the promised day or two",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he would keep his faith with gregson for the promised day or 2",
+ "normalized_hypothesis": "he would keep his faith with gregson for the promised day or 2",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0450",
+ "category": "phonetic_balance",
+ "text": "There was one chance, and only one, of saving Jeanne.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0450.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THERE WAS ONE CHANCE AND ONLY ONE OF SAVING JEAN",
+ "raw": {
+ "normalized_reference": "there was one chance, and only one, of saving jeanne.",
+ "normalized_hypothesis": "there was one chance and only one of saving jean",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ },
+ "semantic": {
+ "normalized_reference": "there was one chance and only one of saving jeanne",
+ "normalized_hypothesis": "there was one chance and only one of saving jean",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ }
+ },
+ {
+ "id": "release500_0451",
+ "category": "phonetic_balance",
+ "text": "I was beneath the water, suffocating and drowning.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0451.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I WAS BENEATH THE WATER SUFFOCATING AND DROWNING",
+ "raw": {
+ "normalized_reference": "i was beneath the water, suffocating and drowning.",
+ "normalized_hypothesis": "i was beneath the water suffocating and drowning",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i was beneath the water suffocating and drowning",
+ "normalized_hypothesis": "i was beneath the water suffocating and drowning",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0452",
+ "category": "phonetic_balance",
+ "text": "Give them their choice between a fine or an official whipping.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0452.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "GIVE THEM THEIR CHOICE BETWEEN A FINE OR AN OFFICIAL WHIPPING",
+ "raw": {
+ "normalized_reference": "give them their choice between a fine or an official whipping.",
+ "normalized_hypothesis": "give them their choice between a fine or an official whipping",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "give them their choice between a fine or an official whipping",
+ "normalized_hypothesis": "give them their choice between a fine or an official whipping",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0453",
+ "category": "phonetic_balance",
+ "text": "They must have been swept away by the chaotic currents.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0453.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THEY MUST HAVE BEEN SWEPT AWAY BY THE CHAOTIC CURRENTS",
+ "raw": {
+ "normalized_reference": "they must have been swept away by the chaotic currents.",
+ "normalized_hypothesis": "they must have been swept away by the chaotic currents",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "they must have been swept away by the chaotic currents",
+ "normalized_hypothesis": "they must have been swept away by the chaotic currents",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0454",
+ "category": "phonetic_balance",
+ "text": "You yellow giant thing of the frost.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0454.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "YOU YELLOW GIANT THING OF THE FROST",
+ "raw": {
+ "normalized_reference": "you yellow giant thing of the frost.",
+ "normalized_hypothesis": "you yellow giant thing of the frost",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "you yellow giant thing of the frost",
+ "normalized_hypothesis": "you yellow giant thing of the frost",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0455",
+ "category": "phonetic_balance",
+ "text": "He had peeled off his shirt and was wildly waving it.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0455.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE HAD PEELED OFF HIS SHIRT AND WAS WILDLY WAVING IT",
+ "raw": {
+ "normalized_reference": "he had peeled off his shirt and was wildly waving it.",
+ "normalized_hypothesis": "he had peeled off his shirt and was wildly waving it",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he had peeled off his shirt and was wildly waving it",
+ "normalized_hypothesis": "he had peeled off his shirt and was wildly waving it",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0456",
+ "category": "phonetic_balance",
+ "text": "To these he gave castor oil.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0456.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "TO THESE HE GAVE CASTOR OIL",
+ "raw": {
+ "normalized_reference": "to these he gave castor oil.",
+ "normalized_hypothesis": "to these he gave castor oil",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "to these he gave castor oil",
+ "normalized_hypothesis": "to these he gave castor oil",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0457",
+ "category": "phonetic_balance",
+ "text": "Tudor surveyed him with withering disgust.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0457.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "TUDOR SURVEYED HIM WITH WITHERING DISGUST",
+ "raw": {
+ "normalized_reference": "tudor surveyed him with withering disgust.",
+ "normalized_hypothesis": "tudor surveyed him with withering disgust",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "tudor surveyed him with withering disgust",
+ "normalized_hypothesis": "tudor surveyed him with withering disgust",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0458",
+ "category": "phonetic_balance",
+ "text": "After all, it was simply a mistake in judgment.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0458.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "AFTER ALL IT WAS SIMPLY A MISTAKE IN JUDGMENT",
+ "raw": {
+ "normalized_reference": "after all, it was simply a mistake in judgment.",
+ "normalized_hypothesis": "after all it was simply a mistake in judgment",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "after all it was simply a mistake in judgment",
+ "normalized_hypothesis": "after all it was simply a mistake in judgment",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0459",
+ "category": "phonetic_balance",
+ "text": "That is the strange part of it.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0459.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THAT IS THE STRANGE PART OF IT",
+ "raw": {
+ "normalized_reference": "that is the strange part of it.",
+ "normalized_hypothesis": "that is the strange part of it",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "that is the strange part of it",
+ "normalized_hypothesis": "that is the strange part of it",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0460",
+ "category": "phonetic_balance",
+ "text": "Not a wheel moved in his empire.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0460.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "NOT A WHEEL MOVED IN HIS EMPIRE",
+ "raw": {
+ "normalized_reference": "not a wheel moved in his empire.",
+ "normalized_hypothesis": "not a wheel moved in his empire",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "not a wheel moved in his empire",
+ "normalized_hypothesis": "not a wheel moved in his empire",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0461",
+ "category": "phonetic_balance",
+ "text": "Its diameter was not more than two hundred yards.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0461.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "ITS DIAMETER WAS NOT MORE THAN TWO HUNDRED YARDS",
+ "raw": {
+ "normalized_reference": "its diameter was not more than two hundred yards.",
+ "normalized_hypothesis": "its diameter was not more than two hundred yards",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "its diameter was not more than 200 yards",
+ "normalized_hypothesis": "its diameter was not more than 200 yards",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0462",
+ "category": "phonetic_balance",
+ "text": "Nimrod replied, with a slight manifestation of sensitiveness.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0462.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "NIMROD REPLIED WITH A SLIGHT MANIFESTATION OF SENSITIVENESS",
+ "raw": {
+ "normalized_reference": "nimrod replied, with a slight manifestation of sensitiveness.",
+ "normalized_hypothesis": "nimrod replied with a slight manifestation of sensitiveness",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "nimrod replied with a slight manifestation of sensitiveness",
+ "normalized_hypothesis": "nimrod replied with a slight manifestation of sensitiveness",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0463",
+ "category": "phonetic_balance",
+ "text": "I tried to read George Moore last night, and was dreadfully bored.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0463.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I TRIED TO READ GEORGE MOORE LAST NIGHT AND WAS DREADFULLY BORED",
+ "raw": {
+ "normalized_reference": "i tried to read george moore last night, and was dreadfully bored.",
+ "normalized_hypothesis": "i tried to read george moore last night and was dreadfully bored",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i tried to read george moore last night and was dreadfully bored",
+ "normalized_hypothesis": "i tried to read george moore last night and was dreadfully bored",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0464",
+ "category": "phonetic_balance",
+ "text": "Yes, it was a man who asked, a stranger.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0464.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "YES IT WAS A MAN WHO ASKED A STRANGER",
+ "raw": {
+ "normalized_reference": "yes, it was a man who asked, a stranger.",
+ "normalized_hypothesis": "yes it was a man who asked a stranger",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "yes it was a man who asked a stranger",
+ "normalized_hypothesis": "yes it was a man who asked a stranger",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0465",
+ "category": "phonetic_balance",
+ "text": "We were now good friends.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0465.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WE WERE NOW GOOD FRIENDS",
+ "raw": {
+ "normalized_reference": "we were now good friends.",
+ "normalized_hypothesis": "we were now good friends",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "we were now good friends",
+ "normalized_hypothesis": "we were now good friends",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0466",
+ "category": "phonetic_balance",
+ "text": "Yet, in accordance with Ernest's test of truth, it worked.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0466.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "YET IN ACCORDANCE WITH ERNEST'S TEST OF TRUTH IT WORKED",
+ "raw": {
+ "normalized_reference": "yet, in accordance with ernest's test of truth, it worked.",
+ "normalized_hypothesis": "yet in accordance with ernest's test of truth it worked",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "yet in accordance with ernest is test of truth it worked",
+ "normalized_hypothesis": "yet in accordance with ernest is test of truth it worked",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0467",
+ "category": "phonetic_balance",
+ "text": "It's the strap hangers that'll keep us from going under.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0467.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT'S THE STRAP HANGERS THAT'LL KEEP US FROM GOING UNDER",
+ "raw": {
+ "normalized_reference": "it's the strap hangers that'll keep us from going under.",
+ "normalized_hypothesis": "it's the strap hangers that'll keep us from going under",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it is the strap hangers that will keep us from going under",
+ "normalized_hypothesis": "it is the strap hangers that will keep us from going under",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0468",
+ "category": "phonetic_balance",
+ "text": "I never allow what can't be changed to annoy me.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0468.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I NEVER ALLOW WHAT CAN'T BE CHANGED TO ANNOY ME",
+ "raw": {
+ "normalized_reference": "i never allow what can't be changed to annoy me.",
+ "normalized_hypothesis": "i never allow what can't be changed to annoy me",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i never allow what can not be changed to annoy me",
+ "normalized_hypothesis": "i never allow what can not be changed to annoy me",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0469",
+ "category": "phonetic_balance",
+ "text": "After all superfluous flesh is gone what is left is stringy and resistant.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0469.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "AFTER ALL SUPERFLUOUS FLESH IS CON WHAT IS LEFT IS STRINGY AND RESISTENT",
+ "raw": {
+ "normalized_reference": "after all superfluous flesh is gone what is left is stringy and resistant.",
+ "normalized_hypothesis": "after all superfluous flesh is con what is left is stringy and resistent",
+ "errors": 2,
+ "words": 13,
+ "wer": 0.15384615384615385
+ },
+ "semantic": {
+ "normalized_reference": "after all superfluous flesh is gone what is left is stringy and resistant",
+ "normalized_hypothesis": "after all superfluous flesh is con what is left is stringy and resistent",
+ "errors": 2,
+ "words": 13,
+ "wer": 0.15384615384615385
+ }
+ },
+ {
+ "id": "release500_0470",
+ "category": "phonetic_balance",
+ "text": "Then he shouted, Shut up.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0470.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THEN HE SHOUTED SHUT UP",
+ "raw": {
+ "normalized_reference": "then he shouted, shut up.",
+ "normalized_hypothesis": "then he shouted shut up",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "then he shouted shut up",
+ "normalized_hypothesis": "then he shouted shut up",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0471",
+ "category": "phonetic_balance",
+ "text": "He had worshiped her, as Dante might have worshiped Beatrice.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0471.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "KIHENN WORSHIPPED HER AS DANTE MIGHT HAVE WORSHIPPED BEATRICE",
+ "raw": {
+ "normalized_reference": "he had worshiped her, as dante might have worshiped beatrice.",
+ "normalized_hypothesis": "kihenn worshipped her as dante might have worshipped beatrice",
+ "errors": 4,
+ "words": 10,
+ "wer": 0.4
+ },
+ "semantic": {
+ "normalized_reference": "he had worshiped her as dante might have worshiped beatrice",
+ "normalized_hypothesis": "kihenn worshipped her as dante might have worshipped beatrice",
+ "errors": 4,
+ "words": 10,
+ "wer": 0.4
+ }
+ },
+ {
+ "id": "release500_0472",
+ "category": "phonetic_balance",
+ "text": "He looked like one who had passed through an uncomfortable hour or two.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0472.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE LOOKED LIKE ONE WHO HAD PASSED THROUGH AN UNCOMFORTABLE HOUR OR TWO",
+ "raw": {
+ "normalized_reference": "he looked like one who had passed through an uncomfortable hour or two.",
+ "normalized_hypothesis": "he looked like one who had passed through an uncomfortable hour or two",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he looked like one who had passed through an uncomfortable hour or 2",
+ "normalized_hypothesis": "he looked like one who had passed through an uncomfortable hour or 2",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0473",
+ "category": "phonetic_balance",
+ "text": "The history of our westward-faring race is written in it.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0473.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE HISTORY OF OUR WESTWARDFARING RACE IS WRITTEN IN IT",
+ "raw": {
+ "normalized_reference": "the history of our westward-faring race is written in it.",
+ "normalized_hypothesis": "the history of our westwardfaring race is written in it",
+ "errors": 2,
+ "words": 11,
+ "wer": 0.18181818181818182
+ },
+ "semantic": {
+ "normalized_reference": "the history of our westward faring race is written in it",
+ "normalized_hypothesis": "the history of our westwardfaring race is written in it",
+ "errors": 2,
+ "words": 11,
+ "wer": 0.18181818181818182
+ }
+ },
+ {
+ "id": "release500_0474",
+ "category": "phonetic_balance",
+ "text": "He may desire to escape pain, or to enjoy pleasure.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0474.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE MAY DESIRE TO ESCAPE PAIN OR TO ENJOY PLEASURE",
+ "raw": {
+ "normalized_reference": "he may desire to escape pain, or to enjoy pleasure.",
+ "normalized_hypothesis": "he may desire to escape pain or to enjoy pleasure",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he may desire to escape pain or to enjoy pleasure",
+ "normalized_hypothesis": "he may desire to escape pain or to enjoy pleasure",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0475",
+ "category": "phonetic_balance",
+ "text": "And the air was growing chilly.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0475.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "AND THE AIR WAS GROWING CHILLY",
+ "raw": {
+ "normalized_reference": "and the air was growing chilly.",
+ "normalized_hypothesis": "and the air was growing chilly",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "and the air was growing chilly",
+ "normalized_hypothesis": "and the air was growing chilly",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0476",
+ "category": "phonetic_balance",
+ "text": "Dig in; you're sure good, was Daylight's answer.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0476.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "DEGIN YOU'RE SURE GOOD WAS DAYLIGHT'S ANSWER",
+ "raw": {
+ "normalized_reference": "dig in; you're sure good, was daylight's answer.",
+ "normalized_hypothesis": "degin you're sure good was daylight's answer",
+ "errors": 2,
+ "words": 8,
+ "wer": 0.25
+ },
+ "semantic": {
+ "normalized_reference": "dig in you are sure good was daylight is answer",
+ "normalized_hypothesis": "degin you are sure good was daylight is answer",
+ "errors": 2,
+ "words": 10,
+ "wer": 0.2
+ }
+ },
+ {
+ "id": "release500_0477",
+ "category": "phonetic_balance",
+ "text": "My idea was that he would have more influence over the natives.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0477.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "MY IDEA WAS THAT HE WOULD HAVE MORE INFLUENCE OVER THE NATIVES",
+ "raw": {
+ "normalized_reference": "my idea was that he would have more influence over the natives.",
+ "normalized_hypothesis": "my idea was that he would have more influence over the natives",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "my idea was that he would have more influence over the natives",
+ "normalized_hypothesis": "my idea was that he would have more influence over the natives",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0478",
+ "category": "phonetic_balance",
+ "text": "Now run along, and tell them to hurry.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0478.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "NOW RUN ALONG AND TELL THEM TO HURRY",
+ "raw": {
+ "normalized_reference": "now run along, and tell them to hurry.",
+ "normalized_hypothesis": "now run along and tell them to hurry",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "now run along and tell them to hurry",
+ "normalized_hypothesis": "now run along and tell them to hurry",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0479",
+ "category": "phonetic_balance",
+ "text": "I have no idea, replied Philip.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0479.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I HAVE NO IDEA REPLIED PHILIP",
+ "raw": {
+ "normalized_reference": "i have no idea, replied philip.",
+ "normalized_hypothesis": "i have no idea replied philip",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i have no idea replied philip",
+ "normalized_hypothesis": "i have no idea replied philip",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0480",
+ "category": "phonetic_balance",
+ "text": "For the rest, he was a mere automaton.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0480.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "FOR THE REST KY WAS A MERE AUTOMATON",
+ "raw": {
+ "normalized_reference": "for the rest, he was a mere automaton.",
+ "normalized_hypothesis": "for the rest ky was a mere automaton",
+ "errors": 1,
+ "words": 8,
+ "wer": 0.125
+ },
+ "semantic": {
+ "normalized_reference": "for the rest he was a mere automaton",
+ "normalized_hypothesis": "for the rest ky was a mere automaton",
+ "errors": 1,
+ "words": 8,
+ "wer": 0.125
+ }
+ },
+ {
+ "id": "release500_0481",
+ "category": "phonetic_balance",
+ "text": "Why, the average review is more nauseating than cod liver oil.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0481.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WHY THE AVERAGE REVIEW IS MORE NAUSEATING THAN COD LIVER OIL",
+ "raw": {
+ "normalized_reference": "why, the average review is more nauseating than cod liver oil.",
+ "normalized_hypothesis": "why the average review is more nauseating than cod liver oil",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "why the average review is more nauseating than cod liver oil",
+ "normalized_hypothesis": "why the average review is more nauseating than cod liver oil",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0482",
+ "category": "phonetic_balance",
+ "text": "I have been robbed, sir, I amended.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0482.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I HAVE BEEN ROBBED SIR I AMMENDED",
+ "raw": {
+ "normalized_reference": "i have been robbed, sir, i amended.",
+ "normalized_hypothesis": "i have been robbed sir i ammended",
+ "errors": 1,
+ "words": 7,
+ "wer": 0.14285714285714285
+ },
+ "semantic": {
+ "normalized_reference": "i have been robbed sir i amended",
+ "normalized_hypothesis": "i have been robbed sir i ammended",
+ "errors": 1,
+ "words": 7,
+ "wer": 0.14285714285714285
+ }
+ },
+ {
+ "id": "release500_0483",
+ "category": "phonetic_balance",
+ "text": "He moved away as quietly as he had come.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0483.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE MOVED AWAY AS QUIETLY AS HE HAD COME",
+ "raw": {
+ "normalized_reference": "he moved away as quietly as he had come.",
+ "normalized_hypothesis": "he moved away as quietly as he had come",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he moved away as quietly as he had come",
+ "normalized_hypothesis": "he moved away as quietly as he had come",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0484",
+ "category": "phonetic_balance",
+ "text": "A trickle of fresh blood ran over his face.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0484.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "A TRICKLE OF FRESH BLOOD RAN OVER HIS FACE",
+ "raw": {
+ "normalized_reference": "a trickle of fresh blood ran over his face.",
+ "normalized_hypothesis": "a trickle of fresh blood ran over his face",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "a trickle of fresh blood ran over his face",
+ "normalized_hypothesis": "a trickle of fresh blood ran over his face",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0485",
+ "category": "phonetic_balance",
+ "text": "Suppose you saw me at work through the window.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0485.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SUPPOSE YOU SAW ME AT WORK THROUGH THE WINDOW",
+ "raw": {
+ "normalized_reference": "suppose you saw me at work through the window.",
+ "normalized_hypothesis": "suppose you saw me at work through the window",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "suppose you saw me at work through the window",
+ "normalized_hypothesis": "suppose you saw me at work through the window",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0486",
+ "category": "phonetic_balance",
+ "text": "But who was Eileen's double.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0486.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "BUT WHO WAS AILEEN'S DOUBLE",
+ "raw": {
+ "normalized_reference": "but who was eileen's double.",
+ "normalized_hypothesis": "but who was aileen's double",
+ "errors": 1,
+ "words": 5,
+ "wer": 0.2
+ },
+ "semantic": {
+ "normalized_reference": "but who was eileen is double",
+ "normalized_hypothesis": "but who was aileen is double",
+ "errors": 1,
+ "words": 6,
+ "wer": 0.16666666666666666
+ }
+ },
+ {
+ "id": "release500_0487",
+ "category": "phonetic_balance",
+ "text": "Darkness hid him from Jeanne.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0487.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "DARKNESS HID HIM FROM JEAN",
+ "raw": {
+ "normalized_reference": "darkness hid him from jeanne.",
+ "normalized_hypothesis": "darkness hid him from jean",
+ "errors": 1,
+ "words": 5,
+ "wer": 0.2
+ },
+ "semantic": {
+ "normalized_reference": "darkness hid him from jeanne",
+ "normalized_hypothesis": "darkness hid him from jean",
+ "errors": 1,
+ "words": 5,
+ "wer": 0.2
+ }
+ },
+ {
+ "id": "release500_0488",
+ "category": "phonetic_balance",
+ "text": "By this time Charley was as enraged as the Greek.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0488.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "BY THIS TIME CHARLEY WAS AS ENRAGED AS THE GREEK",
+ "raw": {
+ "normalized_reference": "by this time charley was as enraged as the greek.",
+ "normalized_hypothesis": "by this time charley was as enraged as the greek",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "by this time charley was as enraged as the greek",
+ "normalized_hypothesis": "by this time charley was as enraged as the greek",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0489",
+ "category": "phonetic_balance",
+ "text": "I am writing these lines in Honolulu, Hawaii.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0489.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I AM WRITING THESE LINES IN HONOLULU KAWAII",
+ "raw": {
+ "normalized_reference": "i am writing these lines in honolulu, hawaii.",
+ "normalized_hypothesis": "i am writing these lines in honolulu kawaii",
+ "errors": 1,
+ "words": 8,
+ "wer": 0.125
+ },
+ "semantic": {
+ "normalized_reference": "i am writing these lines in honolulu hawaii",
+ "normalized_hypothesis": "i am writing these lines in honolulu kawaii",
+ "errors": 1,
+ "words": 8,
+ "wer": 0.125
+ }
+ },
+ {
+ "id": "release500_0490",
+ "category": "phonetic_balance",
+ "text": "Nobody knows how the natives got them.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0490.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "NOBODY KNOWS HOW THE NATIVES CAUGHT THEM",
+ "raw": {
+ "normalized_reference": "nobody knows how the natives got them.",
+ "normalized_hypothesis": "nobody knows how the natives caught them",
+ "errors": 1,
+ "words": 7,
+ "wer": 0.14285714285714285
+ },
+ "semantic": {
+ "normalized_reference": "nobody knows how the natives got them",
+ "normalized_hypothesis": "nobody knows how the natives caught them",
+ "errors": 1,
+ "words": 7,
+ "wer": 0.14285714285714285
+ }
+ },
+ {
+ "id": "release500_0491",
+ "category": "phonetic_balance",
+ "text": "And how would we ever find ourselves.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0491.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "AND HOW WOULD WE EVER FIND OURSELVES",
+ "raw": {
+ "normalized_reference": "and how would we ever find ourselves.",
+ "normalized_hypothesis": "and how would we ever find ourselves",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "and how would we ever find ourselves",
+ "normalized_hypothesis": "and how would we ever find ourselves",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0492",
+ "category": "phonetic_balance",
+ "text": "They are coming ashore, whoever they are.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0492.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THEY ARE COMING ASHORE WHOEVER THEY ARE",
+ "raw": {
+ "normalized_reference": "they are coming ashore, whoever they are.",
+ "normalized_hypothesis": "they are coming ashore whoever they are",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "they are coming ashore whoever they are",
+ "normalized_hypothesis": "they are coming ashore whoever they are",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0493",
+ "category": "phonetic_balance",
+ "text": "This is my fifth voyage.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0493.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THIS IS MY FIFTH VOYAGE",
+ "raw": {
+ "normalized_reference": "this is my fifth voyage.",
+ "normalized_hypothesis": "this is my fifth voyage",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "this is my 5th voyage",
+ "normalized_hypothesis": "this is my 5th voyage",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0494",
+ "category": "phonetic_balance",
+ "text": "And yet if she came he had no words to say.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0494.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "AND YET IF SHE CAME HE HAD NO WORDS TO SAY",
+ "raw": {
+ "normalized_reference": "and yet if she came he had no words to say.",
+ "normalized_hypothesis": "and yet if she came he had no words to say",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "and yet if she came he had no words to say",
+ "normalized_hypothesis": "and yet if she came he had no words to say",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0495",
+ "category": "phonetic_balance",
+ "text": "You must sleep, he urged.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0495.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "YOU MUST SLEEP KEY URGED",
+ "raw": {
+ "normalized_reference": "you must sleep, he urged.",
+ "normalized_hypothesis": "you must sleep key urged",
+ "errors": 1,
+ "words": 5,
+ "wer": 0.2
+ },
+ "semantic": {
+ "normalized_reference": "you must sleep he urged",
+ "normalized_hypothesis": "you must sleep key urged",
+ "errors": 1,
+ "words": 5,
+ "wer": 0.2
+ }
+ },
+ {
+ "id": "release500_0496",
+ "category": "phonetic_balance",
+ "text": "Could the incident have anything to do with Jeanne and Pierre.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0496.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "COULD THE INCIDENT HAVE ANYTHING TO DO WITH JEAN AND BIERRE",
+ "raw": {
+ "normalized_reference": "could the incident have anything to do with jeanne and pierre.",
+ "normalized_hypothesis": "could the incident have anything to do with jean and bierre",
+ "errors": 2,
+ "words": 11,
+ "wer": 0.18181818181818182
+ },
+ "semantic": {
+ "normalized_reference": "could the incident have anything to do with jeanne and pierre",
+ "normalized_hypothesis": "could the incident have anything to do with jean and bierre",
+ "errors": 2,
+ "words": 11,
+ "wer": 0.18181818181818182
+ }
+ },
+ {
+ "id": "release500_0497",
+ "category": "phonetic_balance",
+ "text": "Each day she became a more vital part of him.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0497.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "EACH DAY SHE BECAME A MORE VITAL PART OF HIM",
+ "raw": {
+ "normalized_reference": "each day she became a more vital part of him.",
+ "normalized_hypothesis": "each day she became a more vital part of him",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "each day she became a more vital part of him",
+ "normalized_hypothesis": "each day she became a more vital part of him",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0498",
+ "category": "phonetic_balance",
+ "text": "But why continue the tirade, for tirade it was.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0498.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "BUT WHY CONTINUE THE TIRADE FOR TIRADE IT WAS",
+ "raw": {
+ "normalized_reference": "but why continue the tirade, for tirade it was.",
+ "normalized_hypothesis": "but why continue the tirade for tirade it was",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "but why continue the tirade for tirade it was",
+ "normalized_hypothesis": "but why continue the tirade for tirade it was",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0499",
+ "category": "phonetic_balance",
+ "text": "That is what distinguishes all of us from the lower animals.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0499.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THAT IS WHAT DISTINGUISHES ALL OF US FROM THE LOWER ANIMALS",
+ "raw": {
+ "normalized_reference": "that is what distinguishes all of us from the lower animals.",
+ "normalized_hypothesis": "that is what distinguishes all of us from the lower animals",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "that is what distinguishes all of us from the lower animals",
+ "normalized_hypothesis": "that is what distinguishes all of us from the lower animals",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0500",
+ "category": "phonetic_balance",
+ "text": "There was something pathetic in the girl's attitude now.",
+ "model": "inflect-nano-v2",
+ "audio": "benchmark/audio/inflect-nano-v2/release500_0500.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THERE WAS SOMETHING PATHETIC IN THE GIRL'S ATTITUDE NOW",
+ "raw": {
+ "normalized_reference": "there was something pathetic in the girl's attitude now.",
+ "normalized_hypothesis": "there was something pathetic in the girl's attitude now",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "there was something pathetic in the girl is attitude now",
+ "normalized_hypothesis": "there was something pathetic in the girl is attitude now",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0001",
+ "category": "conversational",
+ "text": "It wasn't until later that I realized what had actually happened.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0001.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT WASN'T UNTIL LATER THAT I REALIZED WHAT HAD ACTUALLY HAPPENED",
+ "raw": {
+ "normalized_reference": "it wasn't until later that i realized what had actually happened.",
+ "normalized_hypothesis": "it wasn't until later that i realized what had actually happened",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it was not until later that i realized what had actually happened",
+ "normalized_hypothesis": "it was not until later that i realized what had actually happened",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0002",
+ "category": "descriptive",
+ "text": "The light at that time of afternoon is perfect for about twenty minutes.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0002.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE LIGHT AT THAT TIME OF AFTERNOON IS PERFECT FOR ABOUT TWENTY MINUTES",
+ "raw": {
+ "normalized_reference": "the light at that time of afternoon is perfect for about twenty minutes.",
+ "normalized_hypothesis": "the light at that time of afternoon is perfect for about twenty minutes",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the light at that time of afternoon is perfect for about 20 minutes",
+ "normalized_hypothesis": "the light at that time of afternoon is perfect for about 20 minutes",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0003",
+ "category": "emotional",
+ "text": "That was the scariest thing I've ever done, and also maybe the best.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0003.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THAT WAS THE SCARIEST THING I'VE EVER DONE AND ALSO MAYBE THE BEST",
+ "raw": {
+ "normalized_reference": "that was the scariest thing i've ever done, and also maybe the best.",
+ "normalized_hypothesis": "that was the scariest thing i've ever done and also maybe the best",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "that was the scariest thing i have ever done and also maybe the best",
+ "normalized_hypothesis": "that was the scariest thing i have ever done and also maybe the best",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0004",
+ "category": "homographs",
+ "text": "After I record the record, please present the present to the bass player.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0004.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "AFTER I RECORD THE RECORD PLEASE PRESENT THE PRESENT TO THE BASE PLAYER",
+ "raw": {
+ "normalized_reference": "after i record the record, please present the present to the bass player.",
+ "normalized_hypothesis": "after i record the record please present the present to the base player",
+ "errors": 1,
+ "words": 13,
+ "wer": 0.07692307692307693
+ },
+ "semantic": {
+ "normalized_reference": "after i record the record please present the present to the bass player",
+ "normalized_hypothesis": "after i record the record please present the present to the base player",
+ "errors": 1,
+ "words": 13,
+ "wer": 0.07692307692307693
+ }
+ },
+ {
+ "id": "release500_0005",
+ "category": "long",
+ "text": "If I'm being completely honest, I had no idea what I was doing for most of it, and I think that might be the most normal thing about the whole experience.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0005.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IF I'M BEING COMPLETELY HONEST I HAD NO IDEA WHAT I WAS DOING FOR MOST OF IT AND I THINK THAT MIGHT BE THE MOST NORMAL THING ABOUT THE WHOLE EXPERIENCE",
+ "raw": {
+ "normalized_reference": "if i'm being completely honest, i had no idea what i was doing for most of it, and i think that might be the most normal thing about the whole experience.",
+ "normalized_hypothesis": "if i'm being completely honest i had no idea what i was doing for most of it and i think that might be the most normal thing about the whole experience",
+ "errors": 0,
+ "words": 31,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "if i am being completely honest i had no idea what i was doing for most of it and i think that might be the most normal thing about the whole experience",
+ "normalized_hypothesis": "if i am being completely honest i had no idea what i was doing for most of it and i think that might be the most normal thing about the whole experience",
+ "errors": 0,
+ "words": 32,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0006",
+ "category": "long_range",
+ "text": "The committee postponed its decision after discovering that three pages of the report had been printed in the wrong order.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0006.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE COMMITTEE POSTPONED ITS DECISION AFTER DISCOVERING THAT THREE PAGES OF THE REPORT HAD BEEN PRINTED IN THE WRONG ORDER",
+ "raw": {
+ "normalized_reference": "the committee postponed its decision after discovering that three pages of the report had been printed in the wrong order.",
+ "normalized_hypothesis": "the committee postponed its decision after discovering that three pages of the report had been printed in the wrong order",
+ "errors": 0,
+ "words": 20,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the committee postponed its decision after discovering that 3 pages of the report had been printed in the wrong order",
+ "normalized_hypothesis": "the committee postponed its decision after discovering that 3 pages of the report had been printed in the wrong order",
+ "errors": 0,
+ "words": 20,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0007",
+ "category": "names_places",
+ "text": "Gwendolyn photographed the eucalyptus trees outside Ljubljana.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0007.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "GWENDOLEN PHOTOGRAPHED THE EUCALYPTUS TREES OUTSIDE LUBLIANA",
+ "raw": {
+ "normalized_reference": "gwendolyn photographed the eucalyptus trees outside ljubljana.",
+ "normalized_hypothesis": "gwendolen photographed the eucalyptus trees outside lubliana",
+ "errors": 2,
+ "words": 7,
+ "wer": 0.2857142857142857
+ },
+ "semantic": {
+ "normalized_reference": "gwendolyn photographed the eucalyptus trees outside ljubljana",
+ "normalized_hypothesis": "gwendolen photographed the eucalyptus trees outside lubliana",
+ "errors": 2,
+ "words": 7,
+ "wer": 0.2857142857142857
+ }
+ },
+ {
+ "id": "release500_0008",
+ "category": "narrative",
+ "text": "He had this theory, and the more time passes, the more I think he was right.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0008.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE HAD THIS THEORY AND THE MORE TIME PASSES THE MORE I THINK HE WAS RIGHT",
+ "raw": {
+ "normalized_reference": "he had this theory, and the more time passes, the more i think he was right.",
+ "normalized_hypothesis": "he had this theory and the more time passes the more i think he was right",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he had this theory and the more time passes the more i think he was right",
+ "normalized_hypothesis": "he had this theory and the more time passes the more i think he was right",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0009",
+ "category": "numbers",
+ "text": "The package weighs twelve point six kilograms and arrived on July twenty-first.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0009.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE PACKAGE WEIGHS TWELVE POINT SIX KILOGRAMS AND ARRIVED ON JULY TWENTY FIRST",
+ "raw": {
+ "normalized_reference": "the package weighs twelve point six kilograms and arrived on july twenty-first.",
+ "normalized_hypothesis": "the package weighs twelve point six kilograms and arrived on july twenty first",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the package weighs 12.6 kilograms and arrived on july 21st",
+ "normalized_hypothesis": "the package weighs 12.6 kilograms and arrived on july 21st",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0010",
+ "category": "punctuation",
+ "text": "First, close the window; second, turn off the lamp; finally, lock the door.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0010.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "FIRST CLOSE THE WINDOW SECOND TURN OFF THE LAMP FINALLY LOCK THE DOOR",
+ "raw": {
+ "normalized_reference": "first, close the window; second, turn off the lamp; finally, lock the door.",
+ "normalized_hypothesis": "first close the window second turn off the lamp finally lock the door",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "1st close the window 2nd turn off the lamp finally lock the door",
+ "normalized_hypothesis": "1st close the window 2nd turn off the lamp finally lock the door",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0011",
+ "category": "question",
+ "text": "When did you first realize something was wrong?",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0011.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WHEN DID YOU FIRST REALIZE SOMETHING WAS WRONG",
+ "raw": {
+ "normalized_reference": "when did you first realize something was wrong?",
+ "normalized_hypothesis": "when did you first realize something was wrong",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "when did you 1st realize something was wrong",
+ "normalized_hypothesis": "when did you 1st realize something was wrong",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0012",
+ "category": "rare_words",
+ "text": "The archaeologist catalogued a Byzantine reliquary and several anachronistic fragments.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0012.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE ARCHEOLOGIST CATALOGUED A BIZANTINE RELIQUARY AND SEVERAL ANACRONISTIC FRAGMENTS",
+ "raw": {
+ "normalized_reference": "the archaeologist catalogued a byzantine reliquary and several anachronistic fragments.",
+ "normalized_hypothesis": "the archeologist catalogued a bizantine reliquary and several anacronistic fragments",
+ "errors": 3,
+ "words": 10,
+ "wer": 0.3
+ },
+ "semantic": {
+ "normalized_reference": "the archaeologist catalogued a byzantine reliquary and several anachronistic fragments",
+ "normalized_hypothesis": "the archeologist catalogued a bizantine reliquary and several anacronistic fragments",
+ "errors": 3,
+ "words": 10,
+ "wer": 0.3
+ }
+ },
+ {
+ "id": "release500_0013",
+ "category": "short",
+ "text": "I have no idea.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0013.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I HAVE NO IDEA",
+ "raw": {
+ "normalized_reference": "i have no idea.",
+ "normalized_hypothesis": "i have no idea",
+ "errors": 0,
+ "words": 4,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i have no idea",
+ "normalized_hypothesis": "i have no idea",
+ "errors": 0,
+ "words": 4,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0014",
+ "category": "technical",
+ "text": "The system runs on three core components that all have to stay in sync.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0014.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE SYSTEM RUNS ON THREE CORPS COMPONENTS THAT ALL HAVE TO STAY AND SINK",
+ "raw": {
+ "normalized_reference": "the system runs on three core components that all have to stay in sync.",
+ "normalized_hypothesis": "the system runs on three corps components that all have to stay and sink",
+ "errors": 3,
+ "words": 14,
+ "wer": 0.21428571428571427
+ },
+ "semantic": {
+ "normalized_reference": "the system runs on 3 core components that all have to stay in sync",
+ "normalized_hypothesis": "the system runs on 3 corps components that all have to stay and sink",
+ "errors": 3,
+ "words": 14,
+ "wer": 0.21428571428571427
+ }
+ },
+ {
+ "id": "release500_0015",
+ "category": "conversational",
+ "text": "We improvised the entire second half and somehow nobody noticed.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0015.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WE IMPROVISED THE ENTIRE SECOND HALF AND SOMEHOW NOBODY NOTICED",
+ "raw": {
+ "normalized_reference": "we improvised the entire second half and somehow nobody noticed.",
+ "normalized_hypothesis": "we improvised the entire second half and somehow nobody noticed",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "we improvised the entire 2nd half and somehow nobody noticed",
+ "normalized_hypothesis": "we improvised the entire 2nd half and somehow nobody noticed",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0016",
+ "category": "descriptive",
+ "text": "There's a smell in old buildings that's impossible to describe but immediately familiar.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0016.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THERE'S A SMELL IN OLD BUILDINGS THAT'S IMPOSSIBLE TO DESCRIBE BUT IMMEDIATELY FAMILIAR",
+ "raw": {
+ "normalized_reference": "there's a smell in old buildings that's impossible to describe but immediately familiar.",
+ "normalized_hypothesis": "there's a smell in old buildings that's impossible to describe but immediately familiar",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "there is a smell in old buildings that is impossible to describe but immediately familiar",
+ "normalized_hypothesis": "there is a smell in old buildings that is impossible to describe but immediately familiar",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0017",
+ "category": "emotional",
+ "text": "I specifically asked for one thing. One thing. And it still didn't happen.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0017.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I SPECIFICALLY ASKED FOR ONE THING ONE THING AND IT STILL DIDN'T HAPPEN",
+ "raw": {
+ "normalized_reference": "i specifically asked for one thing. one thing. and it still didn't happen.",
+ "normalized_hypothesis": "i specifically asked for one thing one thing and it still didn't happen",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i specifically asked for one thing one thing and it still did not happen",
+ "normalized_hypothesis": "i specifically asked for one thing one thing and it still did not happen",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0018",
+ "category": "homographs",
+ "text": "They refuse to place the refuse beside the produce aisle.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0018.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THEY REFUSED TO PLACE THE REFUSE BESIDE THE PRODUCE ISLE",
+ "raw": {
+ "normalized_reference": "they refuse to place the refuse beside the produce aisle.",
+ "normalized_hypothesis": "they refused to place the refuse beside the produce isle",
+ "errors": 2,
+ "words": 10,
+ "wer": 0.2
+ },
+ "semantic": {
+ "normalized_reference": "they refuse to place the refuse beside the produce aisle",
+ "normalized_hypothesis": "they refused to place the refuse beside the produce isle",
+ "errors": 2,
+ "words": 10,
+ "wer": 0.2
+ }
+ },
+ {
+ "id": "release500_0019",
+ "category": "long",
+ "text": "There were days last year when I genuinely didn't know how I was going to get through it, and somehow I did, and I still don't fully understand how.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0019.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THERE WERE DAYS LAST YEAR WHEN I GENUINELY DIDN'T KNOW HOW I WAS GOING TO GET THROUGH IT AND SOMEHOW I DID AND I STILL DON'T FULLY UNDERSTAND HOW",
+ "raw": {
+ "normalized_reference": "there were days last year when i genuinely didn't know how i was going to get through it, and somehow i did, and i still don't fully understand how.",
+ "normalized_hypothesis": "there were days last year when i genuinely didn't know how i was going to get through it and somehow i did and i still don't fully understand how",
+ "errors": 0,
+ "words": 29,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "there were days last year when i genuinely did not know how i was going to get through it and somehow i did and i still do not fully understand how",
+ "normalized_hypothesis": "there were days last year when i genuinely did not know how i was going to get through it and somehow i did and i still do not fully understand how",
+ "errors": 0,
+ "words": 31,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0020",
+ "category": "long_range",
+ "text": "Before announcing the winner, the judge thanked the volunteers who had worked through the storm to repair the damaged stage.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0020.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "BEFORE ANNOUNCING THE WINNER THE JUDGE THANKED THE VOLUNTEERS WHO HAD WORKED THROUGH THE STORM TO REPAIR THE DAMAGED STAGE",
+ "raw": {
+ "normalized_reference": "before announcing the winner, the judge thanked the volunteers who had worked through the storm to repair the damaged stage.",
+ "normalized_hypothesis": "before announcing the winner the judge thanked the volunteers who had worked through the storm to repair the damaged stage",
+ "errors": 0,
+ "words": 20,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "before announcing the winner the judge thanked the volunteers who had worked through the storm to repair the damaged stage",
+ "normalized_hypothesis": "before announcing the winner the judge thanked the volunteers who had worked through the storm to repair the damaged stage",
+ "errors": 0,
+ "words": 20,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0021",
+ "category": "names_places",
+ "text": "Professor Nguyen mailed the fluorescent poster to Reykjavik on Thursday.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0021.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "PROFESSOR NEWYAN MAILED THE FLORESCENT POSTER TO REGJEVIK ON THURSDAY",
+ "raw": {
+ "normalized_reference": "professor nguyen mailed the fluorescent poster to reykjavik on thursday.",
+ "normalized_hypothesis": "professor newyan mailed the florescent poster to regjevik on thursday",
+ "errors": 3,
+ "words": 10,
+ "wer": 0.3
+ },
+ "semantic": {
+ "normalized_reference": "professor nguyen mailed the fluorescent poster to reykjavik on thursday",
+ "normalized_hypothesis": "professor newyan mailed the florescent poster to regjevik on thursday",
+ "errors": 3,
+ "words": 10,
+ "wer": 0.3
+ }
+ },
+ {
+ "id": "release500_0022",
+ "category": "narrative",
+ "text": "The plan was simple. The execution was anything but.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0022.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE PLAN WAS SIMPLE THE EXECUTION WAS ANYTHING BUT",
+ "raw": {
+ "normalized_reference": "the plan was simple. the execution was anything but.",
+ "normalized_hypothesis": "the plan was simple the execution was anything but",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the plan was simple the execution was anything but",
+ "normalized_hypothesis": "the plan was simple the execution was anything but",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0023",
+ "category": "numbers",
+ "text": "Her score improved from eighty-two percent to ninety-seven percent in two weeks.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0023.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HER SCORE IMPROVED FROM EIGHTY TWO PER CENT TO NINETY SEVEN PER CENT IN TWO WEEKS",
+ "raw": {
+ "normalized_reference": "her score improved from eighty-two percent to ninety-seven percent in two weeks.",
+ "normalized_hypothesis": "her score improved from eighty two per cent to ninety seven per cent in two weeks",
+ "errors": 4,
+ "words": 14,
+ "wer": 0.2857142857142857
+ },
+ "semantic": {
+ "normalized_reference": "her score improved from 82% to 97% in 2 weeks",
+ "normalized_hypothesis": "her score improved from 82% to 97% in 2 weeks",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0024",
+ "category": "punctuation",
+ "text": "Why would anyone pack spoons, batteries, and a violin for one short trip?",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0024.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WHY WOULD ANYONE PACK SPOONS BATTERIES AND A VIOLIN FOR ONE SHORT TRIP",
+ "raw": {
+ "normalized_reference": "why would anyone pack spoons, batteries, and a violin for one short trip?",
+ "normalized_hypothesis": "why would anyone pack spoons batteries and a violin for one short trip",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "why would anyone pack spoons batteries and a violin for one short trip",
+ "normalized_hypothesis": "why would anyone pack spoons batteries and a violin for one short trip",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0025",
+ "category": "question",
+ "text": "How long have you been sitting on this?",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0025.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HOW LONG HAVE YOU BEEN SITTING ON THIS",
+ "raw": {
+ "normalized_reference": "how long have you been sitting on this?",
+ "normalized_hypothesis": "how long have you been sitting on this",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "how long have you been sitting on this",
+ "normalized_hypothesis": "how long have you been sitting on this",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0026",
+ "category": "rare_words",
+ "text": "A mischievous chameleon rested beside the chrysanthemum and the bougainvillea.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0026.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "A MISCHIEVOUS CAMELIAN RESTED BESIDE THE CHRYSANTHEMUM AND THE BAUGANVILLIA",
+ "raw": {
+ "normalized_reference": "a mischievous chameleon rested beside the chrysanthemum and the bougainvillea.",
+ "normalized_hypothesis": "a mischievous camelian rested beside the chrysanthemum and the bauganvillia",
+ "errors": 2,
+ "words": 10,
+ "wer": 0.2
+ },
+ "semantic": {
+ "normalized_reference": "a mischievous chameleon rested beside the chrysanthemum and the bougainvillea",
+ "normalized_hypothesis": "a mischievous camelian rested beside the chrysanthemum and the bauganvillia",
+ "errors": 2,
+ "words": 10,
+ "wer": 0.2
+ }
+ },
+ {
+ "id": "release500_0027",
+ "category": "short",
+ "text": "We don't talk about that.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0027.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WE DON'T TALK ABOUT THAT",
+ "raw": {
+ "normalized_reference": "we don't talk about that.",
+ "normalized_hypothesis": "we don't talk about that",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "we do not talk about that",
+ "normalized_hypothesis": "we do not talk about that",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0028",
+ "category": "technical",
+ "text": "Hash the UTF eight payload with S H A two fifty-six before opening the checkpoint.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0028.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HASH THE U T F EIGHT PAY LOAD WITH S H A TWO FIFTY SIX BEFORE OPENING THE CHECK POINT",
+ "raw": {
+ "normalized_reference": "hash the utf eight payload with s h a two fifty-six before opening the checkpoint.",
+ "normalized_hypothesis": "hash the u t f eight pay load with s h a two fifty six before opening the check point",
+ "errors": 7,
+ "words": 16,
+ "wer": 0.4375
+ },
+ "semantic": {
+ "normalized_reference": "hash the utf 8 payload with s h a 256 before opening the checkpoint",
+ "normalized_hypothesis": "hash the u t f 8 pay load with s h a 256 before opening the check point",
+ "errors": 7,
+ "words": 14,
+ "wer": 0.5
+ }
+ },
+ {
+ "id": "release500_0029",
+ "category": "conversational",
+ "text": "I thought I had it all figured out, and then this happened.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0029.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I THOUGHT I HAD IT ALL FIGURED OUT AND THEN THIS HAPPENED",
+ "raw": {
+ "normalized_reference": "i thought i had it all figured out, and then this happened.",
+ "normalized_hypothesis": "i thought i had it all figured out and then this happened",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i thought i had it all figured out and then this happened",
+ "normalized_hypothesis": "i thought i had it all figured out and then this happened",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0030",
+ "category": "descriptive",
+ "text": "The kind of quiet you only get at three in the morning in a city.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0030.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE KIND OF QUIET YOU ONLY GET AT THREE IN THE MORNING IN A CITY",
+ "raw": {
+ "normalized_reference": "the kind of quiet you only get at three in the morning in a city.",
+ "normalized_hypothesis": "the kind of quiet you only get at three in the morning in a city",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the kind of quiet you only get at 3 in the morning in a city",
+ "normalized_hypothesis": "the kind of quiet you only get at 3 in the morning in a city",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0031",
+ "category": "emotional",
+ "text": "The joy of it was almost overwhelming, like too much good all at once.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0031.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE JOY OF IT WAS ALMOST OVERWHELMING LIKE TOO MUCH GOOD ALL AT ONCE",
+ "raw": {
+ "normalized_reference": "the joy of it was almost overwhelming, like too much good all at once.",
+ "normalized_hypothesis": "the joy of it was almost overwhelming like too much good all at once",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the joy of it was almost overwhelming like too much good all at once",
+ "normalized_hypothesis": "the joy of it was almost overwhelming like too much good all at once",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0032",
+ "category": "homographs",
+ "text": "Please close the door while I stand close to the microphone.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0032.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "PLEASE CLOSE THE DOOR WHILE I STAND CLOSE TO THE MICROPHONE",
+ "raw": {
+ "normalized_reference": "please close the door while i stand close to the microphone.",
+ "normalized_hypothesis": "please close the door while i stand close to the microphone",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "please close the door while i stand close to the microphone",
+ "normalized_hypothesis": "please close the door while i stand close to the microphone",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0033",
+ "category": "long",
+ "text": "I think the most underrated skill in the world is knowing when to stop explaining yourself and just let people draw their own conclusions.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0033.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I THINK THE MOST UNDURATED SKILL IN THE WORLD IS KNOWING WHEN TO STOP EXPLAINING YOURSELF AND JUST LET PEOPLE DRAW THEIR OWN CONCLUSIONS",
+ "raw": {
+ "normalized_reference": "i think the most underrated skill in the world is knowing when to stop explaining yourself and just let people draw their own conclusions.",
+ "normalized_hypothesis": "i think the most undurated skill in the world is knowing when to stop explaining yourself and just let people draw their own conclusions",
+ "errors": 1,
+ "words": 24,
+ "wer": 0.041666666666666664
+ },
+ "semantic": {
+ "normalized_reference": "i think the most underrated skill in the world is knowing when to stop explaining yourself and just let people draw their own conclusions",
+ "normalized_hypothesis": "i think the most undurated skill in the world is knowing when to stop explaining yourself and just let people draw their own conclusions",
+ "errors": 1,
+ "words": 24,
+ "wer": 0.041666666666666664
+ }
+ },
+ {
+ "id": "release500_0034",
+ "category": "long_range",
+ "text": "When the final bus disappeared around the corner, Nadia remembered that her umbrella, notebook, and lunch were still inside the quiet office.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0034.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WHEN THE FINAL BUST DISAPPEARED AROUND THE CORNER NODYA REMEMBERED THAT HER UMBRELLA NOTE BOOK AND LUNCH WERE STILL INSIDE THE QUIET OFFICE",
+ "raw": {
+ "normalized_reference": "when the final bus disappeared around the corner, nadia remembered that her umbrella, notebook, and lunch were still inside the quiet office.",
+ "normalized_hypothesis": "when the final bust disappeared around the corner nodya remembered that her umbrella note book and lunch were still inside the quiet office",
+ "errors": 4,
+ "words": 22,
+ "wer": 0.18181818181818182
+ },
+ "semantic": {
+ "normalized_reference": "when the final bus disappeared around the corner nadia remembered that her umbrella notebook and lunch were still inside the quiet office",
+ "normalized_hypothesis": "when the final bust disappeared around the corner nodya remembered that her umbrella note book and lunch were still inside the quiet office",
+ "errors": 4,
+ "words": 22,
+ "wer": 0.18181818181818182
+ }
+ },
+ {
+ "id": "release500_0035",
+ "category": "names_places",
+ "text": "Aurelia and Joaquin crossed the Schuylkill River before noon.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0035.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "ORELIAH AND JOQUIN CROSSED THE SHALKILL RIVER BEFORE NOON",
+ "raw": {
+ "normalized_reference": "aurelia and joaquin crossed the schuylkill river before noon.",
+ "normalized_hypothesis": "oreliah and joquin crossed the shalkill river before noon",
+ "errors": 3,
+ "words": 9,
+ "wer": 0.3333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "aurelia and joaquin crossed the schuylkill river before noon",
+ "normalized_hypothesis": "oreliah and joquin crossed the shalkill river before noon",
+ "errors": 3,
+ "words": 9,
+ "wer": 0.3333333333333333
+ }
+ },
+ {
+ "id": "release500_0036",
+ "category": "narrative",
+ "text": "The village had a name, but nobody used it. Everyone just called it home.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0036.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE VILLAGE HAD A NAME BUT NOBODY USED IT EVERYONE JUST CALLED IT HOME",
+ "raw": {
+ "normalized_reference": "the village had a name, but nobody used it. everyone just called it home.",
+ "normalized_hypothesis": "the village had a name but nobody used it everyone just called it home",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the village had a name but nobody used it everyone just called it home",
+ "normalized_hypothesis": "the village had a name but nobody used it everyone just called it home",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0037",
+ "category": "numbers",
+ "text": "Increase the gain by three decibels, wait five seconds, then reduce it by one.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0037.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "INCREASE THE GAIN BY THREE DESIBLES WEIGHT FIVE SECONDS THEN REDUCE IT BY ONE",
+ "raw": {
+ "normalized_reference": "increase the gain by three decibels, wait five seconds, then reduce it by one.",
+ "normalized_hypothesis": "increase the gain by three desibles weight five seconds then reduce it by one",
+ "errors": 2,
+ "words": 14,
+ "wer": 0.14285714285714285
+ },
+ "semantic": {
+ "normalized_reference": "increase the gain by 3 decibels wait 5 seconds then reduce it by one",
+ "normalized_hypothesis": "increase the gain by 3 desibles weight 5 seconds then reduce it by one",
+ "errors": 2,
+ "words": 14,
+ "wer": 0.14285714285714285
+ }
+ },
+ {
+ "id": "release500_0038",
+ "category": "punctuation",
+ "text": "The result was unexpected: clear, stable, and surprisingly natural.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0038.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE RESULT WAS UNEXPECTED CLEAR STABLE AND SURPRISINGLY NATURAL",
+ "raw": {
+ "normalized_reference": "the result was unexpected: clear, stable, and surprisingly natural.",
+ "normalized_hypothesis": "the result was unexpected clear stable and surprisingly natural",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the result was unexpected clear stable and surprisingly natural",
+ "normalized_hypothesis": "the result was unexpected clear stable and surprisingly natural",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0039",
+ "category": "question",
+ "text": "Who told you that was the right way to do it?",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0039.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WHO TOLD YOU THAT WAS THE RIGHT WAY TO DO IT",
+ "raw": {
+ "normalized_reference": "who told you that was the right way to do it?",
+ "normalized_hypothesis": "who told you that was the right way to do it",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "who told you that was the right way to do it",
+ "normalized_hypothesis": "who told you that was the right way to do it",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0040",
+ "category": "rare_words",
+ "text": "The meticulous cartographer examined an iridescent archipelago through binoculars.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0040.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE METICULOUS CARTOGRAPHER EXAMINED AN IRIDESCENT ARCHIPELAGO THROUGH BINOCULARS",
+ "raw": {
+ "normalized_reference": "the meticulous cartographer examined an iridescent archipelago through binoculars.",
+ "normalized_hypothesis": "the meticulous cartographer examined an iridescent archipelago through binoculars",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the meticulous cartographer examined an iridescent archipelago through binoculars",
+ "normalized_hypothesis": "the meticulous cartographer examined an iridescent archipelago through binoculars",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0041",
+ "category": "short",
+ "text": "I mean. Yeah. Kind of.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0041.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I MEAN YET KIND OF",
+ "raw": {
+ "normalized_reference": "i mean. yeah. kind of.",
+ "normalized_hypothesis": "i mean yet kind of",
+ "errors": 1,
+ "words": 5,
+ "wer": 0.2
+ },
+ "semantic": {
+ "normalized_reference": "i mean yeah kind of",
+ "normalized_hypothesis": "i mean yet kind of",
+ "errors": 1,
+ "words": 5,
+ "wer": 0.2
+ }
+ },
+ {
+ "id": "release500_0042",
+ "category": "technical",
+ "text": "The logs are your best friend when something behaves unexpectedly.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0042.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE LOGS ARE YOUR BEST FRIEND WHEN SOMETHING BEHAVES UNEXPECTEDLY",
+ "raw": {
+ "normalized_reference": "the logs are your best friend when something behaves unexpectedly.",
+ "normalized_hypothesis": "the logs are your best friend when something behaves unexpectedly",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the logs are your best friend when something behaves unexpectedly",
+ "normalized_hypothesis": "the logs are your best friend when something behaves unexpectedly",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0043",
+ "category": "conversational",
+ "text": "He kept talking and I kept nodding and neither of us was listening.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0043.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE KEPT TALKING AND I KEPT NODDING AND NEITHER OF US WAS LISTENING",
+ "raw": {
+ "normalized_reference": "he kept talking and i kept nodding and neither of us was listening.",
+ "normalized_hypothesis": "he kept talking and i kept nodding and neither of us was listening",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he kept talking and i kept nodding and neither of us was listening",
+ "normalized_hypothesis": "he kept talking and i kept nodding and neither of us was listening",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0044",
+ "category": "descriptive",
+ "text": "It's one of those places where everyone seems to know each other already.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0044.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT'S ONE OF THOSE PLACES WHERE EVERYONE SEEMS TO KNOW EACH OTHER ALREADY",
+ "raw": {
+ "normalized_reference": "it's one of those places where everyone seems to know each other already.",
+ "normalized_hypothesis": "it's one of those places where everyone seems to know each other already",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it is one of those places where everyone seems to know each other already",
+ "normalized_hypothesis": "it is one of those places where everyone seems to know each other already",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0045",
+ "category": "emotional",
+ "text": "He said he was proud of me and nobody had ever said that to me before.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0045.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE SAID HE WAS PROUD OF ME AND NOBODY HAD EVER SAID THAT TO ME BEFORE",
+ "raw": {
+ "normalized_reference": "he said he was proud of me and nobody had ever said that to me before.",
+ "normalized_hypothesis": "he said he was proud of me and nobody had ever said that to me before",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he said he was proud of me and nobody had ever said that to me before",
+ "normalized_hypothesis": "he said he was proud of me and nobody had ever said that to me before",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0046",
+ "category": "homographs",
+ "text": "The minute detail took a minute to explain clearly.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0046.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE MINUTE DETAIL TOOK A MINUTE TO EXPLAIN CLEARLY",
+ "raw": {
+ "normalized_reference": "the minute detail took a minute to explain clearly.",
+ "normalized_hypothesis": "the minute detail took a minute to explain clearly",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the minute detail took a minute to explain clearly",
+ "normalized_hypothesis": "the minute detail took a minute to explain clearly",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0047",
+ "category": "long",
+ "text": "The strange thing about really wanting something is that getting it never quite feels the way you imagined it would, and not getting it never quite feels as bad.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0047.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE STRANGE THING ABOUT REALLY WANTING SOMETHING IS THAT GETTING IT NEVER QUITE FEELS THE WAY YOU IMAGINED IT WOULD AND NOT GETTING IT NEVER QUITE FEELS AS BAD",
+ "raw": {
+ "normalized_reference": "the strange thing about really wanting something is that getting it never quite feels the way you imagined it would, and not getting it never quite feels as bad.",
+ "normalized_hypothesis": "the strange thing about really wanting something is that getting it never quite feels the way you imagined it would and not getting it never quite feels as bad",
+ "errors": 0,
+ "words": 29,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the strange thing about really wanting something is that getting it never quite feels the way you imagined it would and not getting it never quite feels as bad",
+ "normalized_hypothesis": "the strange thing about really wanting something is that getting it never quite feels the way you imagined it would and not getting it never quite feels as bad",
+ "errors": 0,
+ "words": 29,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0048",
+ "category": "long_range",
+ "text": "Even though the recipe looked simple at first, preparing the sauce required patience, careful timing, and a much larger pan.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0048.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "EVEN THOUGH THE RECIPE LOOKED SIMPLE AT FIRST PREPARING THE SAUCE REQUIRED PATIENCE CAREFUL TIMING AND A MUCH LARGER PAN",
+ "raw": {
+ "normalized_reference": "even though the recipe looked simple at first, preparing the sauce required patience, careful timing, and a much larger pan.",
+ "normalized_hypothesis": "even though the recipe looked simple at first preparing the sauce required patience careful timing and a much larger pan",
+ "errors": 0,
+ "words": 20,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "even though the recipe looked simple at 1st preparing the sauce required patience careful timing and a much larger pan",
+ "normalized_hypothesis": "even though the recipe looked simple at 1st preparing the sauce required patience careful timing and a much larger pan",
+ "errors": 0,
+ "words": 20,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0049",
+ "category": "names_places",
+ "text": "Xiomara met Siobhan near the old library in Worcestershire.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0049.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "ZIMARA MET CHAVONN NEAR THE OLD LIBRARY IN WORCESTERSHIRE",
+ "raw": {
+ "normalized_reference": "xiomara met siobhan near the old library in worcestershire.",
+ "normalized_hypothesis": "zimara met chavonn near the old library in worcestershire",
+ "errors": 2,
+ "words": 9,
+ "wer": 0.2222222222222222
+ },
+ "semantic": {
+ "normalized_reference": "xiomara met siobhan near the old library in worcestershire",
+ "normalized_hypothesis": "zimara met chavonn near the old library in worcestershire",
+ "errors": 2,
+ "words": 9,
+ "wer": 0.2222222222222222
+ }
+ },
+ {
+ "id": "release500_0050",
+ "category": "narrative",
+ "text": "She walked into the room and everyone went quiet, not because she was loud, but because she wasn't.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0050.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SHE WALKED INTO THE ROOM AND EVERYONE WENT QUIET NOT BECAUSE SHE WAS LOUD BUT BECAUSE SHE WAS",
+ "raw": {
+ "normalized_reference": "she walked into the room and everyone went quiet, not because she was loud, but because she wasn't.",
+ "normalized_hypothesis": "she walked into the room and everyone went quiet not because she was loud but because she was",
+ "errors": 1,
+ "words": 18,
+ "wer": 0.05555555555555555
+ },
+ "semantic": {
+ "normalized_reference": "she walked into the room and everyone went quiet not because she was loud but because she was not",
+ "normalized_hypothesis": "she walked into the room and everyone went quiet not because she was loud but because she was",
+ "errors": 1,
+ "words": 19,
+ "wer": 0.05263157894736842
+ }
+ },
+ {
+ "id": "release500_0051",
+ "category": "numbers",
+ "text": "At exactly seven forty-five, Morgan placed eighteen dollars and fifty cents beside the notebook.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0051.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "AT EXACTLY SEVEN FORTY FIVE MORGAN PLACED EIGHTEEN DOLLARS AND FIFTY CENTS BESIDE THE NOTE BOOK",
+ "raw": {
+ "normalized_reference": "at exactly seven forty-five, morgan placed eighteen dollars and fifty cents beside the notebook.",
+ "normalized_hypothesis": "at exactly seven forty five morgan placed eighteen dollars and fifty cents beside the note book",
+ "errors": 2,
+ "words": 15,
+ "wer": 0.13333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "at exactly 745 morgan placed $18.50 beside the notebook",
+ "normalized_hypothesis": "at exactly 745 morgan placed $18.50 beside the note book",
+ "errors": 2,
+ "words": 10,
+ "wer": 0.2
+ }
+ },
+ {
+ "id": "release500_0052",
+ "category": "punctuation",
+ "text": "Does punctuation work? Yes: commas, semicolons, and periods should create sensible boundaries.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0052.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "DOES PUNCTUATION WORK YES CAMAS SEMICOLANS AND PERIODS SHOULD CREATE SENSIBLE BOUNDARIES",
+ "raw": {
+ "normalized_reference": "does punctuation work? yes: commas, semicolons, and periods should create sensible boundaries.",
+ "normalized_hypothesis": "does punctuation work yes camas semicolans and periods should create sensible boundaries",
+ "errors": 2,
+ "words": 12,
+ "wer": 0.16666666666666666
+ },
+ "semantic": {
+ "normalized_reference": "does punctuation work yes commas semicolons and periods should create sensible boundaries",
+ "normalized_hypothesis": "does punctuation work yes camas semicolans and periods should create sensible boundaries",
+ "errors": 2,
+ "words": 12,
+ "wer": 0.16666666666666666
+ }
+ },
+ {
+ "id": "release500_0053",
+ "category": "question",
+ "text": "Have you ever just stopped and thought about how strange all of this is?",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0053.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HAVE YOU EVER JUST STOPPED AND THOUGHT ABOUT HOW STRANGE ALL OF THIS IS",
+ "raw": {
+ "normalized_reference": "have you ever just stopped and thought about how strange all of this is?",
+ "normalized_hypothesis": "have you ever just stopped and thought about how strange all of this is",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "have you ever just stopped and thought about how strange all of this is",
+ "normalized_hypothesis": "have you ever just stopped and thought about how strange all of this is",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0054",
+ "category": "rare_words",
+ "text": "Her idiosyncratic explanation sounded plausible but remained fundamentally ambiguous.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0054.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HER IDIOSYNCRATIC EXPLANATIONS SOUNDED PLAUSIBLE BUT REMAINED FUNDAMENTALLY AMBIGUOUS",
+ "raw": {
+ "normalized_reference": "her idiosyncratic explanation sounded plausible but remained fundamentally ambiguous.",
+ "normalized_hypothesis": "her idiosyncratic explanations sounded plausible but remained fundamentally ambiguous",
+ "errors": 1,
+ "words": 9,
+ "wer": 0.1111111111111111
+ },
+ "semantic": {
+ "normalized_reference": "her idiosyncratic explanation sounded plausible but remained fundamentally ambiguous",
+ "normalized_hypothesis": "her idiosyncratic explanations sounded plausible but remained fundamentally ambiguous",
+ "errors": 1,
+ "words": 9,
+ "wer": 0.1111111111111111
+ }
+ },
+ {
+ "id": "release500_0055",
+ "category": "short",
+ "text": "That tracks.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0055.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THAT TRACKS",
+ "raw": {
+ "normalized_reference": "that tracks.",
+ "normalized_hypothesis": "that tracks",
+ "errors": 0,
+ "words": 2,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "that tracks",
+ "normalized_hypothesis": "that tracks",
+ "errors": 0,
+ "words": 2,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0056",
+ "category": "technical",
+ "text": "The key thing to remember is that order matters here more than you'd think.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0056.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE KEY THING TO REMEMBER IS THAT ORDER MATTERS HERE MORE THAN YOU'D THINK",
+ "raw": {
+ "normalized_reference": "the key thing to remember is that order matters here more than you'd think.",
+ "normalized_hypothesis": "the key thing to remember is that order matters here more than you'd think",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the key thing to remember is that order matters here more than you would think",
+ "normalized_hypothesis": "the key thing to remember is that order matters here more than you would think",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0057",
+ "category": "conversational",
+ "text": "I would not say that again in this room if I were you.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0057.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I WOULD NOT SAY THAT AGAIN IN THIS ROOM IF I WERE YOU",
+ "raw": {
+ "normalized_reference": "i would not say that again in this room if i were you.",
+ "normalized_hypothesis": "i would not say that again in this room if i were you",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i would not say that again in this room if i were you",
+ "normalized_hypothesis": "i would not say that again in this room if i were you",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0058",
+ "category": "descriptive",
+ "text": "The room was filled with the quiet that means something is about to change.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0058.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE ROOM WAS FILLED WITH THE QUIET THAT MEANS SOMETHING IS ABOUT TO CHANGE",
+ "raw": {
+ "normalized_reference": "the room was filled with the quiet that means something is about to change.",
+ "normalized_hypothesis": "the room was filled with the quiet that means something is about to change",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the room was filled with the quiet that means something is about to change",
+ "normalized_hypothesis": "the room was filled with the quiet that means something is about to change",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0059",
+ "category": "emotional",
+ "text": "I didn't expect to feel that way about it, but I did.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0059.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I DIDN'T EXPECT TO FEEL THAT WAY ABOUT IT BUT I DID",
+ "raw": {
+ "normalized_reference": "i didn't expect to feel that way about it, but i did.",
+ "normalized_hypothesis": "i didn't expect to feel that way about it but i did",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i did not expect to feel that way about it but i did",
+ "normalized_hypothesis": "i did not expect to feel that way about it but i did",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0060",
+ "category": "homographs",
+ "text": "I object to the strange object that was left near the entrance.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0060.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I OBJECT TO THE STRANGE OBJECT THAT WAS LEFT NEAR THE ENTRANCE",
+ "raw": {
+ "normalized_reference": "i object to the strange object that was left near the entrance.",
+ "normalized_hypothesis": "i object to the strange object that was left near the entrance",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i object to the strange object that was left near the entrance",
+ "normalized_hypothesis": "i object to the strange object that was left near the entrance",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0061",
+ "category": "long",
+ "text": "I've learned more from the things that went wrong than from the things that worked, which is a frustrating way to have to learn things but apparently the only way that stuck.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0061.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I'VE LEARNED MORE FROM THE THINGS THAT WENT WRONG THAN FROM THE THINGS THAT WORKED WHICH IS A FRUSTRATING WAY TO HAVE TO LEARN THINGS BUT APPARENTLY THE ONLY WAY THAT STUCK",
+ "raw": {
+ "normalized_reference": "i've learned more from the things that went wrong than from the things that worked, which is a frustrating way to have to learn things but apparently the only way that stuck.",
+ "normalized_hypothesis": "i've learned more from the things that went wrong than from the things that worked which is a frustrating way to have to learn things but apparently the only way that stuck",
+ "errors": 0,
+ "words": 32,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i have learned more from the things that went wrong than from the things that worked which is a frustrating way to have to learn things but apparently the only way that stuck",
+ "normalized_hypothesis": "i have learned more from the things that went wrong than from the things that worked which is a frustrating way to have to learn things but apparently the only way that stuck",
+ "errors": 0,
+ "words": 33,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0062",
+ "category": "long_range",
+ "text": "Because every nearby hotel was full, the exhausted hikers followed the narrow road until they found a farmhouse with one empty room.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0062.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "BECAUSE EVERY NEARBY HOTEL WAS FULL THE EXHAUSTED HYKERS FOLLOWED THE NARROW ROAD UNTIL THEY FOUND A FARMHOUSE WITH ONE EMPTY ROOM",
+ "raw": {
+ "normalized_reference": "because every nearby hotel was full, the exhausted hikers followed the narrow road until they found a farmhouse with one empty room.",
+ "normalized_hypothesis": "because every nearby hotel was full the exhausted hykers followed the narrow road until they found a farmhouse with one empty room",
+ "errors": 1,
+ "words": 22,
+ "wer": 0.045454545454545456
+ },
+ "semantic": {
+ "normalized_reference": "because every nearby hotel was full the exhausted hikers followed the narrow road until they found a farmhouse with one empty room",
+ "normalized_hypothesis": "because every nearby hotel was full the exhausted hykers followed the narrow road until they found a farmhouse with one empty room",
+ "errors": 1,
+ "words": 22,
+ "wer": 0.045454545454545456
+ }
+ },
+ {
+ "id": "release500_0063",
+ "category": "names_places",
+ "text": "The itinerary sends Beatrice from Albuquerque to Des Moines and then to Boise.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0063.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE ITINERARY SENDS BEATRICE FROM ALBAKIRKI TO DEMOIN AND THEN TO BOISEY",
+ "raw": {
+ "normalized_reference": "the itinerary sends beatrice from albuquerque to des moines and then to boise.",
+ "normalized_hypothesis": "the itinerary sends beatrice from albakirki to demoin and then to boisey",
+ "errors": 4,
+ "words": 13,
+ "wer": 0.3076923076923077
+ },
+ "semantic": {
+ "normalized_reference": "the itinerary sends beatrice from albuquerque to des moines and then to boise",
+ "normalized_hypothesis": "the itinerary sends beatrice from albakirki to demoin and then to boisey",
+ "errors": 4,
+ "words": 13,
+ "wer": 0.3076923076923077
+ }
+ },
+ {
+ "id": "release500_0064",
+ "category": "narrative",
+ "text": "We had been driving for six hours when we finally saw the sign for the town.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0064.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WE HAD BEEN DRIVING FOR SIX HOURS WHEN WE FINALLY SAW THE SIGN FOR THE TOWN",
+ "raw": {
+ "normalized_reference": "we had been driving for six hours when we finally saw the sign for the town.",
+ "normalized_hypothesis": "we had been driving for six hours when we finally saw the sign for the town",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "we had been driving for 6 hours when we finally saw the sign for the town",
+ "normalized_hypothesis": "we had been driving for 6 hours when we finally saw the sign for the town",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0065",
+ "category": "numbers",
+ "text": "Call extension zero four nine after eight thirty in the morning.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0065.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "CALL EXTENSION ZERO FOUR NINE AFTER EIGHT THIRTY IN THE MORNING",
+ "raw": {
+ "normalized_reference": "call extension zero four nine after eight thirty in the morning.",
+ "normalized_hypothesis": "call extension zero four nine after eight thirty in the morning",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "call extension 049 after 830 in the morning",
+ "normalized_hypothesis": "call extension 049 after 830 in the morning",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0066",
+ "category": "punctuation",
+ "text": "Stop - listen carefully - then answer the question without guessing.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0066.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "STOP LISTEN CAREFULLY THEN ANSWER THE QUESTION WITHOUT GUESSING",
+ "raw": {
+ "normalized_reference": "stop - listen carefully - then answer the question without guessing.",
+ "normalized_hypothesis": "stop listen carefully then answer the question without guessing",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "stop listen carefully then answer the question without guessing",
+ "normalized_hypothesis": "stop listen carefully then answer the question without guessing",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0067",
+ "category": "question",
+ "text": "Does it matter at this point, honestly?",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0067.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "DOES IT MATTER AT THIS POINT HONESTLY",
+ "raw": {
+ "normalized_reference": "does it matter at this point, honestly?",
+ "normalized_hypothesis": "does it matter at this point honestly",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "does it matter at this point honestly",
+ "normalized_hypothesis": "does it matter at this point honestly",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0068",
+ "category": "rare_words",
+ "text": "An entrepreneurial meteorologist described the unprecedented atmospheric phenomenon.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0068.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "AN ONTRIPRENURIAL METEOROLOGIST DESCRIBED THE UNPRECEDENTED ATMOSPHERIC PHENOMENON",
+ "raw": {
+ "normalized_reference": "an entrepreneurial meteorologist described the unprecedented atmospheric phenomenon.",
+ "normalized_hypothesis": "an ontriprenurial meteorologist described the unprecedented atmospheric phenomenon",
+ "errors": 1,
+ "words": 8,
+ "wer": 0.125
+ },
+ "semantic": {
+ "normalized_reference": "an entrepreneurial meteorologist described the unprecedented atmospheric phenomenon",
+ "normalized_hypothesis": "an ontriprenurial meteorologist described the unprecedented atmospheric phenomenon",
+ "errors": 1,
+ "words": 8,
+ "wer": 0.125
+ }
+ },
+ {
+ "id": "release500_0069",
+ "category": "short",
+ "text": "That's not what I expected.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0069.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THAT'S NOT WHAT I EXPECTED",
+ "raw": {
+ "normalized_reference": "that's not what i expected.",
+ "normalized_hypothesis": "that's not what i expected",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "that is not what i expected",
+ "normalized_hypothesis": "that is not what i expected",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0070",
+ "category": "technical",
+ "text": "There's a known issue with that version that causes failures under load.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0070.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THERE IS A KNOWN ISSUE WITH THAT VERSION THAT CAUSES FAILURES UNDER LOAD",
+ "raw": {
+ "normalized_reference": "there's a known issue with that version that causes failures under load.",
+ "normalized_hypothesis": "there is a known issue with that version that causes failures under load",
+ "errors": 2,
+ "words": 12,
+ "wer": 0.16666666666666666
+ },
+ "semantic": {
+ "normalized_reference": "there is a known issue with that version that causes failures under load",
+ "normalized_hypothesis": "there is a known issue with that version that causes failures under load",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0071",
+ "category": "conversational",
+ "text": "Don't look now but I think that's him over there.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0071.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "DON'T LOOK NOW BUT I THINK THAT'S HIM OVER THERE",
+ "raw": {
+ "normalized_reference": "don't look now but i think that's him over there.",
+ "normalized_hypothesis": "don't look now but i think that's him over there",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "do not look now but i think that is him over there",
+ "normalized_hypothesis": "do not look now but i think that is him over there",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0072",
+ "category": "descriptive",
+ "text": "The city at night looks completely different from how it feels during the day.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0072.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE CITY AT NIGHT LOOKS COMPLETELY DIFFERENT FROM HOW IT FEELS DURING THE DAY",
+ "raw": {
+ "normalized_reference": "the city at night looks completely different from how it feels during the day.",
+ "normalized_hypothesis": "the city at night looks completely different from how it feels during the day",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the city at night looks completely different from how it feels during the day",
+ "normalized_hypothesis": "the city at night looks completely different from how it feels during the day",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0073",
+ "category": "emotional",
+ "text": "It's the kind of happy that makes you nervous because you know it can't last.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0073.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT'S THE KIND OF HAPPY THAT MAKES YOU NERVOUS BECAUSE YOU KNOW IT CAN'T LAST",
+ "raw": {
+ "normalized_reference": "it's the kind of happy that makes you nervous because you know it can't last.",
+ "normalized_hypothesis": "it's the kind of happy that makes you nervous because you know it can't last",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it is the kind of happy that makes you nervous because you know it can not last",
+ "normalized_hypothesis": "it is the kind of happy that makes you nervous because you know it can not last",
+ "errors": 0,
+ "words": 17,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0074",
+ "category": "homographs",
+ "text": "The dove dove beneath the branch while the wind began to wind the loose cable.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0074.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE DOVE DOVED BENEATH THE BRANCH WHILE THE WIND BEGAN TO WIND THE LOOSE CABLE",
+ "raw": {
+ "normalized_reference": "the dove dove beneath the branch while the wind began to wind the loose cable.",
+ "normalized_hypothesis": "the dove doved beneath the branch while the wind began to wind the loose cable",
+ "errors": 1,
+ "words": 15,
+ "wer": 0.06666666666666667
+ },
+ "semantic": {
+ "normalized_reference": "the dove dove beneath the branch while the wind began to wind the loose cable",
+ "normalized_hypothesis": "the dove doved beneath the branch while the wind began to wind the loose cable",
+ "errors": 1,
+ "words": 15,
+ "wer": 0.06666666666666667
+ }
+ },
+ {
+ "id": "release500_0075",
+ "category": "long",
+ "text": "I've thought a lot about what I would do differently if I could go back, and the honest answer is probably not as much as I'd like to think.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0075.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I'VE THOUGHT A LOT ABOUT WHAT I WOULD DO DIFFERENTLY IF I COULD GO BACK AND THE HONEST ANSWER IS PROBABLY NOT AS MUCH AS I'D LIKE TO THINK",
+ "raw": {
+ "normalized_reference": "i've thought a lot about what i would do differently if i could go back, and the honest answer is probably not as much as i'd like to think.",
+ "normalized_hypothesis": "i've thought a lot about what i would do differently if i could go back and the honest answer is probably not as much as i'd like to think",
+ "errors": 0,
+ "words": 29,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i have thought a lot about what i would do differently if i could go back and the honest answer is probably not as much as i would like to think",
+ "normalized_hypothesis": "i have thought a lot about what i would do differently if i could go back and the honest answer is probably not as much as i would like to think",
+ "errors": 0,
+ "words": 31,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0076",
+ "category": "long_range",
+ "text": "Although the weather changed without warning, the musicians finished packing before anyone opened the enormous glass door.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0076.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "ALTHOUGH THE WEATHER CHANGED WITHOUT WARNING THE MUSICIANS FINISHED PACKING BEFORE ANY ONE OPENED THE ENORMOUS GLASS DOOR",
+ "raw": {
+ "normalized_reference": "although the weather changed without warning, the musicians finished packing before anyone opened the enormous glass door.",
+ "normalized_hypothesis": "although the weather changed without warning the musicians finished packing before any one opened the enormous glass door",
+ "errors": 2,
+ "words": 17,
+ "wer": 0.11764705882352941
+ },
+ "semantic": {
+ "normalized_reference": "although the weather changed without warning the musicians finished packing before anyone opened the enormous glass door",
+ "normalized_hypothesis": "although the weather changed without warning the musicians finished packing before any one opened the enormous glass door",
+ "errors": 2,
+ "words": 17,
+ "wer": 0.11764705882352941
+ }
+ },
+ {
+ "id": "release500_0077",
+ "category": "names_places",
+ "text": "Saskatchewan looked unusually bright beneath the pale February sunrise.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0077.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SASKECHUAN LOOKED UNUSUALLY BRIGHT BENEATH THE PALE FEBRUARY SUNRISE",
+ "raw": {
+ "normalized_reference": "saskatchewan looked unusually bright beneath the pale february sunrise.",
+ "normalized_hypothesis": "saskechuan looked unusually bright beneath the pale february sunrise",
+ "errors": 1,
+ "words": 9,
+ "wer": 0.1111111111111111
+ },
+ "semantic": {
+ "normalized_reference": "saskatchewan looked unusually bright beneath the pale february sunrise",
+ "normalized_hypothesis": "saskechuan looked unusually bright beneath the pale february sunrise",
+ "errors": 1,
+ "words": 9,
+ "wer": 0.1111111111111111
+ }
+ },
+ {
+ "id": "release500_0078",
+ "category": "narrative",
+ "text": "She handled it with a grace I genuinely did not expect.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0078.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SHE HANDLED IT WITH A GRACE I GENUINELY DID NOT EXPECT",
+ "raw": {
+ "normalized_reference": "she handled it with a grace i genuinely did not expect.",
+ "normalized_hypothesis": "she handled it with a grace i genuinely did not expect",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "she handled it with a grace i genuinely did not expect",
+ "normalized_hypothesis": "she handled it with a grace i genuinely did not expect",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0079",
+ "category": "numbers",
+ "text": "Apartment four hundred seven is across from room twenty-nine on the third floor.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0079.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "APARTMENT FOUR SEVEN IS ACROSS FROM ROOM TWENTY NINE ON THE THIRD FLOOR",
+ "raw": {
+ "normalized_reference": "apartment four hundred seven is across from room twenty-nine on the third floor.",
+ "normalized_hypothesis": "apartment four seven is across from room twenty nine on the third floor",
+ "errors": 1,
+ "words": 14,
+ "wer": 0.07142857142857142
+ },
+ "semantic": {
+ "normalized_reference": "apartment 407 is across from room 29 on the 3rd floor",
+ "normalized_hypothesis": "apartment 47 is across from room 29 on the 3rd floor",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ }
+ },
+ {
+ "id": "release500_0080",
+ "category": "punctuation",
+ "text": "\"Not yet,\" Rowan whispered. \"Give me another minute... please.\"",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0080.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "NOT YET ROWEN WHISPERED GIVE ME ANOTHER MINUTE PLEASE",
+ "raw": {
+ "normalized_reference": "\"not yet,\" rowan whispered. \"give me another minute... please.\"",
+ "normalized_hypothesis": "not yet rowen whispered give me another minute please",
+ "errors": 1,
+ "words": 9,
+ "wer": 0.1111111111111111
+ },
+ "semantic": {
+ "normalized_reference": "not yet rowan whispered give me another minute please",
+ "normalized_hypothesis": "not yet rowen whispered give me another minute please",
+ "errors": 1,
+ "words": 9,
+ "wer": 0.1111111111111111
+ }
+ },
+ {
+ "id": "release500_0081",
+ "category": "question",
+ "text": "What are we actually trying to accomplish here?",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0081.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WHAT ARE WE ACTUALLY TRYING TO ACCOMPLISH HERE",
+ "raw": {
+ "normalized_reference": "what are we actually trying to accomplish here?",
+ "normalized_hypothesis": "what are we actually trying to accomplish here",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "what are we actually trying to accomplish here",
+ "normalized_hypothesis": "what are we actually trying to accomplish here",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0082",
+ "category": "rare_words",
+ "text": "The physician recommended a noninvasive procedure for the recurring neurological symptoms.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0082.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE PHYSICIAN RECOMMENDED A NON EVASIVE PROCEDURE FOR THE RECURRING NEUROLOGICAL SYMPTOMS",
+ "raw": {
+ "normalized_reference": "the physician recommended a noninvasive procedure for the recurring neurological symptoms.",
+ "normalized_hypothesis": "the physician recommended a non evasive procedure for the recurring neurological symptoms",
+ "errors": 2,
+ "words": 11,
+ "wer": 0.18181818181818182
+ },
+ "semantic": {
+ "normalized_reference": "the physician recommended a noninvasive procedure for the recurring neurological symptoms",
+ "normalized_hypothesis": "the physician recommended a non evasive procedure for the recurring neurological symptoms",
+ "errors": 2,
+ "words": 11,
+ "wer": 0.18181818181818182
+ }
+ },
+ {
+ "id": "release500_0083",
+ "category": "short",
+ "text": "Are you serious right now?",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0083.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "ARE YOU SERIOUS RIGHT NOW",
+ "raw": {
+ "normalized_reference": "are you serious right now?",
+ "normalized_hypothesis": "are you serious right now",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "are you serious right now",
+ "normalized_hypothesis": "are you serious right now",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0084",
+ "category": "technical",
+ "text": "The performance improvement is real, but it comes with tradeoffs.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0084.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE PERFORMANCE IMPROVEMENT IS REAL BUT IT COMES WITH TRADE OFFS",
+ "raw": {
+ "normalized_reference": "the performance improvement is real, but it comes with tradeoffs.",
+ "normalized_hypothesis": "the performance improvement is real but it comes with trade offs",
+ "errors": 2,
+ "words": 10,
+ "wer": 0.2
+ },
+ "semantic": {
+ "normalized_reference": "the performance improvement is real but it comes with tradeoffs",
+ "normalized_hypothesis": "the performance improvement is real but it comes with trade offs",
+ "errors": 2,
+ "words": 10,
+ "wer": 0.2
+ }
+ },
+ {
+ "id": "release500_0085",
+ "category": "conversational",
+ "text": "I don't know. It's just been one of those weeks, you know?",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0085.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I DON'T KNOW IT'S JUST BEEN ONE OF THOSE WEEKS YOU KNOW",
+ "raw": {
+ "normalized_reference": "i don't know. it's just been one of those weeks, you know?",
+ "normalized_hypothesis": "i don't know it's just been one of those weeks you know",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i do not know it is just been one of those weeks you know",
+ "normalized_hypothesis": "i do not know it is just been one of those weeks you know",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0086",
+ "category": "descriptive",
+ "text": "The moment you stop expecting perfection, things get a lot more interesting.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0086.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE MOMENT YOU STOP EXPECTING PERFECTION THINGS GET A LOT MORE INTERESTING",
+ "raw": {
+ "normalized_reference": "the moment you stop expecting perfection, things get a lot more interesting.",
+ "normalized_hypothesis": "the moment you stop expecting perfection things get a lot more interesting",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the moment you stop expecting perfection things get a lot more interesting",
+ "normalized_hypothesis": "the moment you stop expecting perfection things get a lot more interesting",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0087",
+ "category": "emotional",
+ "text": "It's the kind of thing that stays with you even years later.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0087.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT'S THE KIND OF THING THAT STAYS WITH YOU EVEN YEARS LATER",
+ "raw": {
+ "normalized_reference": "it's the kind of thing that stays with you even years later.",
+ "normalized_hypothesis": "it's the kind of thing that stays with you even years later",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it is the kind of thing that stays with you even years later",
+ "normalized_hypothesis": "it is the kind of thing that stays with you even years later",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0088",
+ "category": "long",
+ "text": "She told me once that the most important skill she'd developed was learning to disagree with someone without making them feel they were wrong.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0088.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SHE TOLD ME ONCE THAT THE MOST IMPORTANT SKILL SHE DEVELOPED WAS LEARNING TO DISAGREE WITH SOMEONE WITHOUT MAKING THEM FEEL THEY WERE WRONG",
+ "raw": {
+ "normalized_reference": "she told me once that the most important skill she'd developed was learning to disagree with someone without making them feel they were wrong.",
+ "normalized_hypothesis": "she told me once that the most important skill she developed was learning to disagree with someone without making them feel they were wrong",
+ "errors": 1,
+ "words": 24,
+ "wer": 0.041666666666666664
+ },
+ "semantic": {
+ "normalized_reference": "she told me once that the most important skill she would developed was learning to disagree with someone without making them feel they were wrong",
+ "normalized_hypothesis": "she told me once that the most important skill she developed was learning to disagree with someone without making them feel they were wrong",
+ "errors": 1,
+ "words": 25,
+ "wer": 0.04
+ }
+ },
+ {
+ "id": "release500_0089",
+ "category": "narrative",
+ "text": "He stood at the edge of the cliff and wondered, not for the first time, how he'd gotten here.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0089.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE STOOD AT THE EDGE OF THE CLIFF AND WONDERED NOT FOR THE FIRST TIME HOW HE'D GOT HERE",
+ "raw": {
+ "normalized_reference": "he stood at the edge of the cliff and wondered, not for the first time, how he'd gotten here.",
+ "normalized_hypothesis": "he stood at the edge of the cliff and wondered not for the first time how he'd got here",
+ "errors": 1,
+ "words": 19,
+ "wer": 0.05263157894736842
+ },
+ "semantic": {
+ "normalized_reference": "he stood at the edge of the cliff and wondered not for the 1st time how he would gotten here",
+ "normalized_hypothesis": "he stood at the edge of the cliff and wondered not for the 1st time how he would got here",
+ "errors": 1,
+ "words": 20,
+ "wer": 0.05
+ }
+ },
+ {
+ "id": "release500_0090",
+ "category": "question",
+ "text": "Why does asking for help feel so hard when you actually need it?",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0090.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WHY DOES ASKING FOR HELP FEEL SO HARD WHEN YOU ACTUALLY NEED IT",
+ "raw": {
+ "normalized_reference": "why does asking for help feel so hard when you actually need it?",
+ "normalized_hypothesis": "why does asking for help feel so hard when you actually need it",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "why does asking for help feel so hard when you actually need it",
+ "normalized_hypothesis": "why does asking for help feel so hard when you actually need it",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0091",
+ "category": "short",
+ "text": "That's fair.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0091.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THAT'S FAIR",
+ "raw": {
+ "normalized_reference": "that's fair.",
+ "normalized_hypothesis": "that's fair",
+ "errors": 0,
+ "words": 2,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "that is fair",
+ "normalized_hypothesis": "that is fair",
+ "errors": 0,
+ "words": 3,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0092",
+ "category": "technical",
+ "text": "If you run into that error, the fix is usually just restarting the service.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0092.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IF YOU RUN INTO THAT ERROR THE FIX IS USUALLY JUST RESTARTING THE SERVICE",
+ "raw": {
+ "normalized_reference": "if you run into that error, the fix is usually just restarting the service.",
+ "normalized_hypothesis": "if you run into that error the fix is usually just restarting the service",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "if you run into that error the fix is usually just restarting the service",
+ "normalized_hypothesis": "if you run into that error the fix is usually just restarting the service",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0093",
+ "category": "conversational",
+ "text": "It was a disaster in the best possible way.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0093.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT WAS A DISASTER IN THE BEST POSSIBLE WAY",
+ "raw": {
+ "normalized_reference": "it was a disaster in the best possible way.",
+ "normalized_hypothesis": "it was a disaster in the best possible way",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it was a disaster in the best possible way",
+ "normalized_hypothesis": "it was a disaster in the best possible way",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0094",
+ "category": "descriptive",
+ "text": "It had the look of a place that used to mean something to someone.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0094.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT HAD THE LOOK OF A PLACE THAT USED TO MEAN SOMETHING TO SOMEONE",
+ "raw": {
+ "normalized_reference": "it had the look of a place that used to mean something to someone.",
+ "normalized_hypothesis": "it had the look of a place that used to mean something to someone",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it had the look of a place that used to mean something to someone",
+ "normalized_hypothesis": "it had the look of a place that used to mean something to someone",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0095",
+ "category": "emotional",
+ "text": "There's this feeling you get when something finally works after weeks of failure.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0095.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THERE'S THIS FEELING YOU GET WHEN SOMETHING FINALLY WORKS AFTER WEEKS OF FAILURE",
+ "raw": {
+ "normalized_reference": "there's this feeling you get when something finally works after weeks of failure.",
+ "normalized_hypothesis": "there's this feeling you get when something finally works after weeks of failure",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "there is this feeling you get when something finally works after weeks of failure",
+ "normalized_hypothesis": "there is this feeling you get when something finally works after weeks of failure",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0096",
+ "category": "long",
+ "text": "He had a theory that most arguments aren't actually about what they seem to be about, and the longer I live the more I think he was onto something.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0096.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE HAD A THEORY THAT MOST ARGUMENTS AREN'T ACTUALLY ABOUT WHAT THEY SEEM TO BE ABOUT AND THE LONGER I LIVE THE MORE I THINK HE WAS ON TO SOMETHING",
+ "raw": {
+ "normalized_reference": "he had a theory that most arguments aren't actually about what they seem to be about, and the longer i live the more i think he was onto something.",
+ "normalized_hypothesis": "he had a theory that most arguments aren't actually about what they seem to be about and the longer i live the more i think he was on to something",
+ "errors": 2,
+ "words": 29,
+ "wer": 0.06896551724137931
+ },
+ "semantic": {
+ "normalized_reference": "he had a theory that most arguments are not actually about what they seem to be about and the longer i live the more i think he was onto something",
+ "normalized_hypothesis": "he had a theory that most arguments are not actually about what they seem to be about and the longer i live the more i think he was on to something",
+ "errors": 2,
+ "words": 30,
+ "wer": 0.06666666666666667
+ }
+ },
+ {
+ "id": "release500_0097",
+ "category": "narrative",
+ "text": "The thing nobody tells you about starting over is how ordinary most of it feels.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0097.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE THING NOBODY TELLS YOU ABOUT STARTING OVER IS HOW ORDINARY MOST OF IT FEELS",
+ "raw": {
+ "normalized_reference": "the thing nobody tells you about starting over is how ordinary most of it feels.",
+ "normalized_hypothesis": "the thing nobody tells you about starting over is how ordinary most of it feels",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the thing nobody tells you about starting over is how ordinary most of it feels",
+ "normalized_hypothesis": "the thing nobody tells you about starting over is how ordinary most of it feels",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0098",
+ "category": "question",
+ "text": "What exactly are we waiting for at this point?",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0098.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WHAT EXACTLY ARE WE WAITING FOR AT THIS POINT",
+ "raw": {
+ "normalized_reference": "what exactly are we waiting for at this point?",
+ "normalized_hypothesis": "what exactly are we waiting for at this point",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "what exactly are we waiting for at this point",
+ "normalized_hypothesis": "what exactly are we waiting for at this point",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0099",
+ "category": "short",
+ "text": "I stand by what I said.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0099.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I STAND BY WHAT I SAID",
+ "raw": {
+ "normalized_reference": "i stand by what i said.",
+ "normalized_hypothesis": "i stand by what i said",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i stand by what i said",
+ "normalized_hypothesis": "i stand by what i said",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0100",
+ "category": "technical",
+ "text": "Double-check that value before you push anything to production.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0100.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "DOUBLE CHECK THAT VALUE BEFORE YOU PUSH ANYTHING TO PRODUCTION",
+ "raw": {
+ "normalized_reference": "double-check that value before you push anything to production.",
+ "normalized_hypothesis": "double check that value before you push anything to production",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "double check that value before you push anything to production",
+ "normalized_hypothesis": "double check that value before you push anything to production",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0101",
+ "category": "conversational",
+ "text": "I understand your point; I just do not agree with the conclusion.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0101.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I UNDERSTAND YOUR POINT I JUST DO NOT AGREE WITH THE CONCLUSION",
+ "raw": {
+ "normalized_reference": "i understand your point; i just do not agree with the conclusion.",
+ "normalized_hypothesis": "i understand your point i just do not agree with the conclusion",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i understand your point i just do not agree with the conclusion",
+ "normalized_hypothesis": "i understand your point i just do not agree with the conclusion",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0102",
+ "category": "descriptive",
+ "text": "The sky that day was a strange shade of orange I'd never seen before.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0102.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE SKY THAT DAY WAS A STRANGE SHADE OF ORANGE I'D NEVER SEEN BEFORE",
+ "raw": {
+ "normalized_reference": "the sky that day was a strange shade of orange i'd never seen before.",
+ "normalized_hypothesis": "the sky that day was a strange shade of orange i'd never seen before",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the sky that day was a strange shade of orange i would never seen before",
+ "normalized_hypothesis": "the sky that day was a strange shade of orange i would never seen before",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0103",
+ "category": "emotional",
+ "text": "It hurt more than I expected, and I thought I was prepared.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0103.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT HURT MORE THAN I EXPECTED AND I THOUGHT IT WAS PREPARED",
+ "raw": {
+ "normalized_reference": "it hurt more than i expected, and i thought i was prepared.",
+ "normalized_hypothesis": "it hurt more than i expected and i thought it was prepared",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "it hurt more than i expected and i thought i was prepared",
+ "normalized_hypothesis": "it hurt more than i expected and i thought it was prepared",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ }
+ },
+ {
+ "id": "release500_0104",
+ "category": "long",
+ "text": "The hardest part wasn't making the decision, it was sitting with the uncertainty of not knowing whether it was the right one for months afterward.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0104.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE HARDEST PART WAS MAKING THE DECISION IT WAS SITTING WITH THE UNCERTAINTY OF NOT KNOWING WHETHER IT WAS THE RIGHT ONE FOR MONTHS AFTERWARD",
+ "raw": {
+ "normalized_reference": "the hardest part wasn't making the decision, it was sitting with the uncertainty of not knowing whether it was the right one for months afterward.",
+ "normalized_hypothesis": "the hardest part was making the decision it was sitting with the uncertainty of not knowing whether it was the right one for months afterward",
+ "errors": 1,
+ "words": 25,
+ "wer": 0.04
+ },
+ "semantic": {
+ "normalized_reference": "the hardest part was not making the decision it was sitting with the uncertainty of not knowing whether it was the right one for months afterward",
+ "normalized_hypothesis": "the hardest part was making the decision it was sitting with the uncertainty of not knowing whether it was the right one for months afterward",
+ "errors": 1,
+ "words": 26,
+ "wer": 0.038461538461538464
+ }
+ },
+ {
+ "id": "release500_0105",
+ "category": "narrative",
+ "text": "There was this one day last summer when everything just clicked into place.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0105.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THERE WAS THIS ONE DAY LAST SUMMER WHEN EVERYTHING JUST CLICKED INTO PLACE",
+ "raw": {
+ "normalized_reference": "there was this one day last summer when everything just clicked into place.",
+ "normalized_hypothesis": "there was this one day last summer when everything just clicked into place",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "there was this one day last summer when everything just clicked into place",
+ "normalized_hypothesis": "there was this one day last summer when everything just clicked into place",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0106",
+ "category": "question",
+ "text": "Have you ever thought about why we keep doing things the same way when they stop working?",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0106.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HAVE YOU EVER THOUGHT ABOUT WHY WE KEEP DOING THINGS THE SAME WAY WHEN THEY STOP WORKING",
+ "raw": {
+ "normalized_reference": "have you ever thought about why we keep doing things the same way when they stop working?",
+ "normalized_hypothesis": "have you ever thought about why we keep doing things the same way when they stop working",
+ "errors": 0,
+ "words": 17,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "have you ever thought about why we keep doing things the same way when they stop working",
+ "normalized_hypothesis": "have you ever thought about why we keep doing things the same way when they stop working",
+ "errors": 0,
+ "words": 17,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0107",
+ "category": "short",
+ "text": "I can't believe you just said that.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0107.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I CAN'T BELIEVE YOU JUST SAID THAT",
+ "raw": {
+ "normalized_reference": "i can't believe you just said that.",
+ "normalized_hypothesis": "i can't believe you just said that",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i can not believe you just said that",
+ "normalized_hypothesis": "i can not believe you just said that",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0108",
+ "category": "technical",
+ "text": "You'll want to isolate that variable before drawing any conclusions.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0108.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "YOU'LL WANT TO ISOLATE THAT VARIABLE BEFORE DRAWING ANY CONCLUSIONS",
+ "raw": {
+ "normalized_reference": "you'll want to isolate that variable before drawing any conclusions.",
+ "normalized_hypothesis": "you'll want to isolate that variable before drawing any conclusions",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "you will want to isolate that variable before drawing any conclusions",
+ "normalized_hypothesis": "you will want to isolate that variable before drawing any conclusions",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0109",
+ "category": "conversational",
+ "text": "The meeting that could have been an email lasted two hours.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0109.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE MEETING THAT COULD HAVE BEEN AN EMALE LASTED TWO HOURS",
+ "raw": {
+ "normalized_reference": "the meeting that could have been an email lasted two hours.",
+ "normalized_hypothesis": "the meeting that could have been an emale lasted two hours",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ },
+ "semantic": {
+ "normalized_reference": "the meeting that could have been an email lasted 2 hours",
+ "normalized_hypothesis": "the meeting that could have been an emale lasted 2 hours",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ }
+ },
+ {
+ "id": "release500_0110",
+ "category": "descriptive",
+ "text": "There's a specific kind of tired that sleep doesn't fix.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0110.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THERE'S A SPECIFIC KIND OF TIRED THAT SLEEP DOESN'T FIX",
+ "raw": {
+ "normalized_reference": "there's a specific kind of tired that sleep doesn't fix.",
+ "normalized_hypothesis": "there's a specific kind of tired that sleep doesn't fix",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "there is a specific kind of tired that sleep does not fix",
+ "normalized_hypothesis": "there is a specific kind of tired that sleep does not fix",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0111",
+ "category": "emotional",
+ "text": "Something about that moment broke something open in me that I hadn't realized was closed.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0111.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SOMETHING ABOUT THAT MOMENT BROKE SOMETHING OPEN IN ME THAT I HADN'T REALIZED WAS CLOSED",
+ "raw": {
+ "normalized_reference": "something about that moment broke something open in me that i hadn't realized was closed.",
+ "normalized_hypothesis": "something about that moment broke something open in me that i hadn't realized was closed",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "something about that moment broke something open in me that i had not realized was closed",
+ "normalized_hypothesis": "something about that moment broke something open in me that i had not realized was closed",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0112",
+ "category": "long",
+ "text": "The thing about making something good is that you don't usually know it's good while you're making it. You just know you're not done yet.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0112.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE THING ABOUT MAKING SOMETHING GOOD IS THAT YOU DON'T USUALLY KNOW IT'S GOOD WHILE YOU'RE MAKING IT YOU JUST KNOW YOU'RE NOT DONE YET",
+ "raw": {
+ "normalized_reference": "the thing about making something good is that you don't usually know it's good while you're making it. you just know you're not done yet.",
+ "normalized_hypothesis": "the thing about making something good is that you don't usually know it's good while you're making it you just know you're not done yet",
+ "errors": 0,
+ "words": 25,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the thing about making something good is that you do not usually know it is good while you are making it you just know you are not done yet",
+ "normalized_hypothesis": "the thing about making something good is that you do not usually know it is good while you are making it you just know you are not done yet",
+ "errors": 0,
+ "words": 29,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0113",
+ "category": "narrative",
+ "text": "She opened the door slowly, as if she expected to find something she wasn't ready for.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0113.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SHE OPENED THE DOOR SLOWLY AS IF SHE EXPECTED TO FIND SOMETHING SHE WAS READY FOR",
+ "raw": {
+ "normalized_reference": "she opened the door slowly, as if she expected to find something she wasn't ready for.",
+ "normalized_hypothesis": "she opened the door slowly as if she expected to find something she was ready for",
+ "errors": 1,
+ "words": 16,
+ "wer": 0.0625
+ },
+ "semantic": {
+ "normalized_reference": "she opened the door slowly as if she expected to find something she was not ready for",
+ "normalized_hypothesis": "she opened the door slowly as if she expected to find something she was ready for",
+ "errors": 1,
+ "words": 17,
+ "wer": 0.058823529411764705
+ }
+ },
+ {
+ "id": "release500_0114",
+ "category": "question",
+ "text": "What would you do differently if you could go back?",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0114.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WHAT WOULD YOU DO DIFFERENTLY IF YOU COULD GO BACK",
+ "raw": {
+ "normalized_reference": "what would you do differently if you could go back?",
+ "normalized_hypothesis": "what would you do differently if you could go back",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "what would you do differently if you could go back",
+ "normalized_hypothesis": "what would you do differently if you could go back",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0115",
+ "category": "short",
+ "text": "Okay. Okay okay okay.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0115.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "OQ O K O K O K",
+ "raw": {
+ "normalized_reference": "okay. okay okay okay.",
+ "normalized_hypothesis": "oq o k o k o k",
+ "errors": 7,
+ "words": 4,
+ "wer": 1.75
+ },
+ "semantic": {
+ "normalized_reference": "okay okay okay okay",
+ "normalized_hypothesis": "oq 0 k 0 k 0 k",
+ "errors": 7,
+ "words": 4,
+ "wer": 1.75
+ }
+ },
+ {
+ "id": "release500_0116",
+ "category": "technical",
+ "text": "The latency goes up significantly at scale, which is expected.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0116.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE LATENCY GOES UP SIGNIFICANTLY AT SCALE WHICH IS EXPECTED",
+ "raw": {
+ "normalized_reference": "the latency goes up significantly at scale, which is expected.",
+ "normalized_hypothesis": "the latency goes up significantly at scale which is expected",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the latency goes up significantly at scale which is expected",
+ "normalized_hypothesis": "the latency goes up significantly at scale which is expected",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0117",
+ "category": "conversational",
+ "text": "We ended up staying until two in the morning talking about nothing.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0117.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WE ENDED UP STAYING UNTIL TWO IN THE MORNING TALKING ABOUT NOTHING",
+ "raw": {
+ "normalized_reference": "we ended up staying until two in the morning talking about nothing.",
+ "normalized_hypothesis": "we ended up staying until two in the morning talking about nothing",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "we ended up staying until 2 in the morning talking about nothing",
+ "normalized_hypothesis": "we ended up staying until 2 in the morning talking about nothing",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0118",
+ "category": "descriptive",
+ "text": "The rain started slowly, and then all at once, the way it usually does.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0118.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE RAIN STARTED SLOWLY AND THEN ALL AT ONCE THE WAY IT USUALLY DOES",
+ "raw": {
+ "normalized_reference": "the rain started slowly, and then all at once, the way it usually does.",
+ "normalized_hypothesis": "the rain started slowly and then all at once the way it usually does",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the rain started slowly and then all at once the way it usually does",
+ "normalized_hypothesis": "the rain started slowly and then all at once the way it usually does",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0119",
+ "category": "emotional",
+ "text": "I honestly did not think I was going to make it through that year.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0119.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I HONESTLY DID NOT THINK I WAS GOING TO MAKE IT THROUGH THAT YEAR",
+ "raw": {
+ "normalized_reference": "i honestly did not think i was going to make it through that year.",
+ "normalized_hypothesis": "i honestly did not think i was going to make it through that year",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i honestly did not think i was going to make it through that year",
+ "normalized_hypothesis": "i honestly did not think i was going to make it through that year",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0120",
+ "category": "long",
+ "text": "There's an assumption that people who seem confident all the time actually feel confident, and in my experience that's almost never true.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0120.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THERE'S AN ASSUMPTION THAT PEOPLE WHO SEEM CONFIDENT ALL THE TIME ACTUALLY FEEL CONFIDENT AND IN MY EXPERIENCE THAT'S ALMOST NEVER TRUE",
+ "raw": {
+ "normalized_reference": "there's an assumption that people who seem confident all the time actually feel confident, and in my experience that's almost never true.",
+ "normalized_hypothesis": "there's an assumption that people who seem confident all the time actually feel confident and in my experience that's almost never true",
+ "errors": 0,
+ "words": 22,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "there is an assumption that people who seem confident all the time actually feel confident and in my experience that is almost never true",
+ "normalized_hypothesis": "there is an assumption that people who seem confident all the time actually feel confident and in my experience that is almost never true",
+ "errors": 0,
+ "words": 24,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0121",
+ "category": "narrative",
+ "text": "The moment I opened the door I knew something was different, even if I couldn't say what.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0121.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE MOMENT I OPENED THE DOOR I KNEW SOMETHING WAS DIFFERENT EVEN IF I COULDN'T SAY WHAT",
+ "raw": {
+ "normalized_reference": "the moment i opened the door i knew something was different, even if i couldn't say what.",
+ "normalized_hypothesis": "the moment i opened the door i knew something was different even if i couldn't say what",
+ "errors": 0,
+ "words": 17,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the moment i opened the door i knew something was different even if i could not say what",
+ "normalized_hypothesis": "the moment i opened the door i knew something was different even if i could not say what",
+ "errors": 0,
+ "words": 18,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0122",
+ "category": "question",
+ "text": "How do you explain something that doesn't make sense even to you?",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0122.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HOW DO YOU EXPLAIN SOMETHING THAT DOESN'T MAKE SENSE EVEN TO YOU",
+ "raw": {
+ "normalized_reference": "how do you explain something that doesn't make sense even to you?",
+ "normalized_hypothesis": "how do you explain something that doesn't make sense even to you",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "how do you explain something that does not make sense even to you",
+ "normalized_hypothesis": "how do you explain something that does not make sense even to you",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0123",
+ "category": "short",
+ "text": "Hold on a second.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0123.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HOLD ON A SECOND",
+ "raw": {
+ "normalized_reference": "hold on a second.",
+ "normalized_hypothesis": "hold on a second",
+ "errors": 0,
+ "words": 4,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "hold on a 2nd",
+ "normalized_hypothesis": "hold on a 2nd",
+ "errors": 0,
+ "words": 4,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0124",
+ "category": "technical",
+ "text": "A deterministic split prevents duplicate normalized text from leaking into evaluation.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0124.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "A DETERMINISTIC SPLIT PREVENTS DUPLICATE NORMALIZED TEXT FROM LEAKING INTO EVALUATION",
+ "raw": {
+ "normalized_reference": "a deterministic split prevents duplicate normalized text from leaking into evaluation.",
+ "normalized_hypothesis": "a deterministic split prevents duplicate normalized text from leaking into evaluation",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "a deterministic split prevents duplicate normalized text from leaking into evaluation",
+ "normalized_hypothesis": "a deterministic split prevents duplicate normalized text from leaking into evaluation",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0125",
+ "category": "conversational",
+ "text": "Honestly, that was much better than I expected, but the ending still felt rushed.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0125.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HONESTLY THAT WAS MUCH BETTER THAN EXPECTED BUT THE ENDING STILL FELT RUSHED",
+ "raw": {
+ "normalized_reference": "honestly, that was much better than i expected, but the ending still felt rushed.",
+ "normalized_hypothesis": "honestly that was much better than expected but the ending still felt rushed",
+ "errors": 1,
+ "words": 14,
+ "wer": 0.07142857142857142
+ },
+ "semantic": {
+ "normalized_reference": "honestly that was much better than i expected but the ending still felt rushed",
+ "normalized_hypothesis": "honestly that was much better than expected but the ending still felt rushed",
+ "errors": 1,
+ "words": 14,
+ "wer": 0.07142857142857142
+ }
+ },
+ {
+ "id": "release500_0126",
+ "category": "descriptive",
+ "text": "She had a habit of tilting her head slightly when thinking hard about something.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0126.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SHE HAD A HABIT OF TILTING HER HEAD SLIGHTLY WHEN THINKING HARD ABOUT SOMETHING",
+ "raw": {
+ "normalized_reference": "she had a habit of tilting her head slightly when thinking hard about something.",
+ "normalized_hypothesis": "she had a habit of tilting her head slightly when thinking hard about something",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "she had a habit of tilting her head slightly when thinking hard about something",
+ "normalized_hypothesis": "she had a habit of tilting her head slightly when thinking hard about something",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0127",
+ "category": "emotional",
+ "text": "I cannot stress enough how much I did not want to be right about this.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0127.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I CANNOT STRESS ENOUGH HOW MUCH I DID NOT WANT TO BE RIGHT ABOUT THIS",
+ "raw": {
+ "normalized_reference": "i cannot stress enough how much i did not want to be right about this.",
+ "normalized_hypothesis": "i cannot stress enough how much i did not want to be right about this",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i cannot stress enough how much i did not want to be right about this",
+ "normalized_hypothesis": "i cannot stress enough how much i did not want to be right about this",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0128",
+ "category": "long",
+ "text": "I used to think the goal was to get to a point where nothing was hard anymore, and now I think that point doesn't exist and I'm not sure I'd want it to.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0128.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I USED TO THINK THE GOAL WAS TO GET TO A POINT WHERE NOTHING WAS HARD AN AMORE AND NOW I THINK THAT POINT DOESN'T EXIST AND I'M NOT SURE I'D WANT IT TO",
+ "raw": {
+ "normalized_reference": "i used to think the goal was to get to a point where nothing was hard anymore, and now i think that point doesn't exist and i'm not sure i'd want it to.",
+ "normalized_hypothesis": "i used to think the goal was to get to a point where nothing was hard an amore and now i think that point doesn't exist and i'm not sure i'd want it to",
+ "errors": 2,
+ "words": 33,
+ "wer": 0.06060606060606061
+ },
+ "semantic": {
+ "normalized_reference": "i used to think the goal was to get to a point where nothing was hard anymore and now i think that point does not exist and i am not sure i would want it to",
+ "normalized_hypothesis": "i used to think the goal was to get to a point where nothing was hard an amore and now i think that point does not exist and i am not sure i would want it to",
+ "errors": 2,
+ "words": 36,
+ "wer": 0.05555555555555555
+ }
+ },
+ {
+ "id": "release500_0129",
+ "category": "narrative",
+ "text": "So this is the part of the story where everything goes completely sideways.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0129.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SO THIS IS THE PART OF THE STORY WHERE EVERYTHING GOES COMPLETELY SIDEWAYS",
+ "raw": {
+ "normalized_reference": "so this is the part of the story where everything goes completely sideways.",
+ "normalized_hypothesis": "so this is the part of the story where everything goes completely sideways",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "so this is the part of the story where everything goes completely sideways",
+ "normalized_hypothesis": "so this is the part of the story where everything goes completely sideways",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0130",
+ "category": "question",
+ "text": "Is there a version of this where it actually works out okay?",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0130.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IS THERE A VERSION OF THIS WHERE IT ACTUALLY WORKS OUT O K",
+ "raw": {
+ "normalized_reference": "is there a version of this where it actually works out okay?",
+ "normalized_hypothesis": "is there a version of this where it actually works out o k",
+ "errors": 2,
+ "words": 12,
+ "wer": 0.16666666666666666
+ },
+ "semantic": {
+ "normalized_reference": "is there a version of this where it actually works out okay",
+ "normalized_hypothesis": "is there a version of this where it actually works out 0 k",
+ "errors": 2,
+ "words": 12,
+ "wer": 0.16666666666666666
+ }
+ },
+ {
+ "id": "release500_0131",
+ "category": "short",
+ "text": "Oh no. Oh no no no.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0131.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "OH NO OH NO NO NO",
+ "raw": {
+ "normalized_reference": "oh no. oh no no no.",
+ "normalized_hypothesis": "oh no oh no no no",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "0 no 0 no no no",
+ "normalized_hypothesis": "0 no 0 no no no",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0132",
+ "category": "technical",
+ "text": "Always test on a copy before modifying the original.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0132.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "ALWAYS TEST ON A COPY BEFORE MODIFYING THE ORIGINAL",
+ "raw": {
+ "normalized_reference": "always test on a copy before modifying the original.",
+ "normalized_hypothesis": "always test on a copy before modifying the original",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "always test on a copy before modifying the original",
+ "normalized_hypothesis": "always test on a copy before modifying the original",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0133",
+ "category": "conversational",
+ "text": "I wasn't even supposed to be there. That's the wild part.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0133.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I WASN'T EVEN SUPPOSED TO BE THERE THAT'S THE WILD PART",
+ "raw": {
+ "normalized_reference": "i wasn't even supposed to be there. that's the wild part.",
+ "normalized_hypothesis": "i wasn't even supposed to be there that's the wild part",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i was not even supposed to be there that is the wild part",
+ "normalized_hypothesis": "i was not even supposed to be there that is the wild part",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0134",
+ "category": "descriptive",
+ "text": "It's a small town. Not the kind you leave. The kind you leave and come back to.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0134.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT'S A SMALL TOWN NOT THE KIND YOU LEAVE THE KIND YOU LEAVE AND COME BACK TO",
+ "raw": {
+ "normalized_reference": "it's a small town. not the kind you leave. the kind you leave and come back to.",
+ "normalized_hypothesis": "it's a small town not the kind you leave the kind you leave and come back to",
+ "errors": 0,
+ "words": 17,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it is a small town not the kind you leave the kind you leave and come back to",
+ "normalized_hypothesis": "it is a small town not the kind you leave the kind you leave and come back to",
+ "errors": 0,
+ "words": 18,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0135",
+ "category": "emotional",
+ "text": "She cried. I cried. We all just sort of stood there crying.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0135.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SHE CRIED I CRIED WE ALL JUST SORT OF STOOD THERE CRYING",
+ "raw": {
+ "normalized_reference": "she cried. i cried. we all just sort of stood there crying.",
+ "normalized_hypothesis": "she cried i cried we all just sort of stood there crying",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "she cried i cried we all just sort of stood there crying",
+ "normalized_hypothesis": "she cried i cried we all just sort of stood there crying",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0136",
+ "category": "long",
+ "text": "There's a version of being busy that's actually just a very elaborate way of avoiding the one thing you know you should be doing.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0136.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THERE'S A VERSION OF BEING BUSY THAT'S ACTUALLY JUST A VERY ELABORATE WAY OF AVOIDING THE ONE THING YOU KNOW YOU SHOULD BE DOING",
+ "raw": {
+ "normalized_reference": "there's a version of being busy that's actually just a very elaborate way of avoiding the one thing you know you should be doing.",
+ "normalized_hypothesis": "there's a version of being busy that's actually just a very elaborate way of avoiding the one thing you know you should be doing",
+ "errors": 0,
+ "words": 24,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "there is a version of being busy that is actually just a very elaborate way of avoiding the one thing you know you should be doing",
+ "normalized_hypothesis": "there is a version of being busy that is actually just a very elaborate way of avoiding the one thing you know you should be doing",
+ "errors": 0,
+ "words": 26,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0137",
+ "category": "narrative",
+ "text": "The letter arrived on a Tuesday, which felt appropriate somehow.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0137.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE LETTER ARRIVED ON A TUESDAY WHICH FELT APPROPRIATE SOMEHOW",
+ "raw": {
+ "normalized_reference": "the letter arrived on a tuesday, which felt appropriate somehow.",
+ "normalized_hypothesis": "the letter arrived on a tuesday which felt appropriate somehow",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the letter arrived on a tuesday which felt appropriate somehow",
+ "normalized_hypothesis": "the letter arrived on a tuesday which felt appropriate somehow",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0138",
+ "category": "question",
+ "text": "What exactly did she say when you told her?",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0138.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WHAT EXACTLY DID SHE SAY WHEN YOU TOLD HER",
+ "raw": {
+ "normalized_reference": "what exactly did she say when you told her?",
+ "normalized_hypothesis": "what exactly did she say when you told her",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "what exactly did she say when you told her",
+ "normalized_hypothesis": "what exactly did she say when you told her",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0139",
+ "category": "short",
+ "text": "No way. Absolutely not.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0139.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "NO WAY ABSOLUTELY NOT",
+ "raw": {
+ "normalized_reference": "no way. absolutely not.",
+ "normalized_hypothesis": "no way absolutely not",
+ "errors": 0,
+ "words": 4,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "no way absolutely not",
+ "normalized_hypothesis": "no way absolutely not",
+ "errors": 0,
+ "words": 4,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0140",
+ "category": "technical",
+ "text": "The first thing you want to do is make sure everything is backed up.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0140.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE FIRST THING YOU WANT TO DO IS MAKE SURE EVERYTHING IS BACKED UP",
+ "raw": {
+ "normalized_reference": "the first thing you want to do is make sure everything is backed up.",
+ "normalized_hypothesis": "the first thing you want to do is make sure everything is backed up",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the 1st thing you want to do is make sure everything is backed up",
+ "normalized_hypothesis": "the 1st thing you want to do is make sure everything is backed up",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0141",
+ "category": "conversational",
+ "text": "I have the most insane idea and you need to just hear me out.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0141.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I HAVE THE MOST INSANE IDEA AND YOU NEED TO JUST HEAR ME OUT",
+ "raw": {
+ "normalized_reference": "i have the most insane idea and you need to just hear me out.",
+ "normalized_hypothesis": "i have the most insane idea and you need to just hear me out",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i have the most insane idea and you need to just hear me out",
+ "normalized_hypothesis": "i have the most insane idea and you need to just hear me out",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0142",
+ "category": "descriptive",
+ "text": "There's a coffee shop on the corner that's been there longer than anyone can remember.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0142.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THERE'S A COFFEE SHOP ON THE CORNER THAT'S BEEN THERE LONGER THAN ANYONE CAN REMEMBER",
+ "raw": {
+ "normalized_reference": "there's a coffee shop on the corner that's been there longer than anyone can remember.",
+ "normalized_hypothesis": "there's a coffee shop on the corner that's been there longer than anyone can remember",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "there is a coffee shop on the corner that has been there longer than anyone can remember",
+ "normalized_hypothesis": "there is a coffee shop on the corner that has been there longer than anyone can remember",
+ "errors": 0,
+ "words": 17,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0143",
+ "category": "emotional",
+ "text": "I kept it together until I got to the car, and then I completely fell apart.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0143.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I KEPT IT TOGETHER UNTIL I GOT TO THE CAR AND THEN I COMPLETELY FELL APART",
+ "raw": {
+ "normalized_reference": "i kept it together until i got to the car, and then i completely fell apart.",
+ "normalized_hypothesis": "i kept it together until i got to the car and then i completely fell apart",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i kept it together until i got to the car and then i completely fell apart",
+ "normalized_hypothesis": "i kept it together until i got to the car and then i completely fell apart",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0144",
+ "category": "long",
+ "text": "The older I get, the more I think that most of what we call wisdom is just pattern recognition built from having been wrong enough times to start noticing.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0144.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE OLDER I GET THE MORE I THINK THAT MOST OF WHAT WE CALL WISDOM IS JUST PATTERN RECOGNITION BUILT FROM HAVING BEEN WRONG ENOUGH TIMES TO START NOTICING",
+ "raw": {
+ "normalized_reference": "the older i get, the more i think that most of what we call wisdom is just pattern recognition built from having been wrong enough times to start noticing.",
+ "normalized_hypothesis": "the older i get the more i think that most of what we call wisdom is just pattern recognition built from having been wrong enough times to start noticing",
+ "errors": 0,
+ "words": 29,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the older i get the more i think that most of what we call wisdom is just pattern recognition built from having been wrong enough times to start noticing",
+ "normalized_hypothesis": "the older i get the more i think that most of what we call wisdom is just pattern recognition built from having been wrong enough times to start noticing",
+ "errors": 0,
+ "words": 29,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0145",
+ "category": "narrative",
+ "text": "I remember the exact moment I realized I had made a huge mistake.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0145.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I REMEMBER THE EXACT MOMENT I REALIZED I HAD MADE A HUGE MISTAKE",
+ "raw": {
+ "normalized_reference": "i remember the exact moment i realized i had made a huge mistake.",
+ "normalized_hypothesis": "i remember the exact moment i realized i had made a huge mistake",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i remember the exact moment i realized i had made a huge mistake",
+ "normalized_hypothesis": "i remember the exact moment i realized i had made a huge mistake",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0146",
+ "category": "question",
+ "text": "Do you think it would have gone differently if we'd left earlier?",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0146.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "DO YOU THINK IT WOULD HAVE GONE DIFFERENTLY IF WE'D LEFT EARLIER",
+ "raw": {
+ "normalized_reference": "do you think it would have gone differently if we'd left earlier?",
+ "normalized_hypothesis": "do you think it would have gone differently if we'd left earlier",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "do you think it would have gone differently if we would left earlier",
+ "normalized_hypothesis": "do you think it would have gone differently if we would left earlier",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0147",
+ "category": "short",
+ "text": "You have to be kidding me.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0147.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "YOU HAVE TO BE KIDDING ME",
+ "raw": {
+ "normalized_reference": "you have to be kidding me.",
+ "normalized_hypothesis": "you have to be kidding me",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "you have to be kidding me",
+ "normalized_hypothesis": "you have to be kidding me",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0148",
+ "category": "technical",
+ "text": "If the checksum fails, restart the worker process without deleting the cached tensors.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0148.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IF THE CHEKSUM FAILS RESTART THE WORKER PROCESS WITHOUT DELEADING THE CASHED TENSERS",
+ "raw": {
+ "normalized_reference": "if the checksum fails, restart the worker process without deleting the cached tensors.",
+ "normalized_hypothesis": "if the cheksum fails restart the worker process without deleading the cashed tensers",
+ "errors": 4,
+ "words": 13,
+ "wer": 0.3076923076923077
+ },
+ "semantic": {
+ "normalized_reference": "if the checksum fails restart the worker process without deleting the cached tensors",
+ "normalized_hypothesis": "if the cheksum fails restart the worker process without deleading the cashed tensers",
+ "errors": 4,
+ "words": 13,
+ "wer": 0.3076923076923077
+ }
+ },
+ {
+ "id": "release500_0149",
+ "category": "conversational",
+ "text": "I need you to understand how close that was to not working.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0149.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I NEED YOU TO UNDERSTAND HOW CLOSE THAT WAS TO NOT WORKING",
+ "raw": {
+ "normalized_reference": "i need you to understand how close that was to not working.",
+ "normalized_hypothesis": "i need you to understand how close that was to not working",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i need you to understand how close that was to not working",
+ "normalized_hypothesis": "i need you to understand how close that was to not working",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0150",
+ "category": "descriptive",
+ "text": "It was the sort of morning where the whole world looks rinsed clean.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0150.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT WAS THE SORT OF MORNING WHERE THE WHOLE WORLD LOOKS RINCED CLEAN",
+ "raw": {
+ "normalized_reference": "it was the sort of morning where the whole world looks rinsed clean.",
+ "normalized_hypothesis": "it was the sort of morning where the whole world looks rinced clean",
+ "errors": 1,
+ "words": 13,
+ "wer": 0.07692307692307693
+ },
+ "semantic": {
+ "normalized_reference": "it was the sort of morning where the whole world looks rinsed clean",
+ "normalized_hypothesis": "it was the sort of morning where the whole world looks rinced clean",
+ "errors": 1,
+ "words": 13,
+ "wer": 0.07692307692307693
+ }
+ },
+ {
+ "id": "release500_0151",
+ "category": "emotional",
+ "text": "She started laughing and I started laughing and we couldn't stop for a full minute.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0151.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SHE STARTED LAUGHING AND I STARTED LAUGHING AND WE COULDN'T STOP FOR A FULL MINUTE",
+ "raw": {
+ "normalized_reference": "she started laughing and i started laughing and we couldn't stop for a full minute.",
+ "normalized_hypothesis": "she started laughing and i started laughing and we couldn't stop for a full minute",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "she started laughing and i started laughing and we could not stop for a full minute",
+ "normalized_hypothesis": "she started laughing and i started laughing and we could not stop for a full minute",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0152",
+ "category": "long",
+ "text": "We spent three years building something we were genuinely proud of, and then the market shifted and none of it mattered the way we thought it would.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0152.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WE SPENT THREE YEARS BUILDING SOMETHING WE WERE GENUINELY PROUD OF AND THEN THE MARKET SHIFTED AND NONE OF IT MATTERED THE WAY WE THOUGHT IT WOULD",
+ "raw": {
+ "normalized_reference": "we spent three years building something we were genuinely proud of, and then the market shifted and none of it mattered the way we thought it would.",
+ "normalized_hypothesis": "we spent three years building something we were genuinely proud of and then the market shifted and none of it mattered the way we thought it would",
+ "errors": 0,
+ "words": 27,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "we spent 3 years building something we were genuinely proud of and then the market shifted and none of it mattered the way we thought it would",
+ "normalized_hypothesis": "we spent 3 years building something we were genuinely proud of and then the market shifted and none of it mattered the way we thought it would",
+ "errors": 0,
+ "words": 27,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0153",
+ "category": "narrative",
+ "text": "And somehow, despite all of that, it actually turned out okay.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0153.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "AND SOMEHOW DESPITE ALL OF THAT IT ACTUALLY TURNED OUT O K",
+ "raw": {
+ "normalized_reference": "and somehow, despite all of that, it actually turned out okay.",
+ "normalized_hypothesis": "and somehow despite all of that it actually turned out o k",
+ "errors": 2,
+ "words": 11,
+ "wer": 0.18181818181818182
+ },
+ "semantic": {
+ "normalized_reference": "and somehow despite all of that it actually turned out okay",
+ "normalized_hypothesis": "and somehow despite all of that it actually turned out 0 k",
+ "errors": 2,
+ "words": 11,
+ "wer": 0.18181818181818182
+ }
+ },
+ {
+ "id": "release500_0154",
+ "category": "question",
+ "text": "Is that something you actually want, or something you think you should want?",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0154.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IS THAT SOMETHING YOU ACTUALLY WANT OR SOMETHING YOU THINK YOU SHOULD WANT",
+ "raw": {
+ "normalized_reference": "is that something you actually want, or something you think you should want?",
+ "normalized_hypothesis": "is that something you actually want or something you think you should want",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "is that something you actually want or something you think you should want",
+ "normalized_hypothesis": "is that something you actually want or something you think you should want",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0155",
+ "category": "short",
+ "text": "This is fine.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0155.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THIS IS FINE",
+ "raw": {
+ "normalized_reference": "this is fine.",
+ "normalized_hypothesis": "this is fine",
+ "errors": 0,
+ "words": 3,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "this is fine",
+ "normalized_hypothesis": "this is fine",
+ "errors": 0,
+ "words": 3,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0156",
+ "category": "technical",
+ "text": "The tricky part isn't the setup, it's making sure all dependencies line up.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0156.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE TRICKY PART ISN'T THE SET UP IT'S MAKING SURE ALL DEPENDENCIES LINE UP",
+ "raw": {
+ "normalized_reference": "the tricky part isn't the setup, it's making sure all dependencies line up.",
+ "normalized_hypothesis": "the tricky part isn't the set up it's making sure all dependencies line up",
+ "errors": 2,
+ "words": 13,
+ "wer": 0.15384615384615385
+ },
+ "semantic": {
+ "normalized_reference": "the tricky part is not the setup it is making sure all dependencies line up",
+ "normalized_hypothesis": "the tricky part is not the set up it is making sure all dependencies line up",
+ "errors": 2,
+ "words": 15,
+ "wer": 0.13333333333333333
+ }
+ },
+ {
+ "id": "release500_0157",
+ "category": "conversational",
+ "text": "He called it a minor setback. I called it a disaster.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0157.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE CALLED IT A MINOR SETBACK I CALLED IT A DISASTER",
+ "raw": {
+ "normalized_reference": "he called it a minor setback. i called it a disaster.",
+ "normalized_hypothesis": "he called it a minor setback i called it a disaster",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he called it a minor setback i called it a disaster",
+ "normalized_hypothesis": "he called it a minor setback i called it a disaster",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0158",
+ "category": "descriptive",
+ "text": "The whole place smelled like pine and old wood and cold air.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0158.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE WHOLE PLACE SMELLED LIKE PINE AND OLD WOOD IN COLD AIR",
+ "raw": {
+ "normalized_reference": "the whole place smelled like pine and old wood and cold air.",
+ "normalized_hypothesis": "the whole place smelled like pine and old wood in cold air",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "the whole place smelled like pine and old wood and cold air",
+ "normalized_hypothesis": "the whole place smelled like pine and old wood in cold air",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ }
+ },
+ {
+ "id": "release500_0159",
+ "category": "emotional",
+ "text": "I've never been so relieved in my entire life.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0159.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I'VE NEVER BEEN SO RELIEVED IN MY ENTIRE LIFE",
+ "raw": {
+ "normalized_reference": "i've never been so relieved in my entire life.",
+ "normalized_hypothesis": "i've never been so relieved in my entire life",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i have never been so relieved in my entire life",
+ "normalized_hypothesis": "i have never been so relieved in my entire life",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0160",
+ "category": "narrative",
+ "text": "I wrote it all down that night so I wouldn't forget, and I still haven't looked at it.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0160.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I WROTE IT ALL DOWN THAT NIGHT SO I WOULDN'T FORGET AND I STILL HAVEN'T LOOKED AT IT",
+ "raw": {
+ "normalized_reference": "i wrote it all down that night so i wouldn't forget, and i still haven't looked at it.",
+ "normalized_hypothesis": "i wrote it all down that night so i wouldn't forget and i still haven't looked at it",
+ "errors": 0,
+ "words": 18,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i wrote it all down that night so i would not forget and i still have not looked at it",
+ "normalized_hypothesis": "i wrote it all down that night so i would not forget and i still have not looked at it",
+ "errors": 0,
+ "words": 20,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0161",
+ "category": "short",
+ "text": "That's kind of wild actually.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0161.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THAT'S KIND OF WILD ACTUALLY",
+ "raw": {
+ "normalized_reference": "that's kind of wild actually.",
+ "normalized_hypothesis": "that's kind of wild actually",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "that is kind of wild actually",
+ "normalized_hypothesis": "that is kind of wild actually",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0162",
+ "category": "technical",
+ "text": "Version pinning will save you a lot of pain later down the line.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0162.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "VERSION PINNING WILL SAVE YOU A LOT OF PAIN LATER DOWN THE LINE",
+ "raw": {
+ "normalized_reference": "version pinning will save you a lot of pain later down the line.",
+ "normalized_hypothesis": "version pinning will save you a lot of pain later down the line",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "version pinning will save you a lot of pain later down the line",
+ "normalized_hypothesis": "version pinning will save you a lot of pain later down the line",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0163",
+ "category": "conversational",
+ "text": "At some point you just have to commit and see what happens.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0163.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "AT SOME POINT YOU JUST HAVE TO COMMIT AND SEE WHAT HAPPENS",
+ "raw": {
+ "normalized_reference": "at some point you just have to commit and see what happens.",
+ "normalized_hypothesis": "at some point you just have to commit and see what happens",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "at some point you just have to commit and see what happens",
+ "normalized_hypothesis": "at some point you just have to commit and see what happens",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0164",
+ "category": "emotional",
+ "text": "Nobody told me it was going to be this hard, and honestly, I'm grateful.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0164.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "NOBODY TOLD ME IT WAS GOING TO BE THIS HARD AND HONESTLY I'M GRATEFUL",
+ "raw": {
+ "normalized_reference": "nobody told me it was going to be this hard, and honestly, i'm grateful.",
+ "normalized_hypothesis": "nobody told me it was going to be this hard and honestly i'm grateful",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "nobody told me it was going to be this hard and honestly i am grateful",
+ "normalized_hypothesis": "nobody told me it was going to be this hard and honestly i am grateful",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0165",
+ "category": "narrative",
+ "text": "Chapter seven. In which everything that could go wrong did, and a few things that couldn't.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0165.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "CHAPTER SEVEN IN WHICH EVERYTHING THAT COULD GO WRONG DID AND A FEW THINGS THAT COULDN'T",
+ "raw": {
+ "normalized_reference": "chapter seven. in which everything that could go wrong did, and a few things that couldn't.",
+ "normalized_hypothesis": "chapter seven in which everything that could go wrong did and a few things that couldn't",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "chapter 7 in which everything that could go wrong did and a few things that could not",
+ "normalized_hypothesis": "chapter 7 in which everything that could go wrong did and a few things that could not",
+ "errors": 0,
+ "words": 17,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0166",
+ "category": "short",
+ "text": "Wait, what?",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0166.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WAIT WHAT",
+ "raw": {
+ "normalized_reference": "wait, what?",
+ "normalized_hypothesis": "wait what",
+ "errors": 0,
+ "words": 2,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "wait what",
+ "normalized_hypothesis": "wait what",
+ "errors": 0,
+ "words": 2,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0167",
+ "category": "technical",
+ "text": "The graphics processor remained idle because the data loader exhausted every central processing core.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0167.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE GRAPHICS PROCESSOR REMAINED IDLE BECAUSE THE DATA LOADER EXHAUSTED EVERY CENTRAL PROCESSING CORPS",
+ "raw": {
+ "normalized_reference": "the graphics processor remained idle because the data loader exhausted every central processing core.",
+ "normalized_hypothesis": "the graphics processor remained idle because the data loader exhausted every central processing corps",
+ "errors": 1,
+ "words": 14,
+ "wer": 0.07142857142857142
+ },
+ "semantic": {
+ "normalized_reference": "the graphics processor remained idle because the data loader exhausted every central processing core",
+ "normalized_hypothesis": "the graphics processor remained idle because the data loader exhausted every central processing corps",
+ "errors": 1,
+ "words": 14,
+ "wer": 0.07142857142857142
+ }
+ },
+ {
+ "id": "release500_0168",
+ "category": "conversational",
+ "text": "Could you speak a little more slowly? I am trying to write every detail down.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0168.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "COULD YOU SPEAK A LITTLE MORE SLOWLY I AM TRYING TO WRITE EVERY DETAIL DOWN",
+ "raw": {
+ "normalized_reference": "could you speak a little more slowly? i am trying to write every detail down.",
+ "normalized_hypothesis": "could you speak a little more slowly i am trying to write every detail down",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "could you speak a little more slowly i am trying to write every detail down",
+ "normalized_hypothesis": "could you speak a little more slowly i am trying to write every detail down",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0169",
+ "category": "emotional",
+ "text": "I was so excited I could barely form a complete sentence.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0169.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I WAS SO EXCITED I COULD BARELY FORM A COMPLETE SENTENCE",
+ "raw": {
+ "normalized_reference": "i was so excited i could barely form a complete sentence.",
+ "normalized_hypothesis": "i was so excited i could barely form a complete sentence",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i was so excited i could barely form a complete sentence",
+ "normalized_hypothesis": "i was so excited i could barely form a complete sentence",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0170",
+ "category": "narrative",
+ "text": "He looked at me like I had said something in a language he didn't speak.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0170.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE LOOKED AT ME LIKE I HAD SAID SOMETHING IN A LANGUAGE HE DIDN'T SPEAK",
+ "raw": {
+ "normalized_reference": "he looked at me like i had said something in a language he didn't speak.",
+ "normalized_hypothesis": "he looked at me like i had said something in a language he didn't speak",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he looked at me like i had said something in a language he did not speak",
+ "normalized_hypothesis": "he looked at me like i had said something in a language he did not speak",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0171",
+ "category": "short",
+ "text": "That's actually hilarious.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0171.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THAT'S ACTUALLY HILARIOUS",
+ "raw": {
+ "normalized_reference": "that's actually hilarious.",
+ "normalized_hypothesis": "that's actually hilarious",
+ "errors": 0,
+ "words": 3,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "that is actually hilarious",
+ "normalized_hypothesis": "that is actually hilarious",
+ "errors": 0,
+ "words": 4,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0172",
+ "category": "technical",
+ "text": "The encoder maps phonemes, stress, and phrase boundaries into a shared hidden representation.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0172.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE INCODER MAPS PHONEAMS STRESS AND PHRASE BOUNDARIES INTO A SHARED HIDDEN REPRESENTATION",
+ "raw": {
+ "normalized_reference": "the encoder maps phonemes, stress, and phrase boundaries into a shared hidden representation.",
+ "normalized_hypothesis": "the incoder maps phoneams stress and phrase boundaries into a shared hidden representation",
+ "errors": 2,
+ "words": 13,
+ "wer": 0.15384615384615385
+ },
+ "semantic": {
+ "normalized_reference": "the encoder maps phonemes stress and phrase boundaries into a shared hidden representation",
+ "normalized_hypothesis": "the incoder maps phoneams stress and phrase boundaries into a shared hidden representation",
+ "errors": 2,
+ "words": 13,
+ "wer": 0.15384615384615385
+ }
+ },
+ {
+ "id": "release500_0173",
+ "category": "conversational",
+ "text": "She showed up three hours late and acted like nothing had happened.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0173.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SHE SHOWED UP THREE HOURS LATE AND ACTED LIKE NOTHING HAD HAPPENED",
+ "raw": {
+ "normalized_reference": "she showed up three hours late and acted like nothing had happened.",
+ "normalized_hypothesis": "she showed up three hours late and acted like nothing had happened",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "she showed up 3 hours late and acted like nothing had happened",
+ "normalized_hypothesis": "she showed up 3 hours late and acted like nothing had happened",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0174",
+ "category": "emotional",
+ "text": "Oh my god. This is the best thing I've ever tasted.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0174.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "O MY GOD THIS IS THE BEST THING I'VE EVER TASTED",
+ "raw": {
+ "normalized_reference": "oh my god. this is the best thing i've ever tasted.",
+ "normalized_hypothesis": "o my god this is the best thing i've ever tasted",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ },
+ "semantic": {
+ "normalized_reference": "0 my god this is the best thing i have ever tasted",
+ "normalized_hypothesis": "0 my god this is the best thing i have ever tasted",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0175",
+ "category": "narrative",
+ "text": "The note was short, which was somehow more frightening than a long one would have been.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0175.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE NOTE WAS SHORT WHICH WAS SOMEHOW MORE FRIGHTEN ING THAN A LONG ONE WOULD HAVE BEEN",
+ "raw": {
+ "normalized_reference": "the note was short, which was somehow more frightening than a long one would have been.",
+ "normalized_hypothesis": "the note was short which was somehow more frighten ing than a long one would have been",
+ "errors": 2,
+ "words": 16,
+ "wer": 0.125
+ },
+ "semantic": {
+ "normalized_reference": "the note was short which was somehow more frightening than a long one would have been",
+ "normalized_hypothesis": "the note was short which was somehow more frighten ing than a long one would have been",
+ "errors": 2,
+ "words": 16,
+ "wer": 0.125
+ }
+ },
+ {
+ "id": "release500_0176",
+ "category": "short",
+ "text": "Noted.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0176.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "NOTED",
+ "raw": {
+ "normalized_reference": "noted.",
+ "normalized_hypothesis": "noted",
+ "errors": 0,
+ "words": 1,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "noted",
+ "normalized_hypothesis": "noted",
+ "errors": 0,
+ "words": 1,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0177",
+ "category": "technical",
+ "text": "It's a simple process once you understand the underlying structure.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0177.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT'S A SIMPLE PROCESS ONCE YOU UNDERSTAND THE UNDERLYING STRUCTURE",
+ "raw": {
+ "normalized_reference": "it's a simple process once you understand the underlying structure.",
+ "normalized_hypothesis": "it's a simple process once you understand the underlying structure",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it is a simple process once you understand the underlying structure",
+ "normalized_hypothesis": "it is a simple process once you understand the underlying structure",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0178",
+ "category": "conversational",
+ "text": "You have to hear this story because it's almost too good to be true.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0178.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "YOU HAVE TO HEAR THIS STORY BECAUSE IT'S ALMOST TOO GOOD TO BE TRUE",
+ "raw": {
+ "normalized_reference": "you have to hear this story because it's almost too good to be true.",
+ "normalized_hypothesis": "you have to hear this story because it's almost too good to be true",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "you have to hear this story because it is almost too good to be true",
+ "normalized_hypothesis": "you have to hear this story because it is almost too good to be true",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0179",
+ "category": "emotional",
+ "text": "I have never felt more seen in my entire life than in that moment.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0179.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I HAVE NEVER FELT MORE SEEN IN MY ENTIRE LIFE THAN IN THAT MOMENT",
+ "raw": {
+ "normalized_reference": "i have never felt more seen in my entire life than in that moment.",
+ "normalized_hypothesis": "i have never felt more seen in my entire life than in that moment",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i have never felt more seen in my entire life than in that moment",
+ "normalized_hypothesis": "i have never felt more seen in my entire life than in that moment",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0180",
+ "category": "narrative",
+ "text": "She knew. She had always known. She just waited for the right moment.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0180.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SHE KNEW SHE HAD ALWAYS KNOWN SHE JUST WAITED FOR THE RIGHT MOMENT",
+ "raw": {
+ "normalized_reference": "she knew. she had always known. she just waited for the right moment.",
+ "normalized_hypothesis": "she knew she had always known she just waited for the right moment",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "she knew she had always known she just waited for the right moment",
+ "normalized_hypothesis": "she knew she had always known she just waited for the right moment",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0181",
+ "category": "short",
+ "text": "Yeah, no. Hard pass.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0181.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "YETH NO HARD PASS",
+ "raw": {
+ "normalized_reference": "yeah, no. hard pass.",
+ "normalized_hypothesis": "yeth no hard pass",
+ "errors": 1,
+ "words": 4,
+ "wer": 0.25
+ },
+ "semantic": {
+ "normalized_reference": "yeah no hard pass",
+ "normalized_hypothesis": "yeth no hard pass",
+ "errors": 1,
+ "words": 4,
+ "wer": 0.25
+ }
+ },
+ {
+ "id": "release500_0182",
+ "category": "technical",
+ "text": "Please verify the sample rate, hop length, and mel filter configuration before inference.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0182.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "PLEASE VERIFY THE SAMPLE RATE HOP LENGTH AND MELL FILTER CONFIGURATION BEFORE INFERENCE",
+ "raw": {
+ "normalized_reference": "please verify the sample rate, hop length, and mel filter configuration before inference.",
+ "normalized_hypothesis": "please verify the sample rate hop length and mell filter configuration before inference",
+ "errors": 1,
+ "words": 13,
+ "wer": 0.07692307692307693
+ },
+ "semantic": {
+ "normalized_reference": "please verify the sample rate hop length and mel filter configuration before inference",
+ "normalized_hypothesis": "please verify the sample rate hop length and mell filter configuration before inference",
+ "errors": 1,
+ "words": 13,
+ "wer": 0.07692307692307693
+ }
+ },
+ {
+ "id": "release500_0183",
+ "category": "conversational",
+ "text": "That's the worst idea I've ever heard and I'm completely in.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0183.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THAT'S THE WORST IDEA I'VE EVER HEARD AND I'M COMPLETELY IN",
+ "raw": {
+ "normalized_reference": "that's the worst idea i've ever heard and i'm completely in.",
+ "normalized_hypothesis": "that's the worst idea i've ever heard and i'm completely in",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "that is the worst idea i have ever heard and i am completely in",
+ "normalized_hypothesis": "that is the worst idea i have ever heard and i am completely in",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0184",
+ "category": "emotional",
+ "text": "I remember thinking: this is it. This is the thing I've been waiting for.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0184.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I REMEMBER THINKING THIS IS IT THIS IS THE THING I HAVE BEEN WAITING FOR",
+ "raw": {
+ "normalized_reference": "i remember thinking: this is it. this is the thing i've been waiting for.",
+ "normalized_hypothesis": "i remember thinking this is it this is the thing i have been waiting for",
+ "errors": 2,
+ "words": 14,
+ "wer": 0.14285714285714285
+ },
+ "semantic": {
+ "normalized_reference": "i remember thinking this is it this is the thing i have been waiting for",
+ "normalized_hypothesis": "i remember thinking this is it this is the thing i have been waiting for",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0185",
+ "category": "narrative",
+ "text": "We never talked about it again, which I think was the right call.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0185.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WE NEVER TALKED ABOUT IT AGAIN WHICH I THINK WAS THE RIGHT CALL",
+ "raw": {
+ "normalized_reference": "we never talked about it again, which i think was the right call.",
+ "normalized_hypothesis": "we never talked about it again which i think was the right call",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "we never talked about it again which i think was the right call",
+ "normalized_hypothesis": "we never talked about it again which i think was the right call",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0186",
+ "category": "short",
+ "text": "In hindsight, obvious.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0186.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IN HINDSIGHT OBVIOUS",
+ "raw": {
+ "normalized_reference": "in hindsight, obvious.",
+ "normalized_hypothesis": "in hindsight obvious",
+ "errors": 0,
+ "words": 3,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "in hindsight obvious",
+ "normalized_hypothesis": "in hindsight obvious",
+ "errors": 0,
+ "words": 3,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0187",
+ "category": "technical",
+ "text": "Once it's configured correctly, it basically runs itself.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0187.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "ONCE IT'S CONFIGURED CORRECTLY IT BASICALLY RUNS ITSELF",
+ "raw": {
+ "normalized_reference": "once it's configured correctly, it basically runs itself.",
+ "normalized_hypothesis": "once it's configured correctly it basically runs itself",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "once it is configured correctly it basically runs itself",
+ "normalized_hypothesis": "once it is configured correctly it basically runs itself",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0188",
+ "category": "conversational",
+ "text": "You got engaged? When? How? Tell me everything right now.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0188.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "YOU GOT ENGAGED WHEN COW TELL ME EVERYTHING RIGHT NOW",
+ "raw": {
+ "normalized_reference": "you got engaged? when? how? tell me everything right now.",
+ "normalized_hypothesis": "you got engaged when cow tell me everything right now",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ },
+ "semantic": {
+ "normalized_reference": "you got engaged when how tell me everything right now",
+ "normalized_hypothesis": "you got engaged when cow tell me everything right now",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ }
+ },
+ {
+ "id": "release500_0189",
+ "category": "emotional",
+ "text": "It's hard to explain how much that moment meant to me.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0189.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT'S HARD TO EXPLAIN HOW MUCH THAT MOMENT MEANT TO ME",
+ "raw": {
+ "normalized_reference": "it's hard to explain how much that moment meant to me.",
+ "normalized_hypothesis": "it's hard to explain how much that moment meant to me",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it is hard to explain how much that moment meant to me",
+ "normalized_hypothesis": "it is hard to explain how much that moment meant to me",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0190",
+ "category": "narrative",
+ "text": "He told the story three different times that night and somehow it got better each time.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0190.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE TOLD THE STORY THREE DIFFERENT TIMES THAT NIGHT AND SOMEHOW IT GOT BETTER EACH TIME",
+ "raw": {
+ "normalized_reference": "he told the story three different times that night and somehow it got better each time.",
+ "normalized_hypothesis": "he told the story three different times that night and somehow it got better each time",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he told the story 3 different times that night and somehow it got better each time",
+ "normalized_hypothesis": "he told the story 3 different times that night and somehow it got better each time",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0191",
+ "category": "short",
+ "text": "Right. Moving on.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0191.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "RIGHT MOVING ON",
+ "raw": {
+ "normalized_reference": "right. moving on.",
+ "normalized_hypothesis": "right moving on",
+ "errors": 0,
+ "words": 3,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "right moving on",
+ "normalized_hypothesis": "right moving on",
+ "errors": 0,
+ "words": 3,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0192",
+ "category": "conversational",
+ "text": "No, I did not move your keys; check beside the small ceramic bowl.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0192.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "NO I DID NOT MOVE YOUR KEYS CHECK BESIDE THE SMALL SERAMIC BOWL",
+ "raw": {
+ "normalized_reference": "no, i did not move your keys; check beside the small ceramic bowl.",
+ "normalized_hypothesis": "no i did not move your keys check beside the small seramic bowl",
+ "errors": 1,
+ "words": 13,
+ "wer": 0.07692307692307693
+ },
+ "semantic": {
+ "normalized_reference": "no i did not move your keys check beside the small ceramic bowl",
+ "normalized_hypothesis": "no i did not move your keys check beside the small seramic bowl",
+ "errors": 1,
+ "words": 13,
+ "wer": 0.07692307692307693
+ }
+ },
+ {
+ "id": "release500_0193",
+ "category": "emotional",
+ "text": "I was devastated. I'm not going to pretend otherwise.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0193.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I WAS DEVASTATED I AM NOT GOING TO PRETEND OTHERWISE",
+ "raw": {
+ "normalized_reference": "i was devastated. i'm not going to pretend otherwise.",
+ "normalized_hypothesis": "i was devastated i am not going to pretend otherwise",
+ "errors": 2,
+ "words": 9,
+ "wer": 0.2222222222222222
+ },
+ "semantic": {
+ "normalized_reference": "i was devastated i am not going to pretend otherwise",
+ "normalized_hypothesis": "i was devastated i am not going to pretend otherwise",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0194",
+ "category": "narrative",
+ "text": "That was the last summer before everything changed, though we didn't know it yet.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0194.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THAT WAS THE LAST SUMMER BEFORE EVERYTHING CHANGED THOUGH WE DIDN'T KNOW IT YET",
+ "raw": {
+ "normalized_reference": "that was the last summer before everything changed, though we didn't know it yet.",
+ "normalized_hypothesis": "that was the last summer before everything changed though we didn't know it yet",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "that was the last summer before everything changed though we did not know it yet",
+ "normalized_hypothesis": "that was the last summer before everything changed though we did not know it yet",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0195",
+ "category": "short",
+ "text": "Bold of you to assume.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0195.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "BOLD OF YOU TO ASSUME",
+ "raw": {
+ "normalized_reference": "bold of you to assume.",
+ "normalized_hypothesis": "bold of you to assume",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "bold of you to assume",
+ "normalized_hypothesis": "bold of you to assume",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0196",
+ "category": "conversational",
+ "text": "It's one of those decisions that makes total sense in the moment.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0196.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT'S ONE OF THOSE DECISIONS THAT MAKES TOTAL SENSE IN THE MOMENT",
+ "raw": {
+ "normalized_reference": "it's one of those decisions that makes total sense in the moment.",
+ "normalized_hypothesis": "it's one of those decisions that makes total sense in the moment",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it is one of those decisions that makes total sense in the moment",
+ "normalized_hypothesis": "it is one of those decisions that makes total sense in the moment",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0197",
+ "category": "emotional",
+ "text": "I was furious. Not the loud kind. The quiet kind that lasts for days.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0197.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I WAS FURIOUS NOT THE LOUD KIND THE QUIET KIND THAT LASTS FOR DAYS",
+ "raw": {
+ "normalized_reference": "i was furious. not the loud kind. the quiet kind that lasts for days.",
+ "normalized_hypothesis": "i was furious not the loud kind the quiet kind that lasts for days",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i was furious not the loud kind the quiet kind that lasts for days",
+ "normalized_hypothesis": "i was furious not the loud kind the quiet kind that lasts for days",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0198",
+ "category": "narrative",
+ "text": "That's the part of the story I usually leave out.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0198.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THAT'S THE PART OF THE STORY I USUALLY LEAVE OUT",
+ "raw": {
+ "normalized_reference": "that's the part of the story i usually leave out.",
+ "normalized_hypothesis": "that's the part of the story i usually leave out",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "that is the part of the story i usually leave out",
+ "normalized_hypothesis": "that is the part of the story i usually leave out",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0199",
+ "category": "short",
+ "text": "I've said too much.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0199.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I'VE SAID TOO MUCH",
+ "raw": {
+ "normalized_reference": "i've said too much.",
+ "normalized_hypothesis": "i've said too much",
+ "errors": 0,
+ "words": 4,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i have said too much",
+ "normalized_hypothesis": "i have said too much",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0200",
+ "category": "conversational",
+ "text": "I've been faking that I understood this the entire time.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0200.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I'VE BEEN FAKING THAT I UNDERSTOOD THIS THE ENTIRE TIME",
+ "raw": {
+ "normalized_reference": "i've been faking that i understood this the entire time.",
+ "normalized_hypothesis": "i've been faking that i understood this the entire time",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i have been faking that i understood this the entire time",
+ "normalized_hypothesis": "i have been faking that i understood this the entire time",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0201",
+ "category": "phonetic_balance",
+ "text": "Peace be unto you and grace before the Lord.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0201.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "PEACE BE UNTO YOU IN GRACE BEFORE THE LORD",
+ "raw": {
+ "normalized_reference": "peace be unto you and grace before the lord.",
+ "normalized_hypothesis": "peace be unto you in grace before the lord",
+ "errors": 1,
+ "words": 9,
+ "wer": 0.1111111111111111
+ },
+ "semantic": {
+ "normalized_reference": "peace be unto you and grace before the lord",
+ "normalized_hypothesis": "peace be unto you in grace before the lord",
+ "errors": 1,
+ "words": 9,
+ "wer": 0.1111111111111111
+ }
+ },
+ {
+ "id": "release500_0202",
+ "category": "phonetic_balance",
+ "text": "From now on we're pals.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0202.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "FROM NOW ON WER PALLS",
+ "raw": {
+ "normalized_reference": "from now on we're pals.",
+ "normalized_hypothesis": "from now on wer palls",
+ "errors": 2,
+ "words": 5,
+ "wer": 0.4
+ },
+ "semantic": {
+ "normalized_reference": "from now on we are pals",
+ "normalized_hypothesis": "from now on wer palls",
+ "errors": 3,
+ "words": 6,
+ "wer": 0.5
+ }
+ },
+ {
+ "id": "release500_0203",
+ "category": "phonetic_balance",
+ "text": "Here we allow our solicitors to look after our legal work.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0203.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HERE WE ALLOW OUR SOLICITORS TO LOOK AFTER OUR LEGAL WORK",
+ "raw": {
+ "normalized_reference": "here we allow our solicitors to look after our legal work.",
+ "normalized_hypothesis": "here we allow our solicitors to look after our legal work",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "here we allow our solicitors to look after our legal work",
+ "normalized_hypothesis": "here we allow our solicitors to look after our legal work",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0204",
+ "category": "phonetic_balance",
+ "text": "Her words sent a strange chill through Philip.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0204.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HER WORDS SENT A STRANGE CHILL THROUGH PHILIP",
+ "raw": {
+ "normalized_reference": "her words sent a strange chill through philip.",
+ "normalized_hypothesis": "her words sent a strange chill through philip",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "her words sent a strange chill through philip",
+ "normalized_hypothesis": "her words sent a strange chill through philip",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0205",
+ "category": "phonetic_balance",
+ "text": "It was introduced by Representative Dick of Ohio.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0205.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT WAS INTRODUCED BY REPRESENTATIVE DICK OF OHIO",
+ "raw": {
+ "normalized_reference": "it was introduced by representative dick of ohio.",
+ "normalized_hypothesis": "it was introduced by representative dick of ohio",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it was introduced by representative dick of ohio",
+ "normalized_hypothesis": "it was introduced by representative dick of ohio",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0206",
+ "category": "phonetic_balance",
+ "text": "It's the aurora borealis.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0206.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT'S THE AURORA BORIALIS",
+ "raw": {
+ "normalized_reference": "it's the aurora borealis.",
+ "normalized_hypothesis": "it's the aurora borialis",
+ "errors": 1,
+ "words": 4,
+ "wer": 0.25
+ },
+ "semantic": {
+ "normalized_reference": "it is the aurora borealis",
+ "normalized_hypothesis": "it is the aurora borialis",
+ "errors": 1,
+ "words": 5,
+ "wer": 0.2
+ }
+ },
+ {
+ "id": "release500_0207",
+ "category": "phonetic_balance",
+ "text": "The boy grew and prospered.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0207.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE BOY GREW AND PROSPERED",
+ "raw": {
+ "normalized_reference": "the boy grew and prospered.",
+ "normalized_hypothesis": "the boy grew and prospered",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the boy grew and prospered",
+ "normalized_hypothesis": "the boy grew and prospered",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0208",
+ "category": "phonetic_balance",
+ "text": "His voice was passionately rebellious.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0208.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HIS VOICE WAS PASSIONATELY REBELLIOUS",
+ "raw": {
+ "normalized_reference": "his voice was passionately rebellious.",
+ "normalized_hypothesis": "his voice was passionately rebellious",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "his voice was passionately rebellious",
+ "normalized_hypothesis": "his voice was passionately rebellious",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0209",
+ "category": "phonetic_balance",
+ "text": "That's where they cut off the Scottish Chiefs and killed all hands.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0209.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THAT'S WHERE THEY CUT OFF THE SCOTTISH CHIEFS AND KILLED ALL HANDS",
+ "raw": {
+ "normalized_reference": "that's where they cut off the scottish chiefs and killed all hands.",
+ "normalized_hypothesis": "that's where they cut off the scottish chiefs and killed all hands",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "that is where they cut off the scottish chiefs and killed all hands",
+ "normalized_hypothesis": "that is where they cut off the scottish chiefs and killed all hands",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0210",
+ "category": "phonetic_balance",
+ "text": "And Tom King patiently endured.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0210.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "AND TOM KING PATIENTLY ENDURED",
+ "raw": {
+ "normalized_reference": "and tom king patiently endured.",
+ "normalized_hypothesis": "and tom king patiently endured",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "and tom king patiently endured",
+ "normalized_hypothesis": "and tom king patiently endured",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0211",
+ "category": "phonetic_balance",
+ "text": "From that moment his friendship for Belize turns to hatred and jealousy.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0211.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "FROM THAT MOMENT HIS FRIENDSHIP FOR BELLIS TURNS TO HATRED AND JEALOUSY",
+ "raw": {
+ "normalized_reference": "from that moment his friendship for belize turns to hatred and jealousy.",
+ "normalized_hypothesis": "from that moment his friendship for bellis turns to hatred and jealousy",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "from that moment his friendship for belize turns to hatred and jealousy",
+ "normalized_hypothesis": "from that moment his friendship for bellis turns to hatred and jealousy",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ }
+ },
+ {
+ "id": "release500_0212",
+ "category": "phonetic_balance",
+ "text": "Then you don't believe in altruism.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0212.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THEN YOU DON'T BELIEVE IN ALTRUISM",
+ "raw": {
+ "normalized_reference": "then you don't believe in altruism.",
+ "normalized_hypothesis": "then you don't believe in altruism",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "then you do not believe in altruism",
+ "normalized_hypothesis": "then you do not believe in altruism",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0213",
+ "category": "phonetic_balance",
+ "text": "What Pascal glimpsed with the vision of a seer, I have lived.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0213.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WHAT PASCALL GLIMPSED WITH THE VISION OF A SEER I HAVE LIVED",
+ "raw": {
+ "normalized_reference": "what pascal glimpsed with the vision of a seer, i have lived.",
+ "normalized_hypothesis": "what pascall glimpsed with the vision of a seer i have lived",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "what pascal glimpsed with the vision of a seer i have lived",
+ "normalized_hypothesis": "what pascall glimpsed with the vision of a seer i have lived",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ }
+ },
+ {
+ "id": "release500_0214",
+ "category": "phonetic_balance",
+ "text": "You were making them talk shop, Ruth charged him.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0214.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "YOU WERE MAKING THEM TALK SHOP RUTH CHARGED HIM",
+ "raw": {
+ "normalized_reference": "you were making them talk shop, ruth charged him.",
+ "normalized_hypothesis": "you were making them talk shop ruth charged him",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "you were making them talk shop ruth charged him",
+ "normalized_hypothesis": "you were making them talk shop ruth charged him",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0215",
+ "category": "phonetic_balance",
+ "text": "Philip snatched at the letter which Gregson held out to him.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0215.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "PHILIP SNATCHED AT THE LETTER WHICH GREGSON HELD OUT TO HIM",
+ "raw": {
+ "normalized_reference": "philip snatched at the letter which gregson held out to him.",
+ "normalized_hypothesis": "philip snatched at the letter which gregson held out to him",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "philip snatched at the letter which gregson held out to him",
+ "normalized_hypothesis": "philip snatched at the letter which gregson held out to him",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0216",
+ "category": "phonetic_balance",
+ "text": "For a time the exciting thrill of his adventure was gone.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0216.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "FOR A TIME THE EXCITING THRILL OF HIS ADVENTURE WAS GONE",
+ "raw": {
+ "normalized_reference": "for a time the exciting thrill of his adventure was gone.",
+ "normalized_hypothesis": "for a time the exciting thrill of his adventure was gone",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "for a time the exciting thrill of his adventure was gone",
+ "normalized_hypothesis": "for a time the exciting thrill of his adventure was gone",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0217",
+ "category": "phonetic_balance",
+ "text": "Obviously, it was a disease that could be contracted by contact.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0217.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "OBVIOUSLY IT WAS A DISEASE THAT COULD BE CONTRACTED BY CONTACT",
+ "raw": {
+ "normalized_reference": "obviously, it was a disease that could be contracted by contact.",
+ "normalized_hypothesis": "obviously it was a disease that could be contracted by contact",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "obviously it was a disease that could be contracted by contact",
+ "normalized_hypothesis": "obviously it was a disease that could be contracted by contact",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0218",
+ "category": "phonetic_balance",
+ "text": "Blind with rage, he darted in.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0218.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "BLIND WITH RAGE HE DARTED",
+ "raw": {
+ "normalized_reference": "blind with rage, he darted in.",
+ "normalized_hypothesis": "blind with rage he darted",
+ "errors": 1,
+ "words": 6,
+ "wer": 0.16666666666666666
+ },
+ "semantic": {
+ "normalized_reference": "blind with rage he darted in",
+ "normalized_hypothesis": "blind with rage he darted",
+ "errors": 1,
+ "words": 6,
+ "wer": 0.16666666666666666
+ }
+ },
+ {
+ "id": "release500_0219",
+ "category": "phonetic_balance",
+ "text": "You live on an income which your father earned.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0219.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "YOU LIVE ON AN INCOME WHICH YOUR FATHER EARNED",
+ "raw": {
+ "normalized_reference": "you live on an income which your father earned.",
+ "normalized_hypothesis": "you live on an income which your father earned",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "you live on an income which your father earned",
+ "normalized_hypothesis": "you live on an income which your father earned",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0220",
+ "category": "phonetic_balance",
+ "text": "Something about it seemed to fascinate him, to challenge his presence.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0220.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SOMETHING ABOUT IT SEEMED TO FASCINATE HIM TO CHALLENGE HIS PRESENCE",
+ "raw": {
+ "normalized_reference": "something about it seemed to fascinate him, to challenge his presence.",
+ "normalized_hypothesis": "something about it seemed to fascinate him to challenge his presence",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "something about it seemed to fascinate him to challenge his presence",
+ "normalized_hypothesis": "something about it seemed to fascinate him to challenge his presence",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0221",
+ "category": "phonetic_balance",
+ "text": "Her beautiful hair was done up in shining coils.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0221.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HER BEAUTIFUL HAIR WAS DONE UP IN SHINING COILS",
+ "raw": {
+ "normalized_reference": "her beautiful hair was done up in shining coils.",
+ "normalized_hypothesis": "her beautiful hair was done up in shining coils",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "her beautiful hair was done up in shining coils",
+ "normalized_hypothesis": "her beautiful hair was done up in shining coils",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0222",
+ "category": "phonetic_balance",
+ "text": "It does, was her audacious answer.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0222.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT DOES WAS HER AUDACIOUS ANSWER",
+ "raw": {
+ "normalized_reference": "it does, was her audacious answer.",
+ "normalized_hypothesis": "it does was her audacious answer",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it does was her audacious answer",
+ "normalized_hypothesis": "it does was her audacious answer",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0223",
+ "category": "phonetic_balance",
+ "text": "He caught himself with a jerk.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0223.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE CAUGHT HIMSELF WITH A JERK",
+ "raw": {
+ "normalized_reference": "he caught himself with a jerk.",
+ "normalized_hypothesis": "he caught himself with a jerk",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he caught himself with a jerk",
+ "normalized_hypothesis": "he caught himself with a jerk",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0224",
+ "category": "phonetic_balance",
+ "text": "He did not believe in the burning of daylight for such a luxury.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0224.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE DID NOT BELIEVE IN THE BURNING OF DAYLIGHT FOR SUCH A LUXURY",
+ "raw": {
+ "normalized_reference": "he did not believe in the burning of daylight for such a luxury.",
+ "normalized_hypothesis": "he did not believe in the burning of daylight for such a luxury",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he did not believe in the burning of daylight for such a luxury",
+ "normalized_hypothesis": "he did not believe in the burning of daylight for such a luxury",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0225",
+ "category": "phonetic_balance",
+ "text": "He had comparatively no advantages at first.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0225.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE HAD COMPARATIVELY NO ADVANTAGES AT FIRST",
+ "raw": {
+ "normalized_reference": "he had comparatively no advantages at first.",
+ "normalized_hypothesis": "he had comparatively no advantages at first",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he had comparatively no advantages at 1st",
+ "normalized_hypothesis": "he had comparatively no advantages at 1st",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0226",
+ "category": "phonetic_balance",
+ "text": "Such men believe, when they come together.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0226.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SUCH MEN BELIEVE WHEN THEY COME TOGETHER",
+ "raw": {
+ "normalized_reference": "such men believe, when they come together.",
+ "normalized_hypothesis": "such men believe when they come together",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "such men believe when they come together",
+ "normalized_hypothesis": "such men believe when they come together",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0227",
+ "category": "phonetic_balance",
+ "text": "The ship should be in within a week or ten days.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0227.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE SHIP SHOULD BE IN WITHIN A WEEK OR TEN DAYS",
+ "raw": {
+ "normalized_reference": "the ship should be in within a week or ten days.",
+ "normalized_hypothesis": "the ship should be in within a week or ten days",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the ship should be in within a week or 10 days",
+ "normalized_hypothesis": "the ship should be in within a week or 10 days",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0228",
+ "category": "phonetic_balance",
+ "text": "Such a dog the wise driver kills, or turns loose.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0228.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SUCH A DOG THE WISE DRIVER KILLS OR TURNS LOOSE",
+ "raw": {
+ "normalized_reference": "such a dog the wise driver kills, or turns loose.",
+ "normalized_hypothesis": "such a dog the wise driver kills or turns loose",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "such a dog the wise driver kills or turns loose",
+ "normalized_hypothesis": "such a dog the wise driver kills or turns loose",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0229",
+ "category": "phonetic_balance",
+ "text": "You have heard always how he was the lover of the Princess Naomi.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0229.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "YOU HAVE HEARD ALWAYS HOW HE WAS THE LOVER OF THE PRINCESS NAOMI",
+ "raw": {
+ "normalized_reference": "you have heard always how he was the lover of the princess naomi.",
+ "normalized_hypothesis": "you have heard always how he was the lover of the princess naomi",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "you have heard always how he was the lover of the princess naomi",
+ "normalized_hypothesis": "you have heard always how he was the lover of the princess naomi",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0230",
+ "category": "phonetic_balance",
+ "text": "I want to know how all this is possible.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0230.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I WANT TO KNOW HOW ALL THIS IS POSSIBLE",
+ "raw": {
+ "normalized_reference": "i want to know how all this is possible.",
+ "normalized_hypothesis": "i want to know how all this is possible",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i want to know how all this is possible",
+ "normalized_hypothesis": "i want to know how all this is possible",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0231",
+ "category": "phonetic_balance",
+ "text": "It was my idea to a tee.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0231.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT WAS MY IDEA TO A T",
+ "raw": {
+ "normalized_reference": "it was my idea to a tee.",
+ "normalized_hypothesis": "it was my idea to a t",
+ "errors": 1,
+ "words": 7,
+ "wer": 0.14285714285714285
+ },
+ "semantic": {
+ "normalized_reference": "it was my idea to a tee",
+ "normalized_hypothesis": "it was my idea to a t",
+ "errors": 1,
+ "words": 7,
+ "wer": 0.14285714285714285
+ }
+ },
+ {
+ "id": "release500_0232",
+ "category": "phonetic_balance",
+ "text": "They likewise are disinclined to being eaten.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0232.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THEY LIKEWISE ARE DISINCLINED TO BEING EATEN",
+ "raw": {
+ "normalized_reference": "they likewise are disinclined to being eaten.",
+ "normalized_hypothesis": "they likewise are disinclined to being eaten",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "they likewise are disinclined to being eaten",
+ "normalized_hypothesis": "they likewise are disinclined to being eaten",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0233",
+ "category": "phonetic_balance",
+ "text": "This state of mind comes of an undue prominence of the ego.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0233.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THIS STATE OF MIND COMES OF AN UNDUE PROMINENCE OF THE EGO",
+ "raw": {
+ "normalized_reference": "this state of mind comes of an undue prominence of the ego.",
+ "normalized_hypothesis": "this state of mind comes of an undue prominence of the ego",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "this state of mind comes of an undue prominence of the ego",
+ "normalized_hypothesis": "this state of mind comes of an undue prominence of the ego",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0234",
+ "category": "phonetic_balance",
+ "text": "Then it is as I said, Womble announced with finality.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0234.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THEN IT IS AS I SAID WAMBLLE ANNOUNCED WITH FINALITY",
+ "raw": {
+ "normalized_reference": "then it is as i said, womble announced with finality.",
+ "normalized_hypothesis": "then it is as i said wamblle announced with finality",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ },
+ "semantic": {
+ "normalized_reference": "then it is as i said womble announced with finality",
+ "normalized_hypothesis": "then it is as i said wamblle announced with finality",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ }
+ },
+ {
+ "id": "release500_0235",
+ "category": "phonetic_balance",
+ "text": "I can see that knife now.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0235.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I CAN SEE THAT KNIFE NOW",
+ "raw": {
+ "normalized_reference": "i can see that knife now.",
+ "normalized_hypothesis": "i can see that knife now",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i can see that knife now",
+ "normalized_hypothesis": "i can see that knife now",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0236",
+ "category": "phonetic_balance",
+ "text": "If I meet her again I shall apologize, said Eileen.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0236.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IF I MEET HER AGAIN I SHALL APOLOGIZE SAID ILEENE",
+ "raw": {
+ "normalized_reference": "if i meet her again i shall apologize, said eileen.",
+ "normalized_hypothesis": "if i meet her again i shall apologize said ileene",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ },
+ "semantic": {
+ "normalized_reference": "if i meet her again i shall apologize said eileen",
+ "normalized_hypothesis": "if i meet her again i shall apologize said ileene",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ }
+ },
+ {
+ "id": "release500_0237",
+ "category": "phonetic_balance",
+ "text": "A wildly exciting time was his during the week preceding Thursday the eighteenth.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0237.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "A WILDLY EXCITING TIME WAS HIS DURING THE WEEK PRECEDING THURSDAY THE EIGHTEENTH",
+ "raw": {
+ "normalized_reference": "a wildly exciting time was his during the week preceding thursday the eighteenth.",
+ "normalized_hypothesis": "a wildly exciting time was his during the week preceding thursday the eighteenth",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "a wildly exciting time was his during the week preceding thursday the 18th",
+ "normalized_hypothesis": "a wildly exciting time was his during the week preceding thursday the 18th",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0238",
+ "category": "phonetic_balance",
+ "text": "In what bucolic school of fence he had been taught was beyond imagining.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0238.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IN WHAT BEUCOLIC SCHOOL OF FENCE HE HAD BEEN TAUGHT WAS BEYOND IMAGINING",
+ "raw": {
+ "normalized_reference": "in what bucolic school of fence he had been taught was beyond imagining.",
+ "normalized_hypothesis": "in what beucolic school of fence he had been taught was beyond imagining",
+ "errors": 1,
+ "words": 13,
+ "wer": 0.07692307692307693
+ },
+ "semantic": {
+ "normalized_reference": "in what bucolic school of fence he had been taught was beyond imagining",
+ "normalized_hypothesis": "in what beucolic school of fence he had been taught was beyond imagining",
+ "errors": 1,
+ "words": 13,
+ "wer": 0.07692307692307693
+ }
+ },
+ {
+ "id": "release500_0239",
+ "category": "phonetic_balance",
+ "text": "So she said, the irate skipper dashed on.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0239.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SO SHE SAID THE IRATE SKIPPER DASHED ON",
+ "raw": {
+ "normalized_reference": "so she said, the irate skipper dashed on.",
+ "normalized_hypothesis": "so she said the irate skipper dashed on",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "so she said the irate skipper dashed on",
+ "normalized_hypothesis": "so she said the irate skipper dashed on",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0240",
+ "category": "phonetic_balance",
+ "text": "It is not an attempt to smash the market.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0240.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT IS NOT AN ATTEMPT TO SMASH THE MARKET",
+ "raw": {
+ "normalized_reference": "it is not an attempt to smash the market.",
+ "normalized_hypothesis": "it is not an attempt to smash the market",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it is not an attempt to smash the market",
+ "normalized_hypothesis": "it is not an attempt to smash the market",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0241",
+ "category": "phonetic_balance",
+ "text": "He destroyed everything that had belonged to the woman.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0241.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE DESTROYED EVERYTHING THAT HAD BELONGED TO THE WOMAN",
+ "raw": {
+ "normalized_reference": "he destroyed everything that had belonged to the woman.",
+ "normalized_hypothesis": "he destroyed everything that had belonged to the woman",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he destroyed everything that had belonged to the woman",
+ "normalized_hypothesis": "he destroyed everything that had belonged to the woman",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0242",
+ "category": "phonetic_balance",
+ "text": "Did Thorpe go to see any one in Churchill.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0242.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "DID THORPE GO TO SEE ANY ONE IN CHURCHILL",
+ "raw": {
+ "normalized_reference": "did thorpe go to see any one in churchill.",
+ "normalized_hypothesis": "did thorpe go to see any one in churchill",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "did thorpe go to see any one in churchill",
+ "normalized_hypothesis": "did thorpe go to see any one in churchill",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0243",
+ "category": "phonetic_balance",
+ "text": "Many other similar disconcerting omissions will be noticed in the Manuscript.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0243.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "MANY OTHER SIMILAR DISCONCERTING OMISSIONS WILL BE NOTICED IN THE MANUSCRIPT",
+ "raw": {
+ "normalized_reference": "many other similar disconcerting omissions will be noticed in the manuscript.",
+ "normalized_hypothesis": "many other similar disconcerting omissions will be noticed in the manuscript",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "many other similar disconcerting omissions will be noticed in the manuscript",
+ "normalized_hypothesis": "many other similar disconcerting omissions will be noticed in the manuscript",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0244",
+ "category": "phonetic_balance",
+ "text": "The ringing of the big bell aroused him.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0244.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE RINGING OF THE BIG BELL AROUSED HIM",
+ "raw": {
+ "normalized_reference": "the ringing of the big bell aroused him.",
+ "normalized_hypothesis": "the ringing of the big bell aroused him",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the ringing of the big bell aroused him",
+ "normalized_hypothesis": "the ringing of the big bell aroused him",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0245",
+ "category": "phonetic_balance",
+ "text": "A sinewy hand, dripping with water, was clutching the rail.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0245.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "A SINEWY HAND DRIPPING WITH WATER WAS CLUTCHING THE RAIL",
+ "raw": {
+ "normalized_reference": "a sinewy hand, dripping with water, was clutching the rail.",
+ "normalized_hypothesis": "a sinewy hand dripping with water was clutching the rail",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "a sinewy hand dripping with water was clutching the rail",
+ "normalized_hypothesis": "a sinewy hand dripping with water was clutching the rail",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0246",
+ "category": "phonetic_balance",
+ "text": "Enters now the psychology of the situation.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0246.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "ENTERS NOW THE PSYCHOLOGY OF THE SITUATION",
+ "raw": {
+ "normalized_reference": "enters now the psychology of the situation.",
+ "normalized_hypothesis": "enters now the psychology of the situation",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "enters now the psychology of the situation",
+ "normalized_hypothesis": "enters now the psychology of the situation",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0247",
+ "category": "phonetic_balance",
+ "text": "Wash your hands of me.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0247.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WASH YOUR HANDS OF ME",
+ "raw": {
+ "normalized_reference": "wash your hands of me.",
+ "normalized_hypothesis": "wash your hands of me",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "wash your hands of me",
+ "normalized_hypothesis": "wash your hands of me",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0248",
+ "category": "phonetic_balance",
+ "text": "He confessed that the sketch had startled him.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0248.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE CONFESSED THAT THE SKETCH HAD STARTLED HIM",
+ "raw": {
+ "normalized_reference": "he confessed that the sketch had startled him.",
+ "normalized_hypothesis": "he confessed that the sketch had startled him",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he confessed that the sketch had startled him",
+ "normalized_hypothesis": "he confessed that the sketch had startled him",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0249",
+ "category": "phonetic_balance",
+ "text": "Her face was against his breast.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0249.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HER FACE WAS AGAINST HIS BREAST",
+ "raw": {
+ "normalized_reference": "her face was against his breast.",
+ "normalized_hypothesis": "her face was against his breast",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "her face was against his breast",
+ "normalized_hypothesis": "her face was against his breast",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0250",
+ "category": "phonetic_balance",
+ "text": "He was a pariah; a wanderer without a friend or a home.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0250.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE WAS A PARIAH A WANDERER WITHOUT A FRIEND OR A HOME",
+ "raw": {
+ "normalized_reference": "he was a pariah; a wanderer without a friend or a home.",
+ "normalized_hypothesis": "he was a pariah a wanderer without a friend or a home",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he was a pariah a wanderer without a friend or a home",
+ "normalized_hypothesis": "he was a pariah a wanderer without a friend or a home",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0251",
+ "category": "phonetic_balance",
+ "text": "A cry of joy burst from Philip's lips.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0251.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "A CRY OF JOY BURST FROM PHILIP'S LIPS",
+ "raw": {
+ "normalized_reference": "a cry of joy burst from philip's lips.",
+ "normalized_hypothesis": "a cry of joy burst from philip's lips",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "a cry of joy burst from philip is lips",
+ "normalized_hypothesis": "a cry of joy burst from philip is lips",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0252",
+ "category": "phonetic_balance",
+ "text": "And the big Persian knew of his existence before he did of hers.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0252.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "AND THE BIG PERSIAN KNEW OF HIS EXISTENCE BEFORE HE DID OF HERS",
+ "raw": {
+ "normalized_reference": "and the big persian knew of his existence before he did of hers.",
+ "normalized_hypothesis": "and the big persian knew of his existence before he did of hers",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "and the big persian knew of his existence before he did of hers",
+ "normalized_hypothesis": "and the big persian knew of his existence before he did of hers",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0253",
+ "category": "phonetic_balance",
+ "text": "The men stared into each other's face.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0253.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE MEN STARED INTO EACH OTHER'S FACE",
+ "raw": {
+ "normalized_reference": "the men stared into each other's face.",
+ "normalized_hypothesis": "the men stared into each other's face",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the men stared into each other is face",
+ "normalized_hypothesis": "the men stared into each other is face",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0254",
+ "category": "phonetic_balance",
+ "text": "It is the fire, partly, she said.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0254.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT IS THE FIRE PARTLY SHE SAID",
+ "raw": {
+ "normalized_reference": "it is the fire, partly, she said.",
+ "normalized_hypothesis": "it is the fire partly she said",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it is the fire partly she said",
+ "normalized_hypothesis": "it is the fire partly she said",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0255",
+ "category": "phonetic_balance",
+ "text": "It seemed nearer to him since he had seen and talked with Gregson.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0255.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT SEEMED NEARER TO HIM SINCE HE HAD SEEN AND TALKED WITH GREGSON",
+ "raw": {
+ "normalized_reference": "it seemed nearer to him since he had seen and talked with gregson.",
+ "normalized_hypothesis": "it seemed nearer to him since he had seen and talked with gregson",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it seemed nearer to him since he had seen and talked with gregson",
+ "normalized_hypothesis": "it seemed nearer to him since he had seen and talked with gregson",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0256",
+ "category": "phonetic_balance",
+ "text": "Among my minor afflictions, I may mention a new and mysterious one.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0256.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "AMONG MY MINOR AFFLICTIONS I MAY MENTION A NEW AND MYSTERIOUS ONE",
+ "raw": {
+ "normalized_reference": "among my minor afflictions, i may mention a new and mysterious one.",
+ "normalized_hypothesis": "among my minor afflictions i may mention a new and mysterious one",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "among my minor afflictions i may mention a new and mysterious one",
+ "normalized_hypothesis": "among my minor afflictions i may mention a new and mysterious one",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0257",
+ "category": "phonetic_balance",
+ "text": "Between him and all domestic animals there must be no hostilities.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0257.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "BETWEEN HIM AND ALL DOMESTIC ANIMALS THERE MUST BE NO HOSTILITIES",
+ "raw": {
+ "normalized_reference": "between him and all domestic animals there must be no hostilities.",
+ "normalized_hypothesis": "between him and all domestic animals there must be no hostilities",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "between him and all domestic animals there must be no hostilities",
+ "normalized_hypothesis": "between him and all domestic animals there must be no hostilities",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0258",
+ "category": "phonetic_balance",
+ "text": "With these arguments he convinced himself that he should go on alone.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0258.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WITH THESE ARGUMENTS HE CONVINCED HIMSELF THAT HE SHOULD GO ON ALONE",
+ "raw": {
+ "normalized_reference": "with these arguments he convinced himself that he should go on alone.",
+ "normalized_hypothesis": "with these arguments he convinced himself that he should go on alone",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "with these arguments he convinced himself that he should go on alone",
+ "normalized_hypothesis": "with these arguments he convinced himself that he should go on alone",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0259",
+ "category": "phonetic_balance",
+ "text": "We are consumed in our own flesh-pots.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0259.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WE ARE CONSUMED IN OUR OWN FLESH POTS",
+ "raw": {
+ "normalized_reference": "we are consumed in our own flesh-pots.",
+ "normalized_hypothesis": "we are consumed in our own flesh pots",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "we are consumed in our own flesh pots",
+ "normalized_hypothesis": "we are consumed in our own flesh pots",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0260",
+ "category": "phonetic_balance",
+ "text": "Mercedes screamed, cried, laughed, and manifested the chaotic abandonment of hysteria.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0260.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "MERCEDES SCREAMED CRIED LAUGHED AND MANIFESTED THE CHAOTIC ABANDONMENT OF HYSTERIA",
+ "raw": {
+ "normalized_reference": "mercedes screamed, cried, laughed, and manifested the chaotic abandonment of hysteria.",
+ "normalized_hypothesis": "mercedes screamed cried laughed and manifested the chaotic abandonment of hysteria",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "mercedes screamed cried laughed and manifested the chaotic abandonment of hysteria",
+ "normalized_hypothesis": "mercedes screamed cried laughed and manifested the chaotic abandonment of hysteria",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0261",
+ "category": "phonetic_balance",
+ "text": "Philip made no effort to follow.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0261.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "PHILIP MADE NO EFFORT TO FOLLOW",
+ "raw": {
+ "normalized_reference": "philip made no effort to follow.",
+ "normalized_hypothesis": "philip made no effort to follow",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "philip made no effort to follow",
+ "normalized_hypothesis": "philip made no effort to follow",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0262",
+ "category": "phonetic_balance",
+ "text": "She said with chattering teeth.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0262.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SHE SAID WITH CHATTERING TEETH",
+ "raw": {
+ "normalized_reference": "she said with chattering teeth.",
+ "normalized_hypothesis": "she said with chattering teeth",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "she said with chattering teeth",
+ "normalized_hypothesis": "she said with chattering teeth",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0263",
+ "category": "phonetic_balance",
+ "text": "Halfway around the track one donkey got into an argument with its rider.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0263.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HALF WAY AROUND THE TRACK ONE DONKEY GOT INTO AN ARGUMENT WITH ITS RIDER",
+ "raw": {
+ "normalized_reference": "halfway around the track one donkey got into an argument with its rider.",
+ "normalized_hypothesis": "half way around the track one donkey got into an argument with its rider",
+ "errors": 2,
+ "words": 13,
+ "wer": 0.15384615384615385
+ },
+ "semantic": {
+ "normalized_reference": "halfway around the track one donkey got into an argument with its rider",
+ "normalized_hypothesis": "half way around the track one donkey got into an argument with its rider",
+ "errors": 2,
+ "words": 13,
+ "wer": 0.15384615384615385
+ }
+ },
+ {
+ "id": "release500_0264",
+ "category": "phonetic_balance",
+ "text": "Earth and gravel seemed to fill the pan.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0264.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "EARTH AND GRAVEL SEEMED TO FILL THE PAN",
+ "raw": {
+ "normalized_reference": "earth and gravel seemed to fill the pan.",
+ "normalized_hypothesis": "earth and gravel seemed to fill the pan",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "earth and gravel seemed to fill the pan",
+ "normalized_hypothesis": "earth and gravel seemed to fill the pan",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0265",
+ "category": "phonetic_balance",
+ "text": "Her achievements with cocoanuts were a revelation.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0265.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HER ACHIEVEMENTS WITH COCOA NUTS WERE A REVELATION",
+ "raw": {
+ "normalized_reference": "her achievements with cocoanuts were a revelation.",
+ "normalized_hypothesis": "her achievements with cocoa nuts were a revelation",
+ "errors": 2,
+ "words": 7,
+ "wer": 0.2857142857142857
+ },
+ "semantic": {
+ "normalized_reference": "her achievements with cocoanuts were a revelation",
+ "normalized_hypothesis": "her achievements with cocoa nuts were a revelation",
+ "errors": 2,
+ "words": 7,
+ "wer": 0.2857142857142857
+ }
+ },
+ {
+ "id": "release500_0266",
+ "category": "phonetic_balance",
+ "text": "For a full minute he crouched and listened.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0266.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "FOR A FULL MINUTE HE CROUCHED AND LISTENED",
+ "raw": {
+ "normalized_reference": "for a full minute he crouched and listened.",
+ "normalized_hypothesis": "for a full minute he crouched and listened",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "for a full minute he crouched and listened",
+ "normalized_hypothesis": "for a full minute he crouched and listened",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0267",
+ "category": "phonetic_balance",
+ "text": "It was edged with ice.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0267.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT WAS EDGED WITH ICE",
+ "raw": {
+ "normalized_reference": "it was edged with ice.",
+ "normalized_hypothesis": "it was edged with ice",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it was edged with ice",
+ "normalized_hypothesis": "it was edged with ice",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0268",
+ "category": "phonetic_balance",
+ "text": "The Portuguese boy passed the Hawaiian.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0268.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE PORTUGUESE BOY PASSED THE HAWAIAN",
+ "raw": {
+ "normalized_reference": "the portuguese boy passed the hawaiian.",
+ "normalized_hypothesis": "the portuguese boy passed the hawaian",
+ "errors": 1,
+ "words": 6,
+ "wer": 0.16666666666666666
+ },
+ "semantic": {
+ "normalized_reference": "the portuguese boy passed the hawaiian",
+ "normalized_hypothesis": "the portuguese boy passed the hawaian",
+ "errors": 1,
+ "words": 6,
+ "wer": 0.16666666666666666
+ }
+ },
+ {
+ "id": "release500_0269",
+ "category": "phonetic_balance",
+ "text": "But he did not broach it, preferring to mature it carefully.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0269.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "BUT HE DID NOT BROACH IT PREFERRING TO MATURE IT CAREFULLY",
+ "raw": {
+ "normalized_reference": "but he did not broach it, preferring to mature it carefully.",
+ "normalized_hypothesis": "but he did not broach it preferring to mature it carefully",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "but he did not broach it preferring to mature it carefully",
+ "normalized_hypothesis": "but he did not broach it preferring to mature it carefully",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0270",
+ "category": "phonetic_balance",
+ "text": "Come on, Del Mar challenged.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0270.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "COME ON DELMAR CHALLENGED",
+ "raw": {
+ "normalized_reference": "come on, del mar challenged.",
+ "normalized_hypothesis": "come on delmar challenged",
+ "errors": 2,
+ "words": 5,
+ "wer": 0.4
+ },
+ "semantic": {
+ "normalized_reference": "come on del mar challenged",
+ "normalized_hypothesis": "come on delmar challenged",
+ "errors": 2,
+ "words": 5,
+ "wer": 0.4
+ }
+ },
+ {
+ "id": "release500_0271",
+ "category": "phonetic_balance",
+ "text": "For a full half minute Jeanne looked at him without speaking.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0271.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "FOR A FULL HALF MINUTE JEAN LOOKED AT HIM WITHOUT SPEAKING",
+ "raw": {
+ "normalized_reference": "for a full half minute jeanne looked at him without speaking.",
+ "normalized_hypothesis": "for a full half minute jean looked at him without speaking",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ },
+ "semantic": {
+ "normalized_reference": "for a full half minute jeanne looked at him without speaking",
+ "normalized_hypothesis": "for a full half minute jean looked at him without speaking",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ }
+ },
+ {
+ "id": "release500_0272",
+ "category": "phonetic_balance",
+ "text": "He bore no grudges and had few enemies.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0272.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE BORE NO GRUDGES AND HAD FEW ENEMIES",
+ "raw": {
+ "normalized_reference": "he bore no grudges and had few enemies.",
+ "normalized_hypothesis": "he bore no grudges and had few enemies",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he bore no grudges and had few enemies",
+ "normalized_hypothesis": "he bore no grudges and had few enemies",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0273",
+ "category": "phonetic_balance",
+ "text": "I'll go over tomorrow afternoon.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0273.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I'LL GO OVER TO MORROW AFTERNOON",
+ "raw": {
+ "normalized_reference": "i'll go over tomorrow afternoon.",
+ "normalized_hypothesis": "i'll go over to morrow afternoon",
+ "errors": 2,
+ "words": 5,
+ "wer": 0.4
+ },
+ "semantic": {
+ "normalized_reference": "i will go over tomorrow afternoon",
+ "normalized_hypothesis": "i will go over to morrow afternoon",
+ "errors": 2,
+ "words": 6,
+ "wer": 0.3333333333333333
+ }
+ },
+ {
+ "id": "release500_0274",
+ "category": "phonetic_balance",
+ "text": "Well, there are better men in Hawaii, that's all.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0274.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WELL THERE ARE BETTER MEN IN HAWAI'I THAT'S ALL",
+ "raw": {
+ "normalized_reference": "well, there are better men in hawaii, that's all.",
+ "normalized_hypothesis": "well there are better men in hawai'i that's all",
+ "errors": 1,
+ "words": 9,
+ "wer": 0.1111111111111111
+ },
+ "semantic": {
+ "normalized_reference": "well there are better men in hawaii that is all",
+ "normalized_hypothesis": "well there are better men in hawai i that is all",
+ "errors": 2,
+ "words": 10,
+ "wer": 0.2
+ }
+ },
+ {
+ "id": "release500_0275",
+ "category": "phonetic_balance",
+ "text": "He was an athlete and a giant.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0275.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE WAS AN ATHLETE AND A GIANT",
+ "raw": {
+ "normalized_reference": "he was an athlete and a giant.",
+ "normalized_hypothesis": "he was an athlete and a giant",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he was an athlete and a giant",
+ "normalized_hypothesis": "he was an athlete and a giant",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0276",
+ "category": "phonetic_balance",
+ "text": "McVeigh when he returns from a trip to Honolulu.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0276.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "MAC VAY WHEN HE RETURNS FROM A TRIP TO HONOLULU",
+ "raw": {
+ "normalized_reference": "mcveigh when he returns from a trip to honolulu.",
+ "normalized_hypothesis": "mac vay when he returns from a trip to honolulu",
+ "errors": 2,
+ "words": 9,
+ "wer": 0.2222222222222222
+ },
+ "semantic": {
+ "normalized_reference": "mcveigh when he returns from a trip to honolulu",
+ "normalized_hypothesis": "mac vay when he returns from a trip to honolulu",
+ "errors": 2,
+ "words": 9,
+ "wer": 0.2222222222222222
+ }
+ },
+ {
+ "id": "release500_0277",
+ "category": "phonetic_balance",
+ "text": "You're a devil for fighting, and will surely win.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0277.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "YOU'RE A DEVIL FOR FIGHTING AND WILL SURELY WIN",
+ "raw": {
+ "normalized_reference": "you're a devil for fighting, and will surely win.",
+ "normalized_hypothesis": "you're a devil for fighting and will surely win",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "you are a devil for fighting and will surely win",
+ "normalized_hypothesis": "you are a devil for fighting and will surely win",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0278",
+ "category": "phonetic_balance",
+ "text": "All an appearance can know is mirage.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0278.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "ALL AN APPEARANCE CAN KNOW IS MIRAGE",
+ "raw": {
+ "normalized_reference": "all an appearance can know is mirage.",
+ "normalized_hypothesis": "all an appearance can know is mirage",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "all an appearance can know is mirage",
+ "normalized_hypothesis": "all an appearance can know is mirage",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0279",
+ "category": "phonetic_balance",
+ "text": "You fired me out of your house, in short.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0279.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "YOU FIRED ME OUT OF YOUR HOUSE IN SHORT",
+ "raw": {
+ "normalized_reference": "you fired me out of your house, in short.",
+ "normalized_hypothesis": "you fired me out of your house in short",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "you fired me out of your house in short",
+ "normalized_hypothesis": "you fired me out of your house in short",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0280",
+ "category": "phonetic_balance",
+ "text": "He made sure that the magazine was loaded, and resumed his paddling.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0280.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE MADE SURE THAT THE MAGAZINE WAS LOADED AND RESUMED HIS PADDLING",
+ "raw": {
+ "normalized_reference": "he made sure that the magazine was loaded, and resumed his paddling.",
+ "normalized_hypothesis": "he made sure that the magazine was loaded and resumed his paddling",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he made sure that the magazine was loaded and resumed his paddling",
+ "normalized_hypothesis": "he made sure that the magazine was loaded and resumed his paddling",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0281",
+ "category": "phonetic_balance",
+ "text": "Change chairs, Daylight commanded.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0281.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "CHANGE CHAIRS DAYLIGHT COMMANDED",
+ "raw": {
+ "normalized_reference": "change chairs, daylight commanded.",
+ "normalized_hypothesis": "change chairs daylight commanded",
+ "errors": 0,
+ "words": 4,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "change chairs daylight commanded",
+ "normalized_hypothesis": "change chairs daylight commanded",
+ "errors": 0,
+ "words": 4,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0282",
+ "category": "phonetic_balance",
+ "text": "By that answer my professional medical prestige stood or fell.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0282.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "BY THAT ANSWER MY PROFESSIONAL MEDICAL PRESTIGE STOOD OR FELL",
+ "raw": {
+ "normalized_reference": "by that answer my professional medical prestige stood or fell.",
+ "normalized_hypothesis": "by that answer my professional medical prestige stood or fell",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "by that answer my professional medical prestige stood or fell",
+ "normalized_hypothesis": "by that answer my professional medical prestige stood or fell",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0283",
+ "category": "phonetic_balance",
+ "text": "The moon had already begun its westward decline.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0283.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE MOON HAD ALREADY BEGUN ITS WESTWARD DECLINE",
+ "raw": {
+ "normalized_reference": "the moon had already begun its westward decline.",
+ "normalized_hypothesis": "the moon had already begun its westward decline",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the moon had already begun its westward decline",
+ "normalized_hypothesis": "the moon had already begun its westward decline",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0284",
+ "category": "phonetic_balance",
+ "text": "At sea, Monday, March 16, 1908.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0284.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "AT S MONDAY MARCH SIXTEENTH NINETEEN EIG",
+ "raw": {
+ "normalized_reference": "at sea, monday, march 16, 1908.",
+ "normalized_hypothesis": "at s monday march sixteenth nineteen eig",
+ "errors": 4,
+ "words": 6,
+ "wer": 0.6666666666666666
+ },
+ "semantic": {
+ "normalized_reference": "at sea monday march 16 1908",
+ "normalized_hypothesis": "at s monday march 16th 19 eig",
+ "errors": 4,
+ "words": 6,
+ "wer": 0.6666666666666666
+ }
+ },
+ {
+ "id": "release500_0285",
+ "category": "phonetic_balance",
+ "text": "And wherever I ranged, the way lay along alcohol-drenched roads.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0285.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "AND WHEREVER I RANGED THE WAY LAY ALONG ALCOHOL DRENCHED ROADS",
+ "raw": {
+ "normalized_reference": "and wherever i ranged, the way lay along alcohol-drenched roads.",
+ "normalized_hypothesis": "and wherever i ranged the way lay along alcohol drenched roads",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "and wherever i ranged the way lay along alcohol drenched roads",
+ "normalized_hypothesis": "and wherever i ranged the way lay along alcohol drenched roads",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0286",
+ "category": "phonetic_balance",
+ "text": "The date was nearly eighteen years old.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0286.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE DATE WAS NEARLY EIGHTEEN YEARS OLD",
+ "raw": {
+ "normalized_reference": "the date was nearly eighteen years old.",
+ "normalized_hypothesis": "the date was nearly eighteen years old",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the date was nearly 18 years old",
+ "normalized_hypothesis": "the date was nearly 18 years old",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0287",
+ "category": "phonetic_balance",
+ "text": "Very early in my life, I separated from my mother.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0287.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "VERY EARLY IN MY LIFE I SEPARATED FROM MY MOTHER",
+ "raw": {
+ "normalized_reference": "very early in my life, i separated from my mother.",
+ "normalized_hypothesis": "very early in my life i separated from my mother",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "very early in my life i separated from my mother",
+ "normalized_hypothesis": "very early in my life i separated from my mother",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0288",
+ "category": "phonetic_balance",
+ "text": "There was pride and strength, the ring of triumph in his voice.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0288.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THERE WAS PRIDE AN STRENGTH THE RING OF TRIUMPH IN HIS VOICE",
+ "raw": {
+ "normalized_reference": "there was pride and strength, the ring of triumph in his voice.",
+ "normalized_hypothesis": "there was pride an strength the ring of triumph in his voice",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "there was pride and strength the ring of triumph in his voice",
+ "normalized_hypothesis": "there was pride an strength the ring of triumph in his voice",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ }
+ },
+ {
+ "id": "release500_0289",
+ "category": "phonetic_balance",
+ "text": "From my earliest recollection my sleep was a period of terror.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0289.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "FROM MY EARLIEST RECOLLECTION MY SLEEP WAS A PERIOD OF TERROR",
+ "raw": {
+ "normalized_reference": "from my earliest recollection my sleep was a period of terror.",
+ "normalized_hypothesis": "from my earliest recollection my sleep was a period of terror",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "from my earliest recollection my sleep was a period of terror",
+ "normalized_hypothesis": "from my earliest recollection my sleep was a period of terror",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0290",
+ "category": "phonetic_balance",
+ "text": "We'll have to watch our chances.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0290.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WE'LL HAVE TO WATCH OUR CHANCES",
+ "raw": {
+ "normalized_reference": "we'll have to watch our chances.",
+ "normalized_hypothesis": "we'll have to watch our chances",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "we will have to watch our chances",
+ "normalized_hypothesis": "we will have to watch our chances",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0291",
+ "category": "phonetic_balance",
+ "text": "Close beside him gleamed the white fangs of the wolf-dog.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0291.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "CLOSE BESIDE HIM GLEAMED THE WHITE FANGS OF THE WOLF DOG",
+ "raw": {
+ "normalized_reference": "close beside him gleamed the white fangs of the wolf-dog.",
+ "normalized_hypothesis": "close beside him gleamed the white fangs of the wolf dog",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "close beside him gleamed the white fangs of the wolf dog",
+ "normalized_hypothesis": "close beside him gleamed the white fangs of the wolf dog",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0292",
+ "category": "phonetic_balance",
+ "text": "Would you be satisfied with that one hundredth part of me.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0292.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WOULD YOU BE SATISFIED WITH THAT ONE HUNDREDTH PART OF ME",
+ "raw": {
+ "normalized_reference": "would you be satisfied with that one hundredth part of me.",
+ "normalized_hypothesis": "would you be satisfied with that one hundredth part of me",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "would you be satisfied with that 100th part of me",
+ "normalized_hypothesis": "would you be satisfied with that 100th part of me",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0293",
+ "category": "phonetic_balance",
+ "text": "Two gigantic owls were tearing at the carcass.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0293.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "TWO GIGANTIC OWLS WERE TEARING AT THE CARCASE",
+ "raw": {
+ "normalized_reference": "two gigantic owls were tearing at the carcass.",
+ "normalized_hypothesis": "two gigantic owls were tearing at the carcase",
+ "errors": 1,
+ "words": 8,
+ "wer": 0.125
+ },
+ "semantic": {
+ "normalized_reference": "2 gigantic owls were tearing at the carcass",
+ "normalized_hypothesis": "2 gigantic owls were tearing at the carcase",
+ "errors": 1,
+ "words": 8,
+ "wer": 0.125
+ }
+ },
+ {
+ "id": "release500_0294",
+ "category": "phonetic_balance",
+ "text": "The lush vegetation of that sheltered spot make a natural shield.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0294.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE LUSH VEGETATION OF THAT SHELTERED SPOT MAKE A NATURAL SHIELD",
+ "raw": {
+ "normalized_reference": "the lush vegetation of that sheltered spot make a natural shield.",
+ "normalized_hypothesis": "the lush vegetation of that sheltered spot make a natural shield",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the lush vegetation of that sheltered spot make a natural shield",
+ "normalized_hypothesis": "the lush vegetation of that sheltered spot make a natural shield",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0295",
+ "category": "phonetic_balance",
+ "text": "Below him the shadow was broken into a pool of rippling starlight.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0295.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "BELOW HIM THE SHADOW WAS BROKEN INTO A POOL OF RIPPLING STARLIGHT",
+ "raw": {
+ "normalized_reference": "below him the shadow was broken into a pool of rippling starlight.",
+ "normalized_hypothesis": "below him the shadow was broken into a pool of rippling starlight",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "below him the shadow was broken into a pool of rippling starlight",
+ "normalized_hypothesis": "below him the shadow was broken into a pool of rippling starlight",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0296",
+ "category": "phonetic_balance",
+ "text": "Lots of men take women buggy riding.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0296.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "LOTS OF MEN TAKE WOMEN BUGGY RIDING",
+ "raw": {
+ "normalized_reference": "lots of men take women buggy riding.",
+ "normalized_hypothesis": "lots of men take women buggy riding",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "lots of men take women buggy riding",
+ "normalized_hypothesis": "lots of men take women buggy riding",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0297",
+ "category": "phonetic_balance",
+ "text": "But it contributed to the smash.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0297.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "BUT IT CONTRIBUTED TO THE SMASH",
+ "raw": {
+ "normalized_reference": "but it contributed to the smash.",
+ "normalized_hypothesis": "but it contributed to the smash",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "but it contributed to the smash",
+ "normalized_hypothesis": "but it contributed to the smash",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0298",
+ "category": "phonetic_balance",
+ "text": "No man ate of the seal meat or the oil.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0298.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "NO MAN ATE OF THE SEAL MEAT OR THE OIL",
+ "raw": {
+ "normalized_reference": "no man ate of the seal meat or the oil.",
+ "normalized_hypothesis": "no man ate of the seal meat or the oil",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "no man ate of the seal meat or the oil",
+ "normalized_hypothesis": "no man ate of the seal meat or the oil",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0299",
+ "category": "phonetic_balance",
+ "text": "You're going in for grab sharing.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0299.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "YOU'RE GOING IN FOR GRAB SHARING",
+ "raw": {
+ "normalized_reference": "you're going in for grab sharing.",
+ "normalized_hypothesis": "you're going in for grab sharing",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "you are going in for grab sharing",
+ "normalized_hypothesis": "you are going in for grab sharing",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0300",
+ "category": "phonetic_balance",
+ "text": "He it was that lived to found the family of the Patino.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0300.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE IT WAS THAT LIVED TO FOUND THE FAMILY OF THE PETINO",
+ "raw": {
+ "normalized_reference": "he it was that lived to found the family of the patino.",
+ "normalized_hypothesis": "he it was that lived to found the family of the petino",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "he it was that lived to found the family of the patino",
+ "normalized_hypothesis": "he it was that lived to found the family of the petino",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ }
+ },
+ {
+ "id": "release500_0301",
+ "category": "phonetic_balance",
+ "text": "For such countries nothing remained but reorganization.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0301.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "FOR SUCH COUNTRIES NOTHING REMAINED BUT REORGANIZATION",
+ "raw": {
+ "normalized_reference": "for such countries nothing remained but reorganization.",
+ "normalized_hypothesis": "for such countries nothing remained but reorganization",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "for such countries nothing remained but reorganization",
+ "normalized_hypothesis": "for such countries nothing remained but reorganization",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0302",
+ "category": "phonetic_balance",
+ "text": "Stand off butcher and baker and all the rest.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0302.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "STAND OFF BUTCHER AND BAKER AND ALL THE REST",
+ "raw": {
+ "normalized_reference": "stand off butcher and baker and all the rest.",
+ "normalized_hypothesis": "stand off butcher and baker and all the rest",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "stand off butcher and baker and all the rest",
+ "normalized_hypothesis": "stand off butcher and baker and all the rest",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0303",
+ "category": "phonetic_balance",
+ "text": "All eyes, however, were staring at him in certitude of expectancy.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0303.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "ALL EYES HOWEVER WERE STARING AT HIM IN CERTITUDE OF EXPECTANCY",
+ "raw": {
+ "normalized_reference": "all eyes, however, were staring at him in certitude of expectancy.",
+ "normalized_hypothesis": "all eyes however were staring at him in certitude of expectancy",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "all eyes however were staring at him in certitude of expectancy",
+ "normalized_hypothesis": "all eyes however were staring at him in certitude of expectancy",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0304",
+ "category": "phonetic_balance",
+ "text": "Two days ago Jeanne learned where her father's men were hiding.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0304.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "TWO DAYS AGO JEAN LEARNED WHERE HER FATHER'S MEN WERE HIDING",
+ "raw": {
+ "normalized_reference": "two days ago jeanne learned where her father's men were hiding.",
+ "normalized_hypothesis": "two days ago jean learned where her father's men were hiding",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ },
+ "semantic": {
+ "normalized_reference": "2 days ago jeanne learned where her father is men were hiding",
+ "normalized_hypothesis": "2 days ago jean learned where her father is men were hiding",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ }
+ },
+ {
+ "id": "release500_0305",
+ "category": "phonetic_balance",
+ "text": "His previous wives had never lived long enough to bear him children.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0305.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HIS PREVIOUS WIVES HAD NEVER LIVED LONG ENOUGH TO BEAR HIM CHILDREN",
+ "raw": {
+ "normalized_reference": "his previous wives had never lived long enough to bear him children.",
+ "normalized_hypothesis": "his previous wives had never lived long enough to bear him children",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "his previous wives had never lived long enough to bear him children",
+ "normalized_hypothesis": "his previous wives had never lived long enough to bear him children",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0306",
+ "category": "phonetic_balance",
+ "text": "But this little defect did not worry him.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0306.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "BUT THIS LITTLE DEFECT DID NOT WORRY HIM",
+ "raw": {
+ "normalized_reference": "but this little defect did not worry him.",
+ "normalized_hypothesis": "but this little defect did not worry him",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "but this little defect did not worry him",
+ "normalized_hypothesis": "but this little defect did not worry him",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0307",
+ "category": "phonetic_balance",
+ "text": "Eli Harding asked, as Shunk started to follow.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0307.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "ILY HARDING ASKED AS SHUNK STARTED TO FOLLOW",
+ "raw": {
+ "normalized_reference": "eli harding asked, as shunk started to follow.",
+ "normalized_hypothesis": "ily harding asked as shunk started to follow",
+ "errors": 1,
+ "words": 8,
+ "wer": 0.125
+ },
+ "semantic": {
+ "normalized_reference": "eli harding asked as shunk started to follow",
+ "normalized_hypothesis": "ily harding asked as shunk started to follow",
+ "errors": 1,
+ "words": 8,
+ "wer": 0.125
+ }
+ },
+ {
+ "id": "release500_0308",
+ "category": "phonetic_balance",
+ "text": "And this was their sole conversation throughout the meal.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0308.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "AND THIS WAS THEIR SOLE CONVERSATION THROUGHOUT THE MEAL",
+ "raw": {
+ "normalized_reference": "and this was their sole conversation throughout the meal.",
+ "normalized_hypothesis": "and this was their sole conversation throughout the meal",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "and this was their sole conversation throughout the meal",
+ "normalized_hypothesis": "and this was their sole conversation throughout the meal",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0309",
+ "category": "phonetic_balance",
+ "text": "Then, again, Tudor had such an irritating way about him.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0309.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THEN AGAIN TUDOR HAD SUCH AN IRRITATING WAY ABOUT HIM",
+ "raw": {
+ "normalized_reference": "then, again, tudor had such an irritating way about him.",
+ "normalized_hypothesis": "then again tudor had such an irritating way about him",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "then again tudor had such an irritating way about him",
+ "normalized_hypothesis": "then again tudor had such an irritating way about him",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0310",
+ "category": "phonetic_balance",
+ "text": "In it there was something that was almost tragedy.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0310.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IN IT THERE WAS SOMETHING THAT WAS ALMOST TRAGEDY",
+ "raw": {
+ "normalized_reference": "in it there was something that was almost tragedy.",
+ "normalized_hypothesis": "in it there was something that was almost tragedy",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "in it there was something that was almost tragedy",
+ "normalized_hypothesis": "in it there was something that was almost tragedy",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0311",
+ "category": "phonetic_balance",
+ "text": "Thirty pounds, said the captain with finality.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0311.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THIRTY POUNDS SAID THE CAPTAIN WITH FINALITY",
+ "raw": {
+ "normalized_reference": "thirty pounds, said the captain with finality.",
+ "normalized_hypothesis": "thirty pounds said the captain with finality",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "\u00a330 said the captain with finality",
+ "normalized_hypothesis": "\u00a330 said the captain with finality",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0312",
+ "category": "phonetic_balance",
+ "text": "It was over when he made his way through the ring of spectators.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0312.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT WAS OVER WHEN HE MADE HIS WAY THROUGH THE RING OF SPECTATORS",
+ "raw": {
+ "normalized_reference": "it was over when he made his way through the ring of spectators.",
+ "normalized_hypothesis": "it was over when he made his way through the ring of spectators",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it was over when he made his way through the ring of spectators",
+ "normalized_hypothesis": "it was over when he made his way through the ring of spectators",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0313",
+ "category": "phonetic_balance",
+ "text": "She had died from cold and starvation.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0313.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SHE HAD DIED FROM COLD AND STARVATION",
+ "raw": {
+ "normalized_reference": "she had died from cold and starvation.",
+ "normalized_hypothesis": "she had died from cold and starvation",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "she had died from cold and starvation",
+ "normalized_hypothesis": "she had died from cold and starvation",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0314",
+ "category": "phonetic_balance",
+ "text": "We fished sharks on Niihau together.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0314.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WE FISHED SHARKS ON NEHOW TOGETHER",
+ "raw": {
+ "normalized_reference": "we fished sharks on niihau together.",
+ "normalized_hypothesis": "we fished sharks on nehow together",
+ "errors": 1,
+ "words": 6,
+ "wer": 0.16666666666666666
+ },
+ "semantic": {
+ "normalized_reference": "we fished sharks on niihau together",
+ "normalized_hypothesis": "we fished sharks on nehow together",
+ "errors": 1,
+ "words": 6,
+ "wer": 0.16666666666666666
+ }
+ },
+ {
+ "id": "release500_0315",
+ "category": "phonetic_balance",
+ "text": "His slim hands gripped the edges of the table.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0315.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HIS SLIM HANDS GRIPPED THE EDGES OF THE TABLE",
+ "raw": {
+ "normalized_reference": "his slim hands gripped the edges of the table.",
+ "normalized_hypothesis": "his slim hands gripped the edges of the table",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "his slim hands gripped the edges of the table",
+ "normalized_hypothesis": "his slim hands gripped the edges of the table",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0316",
+ "category": "phonetic_balance",
+ "text": "We had been chased by them ourselves, more than once.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0316.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WE HAD BEEN CHASED BY THEM OURSELVES MORE THAN ONCE",
+ "raw": {
+ "normalized_reference": "we had been chased by them ourselves, more than once.",
+ "normalized_hypothesis": "we had been chased by them ourselves more than once",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "we had been chased by them ourselves more than once",
+ "normalized_hypothesis": "we had been chased by them ourselves more than once",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0317",
+ "category": "phonetic_balance",
+ "text": "Or have they already devised one.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0317.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "OR HAVE THEY ALREADY DEVISED ONE",
+ "raw": {
+ "normalized_reference": "or have they already devised one.",
+ "normalized_hypothesis": "or have they already devised one",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "or have they already devised one",
+ "normalized_hypothesis": "or have they already devised one",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0318",
+ "category": "phonetic_balance",
+ "text": "I cannot follow you, she said.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0318.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I CANNOT FOLLOW YOU SHE SAID",
+ "raw": {
+ "normalized_reference": "i cannot follow you, she said.",
+ "normalized_hypothesis": "i cannot follow you she said",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i cannot follow you she said",
+ "normalized_hypothesis": "i cannot follow you she said",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0319",
+ "category": "phonetic_balance",
+ "text": "He went down in midstream, searching the shadows of both shores.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0319.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE WENT DOWN IN MIDSTREAM SEARCHING THE SHADOWS OF BOTH SHORES",
+ "raw": {
+ "normalized_reference": "he went down in midstream, searching the shadows of both shores.",
+ "normalized_hypothesis": "he went down in midstream searching the shadows of both shores",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he went down in midstream searching the shadows of both shores",
+ "normalized_hypothesis": "he went down in midstream searching the shadows of both shores",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0320",
+ "category": "phonetic_balance",
+ "text": "I was sick once -- typhoid.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0320.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I WAS SICK ONCE TYPHOID",
+ "raw": {
+ "normalized_reference": "i was sick once -- typhoid.",
+ "normalized_hypothesis": "i was sick once typhoid",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i was sick once typhoid",
+ "normalized_hypothesis": "i was sick once typhoid",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0321",
+ "category": "phonetic_balance",
+ "text": "Why not like any railroad station or ferry depot.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0321.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WHY NOT LIKE ANY RAILROAD STATION OR FERRY DEPOT",
+ "raw": {
+ "normalized_reference": "why not like any railroad station or ferry depot.",
+ "normalized_hypothesis": "why not like any railroad station or ferry depot",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "why not like any railroad station or ferry depot",
+ "normalized_hypothesis": "why not like any railroad station or ferry depot",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0322",
+ "category": "phonetic_balance",
+ "text": "Red-Eye swung back and forth on the branch farther down.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0322.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "RED EYES SWUNG BACK AND FORTH ON THE BRANCH FARTHER DOWN",
+ "raw": {
+ "normalized_reference": "red-eye swung back and forth on the branch farther down.",
+ "normalized_hypothesis": "red eyes swung back and forth on the branch farther down",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ },
+ "semantic": {
+ "normalized_reference": "red eye swung back and forth on the branch farther down",
+ "normalized_hypothesis": "red eyes swung back and forth on the branch farther down",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ }
+ },
+ {
+ "id": "release500_0323",
+ "category": "phonetic_balance",
+ "text": "Within himself he called it no longer his own.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0323.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WITHIN HIMSELF HE CALLED IT NO LONGER HIS OWN",
+ "raw": {
+ "normalized_reference": "within himself he called it no longer his own.",
+ "normalized_hypothesis": "within himself he called it no longer his own",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "within himself he called it no longer his own",
+ "normalized_hypothesis": "within himself he called it no longer his own",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0324",
+ "category": "phonetic_balance",
+ "text": "It lasted as a deterrent for two days.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0324.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT LASTED AS A DETERRANT FOR TWO DAYS",
+ "raw": {
+ "normalized_reference": "it lasted as a deterrent for two days.",
+ "normalized_hypothesis": "it lasted as a deterrant for two days",
+ "errors": 1,
+ "words": 8,
+ "wer": 0.125
+ },
+ "semantic": {
+ "normalized_reference": "it lasted as a deterrent for 2 days",
+ "normalized_hypothesis": "it lasted as a deterrant for 2 days",
+ "errors": 1,
+ "words": 8,
+ "wer": 0.125
+ }
+ },
+ {
+ "id": "release500_0325",
+ "category": "phonetic_balance",
+ "text": "He loved to play Chinese lottery.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0325.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE LOVED TO PLAY CHINESE LOTTERY",
+ "raw": {
+ "normalized_reference": "he loved to play chinese lottery.",
+ "normalized_hypothesis": "he loved to play chinese lottery",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he loved to play chinese lottery",
+ "normalized_hypothesis": "he loved to play chinese lottery",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0326",
+ "category": "phonetic_balance",
+ "text": "Lord, but I'm glad to see you again, Phil.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0326.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "LORD BUT I'M GLAD TO SEE YOU AGAIN PHIL",
+ "raw": {
+ "normalized_reference": "lord, but i'm glad to see you again, phil.",
+ "normalized_hypothesis": "lord but i'm glad to see you again phil",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "lord but i am glad to see you again phil",
+ "normalized_hypothesis": "lord but i am glad to see you again phil",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0327",
+ "category": "phonetic_balance",
+ "text": "There is not an iota of truth in it, certainly not.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0327.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THERE IS NOT AN IOTA OF TRUTH IN IT CERTAINLY NOT",
+ "raw": {
+ "normalized_reference": "there is not an iota of truth in it, certainly not.",
+ "normalized_hypothesis": "there is not an iota of truth in it certainly not",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "there is not an iota of truth in it certainly not",
+ "normalized_hypothesis": "there is not an iota of truth in it certainly not",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0328",
+ "category": "phonetic_balance",
+ "text": "She added, with genuine sympathy in her face and voice.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0328.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SHE ADDED WITH GENUINE SYMPATHY IN HER FACE AND VOICE",
+ "raw": {
+ "normalized_reference": "she added, with genuine sympathy in her face and voice.",
+ "normalized_hypothesis": "she added with genuine sympathy in her face and voice",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "she added with genuine sympathy in her face and voice",
+ "normalized_hypothesis": "she added with genuine sympathy in her face and voice",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0329",
+ "category": "phonetic_balance",
+ "text": "We have plenty of capital ourselves, and yet we want more.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0329.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WE HAVE PLENTY OF CAPITAL OURSELVES AND YET WE WANT MORE",
+ "raw": {
+ "normalized_reference": "we have plenty of capital ourselves, and yet we want more.",
+ "normalized_hypothesis": "we have plenty of capital ourselves and yet we want more",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "we have plenty of capital ourselves and yet we want more",
+ "normalized_hypothesis": "we have plenty of capital ourselves and yet we want more",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0330",
+ "category": "phonetic_balance",
+ "text": "And now put yourself in my place for a moment.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0330.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "AND NOW PUT YOURSELF IN MY PLACE FOR A MOMENT",
+ "raw": {
+ "normalized_reference": "and now put yourself in my place for a moment.",
+ "normalized_hypothesis": "and now put yourself in my place for a moment",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "and now put yourself in my place for a moment",
+ "normalized_hypothesis": "and now put yourself in my place for a moment",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0331",
+ "category": "phonetic_balance",
+ "text": "Fast, but endure.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0331.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "FAST BUT ENDURE",
+ "raw": {
+ "normalized_reference": "fast, but endure.",
+ "normalized_hypothesis": "fast but endure",
+ "errors": 0,
+ "words": 3,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "fast but endure",
+ "normalized_hypothesis": "fast but endure",
+ "errors": 0,
+ "words": 3,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0332",
+ "category": "phonetic_balance",
+ "text": "You were looking squeamish this afternoon, he began.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0332.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "YOU WERE LOOKING SQUEAMISH THIS AFTERNOON HE BEGAN",
+ "raw": {
+ "normalized_reference": "you were looking squeamish this afternoon, he began.",
+ "normalized_hypothesis": "you were looking squeamish this afternoon he began",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "you were looking squeamish this afternoon he began",
+ "normalized_hypothesis": "you were looking squeamish this afternoon he began",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0333",
+ "category": "phonetic_balance",
+ "text": "It is the nearest refuge.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0333.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT IS THE NEAREST REFUGE",
+ "raw": {
+ "normalized_reference": "it is the nearest refuge.",
+ "normalized_hypothesis": "it is the nearest refuge",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it is the nearest refuge",
+ "normalized_hypothesis": "it is the nearest refuge",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0334",
+ "category": "phonetic_balance",
+ "text": "I want to die in it.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0334.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I WANT TO DIE IN IT",
+ "raw": {
+ "normalized_reference": "i want to die in it.",
+ "normalized_hypothesis": "i want to die in it",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i want to die in it",
+ "normalized_hypothesis": "i want to die in it",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0335",
+ "category": "phonetic_balance",
+ "text": "He stopped, and Philip nodded at the horrified question in his eyes.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0335.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE STOPPED AND PHILIP NODDED AT THE HORRIFIED QUESTION IN HIS EYES",
+ "raw": {
+ "normalized_reference": "he stopped, and philip nodded at the horrified question in his eyes.",
+ "normalized_hypothesis": "he stopped and philip nodded at the horrified question in his eyes",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he stopped and philip nodded at the horrified question in his eyes",
+ "normalized_hypothesis": "he stopped and philip nodded at the horrified question in his eyes",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0336",
+ "category": "phonetic_balance",
+ "text": "She turned, fearing that Jacques might see what was in her face.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0336.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SHE TURNED FEARING THAT JACQUES MIGHT SEE WHAT WAS IN HER FACE",
+ "raw": {
+ "normalized_reference": "she turned, fearing that jacques might see what was in her face.",
+ "normalized_hypothesis": "she turned fearing that jacques might see what was in her face",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "she turned fearing that jacques might see what was in her face",
+ "normalized_hypothesis": "she turned fearing that jacques might see what was in her face",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0337",
+ "category": "phonetic_balance",
+ "text": "It was a curious coincidence.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0337.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT WAS A CURIOUS COINCIDENCE",
+ "raw": {
+ "normalized_reference": "it was a curious coincidence.",
+ "normalized_hypothesis": "it was a curious coincidence",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it was a curious coincidence",
+ "normalized_hypothesis": "it was a curious coincidence",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0338",
+ "category": "phonetic_balance",
+ "text": "Daylight made no answer, and the door closed behind him.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0338.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "DAYLIGHT MADE NO ANSWER AND THE DOOR CLOSED BEHIND HIM",
+ "raw": {
+ "normalized_reference": "daylight made no answer, and the door closed behind him.",
+ "normalized_hypothesis": "daylight made no answer and the door closed behind him",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "daylight made no answer and the door closed behind him",
+ "normalized_hypothesis": "daylight made no answer and the door closed behind him",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0339",
+ "category": "phonetic_balance",
+ "text": "And he did hurt my arm.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0339.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "AND HE DID HURT MY ARM",
+ "raw": {
+ "normalized_reference": "and he did hurt my arm.",
+ "normalized_hypothesis": "and he did hurt my arm",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "and he did hurt my arm",
+ "normalized_hypothesis": "and he did hurt my arm",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0340",
+ "category": "phonetic_balance",
+ "text": "Dunham, can your boy go along with Jesse.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0340.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "DUNAM CAN YOUR BOY GO ALONG WITH JESSE",
+ "raw": {
+ "normalized_reference": "dunham, can your boy go along with jesse.",
+ "normalized_hypothesis": "dunam can your boy go along with jesse",
+ "errors": 1,
+ "words": 8,
+ "wer": 0.125
+ },
+ "semantic": {
+ "normalized_reference": "dunham can your boy go along with jesse",
+ "normalized_hypothesis": "dunam can your boy go along with jesse",
+ "errors": 1,
+ "words": 8,
+ "wer": 0.125
+ }
+ },
+ {
+ "id": "release500_0341",
+ "category": "phonetic_balance",
+ "text": "The creative joy, I murmured.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0341.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE CREATIVE JOY I MURMURED",
+ "raw": {
+ "normalized_reference": "the creative joy, i murmured.",
+ "normalized_hypothesis": "the creative joy i murmured",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the creative joy i murmured",
+ "normalized_hypothesis": "the creative joy i murmured",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0342",
+ "category": "phonetic_balance",
+ "text": "He pulled, and the log crashed down to break his back.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0342.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE PULLED AND THE LOG CRASHED DOWN TO BREAK HIS BACK",
+ "raw": {
+ "normalized_reference": "he pulled, and the log crashed down to break his back.",
+ "normalized_hypothesis": "he pulled and the log crashed down to break his back",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he pulled and the log crashed down to break his back",
+ "normalized_hypothesis": "he pulled and the log crashed down to break his back",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0343",
+ "category": "phonetic_balance",
+ "text": "By golly, the boy wins.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0343.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "BY GOLLY THE BOY WINS",
+ "raw": {
+ "normalized_reference": "by golly, the boy wins.",
+ "normalized_hypothesis": "by golly the boy wins",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "by golly the boy wins",
+ "normalized_hypothesis": "by golly the boy wins",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0344",
+ "category": "phonetic_balance",
+ "text": "The boy threw back his head with pride.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0344.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE BOY THREW BACK HIS HEAD WITH PRIDE",
+ "raw": {
+ "normalized_reference": "the boy threw back his head with pride.",
+ "normalized_hypothesis": "the boy threw back his head with pride",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the boy threw back his head with pride",
+ "normalized_hypothesis": "the boy threw back his head with pride",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0345",
+ "category": "phonetic_balance",
+ "text": "The last one I knew was an overseer.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0345.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE LAST ONE I KNEW WAS AN OVERSEER",
+ "raw": {
+ "normalized_reference": "the last one i knew was an overseer.",
+ "normalized_hypothesis": "the last one i knew was an overseer",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the last one i knew was an overseer",
+ "normalized_hypothesis": "the last one i knew was an overseer",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0346",
+ "category": "phonetic_balance",
+ "text": "Two years ago I gave up civilization for this.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0346.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "TWO YEARS AGO I GAVE UP CIVILIZATION FOR THIS",
+ "raw": {
+ "normalized_reference": "two years ago i gave up civilization for this.",
+ "normalized_hypothesis": "two years ago i gave up civilization for this",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "2 years ago i gave up civilization for this",
+ "normalized_hypothesis": "2 years ago i gave up civilization for this",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0347",
+ "category": "phonetic_balance",
+ "text": "The river bared its bosom, and snorting steamboats challenged the wilderness.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0347.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE RIVER BARED ITS BOSOM AND SNORTING STEAMBOATS CHALLENGED THE WILDERNESS",
+ "raw": {
+ "normalized_reference": "the river bared its bosom, and snorting steamboats challenged the wilderness.",
+ "normalized_hypothesis": "the river bared its bosom and snorting steamboats challenged the wilderness",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the river bared its bosom and snorting steamboats challenged the wilderness",
+ "normalized_hypothesis": "the river bared its bosom and snorting steamboats challenged the wilderness",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0348",
+ "category": "phonetic_balance",
+ "text": "Mab, she said.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0348.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "MAB SHE SAID",
+ "raw": {
+ "normalized_reference": "mab, she said.",
+ "normalized_hypothesis": "mab she said",
+ "errors": 0,
+ "words": 3,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "mab she said",
+ "normalized_hypothesis": "mab she said",
+ "errors": 0,
+ "words": 3,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0349",
+ "category": "phonetic_balance",
+ "text": "The regret in his voice was provocative of a second burst of laughter.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0349.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE REGRET IN HIS VOICE WAS PROVOCATIVE OF A SECOND BURST OF LAUGHTER",
+ "raw": {
+ "normalized_reference": "the regret in his voice was provocative of a second burst of laughter.",
+ "normalized_hypothesis": "the regret in his voice was provocative of a second burst of laughter",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the regret in his voice was provocative of a 2nd burst of laughter",
+ "normalized_hypothesis": "the regret in his voice was provocative of a 2nd burst of laughter",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0350",
+ "category": "phonetic_balance",
+ "text": "And each year something happened, and I did not go.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0350.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "AND EACH YEAR SOMETHING HAPPENED AND I DID NOT GO",
+ "raw": {
+ "normalized_reference": "and each year something happened, and i did not go.",
+ "normalized_hypothesis": "and each year something happened and i did not go",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "and each year something happened and i did not go",
+ "normalized_hypothesis": "and each year something happened and i did not go",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0351",
+ "category": "phonetic_balance",
+ "text": "Gad, your letter came just in time.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0351.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "GAD YOUR LETTER CAME JUST IN TIME",
+ "raw": {
+ "normalized_reference": "gad, your letter came just in time.",
+ "normalized_hypothesis": "gad your letter came just in time",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "gad your letter came just in time",
+ "normalized_hypothesis": "gad your letter came just in time",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0352",
+ "category": "phonetic_balance",
+ "text": "But if Pierre did not return, until tomorrow.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0352.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "BUT IF PIERRE DID NOT RETURN UNTIL TO MORROW",
+ "raw": {
+ "normalized_reference": "but if pierre did not return, until tomorrow.",
+ "normalized_hypothesis": "but if pierre did not return until to morrow",
+ "errors": 2,
+ "words": 8,
+ "wer": 0.25
+ },
+ "semantic": {
+ "normalized_reference": "but if pierre did not return until tomorrow",
+ "normalized_hypothesis": "but if pierre did not return until to morrow",
+ "errors": 2,
+ "words": 8,
+ "wer": 0.25
+ }
+ },
+ {
+ "id": "release500_0353",
+ "category": "phonetic_balance",
+ "text": "Billinger may arrive in time.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0353.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "BILLINGER MAY ARRIVE IN TIME",
+ "raw": {
+ "normalized_reference": "billinger may arrive in time.",
+ "normalized_hypothesis": "billinger may arrive in time",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "billinger may arrive in time",
+ "normalized_hypothesis": "billinger may arrive in time",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0354",
+ "category": "phonetic_balance",
+ "text": "In such a tumbling of values was no time to sell.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0354.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IN SUCH A TUMBLING OF VALUES WAS NO TIME TO SELL",
+ "raw": {
+ "normalized_reference": "in such a tumbling of values was no time to sell.",
+ "normalized_hypothesis": "in such a tumbling of values was no time to sell",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "in such a tumbling of values was no time to sell",
+ "normalized_hypothesis": "in such a tumbling of values was no time to sell",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0355",
+ "category": "phonetic_balance",
+ "text": "There's not an iota of truth in it.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0355.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THERE'S NOT AN IOTA OF TRUTH IN IT",
+ "raw": {
+ "normalized_reference": "there's not an iota of truth in it.",
+ "normalized_hypothesis": "there's not an iota of truth in it",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "there is not an iota of truth in it",
+ "normalized_hypothesis": "there is not an iota of truth in it",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0356",
+ "category": "phonetic_balance",
+ "text": "It is dog eat dog, and you ate them up.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0356.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT IS DOG EAT DOG AND YOU EAT THEM",
+ "raw": {
+ "normalized_reference": "it is dog eat dog, and you ate them up.",
+ "normalized_hypothesis": "it is dog eat dog and you eat them",
+ "errors": 2,
+ "words": 10,
+ "wer": 0.2
+ },
+ "semantic": {
+ "normalized_reference": "it is dog eat dog and you ate them up",
+ "normalized_hypothesis": "it is dog eat dog and you eat them",
+ "errors": 2,
+ "words": 10,
+ "wer": 0.2
+ }
+ },
+ {
+ "id": "release500_0357",
+ "category": "phonetic_balance",
+ "text": "You can take a vacation on pay.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0357.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "YOU CAN TAKE A VACATION ON PAY",
+ "raw": {
+ "normalized_reference": "you can take a vacation on pay.",
+ "normalized_hypothesis": "you can take a vacation on pay",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "you can take a vacation on pay",
+ "normalized_hypothesis": "you can take a vacation on pay",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0358",
+ "category": "phonetic_balance",
+ "text": "I have seen myself that one man contemplated by Pascal's philosophic eye.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0358.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I HAVE SEEN MYSELF THAT ONE MAN CONTEMPLATED BY PASCELL'S PHILOSOPHIC EYE",
+ "raw": {
+ "normalized_reference": "i have seen myself that one man contemplated by pascal's philosophic eye.",
+ "normalized_hypothesis": "i have seen myself that one man contemplated by pascell's philosophic eye",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "i have seen myself that one man contemplated by pascal is philosophic eye",
+ "normalized_hypothesis": "i have seen myself that one man contemplated by pascell is philosophic eye",
+ "errors": 1,
+ "words": 13,
+ "wer": 0.07692307692307693
+ }
+ },
+ {
+ "id": "release500_0359",
+ "category": "phonetic_balance",
+ "text": "Ah, I had forgotten, he exclaimed.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0359.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "ALL I HAD FORGOTTEN HE EXCLAIMED",
+ "raw": {
+ "normalized_reference": "ah, i had forgotten, he exclaimed.",
+ "normalized_hypothesis": "all i had forgotten he exclaimed",
+ "errors": 1,
+ "words": 6,
+ "wer": 0.16666666666666666
+ },
+ "semantic": {
+ "normalized_reference": "ah i had forgotten he exclaimed",
+ "normalized_hypothesis": "all i had forgotten he exclaimed",
+ "errors": 1,
+ "words": 6,
+ "wer": 0.16666666666666666
+ }
+ },
+ {
+ "id": "release500_0360",
+ "category": "phonetic_balance",
+ "text": "Providence had delivered him through the maelstrom.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0360.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "PROVIDENCE HAD DELIVERED HIM THROUGH THE MAILSTRAM",
+ "raw": {
+ "normalized_reference": "providence had delivered him through the maelstrom.",
+ "normalized_hypothesis": "providence had delivered him through the mailstram",
+ "errors": 1,
+ "words": 7,
+ "wer": 0.14285714285714285
+ },
+ "semantic": {
+ "normalized_reference": "providence had delivered him through the maelstrom",
+ "normalized_hypothesis": "providence had delivered him through the mailstram",
+ "errors": 1,
+ "words": 7,
+ "wer": 0.14285714285714285
+ }
+ },
+ {
+ "id": "release500_0361",
+ "category": "phonetic_balance",
+ "text": "She was his now, forever.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0361.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SHE WAS HIS NOW FOR EVER",
+ "raw": {
+ "normalized_reference": "she was his now, forever.",
+ "normalized_hypothesis": "she was his now for ever",
+ "errors": 2,
+ "words": 5,
+ "wer": 0.4
+ },
+ "semantic": {
+ "normalized_reference": "she was his now forever",
+ "normalized_hypothesis": "she was his now for ever",
+ "errors": 2,
+ "words": 5,
+ "wer": 0.4
+ }
+ },
+ {
+ "id": "release500_0362",
+ "category": "phonetic_balance",
+ "text": "It was unobtrusive, yet it was there.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0362.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT WAS UNOBTRUSIVE YET IT WAS THERE",
+ "raw": {
+ "normalized_reference": "it was unobtrusive, yet it was there.",
+ "normalized_hypothesis": "it was unobtrusive yet it was there",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it was unobtrusive yet it was there",
+ "normalized_hypothesis": "it was unobtrusive yet it was there",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0363",
+ "category": "phonetic_balance",
+ "text": "Men like Joe Goose dated existence from drunk to drunk.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0363.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "MEN LIKE JOE GOOSE DATED EXISTENCE FROM DRUNK TO DRUNK",
+ "raw": {
+ "normalized_reference": "men like joe goose dated existence from drunk to drunk.",
+ "normalized_hypothesis": "men like joe goose dated existence from drunk to drunk",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "men like joe goose dated existence from drunk to drunk",
+ "normalized_hypothesis": "men like joe goose dated existence from drunk to drunk",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0364",
+ "category": "phonetic_balance",
+ "text": "I was completely lost in my work.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0364.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I WAS COMPLETELY LOST IN MY WORK",
+ "raw": {
+ "normalized_reference": "i was completely lost in my work.",
+ "normalized_hypothesis": "i was completely lost in my work",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i was completely lost in my work",
+ "normalized_hypothesis": "i was completely lost in my work",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0365",
+ "category": "phonetic_balance",
+ "text": "The very thing, Ernest agreed.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0365.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE VERY THING ERNEST AGREED",
+ "raw": {
+ "normalized_reference": "the very thing, ernest agreed.",
+ "normalized_hypothesis": "the very thing ernest agreed",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the very thing ernest agreed",
+ "normalized_hypothesis": "the very thing ernest agreed",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0366",
+ "category": "phonetic_balance",
+ "text": "He was the leader, and Tudor was his lieutenant.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0366.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE WAS THE LEADER AND TUDOR WAS HIS LIEUTENANT",
+ "raw": {
+ "normalized_reference": "he was the leader, and tudor was his lieutenant.",
+ "normalized_hypothesis": "he was the leader and tudor was his lieutenant",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he was the leader and tudor was his lieutenant",
+ "normalized_hypothesis": "he was the leader and tudor was his lieutenant",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0367",
+ "category": "phonetic_balance",
+ "text": "The Fire-Men wore animal skins around their waists and across their shoulders.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0367.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE FIREMEN WORE ANIMAL SKINS AROUND THEIR WAISTS AND ACROSS THEIR SHOULDERS",
+ "raw": {
+ "normalized_reference": "the fire-men wore animal skins around their waists and across their shoulders.",
+ "normalized_hypothesis": "the firemen wore animal skins around their waists and across their shoulders",
+ "errors": 2,
+ "words": 13,
+ "wer": 0.15384615384615385
+ },
+ "semantic": {
+ "normalized_reference": "the fire men wore animal skins around their waists and across their shoulders",
+ "normalized_hypothesis": "the firemen wore animal skins around their waists and across their shoulders",
+ "errors": 2,
+ "words": 13,
+ "wer": 0.15384615384615385
+ }
+ },
+ {
+ "id": "release500_0368",
+ "category": "phonetic_balance",
+ "text": "We could throw stones with our feet.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0368.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WE COULD THROW STONES WITH OUR FEET",
+ "raw": {
+ "normalized_reference": "we could throw stones with our feet.",
+ "normalized_hypothesis": "we could throw stones with our feet",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "we could throw stones with our feet",
+ "normalized_hypothesis": "we could throw stones with our feet",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0369",
+ "category": "phonetic_balance",
+ "text": "He turned sharply, and faced Gregson across the table.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0369.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE TURNED SHARPLY AND FACED GREGSON ACROSS THE TABLE",
+ "raw": {
+ "normalized_reference": "he turned sharply, and faced gregson across the table.",
+ "normalized_hypothesis": "he turned sharply and faced gregson across the table",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he turned sharply and faced gregson across the table",
+ "normalized_hypothesis": "he turned sharply and faced gregson across the table",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0370",
+ "category": "phonetic_balance",
+ "text": "They are to attack your camp tomorrow night.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0370.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THEY ARE TO ATTACK YOUR CAMP TO MORROW NIGHT",
+ "raw": {
+ "normalized_reference": "they are to attack your camp tomorrow night.",
+ "normalized_hypothesis": "they are to attack your camp to morrow night",
+ "errors": 2,
+ "words": 8,
+ "wer": 0.25
+ },
+ "semantic": {
+ "normalized_reference": "they are to attack your camp tomorrow night",
+ "normalized_hypothesis": "they are to attack your camp to morrow night",
+ "errors": 2,
+ "words": 8,
+ "wer": 0.25
+ }
+ },
+ {
+ "id": "release500_0371",
+ "category": "phonetic_balance",
+ "text": "And watch out for wet feet, was his parting advice.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0371.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "AND WATCH OUT FOR WET FEET WAS HIS PARTING ADVICE",
+ "raw": {
+ "normalized_reference": "and watch out for wet feet, was his parting advice.",
+ "normalized_hypothesis": "and watch out for wet feet was his parting advice",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "and watch out for wet feet was his parting advice",
+ "normalized_hypothesis": "and watch out for wet feet was his parting advice",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0372",
+ "category": "phonetic_balance",
+ "text": "Two weeks passed, and in that time Thorpe left camp three times.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0372.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "TWO WEEKS PASSED AND IN THAT TIME THORPE LEFT CAMP THREE TIMES",
+ "raw": {
+ "normalized_reference": "two weeks passed, and in that time thorpe left camp three times.",
+ "normalized_hypothesis": "two weeks passed and in that time thorpe left camp three times",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "2 weeks passed and in that time thorpe left camp 3 times",
+ "normalized_hypothesis": "2 weeks passed and in that time thorpe left camp 3 times",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0373",
+ "category": "phonetic_balance",
+ "text": "But ever his gaze returned to that Crouched Venus on the piano.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0373.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "BUT EVER HIS GAZE RETURNED TO THAT CROUCHED VENUS ON THE PIANO",
+ "raw": {
+ "normalized_reference": "but ever his gaze returned to that crouched venus on the piano.",
+ "normalized_hypothesis": "but ever his gaze returned to that crouched venus on the piano",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "but ever his gaze returned to that crouched venus on the piano",
+ "normalized_hypothesis": "but ever his gaze returned to that crouched venus on the piano",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0374",
+ "category": "phonetic_balance",
+ "text": "But she had become an automaton.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0374.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "BUT SHE HAD BECOME AN AUTOMATON",
+ "raw": {
+ "normalized_reference": "but she had become an automaton.",
+ "normalized_hypothesis": "but she had become an automaton",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "but she had become an automaton",
+ "normalized_hypothesis": "but she had become an automaton",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0375",
+ "category": "phonetic_balance",
+ "text": "I do not blame you for anything; remember that.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0375.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I DO NOT BLAME YOU FOR ANYTHING REMEMBER THAT",
+ "raw": {
+ "normalized_reference": "i do not blame you for anything; remember that.",
+ "normalized_hypothesis": "i do not blame you for anything remember that",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i do not blame you for anything remember that",
+ "normalized_hypothesis": "i do not blame you for anything remember that",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0376",
+ "category": "phonetic_balance",
+ "text": "The next thing to watch out for is bed sores.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0376.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE NEXT THING TO WATCH OUT FOR IS BED SORES",
+ "raw": {
+ "normalized_reference": "the next thing to watch out for is bed sores.",
+ "normalized_hypothesis": "the next thing to watch out for is bed sores",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the next thing to watch out for is bed sores",
+ "normalized_hypothesis": "the next thing to watch out for is bed sores",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0377",
+ "category": "phonetic_balance",
+ "text": "He will follow us soon.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0377.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE WILL FOLLOW US SOON",
+ "raw": {
+ "normalized_reference": "he will follow us soon.",
+ "normalized_hypothesis": "he will follow us soon",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he will follow us soon",
+ "normalized_hypothesis": "he will follow us soon",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0378",
+ "category": "phonetic_balance",
+ "text": "His beady black eyes saw bargains where other men saw bankruptcy.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0378.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HIS BEADY BLACK EYES SAW BARGAINS WHERE OTHER MEN SAW BANKRUPTCY",
+ "raw": {
+ "normalized_reference": "his beady black eyes saw bargains where other men saw bankruptcy.",
+ "normalized_hypothesis": "his beady black eyes saw bargains where other men saw bankruptcy",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "his beady black eyes saw bargains where other men saw bankruptcy",
+ "normalized_hypothesis": "his beady black eyes saw bargains where other men saw bankruptcy",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0379",
+ "category": "phonetic_balance",
+ "text": "They robbed me a few years later.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0379.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THEY ROBBED ME A FEW YEARS LATER",
+ "raw": {
+ "normalized_reference": "they robbed me a few years later.",
+ "normalized_hypothesis": "they robbed me a few years later",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "they robbed me a few years later",
+ "normalized_hypothesis": "they robbed me a few years later",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0380",
+ "category": "phonetic_balance",
+ "text": "Gregson held a lighted match until it burnt his fingertips.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0380.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "GREXSEN HELD A LIGHTED MATCH UNTIL IT BURNT HIS FINGER TIPS",
+ "raw": {
+ "normalized_reference": "gregson held a lighted match until it burnt his fingertips.",
+ "normalized_hypothesis": "grexsen held a lighted match until it burnt his finger tips",
+ "errors": 3,
+ "words": 10,
+ "wer": 0.3
+ },
+ "semantic": {
+ "normalized_reference": "gregson held a lighted match until it burnt his fingertips",
+ "normalized_hypothesis": "grexsen held a lighted match until it burnt his finger tips",
+ "errors": 3,
+ "words": 10,
+ "wer": 0.3
+ }
+ },
+ {
+ "id": "release500_0381",
+ "category": "phonetic_balance",
+ "text": "A burst of laughter was his reward.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0381.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "A BURST OF LAUGHTER WAS HIS REWARD",
+ "raw": {
+ "normalized_reference": "a burst of laughter was his reward.",
+ "normalized_hypothesis": "a burst of laughter was his reward",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "a burst of laughter was his reward",
+ "normalized_hypothesis": "a burst of laughter was his reward",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0382",
+ "category": "phonetic_balance",
+ "text": "He cried, and swung the club wildly.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0382.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE CRIED AND SWUNG THE CLUB WILDLY",
+ "raw": {
+ "normalized_reference": "he cried, and swung the club wildly.",
+ "normalized_hypothesis": "he cried and swung the club wildly",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he cried and swung the club wildly",
+ "normalized_hypothesis": "he cried and swung the club wildly",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0383",
+ "category": "phonetic_balance",
+ "text": "Sandel would never become a world champion.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0383.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SANDEL WOULD NEVER BECOME A WORLD CHAMPION",
+ "raw": {
+ "normalized_reference": "sandel would never become a world champion.",
+ "normalized_hypothesis": "sandel would never become a world champion",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "sandel would never become a world champion",
+ "normalized_hypothesis": "sandel would never become a world champion",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0384",
+ "category": "phonetic_balance",
+ "text": "And after the bath a shave would not be bad.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0384.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "AND AFTER THE BATH TE SHAVE WOULD NOT BE BAD",
+ "raw": {
+ "normalized_reference": "and after the bath a shave would not be bad.",
+ "normalized_hypothesis": "and after the bath te shave would not be bad",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ },
+ "semantic": {
+ "normalized_reference": "and after the bath a shave would not be bad",
+ "normalized_hypothesis": "and after the bath te shave would not be bad",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ }
+ },
+ {
+ "id": "release500_0385",
+ "category": "phonetic_balance",
+ "text": "They saw each other for the first time in Boston.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0385.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THEY SAW EACH OTHER FOR THE FIRST TIME IN BOSTON",
+ "raw": {
+ "normalized_reference": "they saw each other for the first time in boston.",
+ "normalized_hypothesis": "they saw each other for the first time in boston",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "they saw each other for the 1st time in boston",
+ "normalized_hypothesis": "they saw each other for the 1st time in boston",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0386",
+ "category": "phonetic_balance",
+ "text": "I had been born with no organic, chemical predisposition toward alcohol.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0386.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I HAD BEEN BORNE WITH NO ORGANIC CHEMICAL PREDISPOSITION TOWARD ALCOHOL",
+ "raw": {
+ "normalized_reference": "i had been born with no organic, chemical predisposition toward alcohol.",
+ "normalized_hypothesis": "i had been borne with no organic chemical predisposition toward alcohol",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ },
+ "semantic": {
+ "normalized_reference": "i had been born with no organic chemical predisposition toward alcohol",
+ "normalized_hypothesis": "i had been borne with no organic chemical predisposition toward alcohol",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ }
+ },
+ {
+ "id": "release500_0387",
+ "category": "phonetic_balance",
+ "text": "They had no fixed values, to be altered by adjectives and adverbs.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0387.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THEY HAD NO FIXED VALUES TO BE ALTERED BY ADJECTIVES AND ADVERBS",
+ "raw": {
+ "normalized_reference": "they had no fixed values, to be altered by adjectives and adverbs.",
+ "normalized_hypothesis": "they had no fixed values to be altered by adjectives and adverbs",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "they had no fixed values to be altered by adjectives and adverbs",
+ "normalized_hypothesis": "they had no fixed values to be altered by adjectives and adverbs",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0388",
+ "category": "phonetic_balance",
+ "text": "Nor was Elam Harnish an exception.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0388.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "NOR WAS THE LAMB HARNESH AN EXCEPTION",
+ "raw": {
+ "normalized_reference": "nor was elam harnish an exception.",
+ "normalized_hypothesis": "nor was the lamb harnesh an exception",
+ "errors": 3,
+ "words": 6,
+ "wer": 0.5
+ },
+ "semantic": {
+ "normalized_reference": "nor was elam harnish an exception",
+ "normalized_hypothesis": "nor was the lamb harnesh an exception",
+ "errors": 3,
+ "words": 6,
+ "wer": 0.5
+ }
+ },
+ {
+ "id": "release500_0389",
+ "category": "phonetic_balance",
+ "text": "He was a wise hyena.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0389.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE WAS A WISE HYENA",
+ "raw": {
+ "normalized_reference": "he was a wise hyena.",
+ "normalized_hypothesis": "he was a wise hyena",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he was a wise hyena",
+ "normalized_hypothesis": "he was a wise hyena",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0390",
+ "category": "phonetic_balance",
+ "text": "And now behold the perversity of things.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0390.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "AND NOW BEHOLD THE PERVERSITY OF THINGS",
+ "raw": {
+ "normalized_reference": "and now behold the perversity of things.",
+ "normalized_hypothesis": "and now behold the perversity of things",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "and now behold the perversity of things",
+ "normalized_hypothesis": "and now behold the perversity of things",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0391",
+ "category": "phonetic_balance",
+ "text": "The Fire People, like ourselves, lived in caves.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0391.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE FIRE PEOPLE LIKE OURSELVES LIVED IN CAVES",
+ "raw": {
+ "normalized_reference": "the fire people, like ourselves, lived in caves.",
+ "normalized_hypothesis": "the fire people like ourselves lived in caves",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the fire people like ourselves lived in caves",
+ "normalized_hypothesis": "the fire people like ourselves lived in caves",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0392",
+ "category": "phonetic_balance",
+ "text": "To her the bridge was tambo, which is the native for taboo.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0392.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "TO HER THE BRIDGE WAS TAMBO WHICH IS THE NATIVE FOR TABOO",
+ "raw": {
+ "normalized_reference": "to her the bridge was tambo, which is the native for taboo.",
+ "normalized_hypothesis": "to her the bridge was tambo which is the native for taboo",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "to her the bridge was tambo which is the native for taboo",
+ "normalized_hypothesis": "to her the bridge was tambo which is the native for taboo",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0393",
+ "category": "phonetic_balance",
+ "text": "The voyage was our idea of a good time.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0393.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE VOYAGE WAS OUR IDEA OF A GOOD TIME",
+ "raw": {
+ "normalized_reference": "the voyage was our idea of a good time.",
+ "normalized_hypothesis": "the voyage was our idea of a good time",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the voyage was our idea of a good time",
+ "normalized_hypothesis": "the voyage was our idea of a good time",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0394",
+ "category": "phonetic_balance",
+ "text": "Of course much grumbling went on, and little outbursts were continually occurring.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0394.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "OF COURSE MUCH GRUMBLING WENT ON AND LITTLE OUTBURSTS WERE CONTINUALLY OCCURRING",
+ "raw": {
+ "normalized_reference": "of course much grumbling went on, and little outbursts were continually occurring.",
+ "normalized_hypothesis": "of course much grumbling went on and little outbursts were continually occurring",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "of course much grumbling went on and little outbursts were continually occurring",
+ "normalized_hypothesis": "of course much grumbling went on and little outbursts were continually occurring",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0395",
+ "category": "phonetic_balance",
+ "text": "Get down and dig in.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0395.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "GET DOWN AN DIG IN",
+ "raw": {
+ "normalized_reference": "get down and dig in.",
+ "normalized_hypothesis": "get down an dig in",
+ "errors": 1,
+ "words": 5,
+ "wer": 0.2
+ },
+ "semantic": {
+ "normalized_reference": "get down and dig in",
+ "normalized_hypothesis": "get down an dig in",
+ "errors": 1,
+ "words": 5,
+ "wer": 0.2
+ }
+ },
+ {
+ "id": "release500_0396",
+ "category": "phonetic_balance",
+ "text": "It was put together in a casual, helter-skelter sort of way.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0396.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT WAS PUT TOGETHER IN A CASUAL HELTERSKELTER SORT OF WAY",
+ "raw": {
+ "normalized_reference": "it was put together in a casual, helter-skelter sort of way.",
+ "normalized_hypothesis": "it was put together in a casual helterskelter sort of way",
+ "errors": 2,
+ "words": 12,
+ "wer": 0.16666666666666666
+ },
+ "semantic": {
+ "normalized_reference": "it was put together in a casual helter skelter sort of way",
+ "normalized_hypothesis": "it was put together in a casual helterskelter sort of way",
+ "errors": 2,
+ "words": 12,
+ "wer": 0.16666666666666666
+ }
+ },
+ {
+ "id": "release500_0397",
+ "category": "phonetic_balance",
+ "text": "He deluged me, overwhelmed me with argument.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0397.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE DELUGED ME OVERWHELMED ME WITH ARGUMENT",
+ "raw": {
+ "normalized_reference": "he deluged me, overwhelmed me with argument.",
+ "normalized_hypothesis": "he deluged me overwhelmed me with argument",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he deluged me overwhelmed me with argument",
+ "normalized_hypothesis": "he deluged me overwhelmed me with argument",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0398",
+ "category": "phonetic_balance",
+ "text": "The girl faced him, her eyes shining with sudden fear.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0398.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE GIRL FACED HIM HER EYES SHINING WITH SUDDEN FEAR",
+ "raw": {
+ "normalized_reference": "the girl faced him, her eyes shining with sudden fear.",
+ "normalized_hypothesis": "the girl faced him her eyes shining with sudden fear",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the girl faced him her eyes shining with sudden fear",
+ "normalized_hypothesis": "the girl faced him her eyes shining with sudden fear",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0399",
+ "category": "phonetic_balance",
+ "text": "It will break our hearts and our backs to hoist anchor by hand.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0399.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT WILL BREAK OUR HEARTS AND OUR BACKS TO HOIST ANCHOR BY HAND",
+ "raw": {
+ "normalized_reference": "it will break our hearts and our backs to hoist anchor by hand.",
+ "normalized_hypothesis": "it will break our hearts and our backs to hoist anchor by hand",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it will break our hearts and our backs to hoist anchor by hand",
+ "normalized_hypothesis": "it will break our hearts and our backs to hoist anchor by hand",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0400",
+ "category": "phonetic_balance",
+ "text": "He made no reply as he waited for Whittemore to continue.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0400.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE MADE NO REPLY AS HE WAITED FOR WHITTAMORE TO CONTINUE",
+ "raw": {
+ "normalized_reference": "he made no reply as he waited for whittemore to continue.",
+ "normalized_hypothesis": "he made no reply as he waited for whittamore to continue",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ },
+ "semantic": {
+ "normalized_reference": "he made no reply as he waited for whittemore to continue",
+ "normalized_hypothesis": "he made no reply as he waited for whittamore to continue",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ }
+ },
+ {
+ "id": "release500_0401",
+ "category": "phonetic_balance",
+ "text": "Who the devil gave it to you to be judge and jury.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0401.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WHO THE DEVIL GAVE IT TO YOU TO BE JUDGE AND JURY",
+ "raw": {
+ "normalized_reference": "who the devil gave it to you to be judge and jury.",
+ "normalized_hypothesis": "who the devil gave it to you to be judge and jury",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "who the devil gave it to you to be judge and jury",
+ "normalized_hypothesis": "who the devil gave it to you to be judge and jury",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0402",
+ "category": "phonetic_balance",
+ "text": "Daylight found himself charmed and made curious by the little man.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0402.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "DAYLIGHT FOUND HIMSELF CHARMED AND MADE CURIOUS BY THE LITTLE MAN",
+ "raw": {
+ "normalized_reference": "daylight found himself charmed and made curious by the little man.",
+ "normalized_hypothesis": "daylight found himself charmed and made curious by the little man",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "daylight found himself charmed and made curious by the little man",
+ "normalized_hypothesis": "daylight found himself charmed and made curious by the little man",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0403",
+ "category": "phonetic_balance",
+ "text": "It's a Yankee, Joan cried.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0403.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT'S A YANKEE JOAN CRIED",
+ "raw": {
+ "normalized_reference": "it's a yankee, joan cried.",
+ "normalized_hypothesis": "it's a yankee joan cried",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it is a yankee joan cried",
+ "normalized_hypothesis": "it is a yankee joan cried",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0404",
+ "category": "phonetic_balance",
+ "text": "But what they want with your toothbrush is more than I can imagine.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0404.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "BUT WHAT THEY WANT WITH YOUR TOOTH BRUSH IS MORE THAN I CAN IMAGINE",
+ "raw": {
+ "normalized_reference": "but what they want with your toothbrush is more than i can imagine.",
+ "normalized_hypothesis": "but what they want with your tooth brush is more than i can imagine",
+ "errors": 2,
+ "words": 13,
+ "wer": 0.15384615384615385
+ },
+ "semantic": {
+ "normalized_reference": "but what they want with your toothbrush is more than i can imagine",
+ "normalized_hypothesis": "but what they want with your tooth brush is more than i can imagine",
+ "errors": 2,
+ "words": 13,
+ "wer": 0.15384615384615385
+ }
+ },
+ {
+ "id": "release500_0405",
+ "category": "phonetic_balance",
+ "text": "They were babbling and chattering all together.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0405.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THEY WERE BABBLING AND CHATTERING ALL TOGETHER",
+ "raw": {
+ "normalized_reference": "they were babbling and chattering all together.",
+ "normalized_hypothesis": "they were babbling and chattering all together",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "they were babbling and chattering all together",
+ "normalized_hypothesis": "they were babbling and chattering all together",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0406",
+ "category": "phonetic_balance",
+ "text": "Men of Selden's stamp don't stop at women and children.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0406.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "MEN OF SELDON'S STAMP DON'T STOP AT WOMEN AND CHILDREN",
+ "raw": {
+ "normalized_reference": "men of selden's stamp don't stop at women and children.",
+ "normalized_hypothesis": "men of seldon's stamp don't stop at women and children",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ },
+ "semantic": {
+ "normalized_reference": "men of selden is stamp do not stop at women and children",
+ "normalized_hypothesis": "men of seldon is stamp do not stop at women and children",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ }
+ },
+ {
+ "id": "release500_0407",
+ "category": "phonetic_balance",
+ "text": "Cameron looked at his hands with their long, sinewy fingers.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0407.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "CAMERON LOOKED AT HIS HANDS WITH THEIR LONG SINEWY FINGERS",
+ "raw": {
+ "normalized_reference": "cameron looked at his hands with their long, sinewy fingers.",
+ "normalized_hypothesis": "cameron looked at his hands with their long sinewy fingers",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "cameron looked at his hands with their long sinewy fingers",
+ "normalized_hypothesis": "cameron looked at his hands with their long sinewy fingers",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0408",
+ "category": "phonetic_balance",
+ "text": "Saxon waited, for she knew a fresh idea had struck Billy.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0408.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SAXON WAITED FOR SHE KNEW A FRESH IDEA HAD STRUCK BILLY",
+ "raw": {
+ "normalized_reference": "saxon waited, for she knew a fresh idea had struck billy.",
+ "normalized_hypothesis": "saxon waited for she knew a fresh idea had struck billy",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "saxon waited for she knew a fresh idea had struck billy",
+ "normalized_hypothesis": "saxon waited for she knew a fresh idea had struck billy",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0409",
+ "category": "phonetic_balance",
+ "text": "He saw only the effect in a general, sketchy way.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0409.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE SAW ONLY THE EFFECT IN A GENERAL SKETCHY WAY",
+ "raw": {
+ "normalized_reference": "he saw only the effect in a general, sketchy way.",
+ "normalized_hypothesis": "he saw only the effect in a general sketchy way",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he saw only the effect in a general sketchy way",
+ "normalized_hypothesis": "he saw only the effect in a general sketchy way",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0410",
+ "category": "phonetic_balance",
+ "text": "He began to follow the footprints of the dog.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0410.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE BEGAN TO FOLLOW THE FOOTPRINTS OF THE DOG",
+ "raw": {
+ "normalized_reference": "he began to follow the footprints of the dog.",
+ "normalized_hypothesis": "he began to follow the footprints of the dog",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he began to follow the footprints of the dog",
+ "normalized_hypothesis": "he began to follow the footprints of the dog",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0411",
+ "category": "phonetic_balance",
+ "text": "Do you value your hide.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0411.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "DO YOU VALUE YOUR HIDE",
+ "raw": {
+ "normalized_reference": "do you value your hide.",
+ "normalized_hypothesis": "do you value your hide",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "do you value your hide",
+ "normalized_hypothesis": "do you value your hide",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0412",
+ "category": "phonetic_balance",
+ "text": "A rising tide of fat had submerged them.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0412.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "A RISING TIDE OF FAT HAD SUBMERGED THEM",
+ "raw": {
+ "normalized_reference": "a rising tide of fat had submerged them.",
+ "normalized_hypothesis": "a rising tide of fat had submerged them",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "a rising tide of fat had submerged them",
+ "normalized_hypothesis": "a rising tide of fat had submerged them",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0413",
+ "category": "phonetic_balance",
+ "text": "There was nothing on the rock.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0413.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THERE WAS NOTHING ON THE ROCK",
+ "raw": {
+ "normalized_reference": "there was nothing on the rock.",
+ "normalized_hypothesis": "there was nothing on the rock",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "there was nothing on the rock",
+ "normalized_hypothesis": "there was nothing on the rock",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0414",
+ "category": "phonetic_balance",
+ "text": "Whatever he guessed he locked away in the taboo room of Naomi.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0414.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WHATEVER HE GUESSED HE LOCKED AWAY IN THE TABOO ROOM OF NAOMI",
+ "raw": {
+ "normalized_reference": "whatever he guessed he locked away in the taboo room of naomi.",
+ "normalized_hypothesis": "whatever he guessed he locked away in the taboo room of naomi",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "whatever he guessed he locked away in the taboo room of naomi",
+ "normalized_hypothesis": "whatever he guessed he locked away in the taboo room of naomi",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0415",
+ "category": "phonetic_balance",
+ "text": "Will we ever forget it.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0415.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WILL WE EVER FORGET IT",
+ "raw": {
+ "normalized_reference": "will we ever forget it.",
+ "normalized_hypothesis": "will we ever forget it",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "will we ever forget it",
+ "normalized_hypothesis": "will we ever forget it",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0416",
+ "category": "phonetic_balance",
+ "text": "Jeanne and Pierre both gazed toward the great rock.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0416.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "JEAN AND PIERRE BOTH GAZED TOWARD THE GREAT ROCK",
+ "raw": {
+ "normalized_reference": "jeanne and pierre both gazed toward the great rock.",
+ "normalized_hypothesis": "jean and pierre both gazed toward the great rock",
+ "errors": 1,
+ "words": 9,
+ "wer": 0.1111111111111111
+ },
+ "semantic": {
+ "normalized_reference": "jeanne and pierre both gazed toward the great rock",
+ "normalized_hypothesis": "jean and pierre both gazed toward the great rock",
+ "errors": 1,
+ "words": 9,
+ "wer": 0.1111111111111111
+ }
+ },
+ {
+ "id": "release500_0417",
+ "category": "phonetic_balance",
+ "text": "Does the old boy often go off at half-cock that way.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0417.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "DOES THE OLD BOY OFTEN GO OFF AT HALF COCK THAT WAY",
+ "raw": {
+ "normalized_reference": "does the old boy often go off at half-cock that way.",
+ "normalized_hypothesis": "does the old boy often go off at half cock that way",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "does the old boy often go off at half cock that way",
+ "normalized_hypothesis": "does the old boy often go off at half cock that way",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0418",
+ "category": "phonetic_balance",
+ "text": "The temperature dropped to fifty below zero and remained there the whole trip.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0418.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE TEMPERATURE DROPPED TO FIFTY BELOW ZERO AND REMAINED THERE THE WHOLE TRIP",
+ "raw": {
+ "normalized_reference": "the temperature dropped to fifty below zero and remained there the whole trip.",
+ "normalized_hypothesis": "the temperature dropped to fifty below zero and remained there the whole trip",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the temperature dropped to 50 below 0 and remained there the whole trip",
+ "normalized_hypothesis": "the temperature dropped to 50 below 0 and remained there the whole trip",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0419",
+ "category": "phonetic_balance",
+ "text": "It's worth eight dollars.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0419.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT'S WORTH EIGHT DOLLARS",
+ "raw": {
+ "normalized_reference": "it's worth eight dollars.",
+ "normalized_hypothesis": "it's worth eight dollars",
+ "errors": 0,
+ "words": 4,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it is worth $8",
+ "normalized_hypothesis": "it is worth $8",
+ "errors": 0,
+ "words": 4,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0420",
+ "category": "phonetic_balance",
+ "text": "But I did not enjoy it long.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0420.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "BUT I DID NOT ENJOY IT LONG",
+ "raw": {
+ "normalized_reference": "but i did not enjoy it long.",
+ "normalized_hypothesis": "but i did not enjoy it long",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "but i did not enjoy it long",
+ "normalized_hypothesis": "but i did not enjoy it long",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0421",
+ "category": "phonetic_balance",
+ "text": "Shorty turned to their employers.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0421.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SHORTY TURNED TO THEIR EMPLOYERS",
+ "raw": {
+ "normalized_reference": "shorty turned to their employers.",
+ "normalized_hypothesis": "shorty turned to their employers",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "shorty turned to their employers",
+ "normalized_hypothesis": "shorty turned to their employers",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0422",
+ "category": "phonetic_balance",
+ "text": "It was the third or fourth time that Philip had heard MacDougall swear.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0422.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT WAS THE THIRD OR FOURTH TIME THAT PHILIP HAD HEARD MACDUGAL SWEAR",
+ "raw": {
+ "normalized_reference": "it was the third or fourth time that philip had heard macdougall swear.",
+ "normalized_hypothesis": "it was the third or fourth time that philip had heard macdugal swear",
+ "errors": 1,
+ "words": 13,
+ "wer": 0.07692307692307693
+ },
+ "semantic": {
+ "normalized_reference": "it was the 3rd or 4th time that philip had heard macdougall swear",
+ "normalized_hypothesis": "it was the 3rd or 4th time that philip had heard macdugal swear",
+ "errors": 1,
+ "words": 13,
+ "wer": 0.07692307692307693
+ }
+ },
+ {
+ "id": "release500_0423",
+ "category": "phonetic_balance",
+ "text": "Ernest saw in the affair the most sinister import.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0423.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "ERNEST SAW IN THE AFFAIR THE MOST SINISTER IMPORT",
+ "raw": {
+ "normalized_reference": "ernest saw in the affair the most sinister import.",
+ "normalized_hypothesis": "ernest saw in the affair the most sinister import",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "ernest saw in the affair the most sinister import",
+ "normalized_hypothesis": "ernest saw in the affair the most sinister import",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0424",
+ "category": "phonetic_balance",
+ "text": "Also, at regular intervals, he would mutter.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0424.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "ALSO AT REGULAR INTERVALS HE WOULD MUTTER",
+ "raw": {
+ "normalized_reference": "also, at regular intervals, he would mutter.",
+ "normalized_hypothesis": "also at regular intervals he would mutter",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "also at regular intervals he would mutter",
+ "normalized_hypothesis": "also at regular intervals he would mutter",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0425",
+ "category": "phonetic_balance",
+ "text": "He may anticipate the day of his death.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0425.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE MAY ANTICIPATE THE DAY OF HIS DEATH",
+ "raw": {
+ "normalized_reference": "he may anticipate the day of his death.",
+ "normalized_hypothesis": "he may anticipate the day of his death",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he may anticipate the day of his death",
+ "normalized_hypothesis": "he may anticipate the day of his death",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0426",
+ "category": "phonetic_balance",
+ "text": "I defy any man to get a Solomon Island sore in California.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0426.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I DEFY ANY MAN TO GET A SULLOMON ISLAND SORE IN CALIFORNIA",
+ "raw": {
+ "normalized_reference": "i defy any man to get a solomon island sore in california.",
+ "normalized_hypothesis": "i defy any man to get a sullomon island sore in california",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "i defy any man to get a solomon island sore in california",
+ "normalized_hypothesis": "i defy any man to get a sullomon island sore in california",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ }
+ },
+ {
+ "id": "release500_0427",
+ "category": "phonetic_balance",
+ "text": "Now our figuring was all right.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0427.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "NOW OUR FIGURING WAS ALL RIGHT",
+ "raw": {
+ "normalized_reference": "now our figuring was all right.",
+ "normalized_hypothesis": "now our figuring was all right",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "now our figuring was all right",
+ "normalized_hypothesis": "now our figuring was all right",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0428",
+ "category": "phonetic_balance",
+ "text": "MacDougall tapped his forehead suspiciously with a stubby forefinger.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0428.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "MACDUGAL TAPPED HIS FOREHEAD SUSPICIOUSLY WITH A STUBBY FOREFINGER",
+ "raw": {
+ "normalized_reference": "macdougall tapped his forehead suspiciously with a stubby forefinger.",
+ "normalized_hypothesis": "macdugal tapped his forehead suspiciously with a stubby forefinger",
+ "errors": 1,
+ "words": 9,
+ "wer": 0.1111111111111111
+ },
+ "semantic": {
+ "normalized_reference": "macdougall tapped his forehead suspiciously with a stubby forefinger",
+ "normalized_hypothesis": "macdugal tapped his forehead suspiciously with a stubby forefinger",
+ "errors": 1,
+ "words": 9,
+ "wer": 0.1111111111111111
+ }
+ },
+ {
+ "id": "release500_0429",
+ "category": "phonetic_balance",
+ "text": "O'Brien took off his coat and bared his right arm.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0429.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "O'BRIAN TOOK OFF HIS COAT AND BARED HIS RIGHT ARM",
+ "raw": {
+ "normalized_reference": "o'brien took off his coat and bared his right arm.",
+ "normalized_hypothesis": "o'brian took off his coat and bared his right arm",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ },
+ "semantic": {
+ "normalized_reference": "0 brien took off his coat and bared his right arm",
+ "normalized_hypothesis": "0 brian took off his coat and bared his right arm",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ }
+ },
+ {
+ "id": "release500_0430",
+ "category": "phonetic_balance",
+ "text": "But it won't continue, she said with easy confidence.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0430.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "BUT IT WON'T CONTINUE SHE SAID WITH EASY CONFIDENCE",
+ "raw": {
+ "normalized_reference": "but it won't continue, she said with easy confidence.",
+ "normalized_hypothesis": "but it won't continue she said with easy confidence",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "but it will not continue she said with easy confidence",
+ "normalized_hypothesis": "but it will not continue she said with easy confidence",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0431",
+ "category": "phonetic_balance",
+ "text": "I was still weak from my prolonged immersion.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0431.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I WAS STILL WEAK FROM MY PROLONGED IMMERSION",
+ "raw": {
+ "normalized_reference": "i was still weak from my prolonged immersion.",
+ "normalized_hypothesis": "i was still weak from my prolonged immersion",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i was still weak from my prolonged immersion",
+ "normalized_hypothesis": "i was still weak from my prolonged immersion",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0432",
+ "category": "phonetic_balance",
+ "text": "Oolong was two hundred and fifty miles from the nearest land.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0432.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "ULONG WAS TWO HUNDRED AND FIFTY MILES FROM THE NEAREST LAND",
+ "raw": {
+ "normalized_reference": "oolong was two hundred and fifty miles from the nearest land.",
+ "normalized_hypothesis": "ulong was two hundred and fifty miles from the nearest land",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ },
+ "semantic": {
+ "normalized_reference": "oolong was 250 miles from the nearest land",
+ "normalized_hypothesis": "ulong was 250 miles from the nearest land",
+ "errors": 1,
+ "words": 8,
+ "wer": 0.125
+ }
+ },
+ {
+ "id": "release500_0433",
+ "category": "phonetic_balance",
+ "text": "I just do appreciate it without being able to express my feelings.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0433.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I JUST DO APPRECIATE IT WITHOUT BEING ABLE TO EXPRESS MY FEELINGS",
+ "raw": {
+ "normalized_reference": "i just do appreciate it without being able to express my feelings.",
+ "normalized_hypothesis": "i just do appreciate it without being able to express my feelings",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i just do appreciate it without being able to express my feelings",
+ "normalized_hypothesis": "i just do appreciate it without being able to express my feelings",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0434",
+ "category": "phonetic_balance",
+ "text": "Now just what do you want to know.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0434.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "NOW JUST WHAT DO YOU WANT TO KNOW",
+ "raw": {
+ "normalized_reference": "now just what do you want to know.",
+ "normalized_hypothesis": "now just what do you want to know",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "now just what do you want to know",
+ "normalized_hypothesis": "now just what do you want to know",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0435",
+ "category": "phonetic_balance",
+ "text": "But she swung obediently on her heel into the wind.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0435.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "BUT SHE SWUNG OBEDIENTLY ON HER HEEL INTO THE WIND",
+ "raw": {
+ "normalized_reference": "but she swung obediently on her heel into the wind.",
+ "normalized_hypothesis": "but she swung obediently on her heel into the wind",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "but she swung obediently on her heel into the wind",
+ "normalized_hypothesis": "but she swung obediently on her heel into the wind",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0436",
+ "category": "phonetic_balance",
+ "text": "Ask him, I laughed, then turned to Pasquini.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0436.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "ASK HIM I LAUGHED THEN TURNED TO PASQUINI",
+ "raw": {
+ "normalized_reference": "ask him, i laughed, then turned to pasquini.",
+ "normalized_hypothesis": "ask him i laughed then turned to pasquini",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "ask him i laughed then turned to pasquini",
+ "normalized_hypothesis": "ask him i laughed then turned to pasquini",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0437",
+ "category": "phonetic_balance",
+ "text": "He did not rush in.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0437.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE DID NOT RUSH IN",
+ "raw": {
+ "normalized_reference": "he did not rush in.",
+ "normalized_hypothesis": "he did not rush in",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he did not rush in",
+ "normalized_hypothesis": "he did not rush in",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0438",
+ "category": "phonetic_balance",
+ "text": "They obeyed him, and went here and there at his commands.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0438.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THEY OBEYED HIM AND WENT HERE AND THERE AT HIS COMMANDS",
+ "raw": {
+ "normalized_reference": "they obeyed him, and went here and there at his commands.",
+ "normalized_hypothesis": "they obeyed him and went here and there at his commands",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "they obeyed him and went here and there at his commands",
+ "normalized_hypothesis": "they obeyed him and went here and there at his commands",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0439",
+ "category": "phonetic_balance",
+ "text": "Tom Spink has a harpoon.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0439.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "TOM SPINK HAS A HARPOON",
+ "raw": {
+ "normalized_reference": "tom spink has a harpoon.",
+ "normalized_hypothesis": "tom spink has a harpoon",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "tom spink has a harpoon",
+ "normalized_hypothesis": "tom spink has a harpoon",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0440",
+ "category": "phonetic_balance",
+ "text": "It was my reports from the north which chiefly induced people to buy.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0440.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT WAS MY REPORTS FROM THE NORTH WHICH CHIEFLY INDUCED PEOPLE TO BUY",
+ "raw": {
+ "normalized_reference": "it was my reports from the north which chiefly induced people to buy.",
+ "normalized_hypothesis": "it was my reports from the north which chiefly induced people to buy",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it was my reports from the north which chiefly induced people to buy",
+ "normalized_hypothesis": "it was my reports from the north which chiefly induced people to buy",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0441",
+ "category": "phonetic_balance",
+ "text": "Cherokee identified himself with his instinct.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0441.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "CHEROKEY IDENTIFIED HIMSELF WITH HIS INSTINCT",
+ "raw": {
+ "normalized_reference": "cherokee identified himself with his instinct.",
+ "normalized_hypothesis": "cherokey identified himself with his instinct",
+ "errors": 1,
+ "words": 6,
+ "wer": 0.16666666666666666
+ },
+ "semantic": {
+ "normalized_reference": "cherokee identified himself with his instinct",
+ "normalized_hypothesis": "cherokey identified himself with his instinct",
+ "errors": 1,
+ "words": 6,
+ "wer": 0.16666666666666666
+ }
+ },
+ {
+ "id": "release500_0442",
+ "category": "phonetic_balance",
+ "text": "I graduated last of my class.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0442.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I GRADUATED LAST OF MY CLASS",
+ "raw": {
+ "normalized_reference": "i graduated last of my class.",
+ "normalized_hypothesis": "i graduated last of my class",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i graduated last of my class",
+ "normalized_hypothesis": "i graduated last of my class",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0443",
+ "category": "phonetic_balance",
+ "text": "It was a gigantic inadequacy.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0443.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT WAS A GIGANTIC INADEQUACY",
+ "raw": {
+ "normalized_reference": "it was a gigantic inadequacy.",
+ "normalized_hypothesis": "it was a gigantic inadequacy",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it was a gigantic inadequacy",
+ "normalized_hypothesis": "it was a gigantic inadequacy",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0444",
+ "category": "phonetic_balance",
+ "text": "Did I possess too much vitality.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0444.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "DID I POSSESS TOO MUCH VITALITY",
+ "raw": {
+ "normalized_reference": "did i possess too much vitality.",
+ "normalized_hypothesis": "did i possess too much vitality",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "did i possess too much vitality",
+ "normalized_hypothesis": "did i possess too much vitality",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0445",
+ "category": "phonetic_balance",
+ "text": "Therefore, hurrah for the game.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0445.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THEREFORE HURRAH FOR THE GAME",
+ "raw": {
+ "normalized_reference": "therefore, hurrah for the game.",
+ "normalized_hypothesis": "therefore hurrah for the game",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "therefore hurrah for the game",
+ "normalized_hypothesis": "therefore hurrah for the game",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0446",
+ "category": "phonetic_balance",
+ "text": "He had fulfilled his duty and paid properly.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0446.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE HAD FULFILLED HIS DUTY AND PAID PROPERLY",
+ "raw": {
+ "normalized_reference": "he had fulfilled his duty and paid properly.",
+ "normalized_hypothesis": "he had fulfilled his duty and paid properly",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he had fulfilled his duty and paid properly",
+ "normalized_hypothesis": "he had fulfilled his duty and paid properly",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0447",
+ "category": "phonetic_balance",
+ "text": "How can you manage all alone, Mr Young.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0447.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HOW CAN YOU MANAGE ALL ALONE MISTER YOUNG",
+ "raw": {
+ "normalized_reference": "how can you manage all alone, mr young.",
+ "normalized_hypothesis": "how can you manage all alone mister young",
+ "errors": 1,
+ "words": 8,
+ "wer": 0.125
+ },
+ "semantic": {
+ "normalized_reference": "how can you manage all alone mister young",
+ "normalized_hypothesis": "how can you manage all alone mister young",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0448",
+ "category": "phonetic_balance",
+ "text": "Red-Eye never committed a more outrageous deed.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0448.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "REDY NEVER COMMITTED A MORE OUTRAGEOUS DEED",
+ "raw": {
+ "normalized_reference": "red-eye never committed a more outrageous deed.",
+ "normalized_hypothesis": "redy never committed a more outrageous deed",
+ "errors": 2,
+ "words": 8,
+ "wer": 0.25
+ },
+ "semantic": {
+ "normalized_reference": "red eye never committed a more outrageous deed",
+ "normalized_hypothesis": "redy never committed a more outrageous deed",
+ "errors": 2,
+ "words": 8,
+ "wer": 0.25
+ }
+ },
+ {
+ "id": "release500_0449",
+ "category": "phonetic_balance",
+ "text": "He would keep his faith with Gregson for the promised day or two.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0449.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE WOULD KEEP HIS FAITH WITH GREGSON FOR THE PROMISED DAY OR TWO",
+ "raw": {
+ "normalized_reference": "he would keep his faith with gregson for the promised day or two.",
+ "normalized_hypothesis": "he would keep his faith with gregson for the promised day or two",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he would keep his faith with gregson for the promised day or 2",
+ "normalized_hypothesis": "he would keep his faith with gregson for the promised day or 2",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0450",
+ "category": "phonetic_balance",
+ "text": "There was one chance, and only one, of saving Jeanne.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0450.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THERE WAS ONE CHANCE AND ONLY ONE OF SAVING JEAN",
+ "raw": {
+ "normalized_reference": "there was one chance, and only one, of saving jeanne.",
+ "normalized_hypothesis": "there was one chance and only one of saving jean",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ },
+ "semantic": {
+ "normalized_reference": "there was one chance and only one of saving jeanne",
+ "normalized_hypothesis": "there was one chance and only one of saving jean",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ }
+ },
+ {
+ "id": "release500_0451",
+ "category": "phonetic_balance",
+ "text": "I was beneath the water, suffocating and drowning.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0451.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I WAS BENEATH THE WATER SUFFOCATING AND DROWNING",
+ "raw": {
+ "normalized_reference": "i was beneath the water, suffocating and drowning.",
+ "normalized_hypothesis": "i was beneath the water suffocating and drowning",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i was beneath the water suffocating and drowning",
+ "normalized_hypothesis": "i was beneath the water suffocating and drowning",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0452",
+ "category": "phonetic_balance",
+ "text": "Give them their choice between a fine or an official whipping.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0452.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "GIVE THEM THEIR CHOICE BETWEEN A FINE OR AN OFFICIAL WHIPPING",
+ "raw": {
+ "normalized_reference": "give them their choice between a fine or an official whipping.",
+ "normalized_hypothesis": "give them their choice between a fine or an official whipping",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "give them their choice between a fine or an official whipping",
+ "normalized_hypothesis": "give them their choice between a fine or an official whipping",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0453",
+ "category": "phonetic_balance",
+ "text": "They must have been swept away by the chaotic currents.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0453.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THEY MUST HAVE BEEN SWEPT AWAY BY THE CHAOTIC CURRENTS",
+ "raw": {
+ "normalized_reference": "they must have been swept away by the chaotic currents.",
+ "normalized_hypothesis": "they must have been swept away by the chaotic currents",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "they must have been swept away by the chaotic currents",
+ "normalized_hypothesis": "they must have been swept away by the chaotic currents",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0454",
+ "category": "phonetic_balance",
+ "text": "You yellow giant thing of the frost.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0454.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "YOU YELLOW GIANT THING OF THE FROST",
+ "raw": {
+ "normalized_reference": "you yellow giant thing of the frost.",
+ "normalized_hypothesis": "you yellow giant thing of the frost",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "you yellow giant thing of the frost",
+ "normalized_hypothesis": "you yellow giant thing of the frost",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0455",
+ "category": "phonetic_balance",
+ "text": "He had peeled off his shirt and was wildly waving it.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0455.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE HAD PEELED OFF HIS SHIRT AND WAS WILDLY WAVING IT",
+ "raw": {
+ "normalized_reference": "he had peeled off his shirt and was wildly waving it.",
+ "normalized_hypothesis": "he had peeled off his shirt and was wildly waving it",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he had peeled off his shirt and was wildly waving it",
+ "normalized_hypothesis": "he had peeled off his shirt and was wildly waving it",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0456",
+ "category": "phonetic_balance",
+ "text": "To these he gave castor oil.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0456.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "TO THESE HE GAVE CASTER OIL",
+ "raw": {
+ "normalized_reference": "to these he gave castor oil.",
+ "normalized_hypothesis": "to these he gave caster oil",
+ "errors": 1,
+ "words": 6,
+ "wer": 0.16666666666666666
+ },
+ "semantic": {
+ "normalized_reference": "to these he gave castor oil",
+ "normalized_hypothesis": "to these he gave caster oil",
+ "errors": 1,
+ "words": 6,
+ "wer": 0.16666666666666666
+ }
+ },
+ {
+ "id": "release500_0457",
+ "category": "phonetic_balance",
+ "text": "Tudor surveyed him with withering disgust.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0457.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "TUDOR SURVEYED HIM WITH WITHERING DISGUST",
+ "raw": {
+ "normalized_reference": "tudor surveyed him with withering disgust.",
+ "normalized_hypothesis": "tudor surveyed him with withering disgust",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "tudor surveyed him with withering disgust",
+ "normalized_hypothesis": "tudor surveyed him with withering disgust",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0458",
+ "category": "phonetic_balance",
+ "text": "After all, it was simply a mistake in judgment.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0458.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "AFTER ALL IT WAS SIMPLY A MISTAKE IN JUDGMENT",
+ "raw": {
+ "normalized_reference": "after all, it was simply a mistake in judgment.",
+ "normalized_hypothesis": "after all it was simply a mistake in judgment",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "after all it was simply a mistake in judgment",
+ "normalized_hypothesis": "after all it was simply a mistake in judgment",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0459",
+ "category": "phonetic_balance",
+ "text": "That is the strange part of it.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0459.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THAT IS THE STRANGE PART OF IT",
+ "raw": {
+ "normalized_reference": "that is the strange part of it.",
+ "normalized_hypothesis": "that is the strange part of it",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "that is the strange part of it",
+ "normalized_hypothesis": "that is the strange part of it",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0460",
+ "category": "phonetic_balance",
+ "text": "Not a wheel moved in his empire.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0460.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "NOT A WHEEL MOVED IN HIS EMPIRE",
+ "raw": {
+ "normalized_reference": "not a wheel moved in his empire.",
+ "normalized_hypothesis": "not a wheel moved in his empire",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "not a wheel moved in his empire",
+ "normalized_hypothesis": "not a wheel moved in his empire",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0461",
+ "category": "phonetic_balance",
+ "text": "Its diameter was not more than two hundred yards.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0461.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "ITS DIAMETER WAS NOT MORE THAN TWO HUNDRED YARDS",
+ "raw": {
+ "normalized_reference": "its diameter was not more than two hundred yards.",
+ "normalized_hypothesis": "its diameter was not more than two hundred yards",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "its diameter was not more than 200 yards",
+ "normalized_hypothesis": "its diameter was not more than 200 yards",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0462",
+ "category": "phonetic_balance",
+ "text": "Nimrod replied, with a slight manifestation of sensitiveness.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0462.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "NIMROD REPLIED WITH A SLIGHT MANIFESTATION OF SENSITIVENESS",
+ "raw": {
+ "normalized_reference": "nimrod replied, with a slight manifestation of sensitiveness.",
+ "normalized_hypothesis": "nimrod replied with a slight manifestation of sensitiveness",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "nimrod replied with a slight manifestation of sensitiveness",
+ "normalized_hypothesis": "nimrod replied with a slight manifestation of sensitiveness",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0463",
+ "category": "phonetic_balance",
+ "text": "I tried to read George Moore last night, and was dreadfully bored.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0463.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I TRIED TO READ GEORGE MOORE LAST NIGHT AND WAS DREADFULLY BORED",
+ "raw": {
+ "normalized_reference": "i tried to read george moore last night, and was dreadfully bored.",
+ "normalized_hypothesis": "i tried to read george moore last night and was dreadfully bored",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i tried to read george moore last night and was dreadfully bored",
+ "normalized_hypothesis": "i tried to read george moore last night and was dreadfully bored",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0464",
+ "category": "phonetic_balance",
+ "text": "Yes, it was a man who asked, a stranger.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0464.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "YES IT WAS A MAN WHO ASKED A STRANGER",
+ "raw": {
+ "normalized_reference": "yes, it was a man who asked, a stranger.",
+ "normalized_hypothesis": "yes it was a man who asked a stranger",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "yes it was a man who asked a stranger",
+ "normalized_hypothesis": "yes it was a man who asked a stranger",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0465",
+ "category": "phonetic_balance",
+ "text": "We were now good friends.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0465.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WE WERE NOW GOOD FRIENDS",
+ "raw": {
+ "normalized_reference": "we were now good friends.",
+ "normalized_hypothesis": "we were now good friends",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "we were now good friends",
+ "normalized_hypothesis": "we were now good friends",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0466",
+ "category": "phonetic_balance",
+ "text": "Yet, in accordance with Ernest's test of truth, it worked.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0466.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "YET IN ACCORDANCE WITH ERNEST'S TEST OF TRUTH IT WORKED",
+ "raw": {
+ "normalized_reference": "yet, in accordance with ernest's test of truth, it worked.",
+ "normalized_hypothesis": "yet in accordance with ernest's test of truth it worked",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "yet in accordance with ernest is test of truth it worked",
+ "normalized_hypothesis": "yet in accordance with ernest is test of truth it worked",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0467",
+ "category": "phonetic_balance",
+ "text": "It's the strap hangers that'll keep us from going under.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0467.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT'S THE STRAP HANGERS THAT'LL KEEP US FROM GOING UNDER",
+ "raw": {
+ "normalized_reference": "it's the strap hangers that'll keep us from going under.",
+ "normalized_hypothesis": "it's the strap hangers that'll keep us from going under",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it is the strap hangers that will keep us from going under",
+ "normalized_hypothesis": "it is the strap hangers that will keep us from going under",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0468",
+ "category": "phonetic_balance",
+ "text": "I never allow what can't be changed to annoy me.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0468.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I NEVER ALLOW WHAT CAN'T BE CHANGED TO ANNOY ME",
+ "raw": {
+ "normalized_reference": "i never allow what can't be changed to annoy me.",
+ "normalized_hypothesis": "i never allow what can't be changed to annoy me",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i never allow what can not be changed to annoy me",
+ "normalized_hypothesis": "i never allow what can not be changed to annoy me",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0469",
+ "category": "phonetic_balance",
+ "text": "After all superfluous flesh is gone what is left is stringy and resistant.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0469.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "AFTER ALL SUPERFLUOUS FLESH IS GONE WHAT IS LEFT IS STRINGY AND RESISTANT",
+ "raw": {
+ "normalized_reference": "after all superfluous flesh is gone what is left is stringy and resistant.",
+ "normalized_hypothesis": "after all superfluous flesh is gone what is left is stringy and resistant",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "after all superfluous flesh is gone what is left is stringy and resistant",
+ "normalized_hypothesis": "after all superfluous flesh is gone what is left is stringy and resistant",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0470",
+ "category": "phonetic_balance",
+ "text": "Then he shouted, Shut up.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0470.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THEN HE SHOUTED SHUT UP",
+ "raw": {
+ "normalized_reference": "then he shouted, shut up.",
+ "normalized_hypothesis": "then he shouted shut up",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "then he shouted shut up",
+ "normalized_hypothesis": "then he shouted shut up",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0471",
+ "category": "phonetic_balance",
+ "text": "He had worshiped her, as Dante might have worshiped Beatrice.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0471.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE HAD WORSHIPPED HER AS DANTES MIGHT HAVE WORSHIPPED BEATRICE",
+ "raw": {
+ "normalized_reference": "he had worshiped her, as dante might have worshiped beatrice.",
+ "normalized_hypothesis": "he had worshipped her as dantes might have worshipped beatrice",
+ "errors": 3,
+ "words": 10,
+ "wer": 0.3
+ },
+ "semantic": {
+ "normalized_reference": "he had worshiped her as dante might have worshiped beatrice",
+ "normalized_hypothesis": "he had worshipped her as dantes might have worshipped beatrice",
+ "errors": 3,
+ "words": 10,
+ "wer": 0.3
+ }
+ },
+ {
+ "id": "release500_0472",
+ "category": "phonetic_balance",
+ "text": "He looked like one who had passed through an uncomfortable hour or two.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0472.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE LOOKED LIKE ONE WHO HAD PASSED THROUGH AN UNCOMFORTABLE HOUR OR TWO",
+ "raw": {
+ "normalized_reference": "he looked like one who had passed through an uncomfortable hour or two.",
+ "normalized_hypothesis": "he looked like one who had passed through an uncomfortable hour or two",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he looked like one who had passed through an uncomfortable hour or 2",
+ "normalized_hypothesis": "he looked like one who had passed through an uncomfortable hour or 2",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0473",
+ "category": "phonetic_balance",
+ "text": "The history of our westward-faring race is written in it.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0473.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE HISTORY OF OUR WESTWARD FARING RACE IS WRIT IN IT",
+ "raw": {
+ "normalized_reference": "the history of our westward-faring race is written in it.",
+ "normalized_hypothesis": "the history of our westward faring race is writ in it",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ },
+ "semantic": {
+ "normalized_reference": "the history of our westward faring race is written in it",
+ "normalized_hypothesis": "the history of our westward faring race is writ in it",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ }
+ },
+ {
+ "id": "release500_0474",
+ "category": "phonetic_balance",
+ "text": "He may desire to escape pain, or to enjoy pleasure.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0474.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE MAY DESIRE TO ESCAPE PAIN OR TO ENJOY PLEASURE",
+ "raw": {
+ "normalized_reference": "he may desire to escape pain, or to enjoy pleasure.",
+ "normalized_hypothesis": "he may desire to escape pain or to enjoy pleasure",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he may desire to escape pain or to enjoy pleasure",
+ "normalized_hypothesis": "he may desire to escape pain or to enjoy pleasure",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0475",
+ "category": "phonetic_balance",
+ "text": "And the air was growing chilly.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0475.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "AND THE AIR WAS GROWING CHILLY",
+ "raw": {
+ "normalized_reference": "and the air was growing chilly.",
+ "normalized_hypothesis": "and the air was growing chilly",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "and the air was growing chilly",
+ "normalized_hypothesis": "and the air was growing chilly",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0476",
+ "category": "phonetic_balance",
+ "text": "Dig in; you're sure good, was Daylight's answer.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0476.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "DIGGIN YO'RE SURE GOOD WAS DAYLIGHT'S ANSWER",
+ "raw": {
+ "normalized_reference": "dig in; you're sure good, was daylight's answer.",
+ "normalized_hypothesis": "diggin yo're sure good was daylight's answer",
+ "errors": 3,
+ "words": 8,
+ "wer": 0.375
+ },
+ "semantic": {
+ "normalized_reference": "dig in you are sure good was daylight is answer",
+ "normalized_hypothesis": "diggin yo are sure good was daylight is answer",
+ "errors": 3,
+ "words": 10,
+ "wer": 0.3
+ }
+ },
+ {
+ "id": "release500_0477",
+ "category": "phonetic_balance",
+ "text": "My idea was that he would have more influence over the natives.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0477.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "MY IDEA WAS THAT HE WOULD HAVE MORE INFLUENCE OVER THE NATIVES",
+ "raw": {
+ "normalized_reference": "my idea was that he would have more influence over the natives.",
+ "normalized_hypothesis": "my idea was that he would have more influence over the natives",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "my idea was that he would have more influence over the natives",
+ "normalized_hypothesis": "my idea was that he would have more influence over the natives",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0478",
+ "category": "phonetic_balance",
+ "text": "Now run along, and tell them to hurry.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0478.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "NOW RUN ALONG AND TELL THEM TO HURRY",
+ "raw": {
+ "normalized_reference": "now run along, and tell them to hurry.",
+ "normalized_hypothesis": "now run along and tell them to hurry",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "now run along and tell them to hurry",
+ "normalized_hypothesis": "now run along and tell them to hurry",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0479",
+ "category": "phonetic_balance",
+ "text": "I have no idea, replied Philip.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0479.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I HAVE NO IDEA REPLIED PHILIP",
+ "raw": {
+ "normalized_reference": "i have no idea, replied philip.",
+ "normalized_hypothesis": "i have no idea replied philip",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i have no idea replied philip",
+ "normalized_hypothesis": "i have no idea replied philip",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0480",
+ "category": "phonetic_balance",
+ "text": "For the rest, he was a mere automaton.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0480.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "FOR THE REST HE WAS A MERE AUTOMATON",
+ "raw": {
+ "normalized_reference": "for the rest, he was a mere automaton.",
+ "normalized_hypothesis": "for the rest he was a mere automaton",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "for the rest he was a mere automaton",
+ "normalized_hypothesis": "for the rest he was a mere automaton",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0481",
+ "category": "phonetic_balance",
+ "text": "Why, the average review is more nauseating than cod liver oil.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0481.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WHY THE AVERAGE REVIEW IS MORE NAUSEATING THAN COD LIVER OIL",
+ "raw": {
+ "normalized_reference": "why, the average review is more nauseating than cod liver oil.",
+ "normalized_hypothesis": "why the average review is more nauseating than cod liver oil",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "why the average review is more nauseating than cod liver oil",
+ "normalized_hypothesis": "why the average review is more nauseating than cod liver oil",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0482",
+ "category": "phonetic_balance",
+ "text": "I have been robbed, sir, I amended.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0482.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I HAVE BEEN ROBBED SIR I AMENDED",
+ "raw": {
+ "normalized_reference": "i have been robbed, sir, i amended.",
+ "normalized_hypothesis": "i have been robbed sir i amended",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i have been robbed sir i amended",
+ "normalized_hypothesis": "i have been robbed sir i amended",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0483",
+ "category": "phonetic_balance",
+ "text": "He moved away as quietly as he had come.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0483.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE MOVED AWAY AS QUIETLY AS HE HAD COME",
+ "raw": {
+ "normalized_reference": "he moved away as quietly as he had come.",
+ "normalized_hypothesis": "he moved away as quietly as he had come",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he moved away as quietly as he had come",
+ "normalized_hypothesis": "he moved away as quietly as he had come",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0484",
+ "category": "phonetic_balance",
+ "text": "A trickle of fresh blood ran over his face.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0484.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "A TRICKLE OF FRESH BLOOD RAN OVER HIS FACE",
+ "raw": {
+ "normalized_reference": "a trickle of fresh blood ran over his face.",
+ "normalized_hypothesis": "a trickle of fresh blood ran over his face",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "a trickle of fresh blood ran over his face",
+ "normalized_hypothesis": "a trickle of fresh blood ran over his face",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0485",
+ "category": "phonetic_balance",
+ "text": "Suppose you saw me at work through the window.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0485.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SUPPOSE YOU SAW ME AT WORK THROUGH THE WINDOW",
+ "raw": {
+ "normalized_reference": "suppose you saw me at work through the window.",
+ "normalized_hypothesis": "suppose you saw me at work through the window",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "suppose you saw me at work through the window",
+ "normalized_hypothesis": "suppose you saw me at work through the window",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0486",
+ "category": "phonetic_balance",
+ "text": "But who was Eileen's double.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0486.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "BUT WHO WAS ILEEN'S DOUBLE",
+ "raw": {
+ "normalized_reference": "but who was eileen's double.",
+ "normalized_hypothesis": "but who was ileen's double",
+ "errors": 1,
+ "words": 5,
+ "wer": 0.2
+ },
+ "semantic": {
+ "normalized_reference": "but who was eileen is double",
+ "normalized_hypothesis": "but who was ileen is double",
+ "errors": 1,
+ "words": 6,
+ "wer": 0.16666666666666666
+ }
+ },
+ {
+ "id": "release500_0487",
+ "category": "phonetic_balance",
+ "text": "Darkness hid him from Jeanne.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0487.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "DARKNESS HID HIM FROM JEAN",
+ "raw": {
+ "normalized_reference": "darkness hid him from jeanne.",
+ "normalized_hypothesis": "darkness hid him from jean",
+ "errors": 1,
+ "words": 5,
+ "wer": 0.2
+ },
+ "semantic": {
+ "normalized_reference": "darkness hid him from jeanne",
+ "normalized_hypothesis": "darkness hid him from jean",
+ "errors": 1,
+ "words": 5,
+ "wer": 0.2
+ }
+ },
+ {
+ "id": "release500_0488",
+ "category": "phonetic_balance",
+ "text": "By this time Charley was as enraged as the Greek.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0488.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "BY THIS TIME CHARLEY WAS AS ENRAGED AS THE GREEK",
+ "raw": {
+ "normalized_reference": "by this time charley was as enraged as the greek.",
+ "normalized_hypothesis": "by this time charley was as enraged as the greek",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "by this time charley was as enraged as the greek",
+ "normalized_hypothesis": "by this time charley was as enraged as the greek",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0489",
+ "category": "phonetic_balance",
+ "text": "I am writing these lines in Honolulu, Hawaii.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0489.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I AM WRITING THESE LINES IN HONOLULU HAWAI",
+ "raw": {
+ "normalized_reference": "i am writing these lines in honolulu, hawaii.",
+ "normalized_hypothesis": "i am writing these lines in honolulu hawai",
+ "errors": 1,
+ "words": 8,
+ "wer": 0.125
+ },
+ "semantic": {
+ "normalized_reference": "i am writing these lines in honolulu hawaii",
+ "normalized_hypothesis": "i am writing these lines in honolulu hawai",
+ "errors": 1,
+ "words": 8,
+ "wer": 0.125
+ }
+ },
+ {
+ "id": "release500_0490",
+ "category": "phonetic_balance",
+ "text": "Nobody knows how the natives got them.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0490.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "NOBODY KNOWS HOW THE NATIVES GOT THEM",
+ "raw": {
+ "normalized_reference": "nobody knows how the natives got them.",
+ "normalized_hypothesis": "nobody knows how the natives got them",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "nobody knows how the natives got them",
+ "normalized_hypothesis": "nobody knows how the natives got them",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0491",
+ "category": "phonetic_balance",
+ "text": "And how would we ever find ourselves.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0491.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "AND HOW WOULD WE EVER FIND OURSELVES",
+ "raw": {
+ "normalized_reference": "and how would we ever find ourselves.",
+ "normalized_hypothesis": "and how would we ever find ourselves",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "and how would we ever find ourselves",
+ "normalized_hypothesis": "and how would we ever find ourselves",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0492",
+ "category": "phonetic_balance",
+ "text": "They are coming ashore, whoever they are.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0492.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THEY ARE COMING ASHORE WHOEVER THEY ARE",
+ "raw": {
+ "normalized_reference": "they are coming ashore, whoever they are.",
+ "normalized_hypothesis": "they are coming ashore whoever they are",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "they are coming ashore whoever they are",
+ "normalized_hypothesis": "they are coming ashore whoever they are",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0493",
+ "category": "phonetic_balance",
+ "text": "This is my fifth voyage.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0493.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THIS IS MY FIFTH VOYAGE",
+ "raw": {
+ "normalized_reference": "this is my fifth voyage.",
+ "normalized_hypothesis": "this is my fifth voyage",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "this is my 5th voyage",
+ "normalized_hypothesis": "this is my 5th voyage",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0494",
+ "category": "phonetic_balance",
+ "text": "And yet if she came he had no words to say.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0494.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "AND YET IF SHE CAME HE HAD NO WORDS TO SAY",
+ "raw": {
+ "normalized_reference": "and yet if she came he had no words to say.",
+ "normalized_hypothesis": "and yet if she came he had no words to say",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "and yet if she came he had no words to say",
+ "normalized_hypothesis": "and yet if she came he had no words to say",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0495",
+ "category": "phonetic_balance",
+ "text": "You must sleep, he urged.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0495.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "YOU MUST SLEEP HE URGED",
+ "raw": {
+ "normalized_reference": "you must sleep, he urged.",
+ "normalized_hypothesis": "you must sleep he urged",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "you must sleep he urged",
+ "normalized_hypothesis": "you must sleep he urged",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0496",
+ "category": "phonetic_balance",
+ "text": "Could the incident have anything to do with Jeanne and Pierre.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0496.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "COULD THE INCIDENT HAVE ANYTHING TO DO WITH JEAN AND PIERRE",
+ "raw": {
+ "normalized_reference": "could the incident have anything to do with jeanne and pierre.",
+ "normalized_hypothesis": "could the incident have anything to do with jean and pierre",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ },
+ "semantic": {
+ "normalized_reference": "could the incident have anything to do with jeanne and pierre",
+ "normalized_hypothesis": "could the incident have anything to do with jean and pierre",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ }
+ },
+ {
+ "id": "release500_0497",
+ "category": "phonetic_balance",
+ "text": "Each day she became a more vital part of him.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0497.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "EACH DAY SHE BECAME A MORE VITAL PART OF HIM",
+ "raw": {
+ "normalized_reference": "each day she became a more vital part of him.",
+ "normalized_hypothesis": "each day she became a more vital part of him",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "each day she became a more vital part of him",
+ "normalized_hypothesis": "each day she became a more vital part of him",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0498",
+ "category": "phonetic_balance",
+ "text": "But why continue the tirade, for tirade it was.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0498.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "BUT WHY CONTINUE THE TIRADE FOR TIRADE IT WAS",
+ "raw": {
+ "normalized_reference": "but why continue the tirade, for tirade it was.",
+ "normalized_hypothesis": "but why continue the tirade for tirade it was",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "but why continue the tirade for tirade it was",
+ "normalized_hypothesis": "but why continue the tirade for tirade it was",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0499",
+ "category": "phonetic_balance",
+ "text": "That is what distinguishes all of us from the lower animals.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0499.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THAT IS WHAT DISTINGUISHES ALL OF US FROM THE LOWER ANIMALS",
+ "raw": {
+ "normalized_reference": "that is what distinguishes all of us from the lower animals.",
+ "normalized_hypothesis": "that is what distinguishes all of us from the lower animals",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "that is what distinguishes all of us from the lower animals",
+ "normalized_hypothesis": "that is what distinguishes all of us from the lower animals",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0500",
+ "category": "phonetic_balance",
+ "text": "There was something pathetic in the girl's attitude now.",
+ "model": "kitten-nano-bruno",
+ "audio": "benchmark/audio/kitten-nano-bruno/release500_0500.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THERE WAS SOMETHING PATHETIC IN THE GIRL'S ATTITUDE NOW",
+ "raw": {
+ "normalized_reference": "there was something pathetic in the girl's attitude now.",
+ "normalized_hypothesis": "there was something pathetic in the girl's attitude now",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "there was something pathetic in the girl is attitude now",
+ "normalized_hypothesis": "there was something pathetic in the girl is attitude now",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0001",
+ "category": "conversational",
+ "text": "It wasn't until later that I realized what had actually happened.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0001.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT WAS UNTIL LATER THAT I REALIZED WHAT HAD ACTUALLY HAPPENED",
+ "raw": {
+ "normalized_reference": "it wasn't until later that i realized what had actually happened.",
+ "normalized_hypothesis": "it was until later that i realized what had actually happened",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ },
+ "semantic": {
+ "normalized_reference": "it was not until later that i realized what had actually happened",
+ "normalized_hypothesis": "it was until later that i realized what had actually happened",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ }
+ },
+ {
+ "id": "release500_0002",
+ "category": "descriptive",
+ "text": "The light at that time of afternoon is perfect for about twenty minutes.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0002.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE LIGHT AT THAT TIME OF AFTERNOON IS PERFECT FOR ABOUT TWENTY MINUTES",
+ "raw": {
+ "normalized_reference": "the light at that time of afternoon is perfect for about twenty minutes.",
+ "normalized_hypothesis": "the light at that time of afternoon is perfect for about twenty minutes",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the light at that time of afternoon is perfect for about 20 minutes",
+ "normalized_hypothesis": "the light at that time of afternoon is perfect for about 20 minutes",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0003",
+ "category": "emotional",
+ "text": "That was the scariest thing I've ever done, and also maybe the best.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0003.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THAT WAS THE SCARIEST THING I'VE EVER DONE AND ALSO MAY BE THE BEST",
+ "raw": {
+ "normalized_reference": "that was the scariest thing i've ever done, and also maybe the best.",
+ "normalized_hypothesis": "that was the scariest thing i've ever done and also may be the best",
+ "errors": 2,
+ "words": 13,
+ "wer": 0.15384615384615385
+ },
+ "semantic": {
+ "normalized_reference": "that was the scariest thing i have ever done and also maybe the best",
+ "normalized_hypothesis": "that was the scariest thing i have ever done and also may be the best",
+ "errors": 2,
+ "words": 14,
+ "wer": 0.14285714285714285
+ }
+ },
+ {
+ "id": "release500_0004",
+ "category": "homographs",
+ "text": "After I record the record, please present the present to the bass player.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0004.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "AFTER I RECORD THE RECORD PLEASE PRESENT THE PRESENT TO THE BASE PLAYER",
+ "raw": {
+ "normalized_reference": "after i record the record, please present the present to the bass player.",
+ "normalized_hypothesis": "after i record the record please present the present to the base player",
+ "errors": 1,
+ "words": 13,
+ "wer": 0.07692307692307693
+ },
+ "semantic": {
+ "normalized_reference": "after i record the record please present the present to the bass player",
+ "normalized_hypothesis": "after i record the record please present the present to the base player",
+ "errors": 1,
+ "words": 13,
+ "wer": 0.07692307692307693
+ }
+ },
+ {
+ "id": "release500_0005",
+ "category": "long",
+ "text": "If I'm being completely honest, I had no idea what I was doing for most of it, and I think that might be the most normal thing about the whole experience.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0005.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IF I'M BEING COMPLETELY HONEST I HAD NO IDEA WHAT I WAS DOING FOR MOST OF IT AND I THINK THAT MIGHT BE THE MOST NORMAL THING ABOUT THE WHOLE EXPERIENCE",
+ "raw": {
+ "normalized_reference": "if i'm being completely honest, i had no idea what i was doing for most of it, and i think that might be the most normal thing about the whole experience.",
+ "normalized_hypothesis": "if i'm being completely honest i had no idea what i was doing for most of it and i think that might be the most normal thing about the whole experience",
+ "errors": 0,
+ "words": 31,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "if i am being completely honest i had no idea what i was doing for most of it and i think that might be the most normal thing about the whole experience",
+ "normalized_hypothesis": "if i am being completely honest i had no idea what i was doing for most of it and i think that might be the most normal thing about the whole experience",
+ "errors": 0,
+ "words": 32,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0006",
+ "category": "long_range",
+ "text": "The committee postponed its decision after discovering that three pages of the report had been printed in the wrong order.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0006.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE COMMITTEE POSTPONED ITS DECISION AFTER DISCOVERING THAT THREE PAGES OF THE REPORT HAD BEEN PRINTED IN THE WRONG ORDER",
+ "raw": {
+ "normalized_reference": "the committee postponed its decision after discovering that three pages of the report had been printed in the wrong order.",
+ "normalized_hypothesis": "the committee postponed its decision after discovering that three pages of the report had been printed in the wrong order",
+ "errors": 0,
+ "words": 20,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the committee postponed its decision after discovering that 3 pages of the report had been printed in the wrong order",
+ "normalized_hypothesis": "the committee postponed its decision after discovering that 3 pages of the report had been printed in the wrong order",
+ "errors": 0,
+ "words": 20,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0007",
+ "category": "names_places",
+ "text": "Gwendolyn photographed the eucalyptus trees outside Ljubljana.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0007.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "GWENDOLEN PHOTOGRAPHED THE EUCALYPTUS TREES OUTSIDE LUBLIONA",
+ "raw": {
+ "normalized_reference": "gwendolyn photographed the eucalyptus trees outside ljubljana.",
+ "normalized_hypothesis": "gwendolen photographed the eucalyptus trees outside lubliona",
+ "errors": 2,
+ "words": 7,
+ "wer": 0.2857142857142857
+ },
+ "semantic": {
+ "normalized_reference": "gwendolyn photographed the eucalyptus trees outside ljubljana",
+ "normalized_hypothesis": "gwendolen photographed the eucalyptus trees outside lubliona",
+ "errors": 2,
+ "words": 7,
+ "wer": 0.2857142857142857
+ }
+ },
+ {
+ "id": "release500_0008",
+ "category": "narrative",
+ "text": "He had this theory, and the more time passes, the more I think he was right.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0008.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE HAD THIS THEORY AND THE MORE TIME PASSES THE MORE I THINK HE WAS RIGHT",
+ "raw": {
+ "normalized_reference": "he had this theory, and the more time passes, the more i think he was right.",
+ "normalized_hypothesis": "he had this theory and the more time passes the more i think he was right",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he had this theory and the more time passes the more i think he was right",
+ "normalized_hypothesis": "he had this theory and the more time passes the more i think he was right",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0009",
+ "category": "numbers",
+ "text": "The package weighs twelve point six kilograms and arrived on July twenty-first.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0009.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE PACKAGE WEIGHS TWELVE POINT SIX KILOGRAMS AND ARRIVED ON JULY TWENTY FIRST",
+ "raw": {
+ "normalized_reference": "the package weighs twelve point six kilograms and arrived on july twenty-first.",
+ "normalized_hypothesis": "the package weighs twelve point six kilograms and arrived on july twenty first",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the package weighs 12.6 kilograms and arrived on july 21st",
+ "normalized_hypothesis": "the package weighs 12.6 kilograms and arrived on july 21st",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0010",
+ "category": "punctuation",
+ "text": "First, close the window; second, turn off the lamp; finally, lock the door.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0010.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "FIRST CLOSE THE WINDOW SECOND TURN OFF THE LAMP FINALLY LOCK THE DOOR",
+ "raw": {
+ "normalized_reference": "first, close the window; second, turn off the lamp; finally, lock the door.",
+ "normalized_hypothesis": "first close the window second turn off the lamp finally lock the door",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "1st close the window 2nd turn off the lamp finally lock the door",
+ "normalized_hypothesis": "1st close the window 2nd turn off the lamp finally lock the door",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0011",
+ "category": "question",
+ "text": "When did you first realize something was wrong?",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0011.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WHEN DID YOU FIRST REALIZE SOMETHING WAS WRONG",
+ "raw": {
+ "normalized_reference": "when did you first realize something was wrong?",
+ "normalized_hypothesis": "when did you first realize something was wrong",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "when did you 1st realize something was wrong",
+ "normalized_hypothesis": "when did you 1st realize something was wrong",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0012",
+ "category": "rare_words",
+ "text": "The archaeologist catalogued a Byzantine reliquary and several anachronistic fragments.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0012.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE ARCHAEOLOGIST CATALOGUED A BIZANTINE RELIQUARY AND SEVERAL ANICRONISTIC FRAGMENTS",
+ "raw": {
+ "normalized_reference": "the archaeologist catalogued a byzantine reliquary and several anachronistic fragments.",
+ "normalized_hypothesis": "the archaeologist catalogued a bizantine reliquary and several anicronistic fragments",
+ "errors": 2,
+ "words": 10,
+ "wer": 0.2
+ },
+ "semantic": {
+ "normalized_reference": "the archaeologist catalogued a byzantine reliquary and several anachronistic fragments",
+ "normalized_hypothesis": "the archaeologist catalogued a bizantine reliquary and several anicronistic fragments",
+ "errors": 2,
+ "words": 10,
+ "wer": 0.2
+ }
+ },
+ {
+ "id": "release500_0013",
+ "category": "short",
+ "text": "I have no idea.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0013.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I HAVE NO IDEA",
+ "raw": {
+ "normalized_reference": "i have no idea.",
+ "normalized_hypothesis": "i have no idea",
+ "errors": 0,
+ "words": 4,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i have no idea",
+ "normalized_hypothesis": "i have no idea",
+ "errors": 0,
+ "words": 4,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0014",
+ "category": "technical",
+ "text": "The system runs on three core components that all have to stay in sync.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0014.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE SYSTEM RUNS ON THREE CORPS COMPONENTS THAT ALL HAVE TO STAY AND SINK",
+ "raw": {
+ "normalized_reference": "the system runs on three core components that all have to stay in sync.",
+ "normalized_hypothesis": "the system runs on three corps components that all have to stay and sink",
+ "errors": 3,
+ "words": 14,
+ "wer": 0.21428571428571427
+ },
+ "semantic": {
+ "normalized_reference": "the system runs on 3 core components that all have to stay in sync",
+ "normalized_hypothesis": "the system runs on 3 corps components that all have to stay and sink",
+ "errors": 3,
+ "words": 14,
+ "wer": 0.21428571428571427
+ }
+ },
+ {
+ "id": "release500_0015",
+ "category": "conversational",
+ "text": "We improvised the entire second half and somehow nobody noticed.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0015.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WE IMPROVISED THE ENTIRE SECOND HALF AND SOMEHOW NOBODY NOTICED",
+ "raw": {
+ "normalized_reference": "we improvised the entire second half and somehow nobody noticed.",
+ "normalized_hypothesis": "we improvised the entire second half and somehow nobody noticed",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "we improvised the entire 2nd half and somehow nobody noticed",
+ "normalized_hypothesis": "we improvised the entire 2nd half and somehow nobody noticed",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0016",
+ "category": "descriptive",
+ "text": "There's a smell in old buildings that's impossible to describe but immediately familiar.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0016.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THERE'S A SMELL IN OLD BUILDINGS THAT'S IMPOSSIBLE TO DESCRIBE BUT IMMEDIATELY FAMILIAR",
+ "raw": {
+ "normalized_reference": "there's a smell in old buildings that's impossible to describe but immediately familiar.",
+ "normalized_hypothesis": "there's a smell in old buildings that's impossible to describe but immediately familiar",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "there is a smell in old buildings that is impossible to describe but immediately familiar",
+ "normalized_hypothesis": "there is a smell in old buildings that is impossible to describe but immediately familiar",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0017",
+ "category": "emotional",
+ "text": "I specifically asked for one thing. One thing. And it still didn't happen.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0017.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I SPECIFICALLY ASKED FOR ONE THING ONE THING AND IT STILL DIDN'T HAPPEN",
+ "raw": {
+ "normalized_reference": "i specifically asked for one thing. one thing. and it still didn't happen.",
+ "normalized_hypothesis": "i specifically asked for one thing one thing and it still didn't happen",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i specifically asked for one thing one thing and it still did not happen",
+ "normalized_hypothesis": "i specifically asked for one thing one thing and it still did not happen",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0018",
+ "category": "homographs",
+ "text": "They refuse to place the refuse beside the produce aisle.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0018.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THEY REFUSED TO PLACE THE REFUSE BESIDE THE PRODUCE ISLE",
+ "raw": {
+ "normalized_reference": "they refuse to place the refuse beside the produce aisle.",
+ "normalized_hypothesis": "they refused to place the refuse beside the produce isle",
+ "errors": 2,
+ "words": 10,
+ "wer": 0.2
+ },
+ "semantic": {
+ "normalized_reference": "they refuse to place the refuse beside the produce aisle",
+ "normalized_hypothesis": "they refused to place the refuse beside the produce isle",
+ "errors": 2,
+ "words": 10,
+ "wer": 0.2
+ }
+ },
+ {
+ "id": "release500_0019",
+ "category": "long",
+ "text": "There were days last year when I genuinely didn't know how I was going to get through it, and somehow I did, and I still don't fully understand how.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0019.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THERE WERE DAYS LAST YEAR WHEN I GENUINELY DIDN'T KNOW HOW I WAS GOING TO GET THROUGH IT AND SOMEHOW I DID AND I STILL DON'T FULLY UNDERSTAND HOW",
+ "raw": {
+ "normalized_reference": "there were days last year when i genuinely didn't know how i was going to get through it, and somehow i did, and i still don't fully understand how.",
+ "normalized_hypothesis": "there were days last year when i genuinely didn't know how i was going to get through it and somehow i did and i still don't fully understand how",
+ "errors": 0,
+ "words": 29,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "there were days last year when i genuinely did not know how i was going to get through it and somehow i did and i still do not fully understand how",
+ "normalized_hypothesis": "there were days last year when i genuinely did not know how i was going to get through it and somehow i did and i still do not fully understand how",
+ "errors": 0,
+ "words": 31,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0020",
+ "category": "long_range",
+ "text": "Before announcing the winner, the judge thanked the volunteers who had worked through the storm to repair the damaged stage.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0020.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "BEFORE ANNOUNCING THE WINNER THE JUDGE THANKED THE VOLUNTEERS WHO HAD WORKED THROUGH THE STORM TO REPAIR THE DAMAGED STAGE",
+ "raw": {
+ "normalized_reference": "before announcing the winner, the judge thanked the volunteers who had worked through the storm to repair the damaged stage.",
+ "normalized_hypothesis": "before announcing the winner the judge thanked the volunteers who had worked through the storm to repair the damaged stage",
+ "errors": 0,
+ "words": 20,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "before announcing the winner the judge thanked the volunteers who had worked through the storm to repair the damaged stage",
+ "normalized_hypothesis": "before announcing the winner the judge thanked the volunteers who had worked through the storm to repair the damaged stage",
+ "errors": 0,
+ "words": 20,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0021",
+ "category": "names_places",
+ "text": "Professor Nguyen mailed the fluorescent poster to Reykjavik on Thursday.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0021.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "PROFESSOR NEWYAN MAILED THE FLORESCENT POSTER TO RAGJEVIK ON THURSDAY",
+ "raw": {
+ "normalized_reference": "professor nguyen mailed the fluorescent poster to reykjavik on thursday.",
+ "normalized_hypothesis": "professor newyan mailed the florescent poster to ragjevik on thursday",
+ "errors": 3,
+ "words": 10,
+ "wer": 0.3
+ },
+ "semantic": {
+ "normalized_reference": "professor nguyen mailed the fluorescent poster to reykjavik on thursday",
+ "normalized_hypothesis": "professor newyan mailed the florescent poster to ragjevik on thursday",
+ "errors": 3,
+ "words": 10,
+ "wer": 0.3
+ }
+ },
+ {
+ "id": "release500_0022",
+ "category": "narrative",
+ "text": "The plan was simple. The execution was anything but.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0022.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE PLAN WAS SIMPLE THE EXECUTION WAS ANYTHING BUT",
+ "raw": {
+ "normalized_reference": "the plan was simple. the execution was anything but.",
+ "normalized_hypothesis": "the plan was simple the execution was anything but",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the plan was simple the execution was anything but",
+ "normalized_hypothesis": "the plan was simple the execution was anything but",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0023",
+ "category": "numbers",
+ "text": "Her score improved from eighty-two percent to ninety-seven percent in two weeks.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0023.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HER SCORE IMPROVED FROM EIGHTY TWO PER CENT TO NINETY SEVEN PER CENT IN TWO WEEKS",
+ "raw": {
+ "normalized_reference": "her score improved from eighty-two percent to ninety-seven percent in two weeks.",
+ "normalized_hypothesis": "her score improved from eighty two per cent to ninety seven per cent in two weeks",
+ "errors": 4,
+ "words": 14,
+ "wer": 0.2857142857142857
+ },
+ "semantic": {
+ "normalized_reference": "her score improved from 82% to 97% in 2 weeks",
+ "normalized_hypothesis": "her score improved from 82% to 97% in 2 weeks",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0024",
+ "category": "punctuation",
+ "text": "Why would anyone pack spoons, batteries, and a violin for one short trip?",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0024.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WHY WOULD ANYONE PACK SPOONS BATTERIES AND A VIOLIN FOR ONE SHORT TRIP",
+ "raw": {
+ "normalized_reference": "why would anyone pack spoons, batteries, and a violin for one short trip?",
+ "normalized_hypothesis": "why would anyone pack spoons batteries and a violin for one short trip",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "why would anyone pack spoons batteries and a violin for one short trip",
+ "normalized_hypothesis": "why would anyone pack spoons batteries and a violin for one short trip",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0025",
+ "category": "question",
+ "text": "How long have you been sitting on this?",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0025.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HOW LONG HAVE YOU BEEN SITTING ON THIS",
+ "raw": {
+ "normalized_reference": "how long have you been sitting on this?",
+ "normalized_hypothesis": "how long have you been sitting on this",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "how long have you been sitting on this",
+ "normalized_hypothesis": "how long have you been sitting on this",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0026",
+ "category": "rare_words",
+ "text": "A mischievous chameleon rested beside the chrysanthemum and the bougainvillea.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0026.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "A MISCHIEVOUS CAMELION RESTED BESIDE THE CHRYSANTHEMUM AND THE BAUGANVILLIA",
+ "raw": {
+ "normalized_reference": "a mischievous chameleon rested beside the chrysanthemum and the bougainvillea.",
+ "normalized_hypothesis": "a mischievous camelion rested beside the chrysanthemum and the bauganvillia",
+ "errors": 2,
+ "words": 10,
+ "wer": 0.2
+ },
+ "semantic": {
+ "normalized_reference": "a mischievous chameleon rested beside the chrysanthemum and the bougainvillea",
+ "normalized_hypothesis": "a mischievous camelion rested beside the chrysanthemum and the bauganvillia",
+ "errors": 2,
+ "words": 10,
+ "wer": 0.2
+ }
+ },
+ {
+ "id": "release500_0027",
+ "category": "short",
+ "text": "We don't talk about that.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0027.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WE DON'T TALK ABOUT THAT",
+ "raw": {
+ "normalized_reference": "we don't talk about that.",
+ "normalized_hypothesis": "we don't talk about that",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "we do not talk about that",
+ "normalized_hypothesis": "we do not talk about that",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0028",
+ "category": "technical",
+ "text": "Hash the UTF eight payload with S H A two fifty-six before opening the checkpoint.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0028.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HASH THE U T F EIGHT PAYLOAD WITH S H A TWO FIFTY SIX BEFORE OPENING THE CHECK POINT",
+ "raw": {
+ "normalized_reference": "hash the utf eight payload with s h a two fifty-six before opening the checkpoint.",
+ "normalized_hypothesis": "hash the u t f eight payload with s h a two fifty six before opening the check point",
+ "errors": 5,
+ "words": 16,
+ "wer": 0.3125
+ },
+ "semantic": {
+ "normalized_reference": "hash the utf 8 payload with s h a 256 before opening the checkpoint",
+ "normalized_hypothesis": "hash the u t f 8 payload with s h a 256 before opening the check point",
+ "errors": 5,
+ "words": 14,
+ "wer": 0.35714285714285715
+ }
+ },
+ {
+ "id": "release500_0029",
+ "category": "conversational",
+ "text": "I thought I had it all figured out, and then this happened.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0029.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I THOUGHT I HAD IT ALL FIGURED OUT AND THEN THIS HAPPENED",
+ "raw": {
+ "normalized_reference": "i thought i had it all figured out, and then this happened.",
+ "normalized_hypothesis": "i thought i had it all figured out and then this happened",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i thought i had it all figured out and then this happened",
+ "normalized_hypothesis": "i thought i had it all figured out and then this happened",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0030",
+ "category": "descriptive",
+ "text": "The kind of quiet you only get at three in the morning in a city.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0030.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE KIND OF QUIET YOU ONLY GET AT THREE IN THE MORNING IN A CITY",
+ "raw": {
+ "normalized_reference": "the kind of quiet you only get at three in the morning in a city.",
+ "normalized_hypothesis": "the kind of quiet you only get at three in the morning in a city",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the kind of quiet you only get at 3 in the morning in a city",
+ "normalized_hypothesis": "the kind of quiet you only get at 3 in the morning in a city",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0031",
+ "category": "emotional",
+ "text": "The joy of it was almost overwhelming, like too much good all at once.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0031.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE JOY OF IT WAS ALMOST OVERWHELMING LIKE TOO MUCH GOOD ALL AT ONCE",
+ "raw": {
+ "normalized_reference": "the joy of it was almost overwhelming, like too much good all at once.",
+ "normalized_hypothesis": "the joy of it was almost overwhelming like too much good all at once",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the joy of it was almost overwhelming like too much good all at once",
+ "normalized_hypothesis": "the joy of it was almost overwhelming like too much good all at once",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0032",
+ "category": "homographs",
+ "text": "Please close the door while I stand close to the microphone.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0032.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "PLEASE CLOSE THE DOOR WHILE I STAND CLOSE TO THE MICROPHONE",
+ "raw": {
+ "normalized_reference": "please close the door while i stand close to the microphone.",
+ "normalized_hypothesis": "please close the door while i stand close to the microphone",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "please close the door while i stand close to the microphone",
+ "normalized_hypothesis": "please close the door while i stand close to the microphone",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0033",
+ "category": "long",
+ "text": "I think the most underrated skill in the world is knowing when to stop explaining yourself and just let people draw their own conclusions.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0033.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I THINK THE MOST UNDERRATED SKILL IN THE WORLD IS KNOWING WHEN TO STOP EXPLAINING YOURSELF AND JUST LET PEOPLE DRAW THEIR OWN CONCLUSIONS",
+ "raw": {
+ "normalized_reference": "i think the most underrated skill in the world is knowing when to stop explaining yourself and just let people draw their own conclusions.",
+ "normalized_hypothesis": "i think the most underrated skill in the world is knowing when to stop explaining yourself and just let people draw their own conclusions",
+ "errors": 0,
+ "words": 24,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i think the most underrated skill in the world is knowing when to stop explaining yourself and just let people draw their own conclusions",
+ "normalized_hypothesis": "i think the most underrated skill in the world is knowing when to stop explaining yourself and just let people draw their own conclusions",
+ "errors": 0,
+ "words": 24,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0034",
+ "category": "long_range",
+ "text": "When the final bus disappeared around the corner, Nadia remembered that her umbrella, notebook, and lunch were still inside the quiet office.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0034.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WHEN THE FINAL BUST DISAPPEARED AROUND THE CORNER NODYA REMEMBERED THAT HER UMBRELLA NOTE BOOK AND LUNCH WERE STILL INSIDE THE QUIET OFFICE",
+ "raw": {
+ "normalized_reference": "when the final bus disappeared around the corner, nadia remembered that her umbrella, notebook, and lunch were still inside the quiet office.",
+ "normalized_hypothesis": "when the final bust disappeared around the corner nodya remembered that her umbrella note book and lunch were still inside the quiet office",
+ "errors": 4,
+ "words": 22,
+ "wer": 0.18181818181818182
+ },
+ "semantic": {
+ "normalized_reference": "when the final bus disappeared around the corner nadia remembered that her umbrella notebook and lunch were still inside the quiet office",
+ "normalized_hypothesis": "when the final bust disappeared around the corner nodya remembered that her umbrella note book and lunch were still inside the quiet office",
+ "errors": 4,
+ "words": 22,
+ "wer": 0.18181818181818182
+ }
+ },
+ {
+ "id": "release500_0035",
+ "category": "names_places",
+ "text": "Aurelia and Joaquin crossed the Schuylkill River before noon.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0035.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "ORELIAH AND JOKWIN CROSSED THE SHALKELL RIVER BEFORE NOON",
+ "raw": {
+ "normalized_reference": "aurelia and joaquin crossed the schuylkill river before noon.",
+ "normalized_hypothesis": "oreliah and jokwin crossed the shalkell river before noon",
+ "errors": 3,
+ "words": 9,
+ "wer": 0.3333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "aurelia and joaquin crossed the schuylkill river before noon",
+ "normalized_hypothesis": "oreliah and jokwin crossed the shalkell river before noon",
+ "errors": 3,
+ "words": 9,
+ "wer": 0.3333333333333333
+ }
+ },
+ {
+ "id": "release500_0036",
+ "category": "narrative",
+ "text": "The village had a name, but nobody used it. Everyone just called it home.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0036.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE VILLAGE HAD A NAME BUT NOBODY USED IT EVERYONE JUST CALLED IT HOME",
+ "raw": {
+ "normalized_reference": "the village had a name, but nobody used it. everyone just called it home.",
+ "normalized_hypothesis": "the village had a name but nobody used it everyone just called it home",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the village had a name but nobody used it everyone just called it home",
+ "normalized_hypothesis": "the village had a name but nobody used it everyone just called it home",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0037",
+ "category": "numbers",
+ "text": "Increase the gain by three decibels, wait five seconds, then reduce it by one.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0037.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "INCREASE THE GAIN BY THREE DESIBALS WEIGHT FIVE SECONDS THEN REDUCE IT BY ONE",
+ "raw": {
+ "normalized_reference": "increase the gain by three decibels, wait five seconds, then reduce it by one.",
+ "normalized_hypothesis": "increase the gain by three desibals weight five seconds then reduce it by one",
+ "errors": 2,
+ "words": 14,
+ "wer": 0.14285714285714285
+ },
+ "semantic": {
+ "normalized_reference": "increase the gain by 3 decibels wait 5 seconds then reduce it by one",
+ "normalized_hypothesis": "increase the gain by 3 desibals weight 5 seconds then reduce it by one",
+ "errors": 2,
+ "words": 14,
+ "wer": 0.14285714285714285
+ }
+ },
+ {
+ "id": "release500_0038",
+ "category": "punctuation",
+ "text": "The result was unexpected: clear, stable, and surprisingly natural.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0038.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE RESULT WAS UNEXPECTED CLEAR STABLE AND SURPRISINGLY NATURAL",
+ "raw": {
+ "normalized_reference": "the result was unexpected: clear, stable, and surprisingly natural.",
+ "normalized_hypothesis": "the result was unexpected clear stable and surprisingly natural",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the result was unexpected clear stable and surprisingly natural",
+ "normalized_hypothesis": "the result was unexpected clear stable and surprisingly natural",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0039",
+ "category": "question",
+ "text": "Who told you that was the right way to do it?",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0039.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WHO TOLD YOU THAT WAS THE RIGHT WAY TO DO IT",
+ "raw": {
+ "normalized_reference": "who told you that was the right way to do it?",
+ "normalized_hypothesis": "who told you that was the right way to do it",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "who told you that was the right way to do it",
+ "normalized_hypothesis": "who told you that was the right way to do it",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0040",
+ "category": "rare_words",
+ "text": "The meticulous cartographer examined an iridescent archipelago through binoculars.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0040.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE MATICULOUS CARTOGRAPHER EXAMINED AN IRIDESCENT ARCHIPELAGO THROUGH BONOCULARS",
+ "raw": {
+ "normalized_reference": "the meticulous cartographer examined an iridescent archipelago through binoculars.",
+ "normalized_hypothesis": "the maticulous cartographer examined an iridescent archipelago through bonoculars",
+ "errors": 2,
+ "words": 9,
+ "wer": 0.2222222222222222
+ },
+ "semantic": {
+ "normalized_reference": "the meticulous cartographer examined an iridescent archipelago through binoculars",
+ "normalized_hypothesis": "the maticulous cartographer examined an iridescent archipelago through bonoculars",
+ "errors": 2,
+ "words": 9,
+ "wer": 0.2222222222222222
+ }
+ },
+ {
+ "id": "release500_0041",
+ "category": "short",
+ "text": "I mean. Yeah. Kind of.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0041.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I MEAN YES KIND OF",
+ "raw": {
+ "normalized_reference": "i mean. yeah. kind of.",
+ "normalized_hypothesis": "i mean yes kind of",
+ "errors": 1,
+ "words": 5,
+ "wer": 0.2
+ },
+ "semantic": {
+ "normalized_reference": "i mean yeah kind of",
+ "normalized_hypothesis": "i mean yes kind of",
+ "errors": 1,
+ "words": 5,
+ "wer": 0.2
+ }
+ },
+ {
+ "id": "release500_0042",
+ "category": "technical",
+ "text": "The logs are your best friend when something behaves unexpectedly.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0042.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE LOGS ARE YOUR BEST FRIEND WHEN SOMETHING BEHAVES UNEXPECTEDLY",
+ "raw": {
+ "normalized_reference": "the logs are your best friend when something behaves unexpectedly.",
+ "normalized_hypothesis": "the logs are your best friend when something behaves unexpectedly",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the logs are your best friend when something behaves unexpectedly",
+ "normalized_hypothesis": "the logs are your best friend when something behaves unexpectedly",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0043",
+ "category": "conversational",
+ "text": "He kept talking and I kept nodding and neither of us was listening.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0043.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE KEPT TALKING AND I KEPT NODDING AND NEITHER OF US WAS LISTENING",
+ "raw": {
+ "normalized_reference": "he kept talking and i kept nodding and neither of us was listening.",
+ "normalized_hypothesis": "he kept talking and i kept nodding and neither of us was listening",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he kept talking and i kept nodding and neither of us was listening",
+ "normalized_hypothesis": "he kept talking and i kept nodding and neither of us was listening",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0044",
+ "category": "descriptive",
+ "text": "It's one of those places where everyone seems to know each other already.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0044.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT'S ONE OF THOSE PLACES WHERE EVERYONE SEEMS TO KNOW EACH OTHER ALREADY",
+ "raw": {
+ "normalized_reference": "it's one of those places where everyone seems to know each other already.",
+ "normalized_hypothesis": "it's one of those places where everyone seems to know each other already",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it is one of those places where everyone seems to know each other already",
+ "normalized_hypothesis": "it is one of those places where everyone seems to know each other already",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0045",
+ "category": "emotional",
+ "text": "He said he was proud of me and nobody had ever said that to me before.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0045.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE SAID HE WAS PROUD OF ME AND NOBODY HAD EVER SAID THAT TO ME BEFORE",
+ "raw": {
+ "normalized_reference": "he said he was proud of me and nobody had ever said that to me before.",
+ "normalized_hypothesis": "he said he was proud of me and nobody had ever said that to me before",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he said he was proud of me and nobody had ever said that to me before",
+ "normalized_hypothesis": "he said he was proud of me and nobody had ever said that to me before",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0046",
+ "category": "homographs",
+ "text": "The minute detail took a minute to explain clearly.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0046.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE MINUTE DETAIL TOOK A MINUTE TO EXPLAIN CLEARLY",
+ "raw": {
+ "normalized_reference": "the minute detail took a minute to explain clearly.",
+ "normalized_hypothesis": "the minute detail took a minute to explain clearly",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the minute detail took a minute to explain clearly",
+ "normalized_hypothesis": "the minute detail took a minute to explain clearly",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0047",
+ "category": "long",
+ "text": "The strange thing about really wanting something is that getting it never quite feels the way you imagined it would, and not getting it never quite feels as bad.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0047.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE STRANGE THING ABOUT REALLY WANTING SOMETHING IS THAT GETTING IT NEVER QUITE FEELS THE WAY YOU IMAGINED IT WOULD AND NOT GETTING IT NEVER QUITE FEELS AS BAD",
+ "raw": {
+ "normalized_reference": "the strange thing about really wanting something is that getting it never quite feels the way you imagined it would, and not getting it never quite feels as bad.",
+ "normalized_hypothesis": "the strange thing about really wanting something is that getting it never quite feels the way you imagined it would and not getting it never quite feels as bad",
+ "errors": 0,
+ "words": 29,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the strange thing about really wanting something is that getting it never quite feels the way you imagined it would and not getting it never quite feels as bad",
+ "normalized_hypothesis": "the strange thing about really wanting something is that getting it never quite feels the way you imagined it would and not getting it never quite feels as bad",
+ "errors": 0,
+ "words": 29,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0048",
+ "category": "long_range",
+ "text": "Even though the recipe looked simple at first, preparing the sauce required patience, careful timing, and a much larger pan.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0048.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "EVEN THOUGH THE RECIPE LOOKED SIMPLE AT FIRST PREPARING THE SAUCE REQUIRED PATIENCE CAREFUL TIMING AND A MUCH LARGER PAN",
+ "raw": {
+ "normalized_reference": "even though the recipe looked simple at first, preparing the sauce required patience, careful timing, and a much larger pan.",
+ "normalized_hypothesis": "even though the recipe looked simple at first preparing the sauce required patience careful timing and a much larger pan",
+ "errors": 0,
+ "words": 20,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "even though the recipe looked simple at 1st preparing the sauce required patience careful timing and a much larger pan",
+ "normalized_hypothesis": "even though the recipe looked simple at 1st preparing the sauce required patience careful timing and a much larger pan",
+ "errors": 0,
+ "words": 20,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0049",
+ "category": "names_places",
+ "text": "Xiomara met Siobhan near the old library in Worcestershire.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0049.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "ZIMARA MET CHAVONN NEAR THE OLD LIBRARY IN WORESTERSHIRE",
+ "raw": {
+ "normalized_reference": "xiomara met siobhan near the old library in worcestershire.",
+ "normalized_hypothesis": "zimara met chavonn near the old library in worestershire",
+ "errors": 3,
+ "words": 9,
+ "wer": 0.3333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "xiomara met siobhan near the old library in worcestershire",
+ "normalized_hypothesis": "zimara met chavonn near the old library in worestershire",
+ "errors": 3,
+ "words": 9,
+ "wer": 0.3333333333333333
+ }
+ },
+ {
+ "id": "release500_0050",
+ "category": "narrative",
+ "text": "She walked into the room and everyone went quiet, not because she was loud, but because she wasn't.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0050.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SHE WALKED INTO THE ROOM AND EVERYONE WENT QUIET NOT BECAUSE SHE WAS LOUD BUT BECAUSE SHE WAS",
+ "raw": {
+ "normalized_reference": "she walked into the room and everyone went quiet, not because she was loud, but because she wasn't.",
+ "normalized_hypothesis": "she walked into the room and everyone went quiet not because she was loud but because she was",
+ "errors": 1,
+ "words": 18,
+ "wer": 0.05555555555555555
+ },
+ "semantic": {
+ "normalized_reference": "she walked into the room and everyone went quiet not because she was loud but because she was not",
+ "normalized_hypothesis": "she walked into the room and everyone went quiet not because she was loud but because she was",
+ "errors": 1,
+ "words": 19,
+ "wer": 0.05263157894736842
+ }
+ },
+ {
+ "id": "release500_0051",
+ "category": "numbers",
+ "text": "At exactly seven forty-five, Morgan placed eighteen dollars and fifty cents beside the notebook.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0051.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "AT EXACTLY SEVEN FORTO FIVE MORGAN PLACED EIGHTEEN DOLLARS AND FIFTY CENTS BESIDE THE NOTE BOOK",
+ "raw": {
+ "normalized_reference": "at exactly seven forty-five, morgan placed eighteen dollars and fifty cents beside the notebook.",
+ "normalized_hypothesis": "at exactly seven forto five morgan placed eighteen dollars and fifty cents beside the note book",
+ "errors": 3,
+ "words": 15,
+ "wer": 0.2
+ },
+ "semantic": {
+ "normalized_reference": "at exactly 745 morgan placed $18.50 beside the notebook",
+ "normalized_hypothesis": "at exactly 7 forto 5 morgan placed $18.50 beside the note book",
+ "errors": 5,
+ "words": 10,
+ "wer": 0.5
+ }
+ },
+ {
+ "id": "release500_0052",
+ "category": "punctuation",
+ "text": "Does punctuation work? Yes: commas, semicolons, and periods should create sensible boundaries.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0052.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "DOES PUNCTUATION WORK YES CAMAS SEMICOLANS AND PERIODS SHOULD CREATE SENSIBLE BOUNDARIES",
+ "raw": {
+ "normalized_reference": "does punctuation work? yes: commas, semicolons, and periods should create sensible boundaries.",
+ "normalized_hypothesis": "does punctuation work yes camas semicolans and periods should create sensible boundaries",
+ "errors": 2,
+ "words": 12,
+ "wer": 0.16666666666666666
+ },
+ "semantic": {
+ "normalized_reference": "does punctuation work yes commas semicolons and periods should create sensible boundaries",
+ "normalized_hypothesis": "does punctuation work yes camas semicolans and periods should create sensible boundaries",
+ "errors": 2,
+ "words": 12,
+ "wer": 0.16666666666666666
+ }
+ },
+ {
+ "id": "release500_0053",
+ "category": "question",
+ "text": "Have you ever just stopped and thought about how strange all of this is?",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0053.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HAVE YOU EVER JUST STOPPED AND THOUGHT ABOUT HOW STRANGE ALL OF THIS IS",
+ "raw": {
+ "normalized_reference": "have you ever just stopped and thought about how strange all of this is?",
+ "normalized_hypothesis": "have you ever just stopped and thought about how strange all of this is",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "have you ever just stopped and thought about how strange all of this is",
+ "normalized_hypothesis": "have you ever just stopped and thought about how strange all of this is",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0054",
+ "category": "rare_words",
+ "text": "Her idiosyncratic explanation sounded plausible but remained fundamentally ambiguous.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0054.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HER IDIOSYNCRATIC EXPLANATIONS SOUNDED PLAUSIBLE BUT REMAINED FUNDAMENTALLY AMBIGUOUS",
+ "raw": {
+ "normalized_reference": "her idiosyncratic explanation sounded plausible but remained fundamentally ambiguous.",
+ "normalized_hypothesis": "her idiosyncratic explanations sounded plausible but remained fundamentally ambiguous",
+ "errors": 1,
+ "words": 9,
+ "wer": 0.1111111111111111
+ },
+ "semantic": {
+ "normalized_reference": "her idiosyncratic explanation sounded plausible but remained fundamentally ambiguous",
+ "normalized_hypothesis": "her idiosyncratic explanations sounded plausible but remained fundamentally ambiguous",
+ "errors": 1,
+ "words": 9,
+ "wer": 0.1111111111111111
+ }
+ },
+ {
+ "id": "release500_0055",
+ "category": "short",
+ "text": "That tracks.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0055.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THAT TRACKS",
+ "raw": {
+ "normalized_reference": "that tracks.",
+ "normalized_hypothesis": "that tracks",
+ "errors": 0,
+ "words": 2,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "that tracks",
+ "normalized_hypothesis": "that tracks",
+ "errors": 0,
+ "words": 2,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0056",
+ "category": "technical",
+ "text": "The key thing to remember is that order matters here more than you'd think.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0056.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE KEY THING TO REMEMBER IS THAT ORDER MATTERS HERE MORE THAN YOU' THINK",
+ "raw": {
+ "normalized_reference": "the key thing to remember is that order matters here more than you'd think.",
+ "normalized_hypothesis": "the key thing to remember is that order matters here more than you' think",
+ "errors": 1,
+ "words": 14,
+ "wer": 0.07142857142857142
+ },
+ "semantic": {
+ "normalized_reference": "the key thing to remember is that order matters here more than you would think",
+ "normalized_hypothesis": "the key thing to remember is that order matters here more than you think",
+ "errors": 1,
+ "words": 15,
+ "wer": 0.06666666666666667
+ }
+ },
+ {
+ "id": "release500_0057",
+ "category": "conversational",
+ "text": "I would not say that again in this room if I were you.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0057.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I WOULD NOT SAY THAT AGAIN IN THIS ROOM IF I WERE YOU",
+ "raw": {
+ "normalized_reference": "i would not say that again in this room if i were you.",
+ "normalized_hypothesis": "i would not say that again in this room if i were you",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i would not say that again in this room if i were you",
+ "normalized_hypothesis": "i would not say that again in this room if i were you",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0058",
+ "category": "descriptive",
+ "text": "The room was filled with the quiet that means something is about to change.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0058.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE ROOM WAS FILLED WITH THE QUIET THAT MEANS SOMETHING IS ABOUT TO CHANGE",
+ "raw": {
+ "normalized_reference": "the room was filled with the quiet that means something is about to change.",
+ "normalized_hypothesis": "the room was filled with the quiet that means something is about to change",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the room was filled with the quiet that means something is about to change",
+ "normalized_hypothesis": "the room was filled with the quiet that means something is about to change",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0059",
+ "category": "emotional",
+ "text": "I didn't expect to feel that way about it, but I did.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0059.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I DIDN'T EXPECT TO FEEL THAT WAY ABOUT IT BUT I DID",
+ "raw": {
+ "normalized_reference": "i didn't expect to feel that way about it, but i did.",
+ "normalized_hypothesis": "i didn't expect to feel that way about it but i did",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i did not expect to feel that way about it but i did",
+ "normalized_hypothesis": "i did not expect to feel that way about it but i did",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0060",
+ "category": "homographs",
+ "text": "I object to the strange object that was left near the entrance.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0060.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I OBJECT TO THE STRANGE OBJECT THAT WAS LEFT NEAR THE ENTRANCE",
+ "raw": {
+ "normalized_reference": "i object to the strange object that was left near the entrance.",
+ "normalized_hypothesis": "i object to the strange object that was left near the entrance",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i object to the strange object that was left near the entrance",
+ "normalized_hypothesis": "i object to the strange object that was left near the entrance",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0061",
+ "category": "long",
+ "text": "I've learned more from the things that went wrong than from the things that worked, which is a frustrating way to have to learn things but apparently the only way that stuck.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0061.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I'VE LEARNED MORE FROM THE THINGS THAT WENT WRONG THAN FROM THE THINGS THAT WORKED WHICH IS A FRUSTRATING WAY TO HAVE TO LEARN THINGS BUT APPARENTLY THE ONLY WAY THAT STUCK",
+ "raw": {
+ "normalized_reference": "i've learned more from the things that went wrong than from the things that worked, which is a frustrating way to have to learn things but apparently the only way that stuck.",
+ "normalized_hypothesis": "i've learned more from the things that went wrong than from the things that worked which is a frustrating way to have to learn things but apparently the only way that stuck",
+ "errors": 0,
+ "words": 32,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i have learned more from the things that went wrong than from the things that worked which is a frustrating way to have to learn things but apparently the only way that stuck",
+ "normalized_hypothesis": "i have learned more from the things that went wrong than from the things that worked which is a frustrating way to have to learn things but apparently the only way that stuck",
+ "errors": 0,
+ "words": 33,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0062",
+ "category": "long_range",
+ "text": "Because every nearby hotel was full, the exhausted hikers followed the narrow road until they found a farmhouse with one empty room.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0062.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "BECAUSE EVERY NEARBY HOTEL WAS FULL THE EXHAUSTED HYKERS FOLLOWED THE NARROW ROAD UNTIL THEY FOUND A FARMHOUSE WITH ONE EMPTY ROOM",
+ "raw": {
+ "normalized_reference": "because every nearby hotel was full, the exhausted hikers followed the narrow road until they found a farmhouse with one empty room.",
+ "normalized_hypothesis": "because every nearby hotel was full the exhausted hykers followed the narrow road until they found a farmhouse with one empty room",
+ "errors": 1,
+ "words": 22,
+ "wer": 0.045454545454545456
+ },
+ "semantic": {
+ "normalized_reference": "because every nearby hotel was full the exhausted hikers followed the narrow road until they found a farmhouse with one empty room",
+ "normalized_hypothesis": "because every nearby hotel was full the exhausted hykers followed the narrow road until they found a farmhouse with one empty room",
+ "errors": 1,
+ "words": 22,
+ "wer": 0.045454545454545456
+ }
+ },
+ {
+ "id": "release500_0063",
+ "category": "names_places",
+ "text": "The itinerary sends Beatrice from Albuquerque to Des Moines and then to Boise.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0063.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE ITINERARY SENDS BEATRICE FROM ALBAKIRKY TO DE MOINE AND THEN TO BOISY",
+ "raw": {
+ "normalized_reference": "the itinerary sends beatrice from albuquerque to des moines and then to boise.",
+ "normalized_hypothesis": "the itinerary sends beatrice from albakirky to de moine and then to boisy",
+ "errors": 4,
+ "words": 13,
+ "wer": 0.3076923076923077
+ },
+ "semantic": {
+ "normalized_reference": "the itinerary sends beatrice from albuquerque to des moines and then to boise",
+ "normalized_hypothesis": "the itinerary sends beatrice from albakirky to de moine and then to boisy",
+ "errors": 4,
+ "words": 13,
+ "wer": 0.3076923076923077
+ }
+ },
+ {
+ "id": "release500_0064",
+ "category": "narrative",
+ "text": "We had been driving for six hours when we finally saw the sign for the town.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0064.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WE HAD BEEN DRIVING FOR SIX HOURS WHEN WE FINALLY SAW THE SIGN FOR THE TOWN",
+ "raw": {
+ "normalized_reference": "we had been driving for six hours when we finally saw the sign for the town.",
+ "normalized_hypothesis": "we had been driving for six hours when we finally saw the sign for the town",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "we had been driving for 6 hours when we finally saw the sign for the town",
+ "normalized_hypothesis": "we had been driving for 6 hours when we finally saw the sign for the town",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0065",
+ "category": "numbers",
+ "text": "Call extension zero four nine after eight thirty in the morning.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0065.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "CALL EXTENSIONS IN FOR NINE AFTER EIGHT THIRTY IN THE MORNING",
+ "raw": {
+ "normalized_reference": "call extension zero four nine after eight thirty in the morning.",
+ "normalized_hypothesis": "call extensions in for nine after eight thirty in the morning",
+ "errors": 3,
+ "words": 11,
+ "wer": 0.2727272727272727
+ },
+ "semantic": {
+ "normalized_reference": "call extension 049 after 830 in the morning",
+ "normalized_hypothesis": "call extensions in for 9 after 830 in the morning",
+ "errors": 4,
+ "words": 8,
+ "wer": 0.5
+ }
+ },
+ {
+ "id": "release500_0066",
+ "category": "punctuation",
+ "text": "Stop - listen carefully - then answer the question without guessing.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0066.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "STOP LISTEN CAREFULLY THEN ANSWER THE QUESTION WITHOUT GUESSING",
+ "raw": {
+ "normalized_reference": "stop - listen carefully - then answer the question without guessing.",
+ "normalized_hypothesis": "stop listen carefully then answer the question without guessing",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "stop listen carefully then answer the question without guessing",
+ "normalized_hypothesis": "stop listen carefully then answer the question without guessing",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0067",
+ "category": "question",
+ "text": "Does it matter at this point, honestly?",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0067.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "DOES IT MATTER AT THIS POINT HONESTLY",
+ "raw": {
+ "normalized_reference": "does it matter at this point, honestly?",
+ "normalized_hypothesis": "does it matter at this point honestly",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "does it matter at this point honestly",
+ "normalized_hypothesis": "does it matter at this point honestly",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0068",
+ "category": "rare_words",
+ "text": "An entrepreneurial meteorologist described the unprecedented atmospheric phenomenon.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0068.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "AN ONTRIPRINURIAL METEOROLOGIST DESCRIBED THE UNPRECEDENTED ATMOSPHERIC PHENOMENON",
+ "raw": {
+ "normalized_reference": "an entrepreneurial meteorologist described the unprecedented atmospheric phenomenon.",
+ "normalized_hypothesis": "an ontriprinurial meteorologist described the unprecedented atmospheric phenomenon",
+ "errors": 1,
+ "words": 8,
+ "wer": 0.125
+ },
+ "semantic": {
+ "normalized_reference": "an entrepreneurial meteorologist described the unprecedented atmospheric phenomenon",
+ "normalized_hypothesis": "an ontriprinurial meteorologist described the unprecedented atmospheric phenomenon",
+ "errors": 1,
+ "words": 8,
+ "wer": 0.125
+ }
+ },
+ {
+ "id": "release500_0069",
+ "category": "short",
+ "text": "That's not what I expected.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0069.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THAT'S NOT WHAT I EXPECTED",
+ "raw": {
+ "normalized_reference": "that's not what i expected.",
+ "normalized_hypothesis": "that's not what i expected",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "that is not what i expected",
+ "normalized_hypothesis": "that is not what i expected",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0070",
+ "category": "technical",
+ "text": "There's a known issue with that version that causes failures under load.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0070.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THERE IS A KNOWN ISSUE WITH THAT VERSION THAT CAUSES FAILURES UNDER LOAD",
+ "raw": {
+ "normalized_reference": "there's a known issue with that version that causes failures under load.",
+ "normalized_hypothesis": "there is a known issue with that version that causes failures under load",
+ "errors": 2,
+ "words": 12,
+ "wer": 0.16666666666666666
+ },
+ "semantic": {
+ "normalized_reference": "there is a known issue with that version that causes failures under load",
+ "normalized_hypothesis": "there is a known issue with that version that causes failures under load",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0071",
+ "category": "conversational",
+ "text": "Don't look now but I think that's him over there.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0071.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "DON'T LOOK NOW BUT I THINK THAT'S HIM OVER THERE",
+ "raw": {
+ "normalized_reference": "don't look now but i think that's him over there.",
+ "normalized_hypothesis": "don't look now but i think that's him over there",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "do not look now but i think that is him over there",
+ "normalized_hypothesis": "do not look now but i think that is him over there",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0072",
+ "category": "descriptive",
+ "text": "The city at night looks completely different from how it feels during the day.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0072.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE CITY AT NIGHT LOOKS COMPLETELY DIFFERENT FROM HOW IT FEELS DURING THE DAY",
+ "raw": {
+ "normalized_reference": "the city at night looks completely different from how it feels during the day.",
+ "normalized_hypothesis": "the city at night looks completely different from how it feels during the day",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the city at night looks completely different from how it feels during the day",
+ "normalized_hypothesis": "the city at night looks completely different from how it feels during the day",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0073",
+ "category": "emotional",
+ "text": "It's the kind of happy that makes you nervous because you know it can't last.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0073.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT'S THE KIND OF HAPPY THAT MAKES YOU NERVOUS BECAUSE YOU KNOW IT CAN'T LAST",
+ "raw": {
+ "normalized_reference": "it's the kind of happy that makes you nervous because you know it can't last.",
+ "normalized_hypothesis": "it's the kind of happy that makes you nervous because you know it can't last",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it is the kind of happy that makes you nervous because you know it can not last",
+ "normalized_hypothesis": "it is the kind of happy that makes you nervous because you know it can not last",
+ "errors": 0,
+ "words": 17,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0074",
+ "category": "homographs",
+ "text": "The dove dove beneath the branch while the wind began to wind the loose cable.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0074.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE DOVE DOVE BENEATH THE BRANCH WHILE THE WIND BEGAN TO WIND THE LOOSE CABLE",
+ "raw": {
+ "normalized_reference": "the dove dove beneath the branch while the wind began to wind the loose cable.",
+ "normalized_hypothesis": "the dove dove beneath the branch while the wind began to wind the loose cable",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the dove dove beneath the branch while the wind began to wind the loose cable",
+ "normalized_hypothesis": "the dove dove beneath the branch while the wind began to wind the loose cable",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0075",
+ "category": "long",
+ "text": "I've thought a lot about what I would do differently if I could go back, and the honest answer is probably not as much as I'd like to think.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0075.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I'VE THOUGHT A LOT ABOUT WHAT I WOULD DO DIFFERENTLY IF I COULD GO BACK AND THE HONEST ANSWER IS PROBABLY NOT AS MUCH AS I'D LIKE TO THINK",
+ "raw": {
+ "normalized_reference": "i've thought a lot about what i would do differently if i could go back, and the honest answer is probably not as much as i'd like to think.",
+ "normalized_hypothesis": "i've thought a lot about what i would do differently if i could go back and the honest answer is probably not as much as i'd like to think",
+ "errors": 0,
+ "words": 29,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i have thought a lot about what i would do differently if i could go back and the honest answer is probably not as much as i would like to think",
+ "normalized_hypothesis": "i have thought a lot about what i would do differently if i could go back and the honest answer is probably not as much as i would like to think",
+ "errors": 0,
+ "words": 31,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0076",
+ "category": "long_range",
+ "text": "Although the weather changed without warning, the musicians finished packing before anyone opened the enormous glass door.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0076.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "ALTHOUGH THE WEATHER CHANGED WITHOUT WARNING THE MUSICIANS FINISHED PACKING BEFORE ANY ONE OPENED THE ENORMOUS GLASS DOOR",
+ "raw": {
+ "normalized_reference": "although the weather changed without warning, the musicians finished packing before anyone opened the enormous glass door.",
+ "normalized_hypothesis": "although the weather changed without warning the musicians finished packing before any one opened the enormous glass door",
+ "errors": 2,
+ "words": 17,
+ "wer": 0.11764705882352941
+ },
+ "semantic": {
+ "normalized_reference": "although the weather changed without warning the musicians finished packing before anyone opened the enormous glass door",
+ "normalized_hypothesis": "although the weather changed without warning the musicians finished packing before any one opened the enormous glass door",
+ "errors": 2,
+ "words": 17,
+ "wer": 0.11764705882352941
+ }
+ },
+ {
+ "id": "release500_0077",
+ "category": "names_places",
+ "text": "Saskatchewan looked unusually bright beneath the pale February sunrise.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0077.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SASKICHUEN LOOKED UNUSUALLY BRIGHT BENEATH THE PALE FEBRUARY SUNRISE",
+ "raw": {
+ "normalized_reference": "saskatchewan looked unusually bright beneath the pale february sunrise.",
+ "normalized_hypothesis": "saskichuen looked unusually bright beneath the pale february sunrise",
+ "errors": 1,
+ "words": 9,
+ "wer": 0.1111111111111111
+ },
+ "semantic": {
+ "normalized_reference": "saskatchewan looked unusually bright beneath the pale february sunrise",
+ "normalized_hypothesis": "saskichuen looked unusually bright beneath the pale february sunrise",
+ "errors": 1,
+ "words": 9,
+ "wer": 0.1111111111111111
+ }
+ },
+ {
+ "id": "release500_0078",
+ "category": "narrative",
+ "text": "She handled it with a grace I genuinely did not expect.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0078.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SHE HANDLED IT WITH A GRACE I GENUINELY DID NOT EXPECT",
+ "raw": {
+ "normalized_reference": "she handled it with a grace i genuinely did not expect.",
+ "normalized_hypothesis": "she handled it with a grace i genuinely did not expect",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "she handled it with a grace i genuinely did not expect",
+ "normalized_hypothesis": "she handled it with a grace i genuinely did not expect",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0079",
+ "category": "numbers",
+ "text": "Apartment four hundred seven is across from room twenty-nine on the third floor.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0079.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "APARTMENT FOUR HAM SEVEN IS ACROSS FROM ROOM TWENTY NINE ON THE THIRD FLOOR",
+ "raw": {
+ "normalized_reference": "apartment four hundred seven is across from room twenty-nine on the third floor.",
+ "normalized_hypothesis": "apartment four ham seven is across from room twenty nine on the third floor",
+ "errors": 1,
+ "words": 14,
+ "wer": 0.07142857142857142
+ },
+ "semantic": {
+ "normalized_reference": "apartment 407 is across from room 29 on the 3rd floor",
+ "normalized_hypothesis": "apartment 4 ham 7 is across from room 29 on the 3rd floor",
+ "errors": 3,
+ "words": 11,
+ "wer": 0.2727272727272727
+ }
+ },
+ {
+ "id": "release500_0080",
+ "category": "punctuation",
+ "text": "\"Not yet,\" Rowan whispered. \"Give me another minute... please.\"",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0080.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "NOT YET ROWEN WHISPERED GIVE ME ANOTHER MINUTE PLEASE",
+ "raw": {
+ "normalized_reference": "\"not yet,\" rowan whispered. \"give me another minute... please.\"",
+ "normalized_hypothesis": "not yet rowen whispered give me another minute please",
+ "errors": 1,
+ "words": 9,
+ "wer": 0.1111111111111111
+ },
+ "semantic": {
+ "normalized_reference": "not yet rowan whispered give me another minute please",
+ "normalized_hypothesis": "not yet rowen whispered give me another minute please",
+ "errors": 1,
+ "words": 9,
+ "wer": 0.1111111111111111
+ }
+ },
+ {
+ "id": "release500_0081",
+ "category": "question",
+ "text": "What are we actually trying to accomplish here?",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0081.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WHAT ARE WE ACTUALLY TRYING TO ACCOMPLISH HERE",
+ "raw": {
+ "normalized_reference": "what are we actually trying to accomplish here?",
+ "normalized_hypothesis": "what are we actually trying to accomplish here",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "what are we actually trying to accomplish here",
+ "normalized_hypothesis": "what are we actually trying to accomplish here",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0082",
+ "category": "rare_words",
+ "text": "The physician recommended a noninvasive procedure for the recurring neurological symptoms.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0082.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE PHYSICIAN RECOMMENDED A NON EVASIVE PROCEDURE FOR THE RECURRING NEUROLOGICAL SYMPTOMS",
+ "raw": {
+ "normalized_reference": "the physician recommended a noninvasive procedure for the recurring neurological symptoms.",
+ "normalized_hypothesis": "the physician recommended a non evasive procedure for the recurring neurological symptoms",
+ "errors": 2,
+ "words": 11,
+ "wer": 0.18181818181818182
+ },
+ "semantic": {
+ "normalized_reference": "the physician recommended a noninvasive procedure for the recurring neurological symptoms",
+ "normalized_hypothesis": "the physician recommended a non evasive procedure for the recurring neurological symptoms",
+ "errors": 2,
+ "words": 11,
+ "wer": 0.18181818181818182
+ }
+ },
+ {
+ "id": "release500_0083",
+ "category": "short",
+ "text": "Are you serious right now?",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0083.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "ARE YOU SERIOUS RIGHT NOW",
+ "raw": {
+ "normalized_reference": "are you serious right now?",
+ "normalized_hypothesis": "are you serious right now",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "are you serious right now",
+ "normalized_hypothesis": "are you serious right now",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0084",
+ "category": "technical",
+ "text": "The performance improvement is real, but it comes with tradeoffs.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0084.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE PERFORMANCE IMPROVEMENT IS REAL BUT IT COMES WITH TRADE OFFS",
+ "raw": {
+ "normalized_reference": "the performance improvement is real, but it comes with tradeoffs.",
+ "normalized_hypothesis": "the performance improvement is real but it comes with trade offs",
+ "errors": 2,
+ "words": 10,
+ "wer": 0.2
+ },
+ "semantic": {
+ "normalized_reference": "the performance improvement is real but it comes with tradeoffs",
+ "normalized_hypothesis": "the performance improvement is real but it comes with trade offs",
+ "errors": 2,
+ "words": 10,
+ "wer": 0.2
+ }
+ },
+ {
+ "id": "release500_0085",
+ "category": "conversational",
+ "text": "I don't know. It's just been one of those weeks, you know?",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0085.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I DON'T KNOW IT'S JUST BEEN ONE OF THOSE WEEKS YOU KNOW",
+ "raw": {
+ "normalized_reference": "i don't know. it's just been one of those weeks, you know?",
+ "normalized_hypothesis": "i don't know it's just been one of those weeks you know",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i do not know it is just been one of those weeks you know",
+ "normalized_hypothesis": "i do not know it is just been one of those weeks you know",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0086",
+ "category": "descriptive",
+ "text": "The moment you stop expecting perfection, things get a lot more interesting.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0086.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE MOMENT YOU STOP EXPECTING PERFECTION THINGS GET A LOT MORE INTERESTING",
+ "raw": {
+ "normalized_reference": "the moment you stop expecting perfection, things get a lot more interesting.",
+ "normalized_hypothesis": "the moment you stop expecting perfection things get a lot more interesting",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the moment you stop expecting perfection things get a lot more interesting",
+ "normalized_hypothesis": "the moment you stop expecting perfection things get a lot more interesting",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0087",
+ "category": "emotional",
+ "text": "It's the kind of thing that stays with you even years later.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0087.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT'S THE KIND OF THING THAT STAYS WITH YOU EVEN YEARS LATER",
+ "raw": {
+ "normalized_reference": "it's the kind of thing that stays with you even years later.",
+ "normalized_hypothesis": "it's the kind of thing that stays with you even years later",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it is the kind of thing that stays with you even years later",
+ "normalized_hypothesis": "it is the kind of thing that stays with you even years later",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0088",
+ "category": "long",
+ "text": "She told me once that the most important skill she'd developed was learning to disagree with someone without making them feel they were wrong.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0088.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SHE TOLD ME ONCE THAT THE MOST IMPORTANT SKILL SHE DEVELOPED WAS LEARNING TO DISAGREE WITH SOME ONE WITHOUT MAKING THEM FEEL THEY WERE WRONG",
+ "raw": {
+ "normalized_reference": "she told me once that the most important skill she'd developed was learning to disagree with someone without making them feel they were wrong.",
+ "normalized_hypothesis": "she told me once that the most important skill she developed was learning to disagree with some one without making them feel they were wrong",
+ "errors": 3,
+ "words": 24,
+ "wer": 0.125
+ },
+ "semantic": {
+ "normalized_reference": "she told me once that the most important skill she would developed was learning to disagree with someone without making them feel they were wrong",
+ "normalized_hypothesis": "she told me once that the most important skill she developed was learning to disagree with some one without making them feel they were wrong",
+ "errors": 3,
+ "words": 25,
+ "wer": 0.12
+ }
+ },
+ {
+ "id": "release500_0089",
+ "category": "narrative",
+ "text": "He stood at the edge of the cliff and wondered, not for the first time, how he'd gotten here.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0089.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE STOOD AT THE EDGE OF THE CLIFF AND WONDERED NOT FOR THE FIRST TIME HOW HE' GOT HERE",
+ "raw": {
+ "normalized_reference": "he stood at the edge of the cliff and wondered, not for the first time, how he'd gotten here.",
+ "normalized_hypothesis": "he stood at the edge of the cliff and wondered not for the first time how he' got here",
+ "errors": 2,
+ "words": 19,
+ "wer": 0.10526315789473684
+ },
+ "semantic": {
+ "normalized_reference": "he stood at the edge of the cliff and wondered not for the 1st time how he would gotten here",
+ "normalized_hypothesis": "he stood at the edge of the cliff and wondered not for the 1st time how he got here",
+ "errors": 2,
+ "words": 20,
+ "wer": 0.1
+ }
+ },
+ {
+ "id": "release500_0090",
+ "category": "question",
+ "text": "Why does asking for help feel so hard when you actually need it?",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0090.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WHY DOES ASKING FOR HELP FEEL SO HARD WHEN YOU ACTUALLY NEED IT",
+ "raw": {
+ "normalized_reference": "why does asking for help feel so hard when you actually need it?",
+ "normalized_hypothesis": "why does asking for help feel so hard when you actually need it",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "why does asking for help feel so hard when you actually need it",
+ "normalized_hypothesis": "why does asking for help feel so hard when you actually need it",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0091",
+ "category": "short",
+ "text": "That's fair.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0091.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THAT'S FAIR",
+ "raw": {
+ "normalized_reference": "that's fair.",
+ "normalized_hypothesis": "that's fair",
+ "errors": 0,
+ "words": 2,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "that is fair",
+ "normalized_hypothesis": "that is fair",
+ "errors": 0,
+ "words": 3,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0092",
+ "category": "technical",
+ "text": "If you run into that error, the fix is usually just restarting the service.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0092.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IF YOU RUN INTO THAT ERROR THE FIX IS USUALLY JUST RESTARTING THE SERVICE",
+ "raw": {
+ "normalized_reference": "if you run into that error, the fix is usually just restarting the service.",
+ "normalized_hypothesis": "if you run into that error the fix is usually just restarting the service",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "if you run into that error the fix is usually just restarting the service",
+ "normalized_hypothesis": "if you run into that error the fix is usually just restarting the service",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0093",
+ "category": "conversational",
+ "text": "It was a disaster in the best possible way.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0093.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT WAS A DISASTER IN THE BEST POSSIBLE WAY",
+ "raw": {
+ "normalized_reference": "it was a disaster in the best possible way.",
+ "normalized_hypothesis": "it was a disaster in the best possible way",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it was a disaster in the best possible way",
+ "normalized_hypothesis": "it was a disaster in the best possible way",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0094",
+ "category": "descriptive",
+ "text": "It had the look of a place that used to mean something to someone.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0094.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT HAD THE LOOK OF A PLACE THAT USED TO MEAN SOMETHING TO SOMEONE",
+ "raw": {
+ "normalized_reference": "it had the look of a place that used to mean something to someone.",
+ "normalized_hypothesis": "it had the look of a place that used to mean something to someone",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it had the look of a place that used to mean something to someone",
+ "normalized_hypothesis": "it had the look of a place that used to mean something to someone",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0095",
+ "category": "emotional",
+ "text": "There's this feeling you get when something finally works after weeks of failure.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0095.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THERE'S THIS FEELING YOU GET WHEN SOMETHING FINALLY WORKS AFTER WEEKS OF FAILURE",
+ "raw": {
+ "normalized_reference": "there's this feeling you get when something finally works after weeks of failure.",
+ "normalized_hypothesis": "there's this feeling you get when something finally works after weeks of failure",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "there is this feeling you get when something finally works after weeks of failure",
+ "normalized_hypothesis": "there is this feeling you get when something finally works after weeks of failure",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0096",
+ "category": "long",
+ "text": "He had a theory that most arguments aren't actually about what they seem to be about, and the longer I live the more I think he was onto something.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0096.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE HAD A THEORY THAT MOST ARGUMENTS AREN'T ACTUALLY ABOUT WHAT THEY SEEM TO BE ABOUT AND THE LONGER I LIVE THE MORE I THINK HE WAS ON TO SOMETHING",
+ "raw": {
+ "normalized_reference": "he had a theory that most arguments aren't actually about what they seem to be about, and the longer i live the more i think he was onto something.",
+ "normalized_hypothesis": "he had a theory that most arguments aren't actually about what they seem to be about and the longer i live the more i think he was on to something",
+ "errors": 2,
+ "words": 29,
+ "wer": 0.06896551724137931
+ },
+ "semantic": {
+ "normalized_reference": "he had a theory that most arguments are not actually about what they seem to be about and the longer i live the more i think he was onto something",
+ "normalized_hypothesis": "he had a theory that most arguments are not actually about what they seem to be about and the longer i live the more i think he was on to something",
+ "errors": 2,
+ "words": 30,
+ "wer": 0.06666666666666667
+ }
+ },
+ {
+ "id": "release500_0097",
+ "category": "narrative",
+ "text": "The thing nobody tells you about starting over is how ordinary most of it feels.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0097.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE THING NOBODY TELLS YOU ABOUT STARTING OVER IS HOW ORDINARY MOST OF IT FEELS",
+ "raw": {
+ "normalized_reference": "the thing nobody tells you about starting over is how ordinary most of it feels.",
+ "normalized_hypothesis": "the thing nobody tells you about starting over is how ordinary most of it feels",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the thing nobody tells you about starting over is how ordinary most of it feels",
+ "normalized_hypothesis": "the thing nobody tells you about starting over is how ordinary most of it feels",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0098",
+ "category": "question",
+ "text": "What exactly are we waiting for at this point?",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0098.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WHAT EXACTLY ARE WE WAITING FOR AT THIS POINT",
+ "raw": {
+ "normalized_reference": "what exactly are we waiting for at this point?",
+ "normalized_hypothesis": "what exactly are we waiting for at this point",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "what exactly are we waiting for at this point",
+ "normalized_hypothesis": "what exactly are we waiting for at this point",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0099",
+ "category": "short",
+ "text": "I stand by what I said.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0099.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I STAND BY WHAT I SAID",
+ "raw": {
+ "normalized_reference": "i stand by what i said.",
+ "normalized_hypothesis": "i stand by what i said",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i stand by what i said",
+ "normalized_hypothesis": "i stand by what i said",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0100",
+ "category": "technical",
+ "text": "Double-check that value before you push anything to production.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0100.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "DOUBLE CHECK THAT VALUE BEFORE YOU PUSH ANYTHING TO PRODUCTION",
+ "raw": {
+ "normalized_reference": "double-check that value before you push anything to production.",
+ "normalized_hypothesis": "double check that value before you push anything to production",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "double check that value before you push anything to production",
+ "normalized_hypothesis": "double check that value before you push anything to production",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0101",
+ "category": "conversational",
+ "text": "I understand your point; I just do not agree with the conclusion.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0101.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I UNDERSTAND YOUR POINT I JUST DO NOT AGREE WITH THE CONCLUSION",
+ "raw": {
+ "normalized_reference": "i understand your point; i just do not agree with the conclusion.",
+ "normalized_hypothesis": "i understand your point i just do not agree with the conclusion",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i understand your point i just do not agree with the conclusion",
+ "normalized_hypothesis": "i understand your point i just do not agree with the conclusion",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0102",
+ "category": "descriptive",
+ "text": "The sky that day was a strange shade of orange I'd never seen before.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0102.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE SKY THAT DAY WAS A STRANGE SHADE OF ORANGE I'D NEVER SEEN BEFORE",
+ "raw": {
+ "normalized_reference": "the sky that day was a strange shade of orange i'd never seen before.",
+ "normalized_hypothesis": "the sky that day was a strange shade of orange i'd never seen before",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the sky that day was a strange shade of orange i would never seen before",
+ "normalized_hypothesis": "the sky that day was a strange shade of orange i would never seen before",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0103",
+ "category": "emotional",
+ "text": "It hurt more than I expected, and I thought I was prepared.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0103.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT HURT MORE THAN I EXPECTED AND I THOUGHT I WAS PREPARED",
+ "raw": {
+ "normalized_reference": "it hurt more than i expected, and i thought i was prepared.",
+ "normalized_hypothesis": "it hurt more than i expected and i thought i was prepared",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it hurt more than i expected and i thought i was prepared",
+ "normalized_hypothesis": "it hurt more than i expected and i thought i was prepared",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0104",
+ "category": "long",
+ "text": "The hardest part wasn't making the decision, it was sitting with the uncertainty of not knowing whether it was the right one for months afterward.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0104.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE HARDEST PART WAS MAKING THE DECISION IT WAS SITTING WITH THE UNCERTAINTY OF NOT KNOWING WHETHER IT WAS THE RIGHT ONE FOR MONTHS AFTERWARD",
+ "raw": {
+ "normalized_reference": "the hardest part wasn't making the decision, it was sitting with the uncertainty of not knowing whether it was the right one for months afterward.",
+ "normalized_hypothesis": "the hardest part was making the decision it was sitting with the uncertainty of not knowing whether it was the right one for months afterward",
+ "errors": 1,
+ "words": 25,
+ "wer": 0.04
+ },
+ "semantic": {
+ "normalized_reference": "the hardest part was not making the decision it was sitting with the uncertainty of not knowing whether it was the right one for months afterward",
+ "normalized_hypothesis": "the hardest part was making the decision it was sitting with the uncertainty of not knowing whether it was the right one for months afterward",
+ "errors": 1,
+ "words": 26,
+ "wer": 0.038461538461538464
+ }
+ },
+ {
+ "id": "release500_0105",
+ "category": "narrative",
+ "text": "There was this one day last summer when everything just clicked into place.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0105.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THERE WAS THIS ONE DAY LAST SUMMER WHEN EVERYTHING JUST CLICKED INTO PLACE",
+ "raw": {
+ "normalized_reference": "there was this one day last summer when everything just clicked into place.",
+ "normalized_hypothesis": "there was this one day last summer when everything just clicked into place",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "there was this one day last summer when everything just clicked into place",
+ "normalized_hypothesis": "there was this one day last summer when everything just clicked into place",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0106",
+ "category": "question",
+ "text": "Have you ever thought about why we keep doing things the same way when they stop working?",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0106.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HAVE YOU EVER THOUGHT ABOUT WHY WE KEEP DOING THINGS THE SAME WAY WHEN THEY STOP WORKING",
+ "raw": {
+ "normalized_reference": "have you ever thought about why we keep doing things the same way when they stop working?",
+ "normalized_hypothesis": "have you ever thought about why we keep doing things the same way when they stop working",
+ "errors": 0,
+ "words": 17,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "have you ever thought about why we keep doing things the same way when they stop working",
+ "normalized_hypothesis": "have you ever thought about why we keep doing things the same way when they stop working",
+ "errors": 0,
+ "words": 17,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0107",
+ "category": "short",
+ "text": "I can't believe you just said that.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0107.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I CAN'T BELIEVE YOU JUST SAID THAT",
+ "raw": {
+ "normalized_reference": "i can't believe you just said that.",
+ "normalized_hypothesis": "i can't believe you just said that",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i can not believe you just said that",
+ "normalized_hypothesis": "i can not believe you just said that",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0108",
+ "category": "technical",
+ "text": "You'll want to isolate that variable before drawing any conclusions.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0108.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "YOU'LL WANT TO ISOLATE THAT VARIABLE BEFORE DRAWING ANY CONCLUSIONS",
+ "raw": {
+ "normalized_reference": "you'll want to isolate that variable before drawing any conclusions.",
+ "normalized_hypothesis": "you'll want to isolate that variable before drawing any conclusions",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "you will want to isolate that variable before drawing any conclusions",
+ "normalized_hypothesis": "you will want to isolate that variable before drawing any conclusions",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0109",
+ "category": "conversational",
+ "text": "The meeting that could have been an email lasted two hours.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0109.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE MEETING THAT COULD HAVE BEEN AN EMALE LASTED TWO HOURS",
+ "raw": {
+ "normalized_reference": "the meeting that could have been an email lasted two hours.",
+ "normalized_hypothesis": "the meeting that could have been an emale lasted two hours",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ },
+ "semantic": {
+ "normalized_reference": "the meeting that could have been an email lasted 2 hours",
+ "normalized_hypothesis": "the meeting that could have been an emale lasted 2 hours",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ }
+ },
+ {
+ "id": "release500_0110",
+ "category": "descriptive",
+ "text": "There's a specific kind of tired that sleep doesn't fix.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0110.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THERE'S A SPECIFIC KIND OF TIRED THAT SLEEP DOESN'T FIX",
+ "raw": {
+ "normalized_reference": "there's a specific kind of tired that sleep doesn't fix.",
+ "normalized_hypothesis": "there's a specific kind of tired that sleep doesn't fix",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "there is a specific kind of tired that sleep does not fix",
+ "normalized_hypothesis": "there is a specific kind of tired that sleep does not fix",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0111",
+ "category": "emotional",
+ "text": "Something about that moment broke something open in me that I hadn't realized was closed.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0111.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SOMETHING ABOUT THAT MOMENT BROKE SOMETHING OPEN IN ME THAT I HADN'T REALIZED WAS CLOSED",
+ "raw": {
+ "normalized_reference": "something about that moment broke something open in me that i hadn't realized was closed.",
+ "normalized_hypothesis": "something about that moment broke something open in me that i hadn't realized was closed",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "something about that moment broke something open in me that i had not realized was closed",
+ "normalized_hypothesis": "something about that moment broke something open in me that i had not realized was closed",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0112",
+ "category": "long",
+ "text": "The thing about making something good is that you don't usually know it's good while you're making it. You just know you're not done yet.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0112.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE THING ABOUT MAKING SOMETHING GOOD IS THAT YOU DON'T USUALLY KNOW IT'S GOOD WHILE YOU'RE MAKING IT YOU JUST KNOW YOU'RE NOT DONE YET",
+ "raw": {
+ "normalized_reference": "the thing about making something good is that you don't usually know it's good while you're making it. you just know you're not done yet.",
+ "normalized_hypothesis": "the thing about making something good is that you don't usually know it's good while you're making it you just know you're not done yet",
+ "errors": 0,
+ "words": 25,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the thing about making something good is that you do not usually know it is good while you are making it you just know you are not done yet",
+ "normalized_hypothesis": "the thing about making something good is that you do not usually know it is good while you are making it you just know you are not done yet",
+ "errors": 0,
+ "words": 29,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0113",
+ "category": "narrative",
+ "text": "She opened the door slowly, as if she expected to find something she wasn't ready for.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0113.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SHE OPENED THE DOOR SLOWLY AS IF SHE EXPECTED TO FIND SOMETHING SHE WAS READY FOR",
+ "raw": {
+ "normalized_reference": "she opened the door slowly, as if she expected to find something she wasn't ready for.",
+ "normalized_hypothesis": "she opened the door slowly as if she expected to find something she was ready for",
+ "errors": 1,
+ "words": 16,
+ "wer": 0.0625
+ },
+ "semantic": {
+ "normalized_reference": "she opened the door slowly as if she expected to find something she was not ready for",
+ "normalized_hypothesis": "she opened the door slowly as if she expected to find something she was ready for",
+ "errors": 1,
+ "words": 17,
+ "wer": 0.058823529411764705
+ }
+ },
+ {
+ "id": "release500_0114",
+ "category": "question",
+ "text": "What would you do differently if you could go back?",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0114.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WHAT WOULD YOU DO DIFFERENTLY IF YOU COULD GO BACK",
+ "raw": {
+ "normalized_reference": "what would you do differently if you could go back?",
+ "normalized_hypothesis": "what would you do differently if you could go back",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "what would you do differently if you could go back",
+ "normalized_hypothesis": "what would you do differently if you could go back",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0115",
+ "category": "short",
+ "text": "Okay. Okay okay okay.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0115.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "O K O K O K O K",
+ "raw": {
+ "normalized_reference": "okay. okay okay okay.",
+ "normalized_hypothesis": "o k o k o k o k",
+ "errors": 8,
+ "words": 4,
+ "wer": 2.0
+ },
+ "semantic": {
+ "normalized_reference": "okay okay okay okay",
+ "normalized_hypothesis": "0 k 0 k 0 k 0 k",
+ "errors": 8,
+ "words": 4,
+ "wer": 2.0
+ }
+ },
+ {
+ "id": "release500_0116",
+ "category": "technical",
+ "text": "The latency goes up significantly at scale, which is expected.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0116.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE LATENCY GOES UP SIGNIFICANTLY AT SCALE WHICH IS EXPECTED",
+ "raw": {
+ "normalized_reference": "the latency goes up significantly at scale, which is expected.",
+ "normalized_hypothesis": "the latency goes up significantly at scale which is expected",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the latency goes up significantly at scale which is expected",
+ "normalized_hypothesis": "the latency goes up significantly at scale which is expected",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0117",
+ "category": "conversational",
+ "text": "We ended up staying until two in the morning talking about nothing.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0117.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WE ENDED UP STAYING UNTIL TWO IN THE MORNING TALKING ABOUT NOTHING",
+ "raw": {
+ "normalized_reference": "we ended up staying until two in the morning talking about nothing.",
+ "normalized_hypothesis": "we ended up staying until two in the morning talking about nothing",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "we ended up staying until 2 in the morning talking about nothing",
+ "normalized_hypothesis": "we ended up staying until 2 in the morning talking about nothing",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0118",
+ "category": "descriptive",
+ "text": "The rain started slowly, and then all at once, the way it usually does.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0118.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE RAIN STARTED SLOWLY AND THEN ALL AT ONCE THE WAY IT USUALLY DOES",
+ "raw": {
+ "normalized_reference": "the rain started slowly, and then all at once, the way it usually does.",
+ "normalized_hypothesis": "the rain started slowly and then all at once the way it usually does",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the rain started slowly and then all at once the way it usually does",
+ "normalized_hypothesis": "the rain started slowly and then all at once the way it usually does",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0119",
+ "category": "emotional",
+ "text": "I honestly did not think I was going to make it through that year.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0119.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I HONESTLY DID NOT THINK I WAS GOING TO MAKE IT THROUGH THAT YEAR",
+ "raw": {
+ "normalized_reference": "i honestly did not think i was going to make it through that year.",
+ "normalized_hypothesis": "i honestly did not think i was going to make it through that year",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i honestly did not think i was going to make it through that year",
+ "normalized_hypothesis": "i honestly did not think i was going to make it through that year",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0120",
+ "category": "long",
+ "text": "There's an assumption that people who seem confident all the time actually feel confident, and in my experience that's almost never true.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0120.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THERE'S AN ASSUMPTION THAT PEOPLE WHO SEEM CONFIDENT ALL THE TIME ACTUALLY FEEL CONFIDENT AND IN MY EXPERIENCE THAT'S ALMOST NEVER TRUE",
+ "raw": {
+ "normalized_reference": "there's an assumption that people who seem confident all the time actually feel confident, and in my experience that's almost never true.",
+ "normalized_hypothesis": "there's an assumption that people who seem confident all the time actually feel confident and in my experience that's almost never true",
+ "errors": 0,
+ "words": 22,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "there is an assumption that people who seem confident all the time actually feel confident and in my experience that is almost never true",
+ "normalized_hypothesis": "there is an assumption that people who seem confident all the time actually feel confident and in my experience that is almost never true",
+ "errors": 0,
+ "words": 24,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0121",
+ "category": "narrative",
+ "text": "The moment I opened the door I knew something was different, even if I couldn't say what.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0121.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE MOMENT I OPENED THE DOOR I KNEW SOMETHING WAS DIFFERENT EVEN IF I COULDN'T SAY WHAT",
+ "raw": {
+ "normalized_reference": "the moment i opened the door i knew something was different, even if i couldn't say what.",
+ "normalized_hypothesis": "the moment i opened the door i knew something was different even if i couldn't say what",
+ "errors": 0,
+ "words": 17,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the moment i opened the door i knew something was different even if i could not say what",
+ "normalized_hypothesis": "the moment i opened the door i knew something was different even if i could not say what",
+ "errors": 0,
+ "words": 18,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0122",
+ "category": "question",
+ "text": "How do you explain something that doesn't make sense even to you?",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0122.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HOW DO YOU EXPLAIN SOMETHING THAT DOESN'T MAKE SENSE EVEN TO YOU",
+ "raw": {
+ "normalized_reference": "how do you explain something that doesn't make sense even to you?",
+ "normalized_hypothesis": "how do you explain something that doesn't make sense even to you",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "how do you explain something that does not make sense even to you",
+ "normalized_hypothesis": "how do you explain something that does not make sense even to you",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0123",
+ "category": "short",
+ "text": "Hold on a second.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0123.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HOLD ON A SECOND",
+ "raw": {
+ "normalized_reference": "hold on a second.",
+ "normalized_hypothesis": "hold on a second",
+ "errors": 0,
+ "words": 4,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "hold on a 2nd",
+ "normalized_hypothesis": "hold on a 2nd",
+ "errors": 0,
+ "words": 4,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0124",
+ "category": "technical",
+ "text": "A deterministic split prevents duplicate normalized text from leaking into evaluation.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0124.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "A DETERMINISTIC SPLIT PREVENTS DUPLICATE NORMALIZED TEXT FROM LEAKING INTO EVALUATION",
+ "raw": {
+ "normalized_reference": "a deterministic split prevents duplicate normalized text from leaking into evaluation.",
+ "normalized_hypothesis": "a deterministic split prevents duplicate normalized text from leaking into evaluation",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "a deterministic split prevents duplicate normalized text from leaking into evaluation",
+ "normalized_hypothesis": "a deterministic split prevents duplicate normalized text from leaking into evaluation",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0125",
+ "category": "conversational",
+ "text": "Honestly, that was much better than I expected, but the ending still felt rushed.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0125.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HONESTLY THAT WAS MUCH BETTER THAN EXPECTED BUT THE ENDING STILL FELT RUSHED",
+ "raw": {
+ "normalized_reference": "honestly, that was much better than i expected, but the ending still felt rushed.",
+ "normalized_hypothesis": "honestly that was much better than expected but the ending still felt rushed",
+ "errors": 1,
+ "words": 14,
+ "wer": 0.07142857142857142
+ },
+ "semantic": {
+ "normalized_reference": "honestly that was much better than i expected but the ending still felt rushed",
+ "normalized_hypothesis": "honestly that was much better than expected but the ending still felt rushed",
+ "errors": 1,
+ "words": 14,
+ "wer": 0.07142857142857142
+ }
+ },
+ {
+ "id": "release500_0126",
+ "category": "descriptive",
+ "text": "She had a habit of tilting her head slightly when thinking hard about something.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0126.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SHE HAD A HABIT OF TILTING HER HEAD SLIGHTLY WHEN THINKING HARD ABOUT SOMETHING",
+ "raw": {
+ "normalized_reference": "she had a habit of tilting her head slightly when thinking hard about something.",
+ "normalized_hypothesis": "she had a habit of tilting her head slightly when thinking hard about something",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "she had a habit of tilting her head slightly when thinking hard about something",
+ "normalized_hypothesis": "she had a habit of tilting her head slightly when thinking hard about something",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0127",
+ "category": "emotional",
+ "text": "I cannot stress enough how much I did not want to be right about this.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0127.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I CANNOT STRESS ENOUGH HOW MUCH I DID NOT WANT TO BE RIGHT ABOUT THIS",
+ "raw": {
+ "normalized_reference": "i cannot stress enough how much i did not want to be right about this.",
+ "normalized_hypothesis": "i cannot stress enough how much i did not want to be right about this",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i cannot stress enough how much i did not want to be right about this",
+ "normalized_hypothesis": "i cannot stress enough how much i did not want to be right about this",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0128",
+ "category": "long",
+ "text": "I used to think the goal was to get to a point where nothing was hard anymore, and now I think that point doesn't exist and I'm not sure I'd want it to.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0128.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I USED TO THINK THE GOAL WAS TO GET TO A POINT WHERE NOTHING WAS HARD AN AMORE AND NOW I THINK THAT POINT DOESN'T EXIST AND I'M NOT SURE I'D WANT IT TO",
+ "raw": {
+ "normalized_reference": "i used to think the goal was to get to a point where nothing was hard anymore, and now i think that point doesn't exist and i'm not sure i'd want it to.",
+ "normalized_hypothesis": "i used to think the goal was to get to a point where nothing was hard an amore and now i think that point doesn't exist and i'm not sure i'd want it to",
+ "errors": 2,
+ "words": 33,
+ "wer": 0.06060606060606061
+ },
+ "semantic": {
+ "normalized_reference": "i used to think the goal was to get to a point where nothing was hard anymore and now i think that point does not exist and i am not sure i would want it to",
+ "normalized_hypothesis": "i used to think the goal was to get to a point where nothing was hard an amore and now i think that point does not exist and i am not sure i would want it to",
+ "errors": 2,
+ "words": 36,
+ "wer": 0.05555555555555555
+ }
+ },
+ {
+ "id": "release500_0129",
+ "category": "narrative",
+ "text": "So this is the part of the story where everything goes completely sideways.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0129.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SO THIS IS THE PART OF THE STORY WHERE EVERYTHING GOES COMPLETELY SIDEWAYS",
+ "raw": {
+ "normalized_reference": "so this is the part of the story where everything goes completely sideways.",
+ "normalized_hypothesis": "so this is the part of the story where everything goes completely sideways",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "so this is the part of the story where everything goes completely sideways",
+ "normalized_hypothesis": "so this is the part of the story where everything goes completely sideways",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0130",
+ "category": "question",
+ "text": "Is there a version of this where it actually works out okay?",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0130.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IS THERE A VERSION OF THIS WHERE IT ACTUALLY WORKS OUT O K",
+ "raw": {
+ "normalized_reference": "is there a version of this where it actually works out okay?",
+ "normalized_hypothesis": "is there a version of this where it actually works out o k",
+ "errors": 2,
+ "words": 12,
+ "wer": 0.16666666666666666
+ },
+ "semantic": {
+ "normalized_reference": "is there a version of this where it actually works out okay",
+ "normalized_hypothesis": "is there a version of this where it actually works out 0 k",
+ "errors": 2,
+ "words": 12,
+ "wer": 0.16666666666666666
+ }
+ },
+ {
+ "id": "release500_0131",
+ "category": "short",
+ "text": "Oh no. Oh no no no.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0131.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "OH NO OH NO NO NO",
+ "raw": {
+ "normalized_reference": "oh no. oh no no no.",
+ "normalized_hypothesis": "oh no oh no no no",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "0 no 0 no no no",
+ "normalized_hypothesis": "0 no 0 no no no",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0132",
+ "category": "technical",
+ "text": "Always test on a copy before modifying the original.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0132.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "ALWAYS TEST ON A COPY BEFORE MODIFYING THE ORIGINAL",
+ "raw": {
+ "normalized_reference": "always test on a copy before modifying the original.",
+ "normalized_hypothesis": "always test on a copy before modifying the original",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "always test on a copy before modifying the original",
+ "normalized_hypothesis": "always test on a copy before modifying the original",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0133",
+ "category": "conversational",
+ "text": "I wasn't even supposed to be there. That's the wild part.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0133.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I WASN'T EVEN SUPPOSED TO BE THERE THAT'S THE WILD PART",
+ "raw": {
+ "normalized_reference": "i wasn't even supposed to be there. that's the wild part.",
+ "normalized_hypothesis": "i wasn't even supposed to be there that's the wild part",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i was not even supposed to be there that is the wild part",
+ "normalized_hypothesis": "i was not even supposed to be there that is the wild part",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0134",
+ "category": "descriptive",
+ "text": "It's a small town. Not the kind you leave. The kind you leave and come back to.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0134.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT'S A SMALL TOWN NOT THE KIND YOU LEAVE THE KIND YOU LEAVE AND COME BACK TO",
+ "raw": {
+ "normalized_reference": "it's a small town. not the kind you leave. the kind you leave and come back to.",
+ "normalized_hypothesis": "it's a small town not the kind you leave the kind you leave and come back to",
+ "errors": 0,
+ "words": 17,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it is a small town not the kind you leave the kind you leave and come back to",
+ "normalized_hypothesis": "it is a small town not the kind you leave the kind you leave and come back to",
+ "errors": 0,
+ "words": 18,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0135",
+ "category": "emotional",
+ "text": "She cried. I cried. We all just sort of stood there crying.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0135.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SHE CRIED I CRIED WE ALL JUST SORT OF STOOD THEIR CRYING",
+ "raw": {
+ "normalized_reference": "she cried. i cried. we all just sort of stood there crying.",
+ "normalized_hypothesis": "she cried i cried we all just sort of stood their crying",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "she cried i cried we all just sort of stood there crying",
+ "normalized_hypothesis": "she cried i cried we all just sort of stood their crying",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ }
+ },
+ {
+ "id": "release500_0136",
+ "category": "long",
+ "text": "There's a version of being busy that's actually just a very elaborate way of avoiding the one thing you know you should be doing.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0136.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THERE'S A VERSION OF BEING BUSY THAT'S ACTUALLY JUST A VERY ELABORATE WAY OF AVOIDING THE ONE THING YOU KNOW YOU SHOULD BE DOING",
+ "raw": {
+ "normalized_reference": "there's a version of being busy that's actually just a very elaborate way of avoiding the one thing you know you should be doing.",
+ "normalized_hypothesis": "there's a version of being busy that's actually just a very elaborate way of avoiding the one thing you know you should be doing",
+ "errors": 0,
+ "words": 24,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "there is a version of being busy that is actually just a very elaborate way of avoiding the one thing you know you should be doing",
+ "normalized_hypothesis": "there is a version of being busy that is actually just a very elaborate way of avoiding the one thing you know you should be doing",
+ "errors": 0,
+ "words": 26,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0137",
+ "category": "narrative",
+ "text": "The letter arrived on a Tuesday, which felt appropriate somehow.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0137.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE LETTER ARRIVED ON A TUESDAY WHICH FELT APPROPRIATE SOMEHOW",
+ "raw": {
+ "normalized_reference": "the letter arrived on a tuesday, which felt appropriate somehow.",
+ "normalized_hypothesis": "the letter arrived on a tuesday which felt appropriate somehow",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the letter arrived on a tuesday which felt appropriate somehow",
+ "normalized_hypothesis": "the letter arrived on a tuesday which felt appropriate somehow",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0138",
+ "category": "question",
+ "text": "What exactly did she say when you told her?",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0138.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WHAT EXACTLY DID SHE SAY WHEN YOU TOLD HER",
+ "raw": {
+ "normalized_reference": "what exactly did she say when you told her?",
+ "normalized_hypothesis": "what exactly did she say when you told her",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "what exactly did she say when you told her",
+ "normalized_hypothesis": "what exactly did she say when you told her",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0139",
+ "category": "short",
+ "text": "No way. Absolutely not.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0139.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "NO WAY ABSOLUTELY NOT",
+ "raw": {
+ "normalized_reference": "no way. absolutely not.",
+ "normalized_hypothesis": "no way absolutely not",
+ "errors": 0,
+ "words": 4,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "no way absolutely not",
+ "normalized_hypothesis": "no way absolutely not",
+ "errors": 0,
+ "words": 4,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0140",
+ "category": "technical",
+ "text": "The first thing you want to do is make sure everything is backed up.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0140.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE FIRST THING YOU WANT TO DO IS MAKE SURE EVERYTHING IS BACKED UP",
+ "raw": {
+ "normalized_reference": "the first thing you want to do is make sure everything is backed up.",
+ "normalized_hypothesis": "the first thing you want to do is make sure everything is backed up",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the 1st thing you want to do is make sure everything is backed up",
+ "normalized_hypothesis": "the 1st thing you want to do is make sure everything is backed up",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0141",
+ "category": "conversational",
+ "text": "I have the most insane idea and you need to just hear me out.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0141.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I HAVE THE MOST INSANE IDEA AND YOU NEED TO JUST HEAR ME OUT",
+ "raw": {
+ "normalized_reference": "i have the most insane idea and you need to just hear me out.",
+ "normalized_hypothesis": "i have the most insane idea and you need to just hear me out",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i have the most insane idea and you need to just hear me out",
+ "normalized_hypothesis": "i have the most insane idea and you need to just hear me out",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0142",
+ "category": "descriptive",
+ "text": "There's a coffee shop on the corner that's been there longer than anyone can remember.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0142.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THERE'S A COFFEE SHOP ON THE CORNER THAT'S BEEN THERE LONGER THAN ANYONE CAN REMEMBER",
+ "raw": {
+ "normalized_reference": "there's a coffee shop on the corner that's been there longer than anyone can remember.",
+ "normalized_hypothesis": "there's a coffee shop on the corner that's been there longer than anyone can remember",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "there is a coffee shop on the corner that has been there longer than anyone can remember",
+ "normalized_hypothesis": "there is a coffee shop on the corner that has been there longer than anyone can remember",
+ "errors": 0,
+ "words": 17,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0143",
+ "category": "emotional",
+ "text": "I kept it together until I got to the car, and then I completely fell apart.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0143.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I KEPT IT TOGETHER UNTIL I GOT TO THE CAR AND THEN I COMPLETELY FELL APART",
+ "raw": {
+ "normalized_reference": "i kept it together until i got to the car, and then i completely fell apart.",
+ "normalized_hypothesis": "i kept it together until i got to the car and then i completely fell apart",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i kept it together until i got to the car and then i completely fell apart",
+ "normalized_hypothesis": "i kept it together until i got to the car and then i completely fell apart",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0144",
+ "category": "long",
+ "text": "The older I get, the more I think that most of what we call wisdom is just pattern recognition built from having been wrong enough times to start noticing.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0144.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE OLDER I GET THE MORE I THINK THAT MOST OF WHAT WE CALL WISDOM IS JUST PATTERN RECOGNITION BUILT FROM HAVING BEEN WRONG ENOUGH TIMES TO START NOTICING",
+ "raw": {
+ "normalized_reference": "the older i get, the more i think that most of what we call wisdom is just pattern recognition built from having been wrong enough times to start noticing.",
+ "normalized_hypothesis": "the older i get the more i think that most of what we call wisdom is just pattern recognition built from having been wrong enough times to start noticing",
+ "errors": 0,
+ "words": 29,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the older i get the more i think that most of what we call wisdom is just pattern recognition built from having been wrong enough times to start noticing",
+ "normalized_hypothesis": "the older i get the more i think that most of what we call wisdom is just pattern recognition built from having been wrong enough times to start noticing",
+ "errors": 0,
+ "words": 29,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0145",
+ "category": "narrative",
+ "text": "I remember the exact moment I realized I had made a huge mistake.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0145.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I REMEMBER THE EXACT MOMENT I REALIZED I HAD MADE A HUGE MISTAKE",
+ "raw": {
+ "normalized_reference": "i remember the exact moment i realized i had made a huge mistake.",
+ "normalized_hypothesis": "i remember the exact moment i realized i had made a huge mistake",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i remember the exact moment i realized i had made a huge mistake",
+ "normalized_hypothesis": "i remember the exact moment i realized i had made a huge mistake",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0146",
+ "category": "question",
+ "text": "Do you think it would have gone differently if we'd left earlier?",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0146.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "DO YOU THINK IT WOULD HAVE GONE DIFFERENTLY IF WE'D LEFT EARLIER",
+ "raw": {
+ "normalized_reference": "do you think it would have gone differently if we'd left earlier?",
+ "normalized_hypothesis": "do you think it would have gone differently if we'd left earlier",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "do you think it would have gone differently if we would left earlier",
+ "normalized_hypothesis": "do you think it would have gone differently if we would left earlier",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0147",
+ "category": "short",
+ "text": "You have to be kidding me.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0147.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "YOU HAVE TO BE KITTING ME",
+ "raw": {
+ "normalized_reference": "you have to be kidding me.",
+ "normalized_hypothesis": "you have to be kitting me",
+ "errors": 1,
+ "words": 6,
+ "wer": 0.16666666666666666
+ },
+ "semantic": {
+ "normalized_reference": "you have to be kidding me",
+ "normalized_hypothesis": "you have to be kitting me",
+ "errors": 1,
+ "words": 6,
+ "wer": 0.16666666666666666
+ }
+ },
+ {
+ "id": "release500_0148",
+ "category": "technical",
+ "text": "If the checksum fails, restart the worker process without deleting the cached tensors.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0148.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IF THE CHEXUM FAILS RESTART THE WORKER PROCESS WITHOUT DELEADING THE CASH TENSERS",
+ "raw": {
+ "normalized_reference": "if the checksum fails, restart the worker process without deleting the cached tensors.",
+ "normalized_hypothesis": "if the chexum fails restart the worker process without deleading the cash tensers",
+ "errors": 4,
+ "words": 13,
+ "wer": 0.3076923076923077
+ },
+ "semantic": {
+ "normalized_reference": "if the checksum fails restart the worker process without deleting the cached tensors",
+ "normalized_hypothesis": "if the chexum fails restart the worker process without deleading the cash tensers",
+ "errors": 4,
+ "words": 13,
+ "wer": 0.3076923076923077
+ }
+ },
+ {
+ "id": "release500_0149",
+ "category": "conversational",
+ "text": "I need you to understand how close that was to not working.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0149.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I NEED YOU TO UNDERSTAND HOW CLOSE THAT WAS TO NOT WORKING",
+ "raw": {
+ "normalized_reference": "i need you to understand how close that was to not working.",
+ "normalized_hypothesis": "i need you to understand how close that was to not working",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i need you to understand how close that was to not working",
+ "normalized_hypothesis": "i need you to understand how close that was to not working",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0150",
+ "category": "descriptive",
+ "text": "It was the sort of morning where the whole world looks rinsed clean.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0150.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT WAS A SORT OF MORNING WHERE THE WHOLE WORLD LOOKS RENCED CLEAN",
+ "raw": {
+ "normalized_reference": "it was the sort of morning where the whole world looks rinsed clean.",
+ "normalized_hypothesis": "it was a sort of morning where the whole world looks renced clean",
+ "errors": 2,
+ "words": 13,
+ "wer": 0.15384615384615385
+ },
+ "semantic": {
+ "normalized_reference": "it was the sort of morning where the whole world looks rinsed clean",
+ "normalized_hypothesis": "it was a sort of morning where the whole world looks renced clean",
+ "errors": 2,
+ "words": 13,
+ "wer": 0.15384615384615385
+ }
+ },
+ {
+ "id": "release500_0151",
+ "category": "emotional",
+ "text": "She started laughing and I started laughing and we couldn't stop for a full minute.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0151.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SHE STARTED LAUGHING AND I STARTED LAUGHING AND WE COULDN'T STOP FOR A FULL MINUTE",
+ "raw": {
+ "normalized_reference": "she started laughing and i started laughing and we couldn't stop for a full minute.",
+ "normalized_hypothesis": "she started laughing and i started laughing and we couldn't stop for a full minute",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "she started laughing and i started laughing and we could not stop for a full minute",
+ "normalized_hypothesis": "she started laughing and i started laughing and we could not stop for a full minute",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0152",
+ "category": "long",
+ "text": "We spent three years building something we were genuinely proud of, and then the market shifted and none of it mattered the way we thought it would.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0152.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WE SPENT THREE YEARS BUILDING SOMETHING WE WERE GENUINELY PROUD OF AND THEN THE MARKET SHIFTED AND NONE OF IT MATTERED THE WAY WE THOUGHT IT WOULD",
+ "raw": {
+ "normalized_reference": "we spent three years building something we were genuinely proud of, and then the market shifted and none of it mattered the way we thought it would.",
+ "normalized_hypothesis": "we spent three years building something we were genuinely proud of and then the market shifted and none of it mattered the way we thought it would",
+ "errors": 0,
+ "words": 27,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "we spent 3 years building something we were genuinely proud of and then the market shifted and none of it mattered the way we thought it would",
+ "normalized_hypothesis": "we spent 3 years building something we were genuinely proud of and then the market shifted and none of it mattered the way we thought it would",
+ "errors": 0,
+ "words": 27,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0153",
+ "category": "narrative",
+ "text": "And somehow, despite all of that, it actually turned out okay.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0153.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "AND SOMEHOW DESPITE ALL OF THAT IT ACTUALLY TURNED OUT O K",
+ "raw": {
+ "normalized_reference": "and somehow, despite all of that, it actually turned out okay.",
+ "normalized_hypothesis": "and somehow despite all of that it actually turned out o k",
+ "errors": 2,
+ "words": 11,
+ "wer": 0.18181818181818182
+ },
+ "semantic": {
+ "normalized_reference": "and somehow despite all of that it actually turned out okay",
+ "normalized_hypothesis": "and somehow despite all of that it actually turned out 0 k",
+ "errors": 2,
+ "words": 11,
+ "wer": 0.18181818181818182
+ }
+ },
+ {
+ "id": "release500_0154",
+ "category": "question",
+ "text": "Is that something you actually want, or something you think you should want?",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0154.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IS THAT SOMETHING YOU ACTUALLY WANT OR SOMETHING YOU THINK YOU SHOULD WANT",
+ "raw": {
+ "normalized_reference": "is that something you actually want, or something you think you should want?",
+ "normalized_hypothesis": "is that something you actually want or something you think you should want",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "is that something you actually want or something you think you should want",
+ "normalized_hypothesis": "is that something you actually want or something you think you should want",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0155",
+ "category": "short",
+ "text": "This is fine.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0155.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THIS IS FINE",
+ "raw": {
+ "normalized_reference": "this is fine.",
+ "normalized_hypothesis": "this is fine",
+ "errors": 0,
+ "words": 3,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "this is fine",
+ "normalized_hypothesis": "this is fine",
+ "errors": 0,
+ "words": 3,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0156",
+ "category": "technical",
+ "text": "The tricky part isn't the setup, it's making sure all dependencies line up.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0156.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE TRICKY PART ISN'T THE SET UP IT'S MAKING SURE ALL DEPENDENCIES LINE UP",
+ "raw": {
+ "normalized_reference": "the tricky part isn't the setup, it's making sure all dependencies line up.",
+ "normalized_hypothesis": "the tricky part isn't the set up it's making sure all dependencies line up",
+ "errors": 2,
+ "words": 13,
+ "wer": 0.15384615384615385
+ },
+ "semantic": {
+ "normalized_reference": "the tricky part is not the setup it is making sure all dependencies line up",
+ "normalized_hypothesis": "the tricky part is not the set up it is making sure all dependencies line up",
+ "errors": 2,
+ "words": 15,
+ "wer": 0.13333333333333333
+ }
+ },
+ {
+ "id": "release500_0157",
+ "category": "conversational",
+ "text": "He called it a minor setback. I called it a disaster.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0157.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE CALLED IT A MINOR SETBACK I CALLED IT A DISASTER",
+ "raw": {
+ "normalized_reference": "he called it a minor setback. i called it a disaster.",
+ "normalized_hypothesis": "he called it a minor setback i called it a disaster",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he called it a minor setback i called it a disaster",
+ "normalized_hypothesis": "he called it a minor setback i called it a disaster",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0158",
+ "category": "descriptive",
+ "text": "The whole place smelled like pine and old wood and cold air.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0158.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE WHOLE PLACE SMELLD LIKE PINE AND OLD WOOD IN COLD AIR",
+ "raw": {
+ "normalized_reference": "the whole place smelled like pine and old wood and cold air.",
+ "normalized_hypothesis": "the whole place smelld like pine and old wood in cold air",
+ "errors": 2,
+ "words": 12,
+ "wer": 0.16666666666666666
+ },
+ "semantic": {
+ "normalized_reference": "the whole place smelled like pine and old wood and cold air",
+ "normalized_hypothesis": "the whole place smelld like pine and old wood in cold air",
+ "errors": 2,
+ "words": 12,
+ "wer": 0.16666666666666666
+ }
+ },
+ {
+ "id": "release500_0159",
+ "category": "emotional",
+ "text": "I've never been so relieved in my entire life.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0159.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I'VE NEVER BEEN SO RELIEVED IN MY ENTIRE LIFE",
+ "raw": {
+ "normalized_reference": "i've never been so relieved in my entire life.",
+ "normalized_hypothesis": "i've never been so relieved in my entire life",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i have never been so relieved in my entire life",
+ "normalized_hypothesis": "i have never been so relieved in my entire life",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0160",
+ "category": "narrative",
+ "text": "I wrote it all down that night so I wouldn't forget, and I still haven't looked at it.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0160.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I WROTE IT ALL DOWN THAT NIGHT SO I WOULDN'T FORGET AND I STILL HAVEN'T LOOKED AT IT",
+ "raw": {
+ "normalized_reference": "i wrote it all down that night so i wouldn't forget, and i still haven't looked at it.",
+ "normalized_hypothesis": "i wrote it all down that night so i wouldn't forget and i still haven't looked at it",
+ "errors": 0,
+ "words": 18,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i wrote it all down that night so i would not forget and i still have not looked at it",
+ "normalized_hypothesis": "i wrote it all down that night so i would not forget and i still have not looked at it",
+ "errors": 0,
+ "words": 20,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0161",
+ "category": "short",
+ "text": "That's kind of wild actually.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0161.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THAT'S KIND OF WILD ACTUALLY",
+ "raw": {
+ "normalized_reference": "that's kind of wild actually.",
+ "normalized_hypothesis": "that's kind of wild actually",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "that is kind of wild actually",
+ "normalized_hypothesis": "that is kind of wild actually",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0162",
+ "category": "technical",
+ "text": "Version pinning will save you a lot of pain later down the line.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0162.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "VERSION PINNING WILL SAVE YOU A LOT OF PAIN LATER DOWN THE LINE",
+ "raw": {
+ "normalized_reference": "version pinning will save you a lot of pain later down the line.",
+ "normalized_hypothesis": "version pinning will save you a lot of pain later down the line",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "version pinning will save you a lot of pain later down the line",
+ "normalized_hypothesis": "version pinning will save you a lot of pain later down the line",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0163",
+ "category": "conversational",
+ "text": "At some point you just have to commit and see what happens.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0163.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "AT SOME POINT YOU JUST HAVE TO COMMIT AND SEE WHAT HAPPENS",
+ "raw": {
+ "normalized_reference": "at some point you just have to commit and see what happens.",
+ "normalized_hypothesis": "at some point you just have to commit and see what happens",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "at some point you just have to commit and see what happens",
+ "normalized_hypothesis": "at some point you just have to commit and see what happens",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0164",
+ "category": "emotional",
+ "text": "Nobody told me it was going to be this hard, and honestly, I'm grateful.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0164.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "NOBODY TOLD ME IT WAS GOING TO BE THIS HARD AND HONESTLY I'M GRATEFUL",
+ "raw": {
+ "normalized_reference": "nobody told me it was going to be this hard, and honestly, i'm grateful.",
+ "normalized_hypothesis": "nobody told me it was going to be this hard and honestly i'm grateful",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "nobody told me it was going to be this hard and honestly i am grateful",
+ "normalized_hypothesis": "nobody told me it was going to be this hard and honestly i am grateful",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0165",
+ "category": "narrative",
+ "text": "Chapter seven. In which everything that could go wrong did, and a few things that couldn't.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0165.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "CHAPTER SEVEN IN WHICH EVERYTHING THAT COULD GO WRONG DID AND A FEW THINGS THAT COULDN'T",
+ "raw": {
+ "normalized_reference": "chapter seven. in which everything that could go wrong did, and a few things that couldn't.",
+ "normalized_hypothesis": "chapter seven in which everything that could go wrong did and a few things that couldn't",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "chapter 7 in which everything that could go wrong did and a few things that could not",
+ "normalized_hypothesis": "chapter 7 in which everything that could go wrong did and a few things that could not",
+ "errors": 0,
+ "words": 17,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0166",
+ "category": "short",
+ "text": "Wait, what?",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0166.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WAIT WHAT",
+ "raw": {
+ "normalized_reference": "wait, what?",
+ "normalized_hypothesis": "wait what",
+ "errors": 0,
+ "words": 2,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "wait what",
+ "normalized_hypothesis": "wait what",
+ "errors": 0,
+ "words": 2,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0167",
+ "category": "technical",
+ "text": "The graphics processor remained idle because the data loader exhausted every central processing core.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0167.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE GRAPHICS PROCESSOR REMAINED IDLE BECAUSE THE DATA LOADER EXHAUSTED EVERY CENTRAL PROCESSING CORPS",
+ "raw": {
+ "normalized_reference": "the graphics processor remained idle because the data loader exhausted every central processing core.",
+ "normalized_hypothesis": "the graphics processor remained idle because the data loader exhausted every central processing corps",
+ "errors": 1,
+ "words": 14,
+ "wer": 0.07142857142857142
+ },
+ "semantic": {
+ "normalized_reference": "the graphics processor remained idle because the data loader exhausted every central processing core",
+ "normalized_hypothesis": "the graphics processor remained idle because the data loader exhausted every central processing corps",
+ "errors": 1,
+ "words": 14,
+ "wer": 0.07142857142857142
+ }
+ },
+ {
+ "id": "release500_0168",
+ "category": "conversational",
+ "text": "Could you speak a little more slowly? I am trying to write every detail down.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0168.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "COULD YOU SPEAK A LITTLE MORE SLOWLY I AM TRYING TO WRITE EVERY DETAIL DOWN",
+ "raw": {
+ "normalized_reference": "could you speak a little more slowly? i am trying to write every detail down.",
+ "normalized_hypothesis": "could you speak a little more slowly i am trying to write every detail down",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "could you speak a little more slowly i am trying to write every detail down",
+ "normalized_hypothesis": "could you speak a little more slowly i am trying to write every detail down",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0169",
+ "category": "emotional",
+ "text": "I was so excited I could barely form a complete sentence.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0169.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I WAS SO EXCITED I COULD BARELY FORM A COMPLETE SENTENCE",
+ "raw": {
+ "normalized_reference": "i was so excited i could barely form a complete sentence.",
+ "normalized_hypothesis": "i was so excited i could barely form a complete sentence",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i was so excited i could barely form a complete sentence",
+ "normalized_hypothesis": "i was so excited i could barely form a complete sentence",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0170",
+ "category": "narrative",
+ "text": "He looked at me like I had said something in a language he didn't speak.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0170.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE LOOKED AT ME LIKE I HAD SAID SOMETHING IN A LANGUAGE HE DIDN'T SPEAK",
+ "raw": {
+ "normalized_reference": "he looked at me like i had said something in a language he didn't speak.",
+ "normalized_hypothesis": "he looked at me like i had said something in a language he didn't speak",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he looked at me like i had said something in a language he did not speak",
+ "normalized_hypothesis": "he looked at me like i had said something in a language he did not speak",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0171",
+ "category": "short",
+ "text": "That's actually hilarious.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0171.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THAT'S ACTUALLY HILARIOUS",
+ "raw": {
+ "normalized_reference": "that's actually hilarious.",
+ "normalized_hypothesis": "that's actually hilarious",
+ "errors": 0,
+ "words": 3,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "that is actually hilarious",
+ "normalized_hypothesis": "that is actually hilarious",
+ "errors": 0,
+ "words": 4,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0172",
+ "category": "technical",
+ "text": "The encoder maps phonemes, stress, and phrase boundaries into a shared hidden representation.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0172.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE INCODER MAPS PHONEIMES STRESS AND PHRASE BOUNDARIES INTO A SHARED HIDDEN REPRESENTATION",
+ "raw": {
+ "normalized_reference": "the encoder maps phonemes, stress, and phrase boundaries into a shared hidden representation.",
+ "normalized_hypothesis": "the incoder maps phoneimes stress and phrase boundaries into a shared hidden representation",
+ "errors": 2,
+ "words": 13,
+ "wer": 0.15384615384615385
+ },
+ "semantic": {
+ "normalized_reference": "the encoder maps phonemes stress and phrase boundaries into a shared hidden representation",
+ "normalized_hypothesis": "the incoder maps phoneimes stress and phrase boundaries into a shared hidden representation",
+ "errors": 2,
+ "words": 13,
+ "wer": 0.15384615384615385
+ }
+ },
+ {
+ "id": "release500_0173",
+ "category": "conversational",
+ "text": "She showed up three hours late and acted like nothing had happened.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0173.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SHE SHOWED UP THREE HOURS LATE AND ACTED LIKE NOTHING HAD HAPPENED",
+ "raw": {
+ "normalized_reference": "she showed up three hours late and acted like nothing had happened.",
+ "normalized_hypothesis": "she showed up three hours late and acted like nothing had happened",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "she showed up 3 hours late and acted like nothing had happened",
+ "normalized_hypothesis": "she showed up 3 hours late and acted like nothing had happened",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0174",
+ "category": "emotional",
+ "text": "Oh my god. This is the best thing I've ever tasted.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0174.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "O MY GOD THIS IS THE BEST THING I'VE EVER TASTED",
+ "raw": {
+ "normalized_reference": "oh my god. this is the best thing i've ever tasted.",
+ "normalized_hypothesis": "o my god this is the best thing i've ever tasted",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ },
+ "semantic": {
+ "normalized_reference": "0 my god this is the best thing i have ever tasted",
+ "normalized_hypothesis": "0 my god this is the best thing i have ever tasted",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0175",
+ "category": "narrative",
+ "text": "The note was short, which was somehow more frightening than a long one would have been.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0175.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE NOTE WAS SHORT WHICH WAS SOMEHOW MORE FRIGHTING THAN A LONG ONE WOULD HAVE BEEN",
+ "raw": {
+ "normalized_reference": "the note was short, which was somehow more frightening than a long one would have been.",
+ "normalized_hypothesis": "the note was short which was somehow more frighting than a long one would have been",
+ "errors": 1,
+ "words": 16,
+ "wer": 0.0625
+ },
+ "semantic": {
+ "normalized_reference": "the note was short which was somehow more frightening than a long one would have been",
+ "normalized_hypothesis": "the note was short which was somehow more frighting than a long one would have been",
+ "errors": 1,
+ "words": 16,
+ "wer": 0.0625
+ }
+ },
+ {
+ "id": "release500_0176",
+ "category": "short",
+ "text": "Noted.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0176.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "NOTED",
+ "raw": {
+ "normalized_reference": "noted.",
+ "normalized_hypothesis": "noted",
+ "errors": 0,
+ "words": 1,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "noted",
+ "normalized_hypothesis": "noted",
+ "errors": 0,
+ "words": 1,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0177",
+ "category": "technical",
+ "text": "It's a simple process once you understand the underlying structure.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0177.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT'S A SIMPLE PROCESS ONCE YOU UNDERSTAND THE UNDERLYING STRUCTURE",
+ "raw": {
+ "normalized_reference": "it's a simple process once you understand the underlying structure.",
+ "normalized_hypothesis": "it's a simple process once you understand the underlying structure",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it is a simple process once you understand the underlying structure",
+ "normalized_hypothesis": "it is a simple process once you understand the underlying structure",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0178",
+ "category": "conversational",
+ "text": "You have to hear this story because it's almost too good to be true.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0178.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "YOU HAVE TO HEAR THIS STORY BECAUSE IT'S ALMOST TOO GOOD TO BE TRUE",
+ "raw": {
+ "normalized_reference": "you have to hear this story because it's almost too good to be true.",
+ "normalized_hypothesis": "you have to hear this story because it's almost too good to be true",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "you have to hear this story because it is almost too good to be true",
+ "normalized_hypothesis": "you have to hear this story because it is almost too good to be true",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0179",
+ "category": "emotional",
+ "text": "I have never felt more seen in my entire life than in that moment.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0179.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I HAVE NEVER FELT MORE SEN IN MY ENTIRE LIFE THAN IN THAT MOMENT",
+ "raw": {
+ "normalized_reference": "i have never felt more seen in my entire life than in that moment.",
+ "normalized_hypothesis": "i have never felt more sen in my entire life than in that moment",
+ "errors": 1,
+ "words": 14,
+ "wer": 0.07142857142857142
+ },
+ "semantic": {
+ "normalized_reference": "i have never felt more seen in my entire life than in that moment",
+ "normalized_hypothesis": "i have never felt more senator in my entire life than in that moment",
+ "errors": 1,
+ "words": 14,
+ "wer": 0.07142857142857142
+ }
+ },
+ {
+ "id": "release500_0180",
+ "category": "narrative",
+ "text": "She knew. She had always known. She just waited for the right moment.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0180.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SHE KNEW SHE HAD ALWAYS KNOWN SHE JUST WAITED FOR THE RIGHT MOMENT",
+ "raw": {
+ "normalized_reference": "she knew. she had always known. she just waited for the right moment.",
+ "normalized_hypothesis": "she knew she had always known she just waited for the right moment",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "she knew she had always known she just waited for the right moment",
+ "normalized_hypothesis": "she knew she had always known she just waited for the right moment",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0181",
+ "category": "short",
+ "text": "Yeah, no. Hard pass.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0181.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "YITH NO HARD PASS",
+ "raw": {
+ "normalized_reference": "yeah, no. hard pass.",
+ "normalized_hypothesis": "yith no hard pass",
+ "errors": 1,
+ "words": 4,
+ "wer": 0.25
+ },
+ "semantic": {
+ "normalized_reference": "yeah no hard pass",
+ "normalized_hypothesis": "yith no hard pass",
+ "errors": 1,
+ "words": 4,
+ "wer": 0.25
+ }
+ },
+ {
+ "id": "release500_0182",
+ "category": "technical",
+ "text": "Please verify the sample rate, hop length, and mel filter configuration before inference.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0182.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "PLEASE VERIFY THE SAMPLE RATE HOP LENGTH AND MEL FILTER CONFIGURATION BEFORE INFERENCE",
+ "raw": {
+ "normalized_reference": "please verify the sample rate, hop length, and mel filter configuration before inference.",
+ "normalized_hypothesis": "please verify the sample rate hop length and mel filter configuration before inference",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "please verify the sample rate hop length and mel filter configuration before inference",
+ "normalized_hypothesis": "please verify the sample rate hop length and mel filter configuration before inference",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0183",
+ "category": "conversational",
+ "text": "That's the worst idea I've ever heard and I'm completely in.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0183.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THAT'S THE WORST IDEA I'VE EVER HEARD AND I'M COMPLETELY IN",
+ "raw": {
+ "normalized_reference": "that's the worst idea i've ever heard and i'm completely in.",
+ "normalized_hypothesis": "that's the worst idea i've ever heard and i'm completely in",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "that is the worst idea i have ever heard and i am completely in",
+ "normalized_hypothesis": "that is the worst idea i have ever heard and i am completely in",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0184",
+ "category": "emotional",
+ "text": "I remember thinking: this is it. This is the thing I've been waiting for.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0184.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I REMEMBER THINKING THIS IS IT THIS IS THE THING I'VE BEEN WAITING FOR",
+ "raw": {
+ "normalized_reference": "i remember thinking: this is it. this is the thing i've been waiting for.",
+ "normalized_hypothesis": "i remember thinking this is it this is the thing i've been waiting for",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i remember thinking this is it this is the thing i have been waiting for",
+ "normalized_hypothesis": "i remember thinking this is it this is the thing i have been waiting for",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0185",
+ "category": "narrative",
+ "text": "We never talked about it again, which I think was the right call.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0185.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WE NEVER TALKED ABOUT IT AGAIN WHICH I THINK WAS THE RIGHT CALL",
+ "raw": {
+ "normalized_reference": "we never talked about it again, which i think was the right call.",
+ "normalized_hypothesis": "we never talked about it again which i think was the right call",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "we never talked about it again which i think was the right call",
+ "normalized_hypothesis": "we never talked about it again which i think was the right call",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0186",
+ "category": "short",
+ "text": "In hindsight, obvious.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0186.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IN HINDSIGHT OBVIOUS",
+ "raw": {
+ "normalized_reference": "in hindsight, obvious.",
+ "normalized_hypothesis": "in hindsight obvious",
+ "errors": 0,
+ "words": 3,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "in hindsight obvious",
+ "normalized_hypothesis": "in hindsight obvious",
+ "errors": 0,
+ "words": 3,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0187",
+ "category": "technical",
+ "text": "Once it's configured correctly, it basically runs itself.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0187.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "ONCE IT'S CONFIGURED CORRECTLY IT BASICALLY RUNS ITSELF",
+ "raw": {
+ "normalized_reference": "once it's configured correctly, it basically runs itself.",
+ "normalized_hypothesis": "once it's configured correctly it basically runs itself",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "once it is configured correctly it basically runs itself",
+ "normalized_hypothesis": "once it is configured correctly it basically runs itself",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0188",
+ "category": "conversational",
+ "text": "You got engaged? When? How? Tell me everything right now.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0188.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "YOU GOT ENGAGED WHEN HOW TELL ME EVERYTHING RIGHT NOW",
+ "raw": {
+ "normalized_reference": "you got engaged? when? how? tell me everything right now.",
+ "normalized_hypothesis": "you got engaged when how tell me everything right now",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "you got engaged when how tell me everything right now",
+ "normalized_hypothesis": "you got engaged when how tell me everything right now",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0189",
+ "category": "emotional",
+ "text": "It's hard to explain how much that moment meant to me.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0189.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT'S HARD TO EXPLAIN HOW MUCH THAT MOMENT MEANT TO ME",
+ "raw": {
+ "normalized_reference": "it's hard to explain how much that moment meant to me.",
+ "normalized_hypothesis": "it's hard to explain how much that moment meant to me",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it is hard to explain how much that moment meant to me",
+ "normalized_hypothesis": "it is hard to explain how much that moment meant to me",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0190",
+ "category": "narrative",
+ "text": "He told the story three different times that night and somehow it got better each time.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0190.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE TOLD THI STORY THREE DIFFERENT TIMES THAT NIGHT AND SOMEHOW IT GOT BETTER EACH TIME",
+ "raw": {
+ "normalized_reference": "he told the story three different times that night and somehow it got better each time.",
+ "normalized_hypothesis": "he told thi story three different times that night and somehow it got better each time",
+ "errors": 1,
+ "words": 16,
+ "wer": 0.0625
+ },
+ "semantic": {
+ "normalized_reference": "he told the story 3 different times that night and somehow it got better each time",
+ "normalized_hypothesis": "he told thi story 3 different times that night and somehow it got better each time",
+ "errors": 1,
+ "words": 16,
+ "wer": 0.0625
+ }
+ },
+ {
+ "id": "release500_0191",
+ "category": "short",
+ "text": "Right. Moving on.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0191.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "RIGHT MOVING ON",
+ "raw": {
+ "normalized_reference": "right. moving on.",
+ "normalized_hypothesis": "right moving on",
+ "errors": 0,
+ "words": 3,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "right moving on",
+ "normalized_hypothesis": "right moving on",
+ "errors": 0,
+ "words": 3,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0192",
+ "category": "conversational",
+ "text": "No, I did not move your keys; check beside the small ceramic bowl.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0192.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "NO I DID NOT MOVE YOUR KEYS CHECK BESIDE THE SMALL SERAMIC BOWL",
+ "raw": {
+ "normalized_reference": "no, i did not move your keys; check beside the small ceramic bowl.",
+ "normalized_hypothesis": "no i did not move your keys check beside the small seramic bowl",
+ "errors": 1,
+ "words": 13,
+ "wer": 0.07692307692307693
+ },
+ "semantic": {
+ "normalized_reference": "no i did not move your keys check beside the small ceramic bowl",
+ "normalized_hypothesis": "no i did not move your keys check beside the small seramic bowl",
+ "errors": 1,
+ "words": 13,
+ "wer": 0.07692307692307693
+ }
+ },
+ {
+ "id": "release500_0193",
+ "category": "emotional",
+ "text": "I was devastated. I'm not going to pretend otherwise.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0193.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I WAS DEVASTATED I'M NOT GOING TO PRETEND OTHERWISE",
+ "raw": {
+ "normalized_reference": "i was devastated. i'm not going to pretend otherwise.",
+ "normalized_hypothesis": "i was devastated i'm not going to pretend otherwise",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i was devastated i am not going to pretend otherwise",
+ "normalized_hypothesis": "i was devastated i am not going to pretend otherwise",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0194",
+ "category": "narrative",
+ "text": "That was the last summer before everything changed, though we didn't know it yet.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0194.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THAT WAS THE LAST SUMMER BEFORE EVERYTHING CHANGED THOUGH WE DIDN'T KNOW IT YET",
+ "raw": {
+ "normalized_reference": "that was the last summer before everything changed, though we didn't know it yet.",
+ "normalized_hypothesis": "that was the last summer before everything changed though we didn't know it yet",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "that was the last summer before everything changed though we did not know it yet",
+ "normalized_hypothesis": "that was the last summer before everything changed though we did not know it yet",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0195",
+ "category": "short",
+ "text": "Bold of you to assume.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0195.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "BOLD OF YOU TO ASSUME",
+ "raw": {
+ "normalized_reference": "bold of you to assume.",
+ "normalized_hypothesis": "bold of you to assume",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "bold of you to assume",
+ "normalized_hypothesis": "bold of you to assume",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0196",
+ "category": "conversational",
+ "text": "It's one of those decisions that makes total sense in the moment.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0196.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT'S ONE OF THOSE DECISIONS THAT MAKES TOTAL SENSE IN THE MOMENT",
+ "raw": {
+ "normalized_reference": "it's one of those decisions that makes total sense in the moment.",
+ "normalized_hypothesis": "it's one of those decisions that makes total sense in the moment",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it is one of those decisions that makes total sense in the moment",
+ "normalized_hypothesis": "it is one of those decisions that makes total sense in the moment",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0197",
+ "category": "emotional",
+ "text": "I was furious. Not the loud kind. The quiet kind that lasts for days.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0197.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I WAS FURIOUS NOT THE LOUD KIND THE QUIET KIND THAT LASTS FOR DAYS",
+ "raw": {
+ "normalized_reference": "i was furious. not the loud kind. the quiet kind that lasts for days.",
+ "normalized_hypothesis": "i was furious not the loud kind the quiet kind that lasts for days",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i was furious not the loud kind the quiet kind that lasts for days",
+ "normalized_hypothesis": "i was furious not the loud kind the quiet kind that lasts for days",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0198",
+ "category": "narrative",
+ "text": "That's the part of the story I usually leave out.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0198.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THAT'S THE PART OF THE STORY I USUALLY LEAVE OUT",
+ "raw": {
+ "normalized_reference": "that's the part of the story i usually leave out.",
+ "normalized_hypothesis": "that's the part of the story i usually leave out",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "that is the part of the story i usually leave out",
+ "normalized_hypothesis": "that is the part of the story i usually leave out",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0199",
+ "category": "short",
+ "text": "I've said too much.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0199.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I'VE SAID TOO MUCH",
+ "raw": {
+ "normalized_reference": "i've said too much.",
+ "normalized_hypothesis": "i've said too much",
+ "errors": 0,
+ "words": 4,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i have said too much",
+ "normalized_hypothesis": "i have said too much",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0200",
+ "category": "conversational",
+ "text": "I've been faking that I understood this the entire time.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0200.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I'VE BEEN FAKING THAT I UNDERSTOOD THIS THE ENTIRE TIME",
+ "raw": {
+ "normalized_reference": "i've been faking that i understood this the entire time.",
+ "normalized_hypothesis": "i've been faking that i understood this the entire time",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i have been faking that i understood this the entire time",
+ "normalized_hypothesis": "i have been faking that i understood this the entire time",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0201",
+ "category": "phonetic_balance",
+ "text": "Peace be unto you and grace before the Lord.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0201.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "PEACE BE UNTO YOU IN GRACE BEFORE THE LORD",
+ "raw": {
+ "normalized_reference": "peace be unto you and grace before the lord.",
+ "normalized_hypothesis": "peace be unto you in grace before the lord",
+ "errors": 1,
+ "words": 9,
+ "wer": 0.1111111111111111
+ },
+ "semantic": {
+ "normalized_reference": "peace be unto you and grace before the lord",
+ "normalized_hypothesis": "peace be unto you in grace before the lord",
+ "errors": 1,
+ "words": 9,
+ "wer": 0.1111111111111111
+ }
+ },
+ {
+ "id": "release500_0202",
+ "category": "phonetic_balance",
+ "text": "From now on we're pals.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0202.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "FROM NOW ON WEIR PALLS",
+ "raw": {
+ "normalized_reference": "from now on we're pals.",
+ "normalized_hypothesis": "from now on weir palls",
+ "errors": 2,
+ "words": 5,
+ "wer": 0.4
+ },
+ "semantic": {
+ "normalized_reference": "from now on we are pals",
+ "normalized_hypothesis": "from now on weir palls",
+ "errors": 3,
+ "words": 6,
+ "wer": 0.5
+ }
+ },
+ {
+ "id": "release500_0203",
+ "category": "phonetic_balance",
+ "text": "Here we allow our solicitors to look after our legal work.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0203.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HERE WE ALLOW OUR SOLICITORS TO LOOK AFTER OUR LEGAL WORK",
+ "raw": {
+ "normalized_reference": "here we allow our solicitors to look after our legal work.",
+ "normalized_hypothesis": "here we allow our solicitors to look after our legal work",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "here we allow our solicitors to look after our legal work",
+ "normalized_hypothesis": "here we allow our solicitors to look after our legal work",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0204",
+ "category": "phonetic_balance",
+ "text": "Her words sent a strange chill through Philip.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0204.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HER WORDS SENT A STRANGE CHILL THROUGH PHILIP",
+ "raw": {
+ "normalized_reference": "her words sent a strange chill through philip.",
+ "normalized_hypothesis": "her words sent a strange chill through philip",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "her words sent a strange chill through philip",
+ "normalized_hypothesis": "her words sent a strange chill through philip",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0205",
+ "category": "phonetic_balance",
+ "text": "It was introduced by Representative Dick of Ohio.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0205.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT WAS INTRODUCED BY REPRESENTATIVE DICK OF OHIO",
+ "raw": {
+ "normalized_reference": "it was introduced by representative dick of ohio.",
+ "normalized_hypothesis": "it was introduced by representative dick of ohio",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it was introduced by representative dick of ohio",
+ "normalized_hypothesis": "it was introduced by representative dick of ohio",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0206",
+ "category": "phonetic_balance",
+ "text": "It's the aurora borealis.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0206.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT'S THE AURORA BORIALIS",
+ "raw": {
+ "normalized_reference": "it's the aurora borealis.",
+ "normalized_hypothesis": "it's the aurora borialis",
+ "errors": 1,
+ "words": 4,
+ "wer": 0.25
+ },
+ "semantic": {
+ "normalized_reference": "it is the aurora borealis",
+ "normalized_hypothesis": "it is the aurora borialis",
+ "errors": 1,
+ "words": 5,
+ "wer": 0.2
+ }
+ },
+ {
+ "id": "release500_0207",
+ "category": "phonetic_balance",
+ "text": "The boy grew and prospered.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0207.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE BOY GREW AND PROSPERED",
+ "raw": {
+ "normalized_reference": "the boy grew and prospered.",
+ "normalized_hypothesis": "the boy grew and prospered",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the boy grew and prospered",
+ "normalized_hypothesis": "the boy grew and prospered",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0208",
+ "category": "phonetic_balance",
+ "text": "His voice was passionately rebellious.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0208.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HIS VOICE WAS PASSIONATELY REBELLIOUS",
+ "raw": {
+ "normalized_reference": "his voice was passionately rebellious.",
+ "normalized_hypothesis": "his voice was passionately rebellious",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "his voice was passionately rebellious",
+ "normalized_hypothesis": "his voice was passionately rebellious",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0209",
+ "category": "phonetic_balance",
+ "text": "That's where they cut off the Scottish Chiefs and killed all hands.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0209.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THAT'S WHERE THEY CUT OFF THE SCOTTISH CHIEFS AND KILLED ALL HANDS",
+ "raw": {
+ "normalized_reference": "that's where they cut off the scottish chiefs and killed all hands.",
+ "normalized_hypothesis": "that's where they cut off the scottish chiefs and killed all hands",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "that is where they cut off the scottish chiefs and killed all hands",
+ "normalized_hypothesis": "that is where they cut off the scottish chiefs and killed all hands",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0210",
+ "category": "phonetic_balance",
+ "text": "And Tom King patiently endured.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0210.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "AND TOM KING PATIENTLY ENDURED",
+ "raw": {
+ "normalized_reference": "and tom king patiently endured.",
+ "normalized_hypothesis": "and tom king patiently endured",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "and tom king patiently endured",
+ "normalized_hypothesis": "and tom king patiently endured",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0211",
+ "category": "phonetic_balance",
+ "text": "From that moment his friendship for Belize turns to hatred and jealousy.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0211.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "FROM THAT MOMENT HIS FRIENDSHIP FOR BELLIS TURNS TO HATRED AND JEALOUSY",
+ "raw": {
+ "normalized_reference": "from that moment his friendship for belize turns to hatred and jealousy.",
+ "normalized_hypothesis": "from that moment his friendship for bellis turns to hatred and jealousy",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "from that moment his friendship for belize turns to hatred and jealousy",
+ "normalized_hypothesis": "from that moment his friendship for bellis turns to hatred and jealousy",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ }
+ },
+ {
+ "id": "release500_0212",
+ "category": "phonetic_balance",
+ "text": "Then you don't believe in altruism.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0212.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THEN YOU DON'T BELIEVE IN ALTRUISM",
+ "raw": {
+ "normalized_reference": "then you don't believe in altruism.",
+ "normalized_hypothesis": "then you don't believe in altruism",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "then you do not believe in altruism",
+ "normalized_hypothesis": "then you do not believe in altruism",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0213",
+ "category": "phonetic_balance",
+ "text": "What Pascal glimpsed with the vision of a seer, I have lived.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0213.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WHAT PASCAELL GLIMPSED WITH THE VISION OF A SEER I HAVE LIVED",
+ "raw": {
+ "normalized_reference": "what pascal glimpsed with the vision of a seer, i have lived.",
+ "normalized_hypothesis": "what pascaell glimpsed with the vision of a seer i have lived",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "what pascal glimpsed with the vision of a seer i have lived",
+ "normalized_hypothesis": "what pascaell glimpsed with the vision of a seer i have lived",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ }
+ },
+ {
+ "id": "release500_0214",
+ "category": "phonetic_balance",
+ "text": "You were making them talk shop, Ruth charged him.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0214.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "YOU WERE MAKING THEM TALK SHOP RUTH CHARGED",
+ "raw": {
+ "normalized_reference": "you were making them talk shop, ruth charged him.",
+ "normalized_hypothesis": "you were making them talk shop ruth charged",
+ "errors": 1,
+ "words": 9,
+ "wer": 0.1111111111111111
+ },
+ "semantic": {
+ "normalized_reference": "you were making them talk shop ruth charged him",
+ "normalized_hypothesis": "you were making them talk shop ruth charged",
+ "errors": 1,
+ "words": 9,
+ "wer": 0.1111111111111111
+ }
+ },
+ {
+ "id": "release500_0215",
+ "category": "phonetic_balance",
+ "text": "Philip snatched at the letter which Gregson held out to him.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0215.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "PHILIP SNATCHED AT THE LETTER WHICH GREGSON HELD OUT TO HIM",
+ "raw": {
+ "normalized_reference": "philip snatched at the letter which gregson held out to him.",
+ "normalized_hypothesis": "philip snatched at the letter which gregson held out to him",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "philip snatched at the letter which gregson held out to him",
+ "normalized_hypothesis": "philip snatched at the letter which gregson held out to him",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0216",
+ "category": "phonetic_balance",
+ "text": "For a time the exciting thrill of his adventure was gone.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0216.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "FOR A TIME THE EXCITING THRILL OF HIS ADVENTURE WAS GONE",
+ "raw": {
+ "normalized_reference": "for a time the exciting thrill of his adventure was gone.",
+ "normalized_hypothesis": "for a time the exciting thrill of his adventure was gone",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "for a time the exciting thrill of his adventure was gone",
+ "normalized_hypothesis": "for a time the exciting thrill of his adventure was gone",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0217",
+ "category": "phonetic_balance",
+ "text": "Obviously, it was a disease that could be contracted by contact.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0217.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "OBVIOUSLY IT WAS A DISEASE THAT COULD BE CONTRACTED BY CONTACT",
+ "raw": {
+ "normalized_reference": "obviously, it was a disease that could be contracted by contact.",
+ "normalized_hypothesis": "obviously it was a disease that could be contracted by contact",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "obviously it was a disease that could be contracted by contact",
+ "normalized_hypothesis": "obviously it was a disease that could be contracted by contact",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0218",
+ "category": "phonetic_balance",
+ "text": "Blind with rage, he darted in.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0218.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "BLIND WITH RAGE HE DARTED I",
+ "raw": {
+ "normalized_reference": "blind with rage, he darted in.",
+ "normalized_hypothesis": "blind with rage he darted i",
+ "errors": 1,
+ "words": 6,
+ "wer": 0.16666666666666666
+ },
+ "semantic": {
+ "normalized_reference": "blind with rage he darted in",
+ "normalized_hypothesis": "blind with rage he darted i",
+ "errors": 1,
+ "words": 6,
+ "wer": 0.16666666666666666
+ }
+ },
+ {
+ "id": "release500_0219",
+ "category": "phonetic_balance",
+ "text": "You live on an income which your father earned.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0219.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "YOU LIVE ON AN INCOME WHICH YOUR FATHER EARNED",
+ "raw": {
+ "normalized_reference": "you live on an income which your father earned.",
+ "normalized_hypothesis": "you live on an income which your father earned",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "you live on an income which your father earned",
+ "normalized_hypothesis": "you live on an income which your father earned",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0220",
+ "category": "phonetic_balance",
+ "text": "Something about it seemed to fascinate him, to challenge his presence.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0220.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SOMETHING ABOUT IT SEEMED TO FASCINATE HIM TO CHALLENGE HIS PRESENCE",
+ "raw": {
+ "normalized_reference": "something about it seemed to fascinate him, to challenge his presence.",
+ "normalized_hypothesis": "something about it seemed to fascinate him to challenge his presence",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "something about it seemed to fascinate him to challenge his presence",
+ "normalized_hypothesis": "something about it seemed to fascinate him to challenge his presence",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0221",
+ "category": "phonetic_balance",
+ "text": "Her beautiful hair was done up in shining coils.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0221.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HER BEAUTIFUL HAIR WAS DONE UP IN SHINING COILS",
+ "raw": {
+ "normalized_reference": "her beautiful hair was done up in shining coils.",
+ "normalized_hypothesis": "her beautiful hair was done up in shining coils",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "her beautiful hair was done up in shining coils",
+ "normalized_hypothesis": "her beautiful hair was done up in shining coils",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0222",
+ "category": "phonetic_balance",
+ "text": "It does, was her audacious answer.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0222.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT DOES WAS HER AUDACIOUS ANSWER",
+ "raw": {
+ "normalized_reference": "it does, was her audacious answer.",
+ "normalized_hypothesis": "it does was her audacious answer",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it does was her audacious answer",
+ "normalized_hypothesis": "it does was her audacious answer",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0223",
+ "category": "phonetic_balance",
+ "text": "He caught himself with a jerk.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0223.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE CAUGHT HIMSELF WITH A JERK",
+ "raw": {
+ "normalized_reference": "he caught himself with a jerk.",
+ "normalized_hypothesis": "he caught himself with a jerk",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he caught himself with a jerk",
+ "normalized_hypothesis": "he caught himself with a jerk",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0224",
+ "category": "phonetic_balance",
+ "text": "He did not believe in the burning of daylight for such a luxury.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0224.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE DID NOT BELIEVE IN THE BURNING OF DAYLIGHT FOR SUCH A LUXURY",
+ "raw": {
+ "normalized_reference": "he did not believe in the burning of daylight for such a luxury.",
+ "normalized_hypothesis": "he did not believe in the burning of daylight for such a luxury",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he did not believe in the burning of daylight for such a luxury",
+ "normalized_hypothesis": "he did not believe in the burning of daylight for such a luxury",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0225",
+ "category": "phonetic_balance",
+ "text": "He had comparatively no advantages at first.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0225.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE HAD COMPARATIVELY NO ADVANTAGES AT FIRST",
+ "raw": {
+ "normalized_reference": "he had comparatively no advantages at first.",
+ "normalized_hypothesis": "he had comparatively no advantages at first",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he had comparatively no advantages at 1st",
+ "normalized_hypothesis": "he had comparatively no advantages at 1st",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0226",
+ "category": "phonetic_balance",
+ "text": "Such men believe, when they come together.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0226.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SUCH MEN BELIEVE WHEN THEY COME TOGETHER",
+ "raw": {
+ "normalized_reference": "such men believe, when they come together.",
+ "normalized_hypothesis": "such men believe when they come together",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "such men believe when they come together",
+ "normalized_hypothesis": "such men believe when they come together",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0227",
+ "category": "phonetic_balance",
+ "text": "The ship should be in within a week or ten days.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0227.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE SHIP SHOULD BE IN WITHIN A WEEK OR TEN DAYS",
+ "raw": {
+ "normalized_reference": "the ship should be in within a week or ten days.",
+ "normalized_hypothesis": "the ship should be in within a week or ten days",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the ship should be in within a week or 10 days",
+ "normalized_hypothesis": "the ship should be in within a week or 10 days",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0228",
+ "category": "phonetic_balance",
+ "text": "Such a dog the wise driver kills, or turns loose.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0228.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SUCH A DOG THE WISE DRIVER KILLS OR TURNS LOOSE",
+ "raw": {
+ "normalized_reference": "such a dog the wise driver kills, or turns loose.",
+ "normalized_hypothesis": "such a dog the wise driver kills or turns loose",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "such a dog the wise driver kills or turns loose",
+ "normalized_hypothesis": "such a dog the wise driver kills or turns loose",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0229",
+ "category": "phonetic_balance",
+ "text": "You have heard always how he was the lover of the Princess Naomi.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0229.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "YOU HAVE HEARD ALWAYS HOW HE WAS THE LOVER OF THE PRINCESS NAOMI",
+ "raw": {
+ "normalized_reference": "you have heard always how he was the lover of the princess naomi.",
+ "normalized_hypothesis": "you have heard always how he was the lover of the princess naomi",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "you have heard always how he was the lover of the princess naomi",
+ "normalized_hypothesis": "you have heard always how he was the lover of the princess naomi",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0230",
+ "category": "phonetic_balance",
+ "text": "I want to know how all this is possible.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0230.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I WANT TO KNOW HOW ALL THIS IS POSSIBLE",
+ "raw": {
+ "normalized_reference": "i want to know how all this is possible.",
+ "normalized_hypothesis": "i want to know how all this is possible",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i want to know how all this is possible",
+ "normalized_hypothesis": "i want to know how all this is possible",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0231",
+ "category": "phonetic_balance",
+ "text": "It was my idea to a tee.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0231.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT WAS MY IDEA TO A TA",
+ "raw": {
+ "normalized_reference": "it was my idea to a tee.",
+ "normalized_hypothesis": "it was my idea to a ta",
+ "errors": 1,
+ "words": 7,
+ "wer": 0.14285714285714285
+ },
+ "semantic": {
+ "normalized_reference": "it was my idea to a tee",
+ "normalized_hypothesis": "it was my idea to a ta",
+ "errors": 1,
+ "words": 7,
+ "wer": 0.14285714285714285
+ }
+ },
+ {
+ "id": "release500_0232",
+ "category": "phonetic_balance",
+ "text": "They likewise are disinclined to being eaten.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0232.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THEY LIKEWISE ARE DISINCLINED TO BEING EATEN",
+ "raw": {
+ "normalized_reference": "they likewise are disinclined to being eaten.",
+ "normalized_hypothesis": "they likewise are disinclined to being eaten",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "they likewise are disinclined to being eaten",
+ "normalized_hypothesis": "they likewise are disinclined to being eaten",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0233",
+ "category": "phonetic_balance",
+ "text": "This state of mind comes of an undue prominence of the ego.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0233.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THIS STATE OF MIND COMES OF AN UNDUE PROMINENCE OF THE EGO",
+ "raw": {
+ "normalized_reference": "this state of mind comes of an undue prominence of the ego.",
+ "normalized_hypothesis": "this state of mind comes of an undue prominence of the ego",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "this state of mind comes of an undue prominence of the ego",
+ "normalized_hypothesis": "this state of mind comes of an undue prominence of the ego",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0234",
+ "category": "phonetic_balance",
+ "text": "Then it is as I said, Womble announced with finality.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0234.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THEN IT IS AS I SAID WAMBLE ANNOUNCED WITH FINALITY",
+ "raw": {
+ "normalized_reference": "then it is as i said, womble announced with finality.",
+ "normalized_hypothesis": "then it is as i said wamble announced with finality",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ },
+ "semantic": {
+ "normalized_reference": "then it is as i said womble announced with finality",
+ "normalized_hypothesis": "then it is as i said wamble announced with finality",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ }
+ },
+ {
+ "id": "release500_0235",
+ "category": "phonetic_balance",
+ "text": "I can see that knife now.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0235.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I CAN SEE THAT KNIFE NOW",
+ "raw": {
+ "normalized_reference": "i can see that knife now.",
+ "normalized_hypothesis": "i can see that knife now",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i can see that knife now",
+ "normalized_hypothesis": "i can see that knife now",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0236",
+ "category": "phonetic_balance",
+ "text": "If I meet her again I shall apologize, said Eileen.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0236.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IF I MEET HER AGAIN I SHALL APOLOGIZE SAID ILEN",
+ "raw": {
+ "normalized_reference": "if i meet her again i shall apologize, said eileen.",
+ "normalized_hypothesis": "if i meet her again i shall apologize said ilen",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ },
+ "semantic": {
+ "normalized_reference": "if i meet her again i shall apologize said eileen",
+ "normalized_hypothesis": "if i meet her again i shall apologize said ilen",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ }
+ },
+ {
+ "id": "release500_0237",
+ "category": "phonetic_balance",
+ "text": "A wildly exciting time was his during the week preceding Thursday the eighteenth.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0237.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "A WILDLY EXCITING TIME WAS HIS DURING THE WEEK PRECEDING THURSDAY THE EIGHTEENTH",
+ "raw": {
+ "normalized_reference": "a wildly exciting time was his during the week preceding thursday the eighteenth.",
+ "normalized_hypothesis": "a wildly exciting time was his during the week preceding thursday the eighteenth",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "a wildly exciting time was his during the week preceding thursday the 18th",
+ "normalized_hypothesis": "a wildly exciting time was his during the week preceding thursday the 18th",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0238",
+ "category": "phonetic_balance",
+ "text": "In what bucolic school of fence he had been taught was beyond imagining.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0238.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IN WHAT BEUCALIC SCHOOL OF FENCE HE HAD BEEN TAUGHT WAS BEYOND IMAGINING",
+ "raw": {
+ "normalized_reference": "in what bucolic school of fence he had been taught was beyond imagining.",
+ "normalized_hypothesis": "in what beucalic school of fence he had been taught was beyond imagining",
+ "errors": 1,
+ "words": 13,
+ "wer": 0.07692307692307693
+ },
+ "semantic": {
+ "normalized_reference": "in what bucolic school of fence he had been taught was beyond imagining",
+ "normalized_hypothesis": "in what beucalic school of fence he had been taught was beyond imagining",
+ "errors": 1,
+ "words": 13,
+ "wer": 0.07692307692307693
+ }
+ },
+ {
+ "id": "release500_0239",
+ "category": "phonetic_balance",
+ "text": "So she said, the irate skipper dashed on.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0239.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SO SHE SAID THE IRATE SKIPPER DASHED ON",
+ "raw": {
+ "normalized_reference": "so she said, the irate skipper dashed on.",
+ "normalized_hypothesis": "so she said the irate skipper dashed on",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "so she said the irate skipper dashed on",
+ "normalized_hypothesis": "so she said the irate skipper dashed on",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0240",
+ "category": "phonetic_balance",
+ "text": "It is not an attempt to smash the market.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0240.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT IS NOT AN ATTEMPT TO SMASH THE MARKET",
+ "raw": {
+ "normalized_reference": "it is not an attempt to smash the market.",
+ "normalized_hypothesis": "it is not an attempt to smash the market",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it is not an attempt to smash the market",
+ "normalized_hypothesis": "it is not an attempt to smash the market",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0241",
+ "category": "phonetic_balance",
+ "text": "He destroyed everything that had belonged to the woman.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0241.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE DESTROYED EVERYTHING THAT HAD BELONGED TO THE WOMAN",
+ "raw": {
+ "normalized_reference": "he destroyed everything that had belonged to the woman.",
+ "normalized_hypothesis": "he destroyed everything that had belonged to the woman",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he destroyed everything that had belonged to the woman",
+ "normalized_hypothesis": "he destroyed everything that had belonged to the woman",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0242",
+ "category": "phonetic_balance",
+ "text": "Did Thorpe go to see any one in Churchill.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0242.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "DID THORPE GO TO SEE ANY ONE IN CHURCHILL",
+ "raw": {
+ "normalized_reference": "did thorpe go to see any one in churchill.",
+ "normalized_hypothesis": "did thorpe go to see any one in churchill",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "did thorpe go to see any one in churchill",
+ "normalized_hypothesis": "did thorpe go to see any one in churchill",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0243",
+ "category": "phonetic_balance",
+ "text": "Many other similar disconcerting omissions will be noticed in the Manuscript.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0243.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "MANY OTHER SIMILAR DISCONCERTING OMISSIONS WILL BE NOTICED IN THE MANUSCRIPT",
+ "raw": {
+ "normalized_reference": "many other similar disconcerting omissions will be noticed in the manuscript.",
+ "normalized_hypothesis": "many other similar disconcerting omissions will be noticed in the manuscript",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "many other similar disconcerting omissions will be noticed in the manuscript",
+ "normalized_hypothesis": "many other similar disconcerting omissions will be noticed in the manuscript",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0244",
+ "category": "phonetic_balance",
+ "text": "The ringing of the big bell aroused him.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0244.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE RINGING OF THE BIG BELL AROUSED HIM",
+ "raw": {
+ "normalized_reference": "the ringing of the big bell aroused him.",
+ "normalized_hypothesis": "the ringing of the big bell aroused him",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the ringing of the big bell aroused him",
+ "normalized_hypothesis": "the ringing of the big bell aroused him",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0245",
+ "category": "phonetic_balance",
+ "text": "A sinewy hand, dripping with water, was clutching the rail.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0245.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "A SINEWY HAND DRIPPING WITH WATER WAS CLUTCHING THE RAIL",
+ "raw": {
+ "normalized_reference": "a sinewy hand, dripping with water, was clutching the rail.",
+ "normalized_hypothesis": "a sinewy hand dripping with water was clutching the rail",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "a sinewy hand dripping with water was clutching the rail",
+ "normalized_hypothesis": "a sinewy hand dripping with water was clutching the rail",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0246",
+ "category": "phonetic_balance",
+ "text": "Enters now the psychology of the situation.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0246.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "ENTERS NOW THE PSYCHOLOGY OF THE SITUATION",
+ "raw": {
+ "normalized_reference": "enters now the psychology of the situation.",
+ "normalized_hypothesis": "enters now the psychology of the situation",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "enters now the psychology of the situation",
+ "normalized_hypothesis": "enters now the psychology of the situation",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0247",
+ "category": "phonetic_balance",
+ "text": "Wash your hands of me.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0247.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WASH YOUR HANDS OF ME",
+ "raw": {
+ "normalized_reference": "wash your hands of me.",
+ "normalized_hypothesis": "wash your hands of me",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "wash your hands of me",
+ "normalized_hypothesis": "wash your hands of me",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0248",
+ "category": "phonetic_balance",
+ "text": "He confessed that the sketch had startled him.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0248.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE CONFESSED THAT THE SKETCH HAD STARTLED HIM",
+ "raw": {
+ "normalized_reference": "he confessed that the sketch had startled him.",
+ "normalized_hypothesis": "he confessed that the sketch had startled him",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he confessed that the sketch had startled him",
+ "normalized_hypothesis": "he confessed that the sketch had startled him",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0249",
+ "category": "phonetic_balance",
+ "text": "Her face was against his breast.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0249.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HER FACE WAS AGAINST HIS BREAST",
+ "raw": {
+ "normalized_reference": "her face was against his breast.",
+ "normalized_hypothesis": "her face was against his breast",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "her face was against his breast",
+ "normalized_hypothesis": "her face was against his breast",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0250",
+ "category": "phonetic_balance",
+ "text": "He was a pariah; a wanderer without a friend or a home.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0250.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE WAS A PARIAH A WANDERER WITHOUT A FRIEND OR A HOME",
+ "raw": {
+ "normalized_reference": "he was a pariah; a wanderer without a friend or a home.",
+ "normalized_hypothesis": "he was a pariah a wanderer without a friend or a home",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he was a pariah a wanderer without a friend or a home",
+ "normalized_hypothesis": "he was a pariah a wanderer without a friend or a home",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0251",
+ "category": "phonetic_balance",
+ "text": "A cry of joy burst from Philip's lips.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0251.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "A CRY OF JOY BURST FROM PHILIP'S LIPS",
+ "raw": {
+ "normalized_reference": "a cry of joy burst from philip's lips.",
+ "normalized_hypothesis": "a cry of joy burst from philip's lips",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "a cry of joy burst from philip is lips",
+ "normalized_hypothesis": "a cry of joy burst from philip is lips",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0252",
+ "category": "phonetic_balance",
+ "text": "And the big Persian knew of his existence before he did of hers.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0252.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "AND THE BIG PERSIAN KNEW OF HIS EXISTENCE BEFORE HE DID OF HERS",
+ "raw": {
+ "normalized_reference": "and the big persian knew of his existence before he did of hers.",
+ "normalized_hypothesis": "and the big persian knew of his existence before he did of hers",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "and the big persian knew of his existence before he did of hers",
+ "normalized_hypothesis": "and the big persian knew of his existence before he did of hers",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0253",
+ "category": "phonetic_balance",
+ "text": "The men stared into each other's face.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0253.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE MEN STARED INTO EACH OTHER'S FACE",
+ "raw": {
+ "normalized_reference": "the men stared into each other's face.",
+ "normalized_hypothesis": "the men stared into each other's face",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the men stared into each other is face",
+ "normalized_hypothesis": "the men stared into each other is face",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0254",
+ "category": "phonetic_balance",
+ "text": "It is the fire, partly, she said.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0254.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT IS THE FIRE PARTLY SHE SAID",
+ "raw": {
+ "normalized_reference": "it is the fire, partly, she said.",
+ "normalized_hypothesis": "it is the fire partly she said",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it is the fire partly she said",
+ "normalized_hypothesis": "it is the fire partly she said",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0255",
+ "category": "phonetic_balance",
+ "text": "It seemed nearer to him since he had seen and talked with Gregson.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0255.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT SEEMED NEARER TO HIM SINCE HE HAD SEEN AND TALKED WITH GREGSON",
+ "raw": {
+ "normalized_reference": "it seemed nearer to him since he had seen and talked with gregson.",
+ "normalized_hypothesis": "it seemed nearer to him since he had seen and talked with gregson",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it seemed nearer to him since he had seen and talked with gregson",
+ "normalized_hypothesis": "it seemed nearer to him since he had seen and talked with gregson",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0256",
+ "category": "phonetic_balance",
+ "text": "Among my minor afflictions, I may mention a new and mysterious one.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0256.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "AMONG MY MINOR AFFLICTIONS I MAY MENTION A NEW AND MYSTERIOUS ONE",
+ "raw": {
+ "normalized_reference": "among my minor afflictions, i may mention a new and mysterious one.",
+ "normalized_hypothesis": "among my minor afflictions i may mention a new and mysterious one",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "among my minor afflictions i may mention a new and mysterious one",
+ "normalized_hypothesis": "among my minor afflictions i may mention a new and mysterious one",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0257",
+ "category": "phonetic_balance",
+ "text": "Between him and all domestic animals there must be no hostilities.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0257.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "BETWEEN HIM AND ALL DOMESTIC ANIMALS THERE MUST BE NO HOSTILITIES",
+ "raw": {
+ "normalized_reference": "between him and all domestic animals there must be no hostilities.",
+ "normalized_hypothesis": "between him and all domestic animals there must be no hostilities",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "between him and all domestic animals there must be no hostilities",
+ "normalized_hypothesis": "between him and all domestic animals there must be no hostilities",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0258",
+ "category": "phonetic_balance",
+ "text": "With these arguments he convinced himself that he should go on alone.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0258.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WITH THESE ARGUMENTS HE CONVINCED HIMSELF THAT HE SHOULD GO ON ALONE",
+ "raw": {
+ "normalized_reference": "with these arguments he convinced himself that he should go on alone.",
+ "normalized_hypothesis": "with these arguments he convinced himself that he should go on alone",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "with these arguments he convinced himself that he should go on alone",
+ "normalized_hypothesis": "with these arguments he convinced himself that he should go on alone",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0259",
+ "category": "phonetic_balance",
+ "text": "We are consumed in our own flesh-pots.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0259.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WE ARE CONSUMED IN OUR OWN FLESH POTS",
+ "raw": {
+ "normalized_reference": "we are consumed in our own flesh-pots.",
+ "normalized_hypothesis": "we are consumed in our own flesh pots",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "we are consumed in our own flesh pots",
+ "normalized_hypothesis": "we are consumed in our own flesh pots",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0260",
+ "category": "phonetic_balance",
+ "text": "Mercedes screamed, cried, laughed, and manifested the chaotic abandonment of hysteria.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0260.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "MERCEDES SCREAMED CRIED LAUGHED AND MANIFESTED THE CHAOTIC ABANDONMENT OF HYSTERIA",
+ "raw": {
+ "normalized_reference": "mercedes screamed, cried, laughed, and manifested the chaotic abandonment of hysteria.",
+ "normalized_hypothesis": "mercedes screamed cried laughed and manifested the chaotic abandonment of hysteria",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "mercedes screamed cried laughed and manifested the chaotic abandonment of hysteria",
+ "normalized_hypothesis": "mercedes screamed cried laughed and manifested the chaotic abandonment of hysteria",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0261",
+ "category": "phonetic_balance",
+ "text": "Philip made no effort to follow.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0261.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "PHILIP MADE NO EFFORT TO FOLLOW",
+ "raw": {
+ "normalized_reference": "philip made no effort to follow.",
+ "normalized_hypothesis": "philip made no effort to follow",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "philip made no effort to follow",
+ "normalized_hypothesis": "philip made no effort to follow",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0262",
+ "category": "phonetic_balance",
+ "text": "She said with chattering teeth.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0262.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SHE SAID WITH CHATTERING TEETH",
+ "raw": {
+ "normalized_reference": "she said with chattering teeth.",
+ "normalized_hypothesis": "she said with chattering teeth",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "she said with chattering teeth",
+ "normalized_hypothesis": "she said with chattering teeth",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0263",
+ "category": "phonetic_balance",
+ "text": "Halfway around the track one donkey got into an argument with its rider.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0263.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HALF WAY AROUND THE TRACK ONE DONKEY GOT INTO AN ARGUMENT WITH ITS RIDER",
+ "raw": {
+ "normalized_reference": "halfway around the track one donkey got into an argument with its rider.",
+ "normalized_hypothesis": "half way around the track one donkey got into an argument with its rider",
+ "errors": 2,
+ "words": 13,
+ "wer": 0.15384615384615385
+ },
+ "semantic": {
+ "normalized_reference": "halfway around the track one donkey got into an argument with its rider",
+ "normalized_hypothesis": "half way around the track one donkey got into an argument with its rider",
+ "errors": 2,
+ "words": 13,
+ "wer": 0.15384615384615385
+ }
+ },
+ {
+ "id": "release500_0264",
+ "category": "phonetic_balance",
+ "text": "Earth and gravel seemed to fill the pan.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0264.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "EARTH AND GRAVEL SEEMED TO FILL THE PAN",
+ "raw": {
+ "normalized_reference": "earth and gravel seemed to fill the pan.",
+ "normalized_hypothesis": "earth and gravel seemed to fill the pan",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "earth and gravel seemed to fill the pan",
+ "normalized_hypothesis": "earth and gravel seemed to fill the pan",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0265",
+ "category": "phonetic_balance",
+ "text": "Her achievements with cocoanuts were a revelation.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0265.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HER ACHIEVEMENTS WITH COCOANUTS WERE A REVELATION",
+ "raw": {
+ "normalized_reference": "her achievements with cocoanuts were a revelation.",
+ "normalized_hypothesis": "her achievements with cocoanuts were a revelation",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "her achievements with cocoanuts were a revelation",
+ "normalized_hypothesis": "her achievements with cocoanuts were a revelation",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0266",
+ "category": "phonetic_balance",
+ "text": "For a full minute he crouched and listened.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0266.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "FOR A FULL MINUTE HE CROUCHED AND LISTENED",
+ "raw": {
+ "normalized_reference": "for a full minute he crouched and listened.",
+ "normalized_hypothesis": "for a full minute he crouched and listened",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "for a full minute he crouched and listened",
+ "normalized_hypothesis": "for a full minute he crouched and listened",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0267",
+ "category": "phonetic_balance",
+ "text": "It was edged with ice.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0267.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT WAS EDGED WITH ICE",
+ "raw": {
+ "normalized_reference": "it was edged with ice.",
+ "normalized_hypothesis": "it was edged with ice",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it was edged with ice",
+ "normalized_hypothesis": "it was edged with ice",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0268",
+ "category": "phonetic_balance",
+ "text": "The Portuguese boy passed the Hawaiian.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0268.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE PORTUGUESE BOY PASSED THE HAWAIAN",
+ "raw": {
+ "normalized_reference": "the portuguese boy passed the hawaiian.",
+ "normalized_hypothesis": "the portuguese boy passed the hawaian",
+ "errors": 1,
+ "words": 6,
+ "wer": 0.16666666666666666
+ },
+ "semantic": {
+ "normalized_reference": "the portuguese boy passed the hawaiian",
+ "normalized_hypothesis": "the portuguese boy passed the hawaian",
+ "errors": 1,
+ "words": 6,
+ "wer": 0.16666666666666666
+ }
+ },
+ {
+ "id": "release500_0269",
+ "category": "phonetic_balance",
+ "text": "But he did not broach it, preferring to mature it carefully.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0269.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "BUT HE DID NOT BROACH IT PREFERRING TO MATURE IT CAREFULLY",
+ "raw": {
+ "normalized_reference": "but he did not broach it, preferring to mature it carefully.",
+ "normalized_hypothesis": "but he did not broach it preferring to mature it carefully",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "but he did not broach it preferring to mature it carefully",
+ "normalized_hypothesis": "but he did not broach it preferring to mature it carefully",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0270",
+ "category": "phonetic_balance",
+ "text": "Come on, Del Mar challenged.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0270.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "COME ON DELMAR CHALLENGED",
+ "raw": {
+ "normalized_reference": "come on, del mar challenged.",
+ "normalized_hypothesis": "come on delmar challenged",
+ "errors": 2,
+ "words": 5,
+ "wer": 0.4
+ },
+ "semantic": {
+ "normalized_reference": "come on del mar challenged",
+ "normalized_hypothesis": "come on delmar challenged",
+ "errors": 2,
+ "words": 5,
+ "wer": 0.4
+ }
+ },
+ {
+ "id": "release500_0271",
+ "category": "phonetic_balance",
+ "text": "For a full half minute Jeanne looked at him without speaking.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0271.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "FOR A FULL HALF MINUTE JEAN LOOKED AT HIM WITHOUT SPEAKING",
+ "raw": {
+ "normalized_reference": "for a full half minute jeanne looked at him without speaking.",
+ "normalized_hypothesis": "for a full half minute jean looked at him without speaking",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ },
+ "semantic": {
+ "normalized_reference": "for a full half minute jeanne looked at him without speaking",
+ "normalized_hypothesis": "for a full half minute jean looked at him without speaking",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ }
+ },
+ {
+ "id": "release500_0272",
+ "category": "phonetic_balance",
+ "text": "He bore no grudges and had few enemies.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0272.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE BORE NO GRUDGES AND HAD FEW ENEMIES",
+ "raw": {
+ "normalized_reference": "he bore no grudges and had few enemies.",
+ "normalized_hypothesis": "he bore no grudges and had few enemies",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he bore no grudges and had few enemies",
+ "normalized_hypothesis": "he bore no grudges and had few enemies",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0273",
+ "category": "phonetic_balance",
+ "text": "I'll go over tomorrow afternoon.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0273.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I'LL GO OVER TO MORROW AFTERNOON",
+ "raw": {
+ "normalized_reference": "i'll go over tomorrow afternoon.",
+ "normalized_hypothesis": "i'll go over to morrow afternoon",
+ "errors": 2,
+ "words": 5,
+ "wer": 0.4
+ },
+ "semantic": {
+ "normalized_reference": "i will go over tomorrow afternoon",
+ "normalized_hypothesis": "i will go over to morrow afternoon",
+ "errors": 2,
+ "words": 6,
+ "wer": 0.3333333333333333
+ }
+ },
+ {
+ "id": "release500_0274",
+ "category": "phonetic_balance",
+ "text": "Well, there are better men in Hawaii, that's all.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0274.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WELL THERE ARE BETTER MEN IN HAWAI'I THAT'S ALL",
+ "raw": {
+ "normalized_reference": "well, there are better men in hawaii, that's all.",
+ "normalized_hypothesis": "well there are better men in hawai'i that's all",
+ "errors": 1,
+ "words": 9,
+ "wer": 0.1111111111111111
+ },
+ "semantic": {
+ "normalized_reference": "well there are better men in hawaii that is all",
+ "normalized_hypothesis": "well there are better men in hawai i that is all",
+ "errors": 2,
+ "words": 10,
+ "wer": 0.2
+ }
+ },
+ {
+ "id": "release500_0275",
+ "category": "phonetic_balance",
+ "text": "He was an athlete and a giant.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0275.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE WAS AN ATHLETE AND A GIANT",
+ "raw": {
+ "normalized_reference": "he was an athlete and a giant.",
+ "normalized_hypothesis": "he was an athlete and a giant",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he was an athlete and a giant",
+ "normalized_hypothesis": "he was an athlete and a giant",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0276",
+ "category": "phonetic_balance",
+ "text": "McVeigh when he returns from a trip to Honolulu.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0276.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "MAC VAY WHEN HE RETURNS FROM A TRIP TO HONOLULU",
+ "raw": {
+ "normalized_reference": "mcveigh when he returns from a trip to honolulu.",
+ "normalized_hypothesis": "mac vay when he returns from a trip to honolulu",
+ "errors": 2,
+ "words": 9,
+ "wer": 0.2222222222222222
+ },
+ "semantic": {
+ "normalized_reference": "mcveigh when he returns from a trip to honolulu",
+ "normalized_hypothesis": "mac vay when he returns from a trip to honolulu",
+ "errors": 2,
+ "words": 9,
+ "wer": 0.2222222222222222
+ }
+ },
+ {
+ "id": "release500_0277",
+ "category": "phonetic_balance",
+ "text": "You're a devil for fighting, and will surely win.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0277.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "YOU'RE A DEVIL FOR FIGHTING AND WILL SURELY WIN",
+ "raw": {
+ "normalized_reference": "you're a devil for fighting, and will surely win.",
+ "normalized_hypothesis": "you're a devil for fighting and will surely win",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "you are a devil for fighting and will surely win",
+ "normalized_hypothesis": "you are a devil for fighting and will surely win",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0278",
+ "category": "phonetic_balance",
+ "text": "All an appearance can know is mirage.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0278.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "ALL AN APPEARANCE CAN KNOW IS MIRAGE",
+ "raw": {
+ "normalized_reference": "all an appearance can know is mirage.",
+ "normalized_hypothesis": "all an appearance can know is mirage",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "all an appearance can know is mirage",
+ "normalized_hypothesis": "all an appearance can know is mirage",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0279",
+ "category": "phonetic_balance",
+ "text": "You fired me out of your house, in short.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0279.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "YOU FIRED ME OUT OF YOUR HOUSE IN SHORT",
+ "raw": {
+ "normalized_reference": "you fired me out of your house, in short.",
+ "normalized_hypothesis": "you fired me out of your house in short",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "you fired me out of your house in short",
+ "normalized_hypothesis": "you fired me out of your house in short",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0280",
+ "category": "phonetic_balance",
+ "text": "He made sure that the magazine was loaded, and resumed his paddling.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0280.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE MADE SURE THAT THE MAGAZINE WAS LOADED AND RESUMED HIS PADDLING",
+ "raw": {
+ "normalized_reference": "he made sure that the magazine was loaded, and resumed his paddling.",
+ "normalized_hypothesis": "he made sure that the magazine was loaded and resumed his paddling",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he made sure that the magazine was loaded and resumed his paddling",
+ "normalized_hypothesis": "he made sure that the magazine was loaded and resumed his paddling",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0281",
+ "category": "phonetic_balance",
+ "text": "Change chairs, Daylight commanded.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0281.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "CHANGE CHAIRS DAYLIGHT COMMANDED",
+ "raw": {
+ "normalized_reference": "change chairs, daylight commanded.",
+ "normalized_hypothesis": "change chairs daylight commanded",
+ "errors": 0,
+ "words": 4,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "change chairs daylight commanded",
+ "normalized_hypothesis": "change chairs daylight commanded",
+ "errors": 0,
+ "words": 4,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0282",
+ "category": "phonetic_balance",
+ "text": "By that answer my professional medical prestige stood or fell.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0282.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "BY THAT ANSWER MY PROFESSIONAL MEDICAL PRESTIGE STOOD OR FELL",
+ "raw": {
+ "normalized_reference": "by that answer my professional medical prestige stood or fell.",
+ "normalized_hypothesis": "by that answer my professional medical prestige stood or fell",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "by that answer my professional medical prestige stood or fell",
+ "normalized_hypothesis": "by that answer my professional medical prestige stood or fell",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0283",
+ "category": "phonetic_balance",
+ "text": "The moon had already begun its westward decline.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0283.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE MOON HAD ALREADY BEGUN ITS WESTWARD DECLINE",
+ "raw": {
+ "normalized_reference": "the moon had already begun its westward decline.",
+ "normalized_hypothesis": "the moon had already begun its westward decline",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the moon had already begun its westward decline",
+ "normalized_hypothesis": "the moon had already begun its westward decline",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0284",
+ "category": "phonetic_balance",
+ "text": "At sea, Monday, March 16, 1908.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0284.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "AT SEA MONDAY MARCH SIXTEENTH NINETY EIGHT",
+ "raw": {
+ "normalized_reference": "at sea, monday, march 16, 1908.",
+ "normalized_hypothesis": "at sea monday march sixteenth ninety eight",
+ "errors": 3,
+ "words": 6,
+ "wer": 0.5
+ },
+ "semantic": {
+ "normalized_reference": "at sea monday march 16 1908",
+ "normalized_hypothesis": "at sea monday march 16th 98",
+ "errors": 2,
+ "words": 6,
+ "wer": 0.3333333333333333
+ }
+ },
+ {
+ "id": "release500_0285",
+ "category": "phonetic_balance",
+ "text": "And wherever I ranged, the way lay along alcohol-drenched roads.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0285.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "AND WHEREVER I RANGED THE WAY LAY ALONG ALCOHOL DRENCHED ROADS",
+ "raw": {
+ "normalized_reference": "and wherever i ranged, the way lay along alcohol-drenched roads.",
+ "normalized_hypothesis": "and wherever i ranged the way lay along alcohol drenched roads",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "and wherever i ranged the way lay along alcohol drenched roads",
+ "normalized_hypothesis": "and wherever i ranged the way lay along alcohol drenched roads",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0286",
+ "category": "phonetic_balance",
+ "text": "The date was nearly eighteen years old.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0286.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE DATE WAS NEARLY EIGHTEEN YEARS OLD",
+ "raw": {
+ "normalized_reference": "the date was nearly eighteen years old.",
+ "normalized_hypothesis": "the date was nearly eighteen years old",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the date was nearly 18 years old",
+ "normalized_hypothesis": "the date was nearly 18 years old",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0287",
+ "category": "phonetic_balance",
+ "text": "Very early in my life, I separated from my mother.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0287.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "VERY EARLY IN MY LIFE I SEPARATED FROM MY MOTHER",
+ "raw": {
+ "normalized_reference": "very early in my life, i separated from my mother.",
+ "normalized_hypothesis": "very early in my life i separated from my mother",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "very early in my life i separated from my mother",
+ "normalized_hypothesis": "very early in my life i separated from my mother",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0288",
+ "category": "phonetic_balance",
+ "text": "There was pride and strength, the ring of triumph in his voice.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0288.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THERE WAS PRIDE AND STRENGTH THE RING OF TRIUMPH IN HIS VOICE",
+ "raw": {
+ "normalized_reference": "there was pride and strength, the ring of triumph in his voice.",
+ "normalized_hypothesis": "there was pride and strength the ring of triumph in his voice",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "there was pride and strength the ring of triumph in his voice",
+ "normalized_hypothesis": "there was pride and strength the ring of triumph in his voice",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0289",
+ "category": "phonetic_balance",
+ "text": "From my earliest recollection my sleep was a period of terror.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0289.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "FROM MY EARLIEST RECOLLECTION MY SLEEP WAS A PERIOD OF TERROR",
+ "raw": {
+ "normalized_reference": "from my earliest recollection my sleep was a period of terror.",
+ "normalized_hypothesis": "from my earliest recollection my sleep was a period of terror",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "from my earliest recollection my sleep was a period of terror",
+ "normalized_hypothesis": "from my earliest recollection my sleep was a period of terror",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0290",
+ "category": "phonetic_balance",
+ "text": "We'll have to watch our chances.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0290.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WE'LL HAVE TO WATCH OUR CHANCES",
+ "raw": {
+ "normalized_reference": "we'll have to watch our chances.",
+ "normalized_hypothesis": "we'll have to watch our chances",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "we will have to watch our chances",
+ "normalized_hypothesis": "we will have to watch our chances",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0291",
+ "category": "phonetic_balance",
+ "text": "Close beside him gleamed the white fangs of the wolf-dog.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0291.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "CLOSE BESIDE HIM GLEAMED THE WHITE FANGS OF THE WOLF DOG",
+ "raw": {
+ "normalized_reference": "close beside him gleamed the white fangs of the wolf-dog.",
+ "normalized_hypothesis": "close beside him gleamed the white fangs of the wolf dog",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "close beside him gleamed the white fangs of the wolf dog",
+ "normalized_hypothesis": "close beside him gleamed the white fangs of the wolf dog",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0292",
+ "category": "phonetic_balance",
+ "text": "Would you be satisfied with that one hundredth part of me.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0292.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WOULD YOU BE SATISFIED WITH THAT ONE HUNDREDTH PART OF ME",
+ "raw": {
+ "normalized_reference": "would you be satisfied with that one hundredth part of me.",
+ "normalized_hypothesis": "would you be satisfied with that one hundredth part of me",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "would you be satisfied with that 100th part of me",
+ "normalized_hypothesis": "would you be satisfied with that 100th part of me",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0293",
+ "category": "phonetic_balance",
+ "text": "Two gigantic owls were tearing at the carcass.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0293.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "TWO GIGANTIC OWLS WERE TEARING AT THE CARCASE",
+ "raw": {
+ "normalized_reference": "two gigantic owls were tearing at the carcass.",
+ "normalized_hypothesis": "two gigantic owls were tearing at the carcase",
+ "errors": 1,
+ "words": 8,
+ "wer": 0.125
+ },
+ "semantic": {
+ "normalized_reference": "2 gigantic owls were tearing at the carcass",
+ "normalized_hypothesis": "2 gigantic owls were tearing at the carcase",
+ "errors": 1,
+ "words": 8,
+ "wer": 0.125
+ }
+ },
+ {
+ "id": "release500_0294",
+ "category": "phonetic_balance",
+ "text": "The lush vegetation of that sheltered spot make a natural shield.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0294.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE LUSH VEGETATION OF THAT SHELTERED SPOT MAKE A NATURAL SHIELD",
+ "raw": {
+ "normalized_reference": "the lush vegetation of that sheltered spot make a natural shield.",
+ "normalized_hypothesis": "the lush vegetation of that sheltered spot make a natural shield",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the lush vegetation of that sheltered spot make a natural shield",
+ "normalized_hypothesis": "the lush vegetation of that sheltered spot make a natural shield",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0295",
+ "category": "phonetic_balance",
+ "text": "Below him the shadow was broken into a pool of rippling starlight.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0295.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "BELOW HIM THE SHADOW WAS BROKEN INTO A POOL OF RIPPLING STARLIGHT",
+ "raw": {
+ "normalized_reference": "below him the shadow was broken into a pool of rippling starlight.",
+ "normalized_hypothesis": "below him the shadow was broken into a pool of rippling starlight",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "below him the shadow was broken into a pool of rippling starlight",
+ "normalized_hypothesis": "below him the shadow was broken into a pool of rippling starlight",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0296",
+ "category": "phonetic_balance",
+ "text": "Lots of men take women buggy riding.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0296.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "LOTS OF MEN TAKE WOMEN BUGGY RIDING",
+ "raw": {
+ "normalized_reference": "lots of men take women buggy riding.",
+ "normalized_hypothesis": "lots of men take women buggy riding",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "lots of men take women buggy riding",
+ "normalized_hypothesis": "lots of men take women buggy riding",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0297",
+ "category": "phonetic_balance",
+ "text": "But it contributed to the smash.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0297.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "BUT IT CONTRIBUTED TO THE SMASH",
+ "raw": {
+ "normalized_reference": "but it contributed to the smash.",
+ "normalized_hypothesis": "but it contributed to the smash",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "but it contributed to the smash",
+ "normalized_hypothesis": "but it contributed to the smash",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0298",
+ "category": "phonetic_balance",
+ "text": "No man ate of the seal meat or the oil.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0298.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "NO MAN ATE OF THE SEAL MEAT OR THE OIL",
+ "raw": {
+ "normalized_reference": "no man ate of the seal meat or the oil.",
+ "normalized_hypothesis": "no man ate of the seal meat or the oil",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "no man ate of the seal meat or the oil",
+ "normalized_hypothesis": "no man ate of the seal meat or the oil",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0299",
+ "category": "phonetic_balance",
+ "text": "You're going in for grab sharing.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0299.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "YOU'RE GOING IN FOR GRAB SHARING",
+ "raw": {
+ "normalized_reference": "you're going in for grab sharing.",
+ "normalized_hypothesis": "you're going in for grab sharing",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "you are going in for grab sharing",
+ "normalized_hypothesis": "you are going in for grab sharing",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0300",
+ "category": "phonetic_balance",
+ "text": "He it was that lived to found the family of the Patino.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0300.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE IT WAS THAT LIVED TO FOUND THE FAMILY OF THE PETTINO",
+ "raw": {
+ "normalized_reference": "he it was that lived to found the family of the patino.",
+ "normalized_hypothesis": "he it was that lived to found the family of the pettino",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "he it was that lived to found the family of the patino",
+ "normalized_hypothesis": "he it was that lived to found the family of the pettino",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ }
+ },
+ {
+ "id": "release500_0301",
+ "category": "phonetic_balance",
+ "text": "For such countries nothing remained but reorganization.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0301.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "FOR SUCH COUNTRIES NOTHING REMAINED BUT REORGANIZATION",
+ "raw": {
+ "normalized_reference": "for such countries nothing remained but reorganization.",
+ "normalized_hypothesis": "for such countries nothing remained but reorganization",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "for such countries nothing remained but reorganization",
+ "normalized_hypothesis": "for such countries nothing remained but reorganization",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0302",
+ "category": "phonetic_balance",
+ "text": "Stand off butcher and baker and all the rest.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0302.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "STAND OFF BUTCHER AND BAKER AND ALL THE REST",
+ "raw": {
+ "normalized_reference": "stand off butcher and baker and all the rest.",
+ "normalized_hypothesis": "stand off butcher and baker and all the rest",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "stand off butcher and baker and all the rest",
+ "normalized_hypothesis": "stand off butcher and baker and all the rest",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0303",
+ "category": "phonetic_balance",
+ "text": "All eyes, however, were staring at him in certitude of expectancy.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0303.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "ALL EYES HOWEVER WERE STARING AT HIM IN CERTITUDE OF EXPECTANCY",
+ "raw": {
+ "normalized_reference": "all eyes, however, were staring at him in certitude of expectancy.",
+ "normalized_hypothesis": "all eyes however were staring at him in certitude of expectancy",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "all eyes however were staring at him in certitude of expectancy",
+ "normalized_hypothesis": "all eyes however were staring at him in certitude of expectancy",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0304",
+ "category": "phonetic_balance",
+ "text": "Two days ago Jeanne learned where her father's men were hiding.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0304.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "TWO DAYS AGO JEAN LEARNED WHERE HER FATHER'S MEN WERE HIDING",
+ "raw": {
+ "normalized_reference": "two days ago jeanne learned where her father's men were hiding.",
+ "normalized_hypothesis": "two days ago jean learned where her father's men were hiding",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ },
+ "semantic": {
+ "normalized_reference": "2 days ago jeanne learned where her father is men were hiding",
+ "normalized_hypothesis": "2 days ago jean learned where her father is men were hiding",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ }
+ },
+ {
+ "id": "release500_0305",
+ "category": "phonetic_balance",
+ "text": "His previous wives had never lived long enough to bear him children.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0305.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HIS PREVIOUS WIVES HAD NEVER LIVED LONG ENOUGH TO BEAR HIM CHILDREN",
+ "raw": {
+ "normalized_reference": "his previous wives had never lived long enough to bear him children.",
+ "normalized_hypothesis": "his previous wives had never lived long enough to bear him children",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "his previous wives had never lived long enough to bear him children",
+ "normalized_hypothesis": "his previous wives had never lived long enough to bear him children",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0306",
+ "category": "phonetic_balance",
+ "text": "But this little defect did not worry him.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0306.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "BUT THIS LITTLE DEFECT DID NOT WORRY HIM",
+ "raw": {
+ "normalized_reference": "but this little defect did not worry him.",
+ "normalized_hypothesis": "but this little defect did not worry him",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "but this little defect did not worry him",
+ "normalized_hypothesis": "but this little defect did not worry him",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0307",
+ "category": "phonetic_balance",
+ "text": "Eli Harding asked, as Shunk started to follow.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0307.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "ILLY HARDING ASKED AS SHUNK STARTED TO FOLLOW",
+ "raw": {
+ "normalized_reference": "eli harding asked, as shunk started to follow.",
+ "normalized_hypothesis": "illy harding asked as shunk started to follow",
+ "errors": 1,
+ "words": 8,
+ "wer": 0.125
+ },
+ "semantic": {
+ "normalized_reference": "eli harding asked as shunk started to follow",
+ "normalized_hypothesis": "illy harding asked as shunk started to follow",
+ "errors": 1,
+ "words": 8,
+ "wer": 0.125
+ }
+ },
+ {
+ "id": "release500_0308",
+ "category": "phonetic_balance",
+ "text": "And this was their sole conversation throughout the meal.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0308.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "AND THIS WAS THEIR SOLE CONVERSATION THROUGHOUT THE MEAL",
+ "raw": {
+ "normalized_reference": "and this was their sole conversation throughout the meal.",
+ "normalized_hypothesis": "and this was their sole conversation throughout the meal",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "and this was their sole conversation throughout the meal",
+ "normalized_hypothesis": "and this was their sole conversation throughout the meal",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0309",
+ "category": "phonetic_balance",
+ "text": "Then, again, Tudor had such an irritating way about him.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0309.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THEN AGAIN TUDOR HAD SUCH AN IRRITATING WAY ABOUT HIM",
+ "raw": {
+ "normalized_reference": "then, again, tudor had such an irritating way about him.",
+ "normalized_hypothesis": "then again tudor had such an irritating way about him",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "then again tudor had such an irritating way about him",
+ "normalized_hypothesis": "then again tudor had such an irritating way about him",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0310",
+ "category": "phonetic_balance",
+ "text": "In it there was something that was almost tragedy.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0310.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IN IT THERE WAS SOMETHING THAT WAS ALMOST TRAGEDY",
+ "raw": {
+ "normalized_reference": "in it there was something that was almost tragedy.",
+ "normalized_hypothesis": "in it there was something that was almost tragedy",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "in it there was something that was almost tragedy",
+ "normalized_hypothesis": "in it there was something that was almost tragedy",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0311",
+ "category": "phonetic_balance",
+ "text": "Thirty pounds, said the captain with finality.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0311.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THIRTY POUNDS SAID THE CAPTAIN WITH FINALITY",
+ "raw": {
+ "normalized_reference": "thirty pounds, said the captain with finality.",
+ "normalized_hypothesis": "thirty pounds said the captain with finality",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "\u00a330 said the captain with finality",
+ "normalized_hypothesis": "\u00a330 said the captain with finality",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0312",
+ "category": "phonetic_balance",
+ "text": "It was over when he made his way through the ring of spectators.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0312.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT WAS OVER WHEN HE MADE HIS WAY THROUGH THE RING OF SPECTATORS",
+ "raw": {
+ "normalized_reference": "it was over when he made his way through the ring of spectators.",
+ "normalized_hypothesis": "it was over when he made his way through the ring of spectators",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it was over when he made his way through the ring of spectators",
+ "normalized_hypothesis": "it was over when he made his way through the ring of spectators",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0313",
+ "category": "phonetic_balance",
+ "text": "She had died from cold and starvation.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0313.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SHE HAD DIED FROM COLD AND STARVATION",
+ "raw": {
+ "normalized_reference": "she had died from cold and starvation.",
+ "normalized_hypothesis": "she had died from cold and starvation",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "she had died from cold and starvation",
+ "normalized_hypothesis": "she had died from cold and starvation",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0314",
+ "category": "phonetic_balance",
+ "text": "We fished sharks on Niihau together.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0314.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WE FISHED SHARKS ON MEHOW TOGETHER",
+ "raw": {
+ "normalized_reference": "we fished sharks on niihau together.",
+ "normalized_hypothesis": "we fished sharks on mehow together",
+ "errors": 1,
+ "words": 6,
+ "wer": 0.16666666666666666
+ },
+ "semantic": {
+ "normalized_reference": "we fished sharks on niihau together",
+ "normalized_hypothesis": "we fished sharks on mehow together",
+ "errors": 1,
+ "words": 6,
+ "wer": 0.16666666666666666
+ }
+ },
+ {
+ "id": "release500_0315",
+ "category": "phonetic_balance",
+ "text": "His slim hands gripped the edges of the table.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0315.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HIS SLIM HANDS GRIPPED THE EDGES OF THE TABLE",
+ "raw": {
+ "normalized_reference": "his slim hands gripped the edges of the table.",
+ "normalized_hypothesis": "his slim hands gripped the edges of the table",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "his slim hands gripped the edges of the table",
+ "normalized_hypothesis": "his slim hands gripped the edges of the table",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0316",
+ "category": "phonetic_balance",
+ "text": "We had been chased by them ourselves, more than once.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0316.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WE HAD BEEN CHASED BY THEM OURSELVES MORE THAN ONCE",
+ "raw": {
+ "normalized_reference": "we had been chased by them ourselves, more than once.",
+ "normalized_hypothesis": "we had been chased by them ourselves more than once",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "we had been chased by them ourselves more than once",
+ "normalized_hypothesis": "we had been chased by them ourselves more than once",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0317",
+ "category": "phonetic_balance",
+ "text": "Or have they already devised one.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0317.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "OR HAVE THEY ALREADY DEVISED ONE",
+ "raw": {
+ "normalized_reference": "or have they already devised one.",
+ "normalized_hypothesis": "or have they already devised one",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "or have they already devised one",
+ "normalized_hypothesis": "or have they already devised one",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0318",
+ "category": "phonetic_balance",
+ "text": "I cannot follow you, she said.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0318.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I CANNOT FOLLOW YOU SHE SAID",
+ "raw": {
+ "normalized_reference": "i cannot follow you, she said.",
+ "normalized_hypothesis": "i cannot follow you she said",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i cannot follow you she said",
+ "normalized_hypothesis": "i cannot follow you she said",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0319",
+ "category": "phonetic_balance",
+ "text": "He went down in midstream, searching the shadows of both shores.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0319.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE WENT DOWN IN MIDSTREAM SEARCHING THE SHADOWS OF BOTH SHORES",
+ "raw": {
+ "normalized_reference": "he went down in midstream, searching the shadows of both shores.",
+ "normalized_hypothesis": "he went down in midstream searching the shadows of both shores",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he went down in midstream searching the shadows of both shores",
+ "normalized_hypothesis": "he went down in midstream searching the shadows of both shores",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0320",
+ "category": "phonetic_balance",
+ "text": "I was sick once -- typhoid.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0320.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I WAS SICK ONCE TYPHOID",
+ "raw": {
+ "normalized_reference": "i was sick once -- typhoid.",
+ "normalized_hypothesis": "i was sick once typhoid",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i was sick once typhoid",
+ "normalized_hypothesis": "i was sick once typhoid",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0321",
+ "category": "phonetic_balance",
+ "text": "Why not like any railroad station or ferry depot.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0321.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WHY NOT LIKE ANY RAILROAD STATION OR FERRY DEPOT",
+ "raw": {
+ "normalized_reference": "why not like any railroad station or ferry depot.",
+ "normalized_hypothesis": "why not like any railroad station or ferry depot",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "why not like any railroad station or ferry depot",
+ "normalized_hypothesis": "why not like any railroad station or ferry depot",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0322",
+ "category": "phonetic_balance",
+ "text": "Red-Eye swung back and forth on the branch farther down.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0322.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "RED EYE SWUNG BACK AND FORTH ON THE BRANCH FARTHER DOWN",
+ "raw": {
+ "normalized_reference": "red-eye swung back and forth on the branch farther down.",
+ "normalized_hypothesis": "red eye swung back and forth on the branch farther down",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "red eye swung back and forth on the branch farther down",
+ "normalized_hypothesis": "red eye swung back and forth on the branch farther down",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0323",
+ "category": "phonetic_balance",
+ "text": "Within himself he called it no longer his own.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0323.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WITHIN HIMSELF HE CALLED IT NO LONGER HIS OWN",
+ "raw": {
+ "normalized_reference": "within himself he called it no longer his own.",
+ "normalized_hypothesis": "within himself he called it no longer his own",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "within himself he called it no longer his own",
+ "normalized_hypothesis": "within himself he called it no longer his own",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0324",
+ "category": "phonetic_balance",
+ "text": "It lasted as a deterrent for two days.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0324.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT LASTED AS A DETERRANT FOR TWO DAYS",
+ "raw": {
+ "normalized_reference": "it lasted as a deterrent for two days.",
+ "normalized_hypothesis": "it lasted as a deterrant for two days",
+ "errors": 1,
+ "words": 8,
+ "wer": 0.125
+ },
+ "semantic": {
+ "normalized_reference": "it lasted as a deterrent for 2 days",
+ "normalized_hypothesis": "it lasted as a deterrant for 2 days",
+ "errors": 1,
+ "words": 8,
+ "wer": 0.125
+ }
+ },
+ {
+ "id": "release500_0325",
+ "category": "phonetic_balance",
+ "text": "He loved to play Chinese lottery.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0325.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE LOVED TO PLAY CHINESE LOTTERY",
+ "raw": {
+ "normalized_reference": "he loved to play chinese lottery.",
+ "normalized_hypothesis": "he loved to play chinese lottery",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he loved to play chinese lottery",
+ "normalized_hypothesis": "he loved to play chinese lottery",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0326",
+ "category": "phonetic_balance",
+ "text": "Lord, but I'm glad to see you again, Phil.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0326.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "LORD BUT I'M GLAD TO SEE YOU AGAIN PHIL",
+ "raw": {
+ "normalized_reference": "lord, but i'm glad to see you again, phil.",
+ "normalized_hypothesis": "lord but i'm glad to see you again phil",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "lord but i am glad to see you again phil",
+ "normalized_hypothesis": "lord but i am glad to see you again phil",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0327",
+ "category": "phonetic_balance",
+ "text": "There is not an iota of truth in it, certainly not.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0327.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THERE IS NOT AN IODA OF TRUTH IN IT CERTAINLY NOT",
+ "raw": {
+ "normalized_reference": "there is not an iota of truth in it, certainly not.",
+ "normalized_hypothesis": "there is not an ioda of truth in it certainly not",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ },
+ "semantic": {
+ "normalized_reference": "there is not an iota of truth in it certainly not",
+ "normalized_hypothesis": "there is not an ioda of truth in it certainly not",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ }
+ },
+ {
+ "id": "release500_0328",
+ "category": "phonetic_balance",
+ "text": "She added, with genuine sympathy in her face and voice.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0328.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SHE ADDED WITH GENUINE SYMPATHY IN HER FACE AND VOICE",
+ "raw": {
+ "normalized_reference": "she added, with genuine sympathy in her face and voice.",
+ "normalized_hypothesis": "she added with genuine sympathy in her face and voice",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "she added with genuine sympathy in her face and voice",
+ "normalized_hypothesis": "she added with genuine sympathy in her face and voice",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0329",
+ "category": "phonetic_balance",
+ "text": "We have plenty of capital ourselves, and yet we want more.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0329.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WE HAVE PLENTY OF CAPITAL OURSELVES AND YET WE WANT MORE",
+ "raw": {
+ "normalized_reference": "we have plenty of capital ourselves, and yet we want more.",
+ "normalized_hypothesis": "we have plenty of capital ourselves and yet we want more",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "we have plenty of capital ourselves and yet we want more",
+ "normalized_hypothesis": "we have plenty of capital ourselves and yet we want more",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0330",
+ "category": "phonetic_balance",
+ "text": "And now put yourself in my place for a moment.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0330.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "AND NOW PUT YOURSELF IN MY PLACE FOR A MOMENT",
+ "raw": {
+ "normalized_reference": "and now put yourself in my place for a moment.",
+ "normalized_hypothesis": "and now put yourself in my place for a moment",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "and now put yourself in my place for a moment",
+ "normalized_hypothesis": "and now put yourself in my place for a moment",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0331",
+ "category": "phonetic_balance",
+ "text": "Fast, but endure.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0331.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "FAST BUT ENDURE",
+ "raw": {
+ "normalized_reference": "fast, but endure.",
+ "normalized_hypothesis": "fast but endure",
+ "errors": 0,
+ "words": 3,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "fast but endure",
+ "normalized_hypothesis": "fast but endure",
+ "errors": 0,
+ "words": 3,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0332",
+ "category": "phonetic_balance",
+ "text": "You were looking squeamish this afternoon, he began.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0332.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "YOU WERE LOOKING SQUEAMISH THIS AFTERNOON HE BEGAN",
+ "raw": {
+ "normalized_reference": "you were looking squeamish this afternoon, he began.",
+ "normalized_hypothesis": "you were looking squeamish this afternoon he began",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "you were looking squeamish this afternoon he began",
+ "normalized_hypothesis": "you were looking squeamish this afternoon he began",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0333",
+ "category": "phonetic_balance",
+ "text": "It is the nearest refuge.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0333.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT IS THE NEAREST REFUGE",
+ "raw": {
+ "normalized_reference": "it is the nearest refuge.",
+ "normalized_hypothesis": "it is the nearest refuge",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it is the nearest refuge",
+ "normalized_hypothesis": "it is the nearest refuge",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0334",
+ "category": "phonetic_balance",
+ "text": "I want to die in it.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0334.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I WANT TO DIE IN IT",
+ "raw": {
+ "normalized_reference": "i want to die in it.",
+ "normalized_hypothesis": "i want to die in it",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i want to die in it",
+ "normalized_hypothesis": "i want to die in it",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0335",
+ "category": "phonetic_balance",
+ "text": "He stopped, and Philip nodded at the horrified question in his eyes.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0335.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE STOPPED AND PHILIP NODDED AT THE HORRIFIED QUESTION IN HIS EYES",
+ "raw": {
+ "normalized_reference": "he stopped, and philip nodded at the horrified question in his eyes.",
+ "normalized_hypothesis": "he stopped and philip nodded at the horrified question in his eyes",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he stopped and philip nodded at the horrified question in his eyes",
+ "normalized_hypothesis": "he stopped and philip nodded at the horrified question in his eyes",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0336",
+ "category": "phonetic_balance",
+ "text": "She turned, fearing that Jacques might see what was in her face.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0336.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SHE TURNED FEARING THAT JACK MIGHT SEE WHAT WAS IN HER FACE",
+ "raw": {
+ "normalized_reference": "she turned, fearing that jacques might see what was in her face.",
+ "normalized_hypothesis": "she turned fearing that jack might see what was in her face",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "she turned fearing that jacques might see what was in her face",
+ "normalized_hypothesis": "she turned fearing that jack might see what was in her face",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ }
+ },
+ {
+ "id": "release500_0337",
+ "category": "phonetic_balance",
+ "text": "It was a curious coincidence.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0337.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT WAS A CURIOUS COINCIDENCE",
+ "raw": {
+ "normalized_reference": "it was a curious coincidence.",
+ "normalized_hypothesis": "it was a curious coincidence",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it was a curious coincidence",
+ "normalized_hypothesis": "it was a curious coincidence",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0338",
+ "category": "phonetic_balance",
+ "text": "Daylight made no answer, and the door closed behind him.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0338.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "DAYLIGHT MADE NO ANSWER AND THE DOOR CLOSED BEHIND HIM",
+ "raw": {
+ "normalized_reference": "daylight made no answer, and the door closed behind him.",
+ "normalized_hypothesis": "daylight made no answer and the door closed behind him",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "daylight made no answer and the door closed behind him",
+ "normalized_hypothesis": "daylight made no answer and the door closed behind him",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0339",
+ "category": "phonetic_balance",
+ "text": "And he did hurt my arm.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0339.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "AND HE DID HURT MY ARM",
+ "raw": {
+ "normalized_reference": "and he did hurt my arm.",
+ "normalized_hypothesis": "and he did hurt my arm",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "and he did hurt my arm",
+ "normalized_hypothesis": "and he did hurt my arm",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0340",
+ "category": "phonetic_balance",
+ "text": "Dunham, can your boy go along with Jesse.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0340.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "DUNHAM CAN YOUR BOY GO ALONG WITH JESSIE",
+ "raw": {
+ "normalized_reference": "dunham, can your boy go along with jesse.",
+ "normalized_hypothesis": "dunham can your boy go along with jessie",
+ "errors": 1,
+ "words": 8,
+ "wer": 0.125
+ },
+ "semantic": {
+ "normalized_reference": "dunham can your boy go along with jesse",
+ "normalized_hypothesis": "dunham can your boy go along with jessie",
+ "errors": 1,
+ "words": 8,
+ "wer": 0.125
+ }
+ },
+ {
+ "id": "release500_0341",
+ "category": "phonetic_balance",
+ "text": "The creative joy, I murmured.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0341.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE CREATIVE JOY I MURMURED",
+ "raw": {
+ "normalized_reference": "the creative joy, i murmured.",
+ "normalized_hypothesis": "the creative joy i murmured",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the creative joy i murmured",
+ "normalized_hypothesis": "the creative joy i murmured",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0342",
+ "category": "phonetic_balance",
+ "text": "He pulled, and the log crashed down to break his back.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0342.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE PULLED AND THE LOG CRASHED DOWN TO BREAK HIS BACK",
+ "raw": {
+ "normalized_reference": "he pulled, and the log crashed down to break his back.",
+ "normalized_hypothesis": "he pulled and the log crashed down to break his back",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he pulled and the log crashed down to break his back",
+ "normalized_hypothesis": "he pulled and the log crashed down to break his back",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0343",
+ "category": "phonetic_balance",
+ "text": "By golly, the boy wins.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0343.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "BY GOLLY THE BOY WINS",
+ "raw": {
+ "normalized_reference": "by golly, the boy wins.",
+ "normalized_hypothesis": "by golly the boy wins",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "by golly the boy wins",
+ "normalized_hypothesis": "by golly the boy wins",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0344",
+ "category": "phonetic_balance",
+ "text": "The boy threw back his head with pride.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0344.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE BOY THREW BACK HIS HEAD WITH PRIDE",
+ "raw": {
+ "normalized_reference": "the boy threw back his head with pride.",
+ "normalized_hypothesis": "the boy threw back his head with pride",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the boy threw back his head with pride",
+ "normalized_hypothesis": "the boy threw back his head with pride",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0345",
+ "category": "phonetic_balance",
+ "text": "The last one I knew was an overseer.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0345.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE LAST ONE I KNEW WAS AN OVERSEER",
+ "raw": {
+ "normalized_reference": "the last one i knew was an overseer.",
+ "normalized_hypothesis": "the last one i knew was an overseer",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the last one i knew was an overseer",
+ "normalized_hypothesis": "the last one i knew was an overseer",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0346",
+ "category": "phonetic_balance",
+ "text": "Two years ago I gave up civilization for this.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0346.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "TWO YEARS AGO I GAVE UP CIVILIZATION FOR THIS",
+ "raw": {
+ "normalized_reference": "two years ago i gave up civilization for this.",
+ "normalized_hypothesis": "two years ago i gave up civilization for this",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "2 years ago i gave up civilization for this",
+ "normalized_hypothesis": "2 years ago i gave up civilization for this",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0347",
+ "category": "phonetic_balance",
+ "text": "The river bared its bosom, and snorting steamboats challenged the wilderness.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0347.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE RIVER BARED ITS BOSOM AND SNORTING STEAMBOATS CHALLENGED THE WILDERNESS",
+ "raw": {
+ "normalized_reference": "the river bared its bosom, and snorting steamboats challenged the wilderness.",
+ "normalized_hypothesis": "the river bared its bosom and snorting steamboats challenged the wilderness",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the river bared its bosom and snorting steamboats challenged the wilderness",
+ "normalized_hypothesis": "the river bared its bosom and snorting steamboats challenged the wilderness",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0348",
+ "category": "phonetic_balance",
+ "text": "Mab, she said.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0348.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "MAB SHE SAID",
+ "raw": {
+ "normalized_reference": "mab, she said.",
+ "normalized_hypothesis": "mab she said",
+ "errors": 0,
+ "words": 3,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "mab she said",
+ "normalized_hypothesis": "mab she said",
+ "errors": 0,
+ "words": 3,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0349",
+ "category": "phonetic_balance",
+ "text": "The regret in his voice was provocative of a second burst of laughter.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0349.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE REGRET IN HIS VOICE WAS PROVOCATIVE OF A SECOND BURST OF LAUGHTER",
+ "raw": {
+ "normalized_reference": "the regret in his voice was provocative of a second burst of laughter.",
+ "normalized_hypothesis": "the regret in his voice was provocative of a second burst of laughter",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the regret in his voice was provocative of a 2nd burst of laughter",
+ "normalized_hypothesis": "the regret in his voice was provocative of a 2nd burst of laughter",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0350",
+ "category": "phonetic_balance",
+ "text": "And each year something happened, and I did not go.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0350.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "AND EACH YEAR SOMETHING HAPPENED AND I DID NOT GO",
+ "raw": {
+ "normalized_reference": "and each year something happened, and i did not go.",
+ "normalized_hypothesis": "and each year something happened and i did not go",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "and each year something happened and i did not go",
+ "normalized_hypothesis": "and each year something happened and i did not go",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0351",
+ "category": "phonetic_balance",
+ "text": "Gad, your letter came just in time.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0351.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "GAD YOUR LETTER CAME JUST IN TIME",
+ "raw": {
+ "normalized_reference": "gad, your letter came just in time.",
+ "normalized_hypothesis": "gad your letter came just in time",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "gad your letter came just in time",
+ "normalized_hypothesis": "gad your letter came just in time",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0352",
+ "category": "phonetic_balance",
+ "text": "But if Pierre did not return, until tomorrow.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0352.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "BUT IF PIERRE DID NOT RETURN UNTIL TO MORROW",
+ "raw": {
+ "normalized_reference": "but if pierre did not return, until tomorrow.",
+ "normalized_hypothesis": "but if pierre did not return until to morrow",
+ "errors": 2,
+ "words": 8,
+ "wer": 0.25
+ },
+ "semantic": {
+ "normalized_reference": "but if pierre did not return until tomorrow",
+ "normalized_hypothesis": "but if pierre did not return until to morrow",
+ "errors": 2,
+ "words": 8,
+ "wer": 0.25
+ }
+ },
+ {
+ "id": "release500_0353",
+ "category": "phonetic_balance",
+ "text": "Billinger may arrive in time.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0353.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "BILLINGER MAY ARRIVE IN TIME",
+ "raw": {
+ "normalized_reference": "billinger may arrive in time.",
+ "normalized_hypothesis": "billinger may arrive in time",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "billinger may arrive in time",
+ "normalized_hypothesis": "billinger may arrive in time",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0354",
+ "category": "phonetic_balance",
+ "text": "In such a tumbling of values was no time to sell.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0354.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IN SUCH A TUMBLING OF VALUES WAS NO TIME TO SELL",
+ "raw": {
+ "normalized_reference": "in such a tumbling of values was no time to sell.",
+ "normalized_hypothesis": "in such a tumbling of values was no time to sell",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "in such a tumbling of values was no time to sell",
+ "normalized_hypothesis": "in such a tumbling of values was no time to sell",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0355",
+ "category": "phonetic_balance",
+ "text": "There's not an iota of truth in it.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0355.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THERE'S NOT AN IODA OF TRUTH IN IT",
+ "raw": {
+ "normalized_reference": "there's not an iota of truth in it.",
+ "normalized_hypothesis": "there's not an ioda of truth in it",
+ "errors": 1,
+ "words": 8,
+ "wer": 0.125
+ },
+ "semantic": {
+ "normalized_reference": "there is not an iota of truth in it",
+ "normalized_hypothesis": "there is not an ioda of truth in it",
+ "errors": 1,
+ "words": 9,
+ "wer": 0.1111111111111111
+ }
+ },
+ {
+ "id": "release500_0356",
+ "category": "phonetic_balance",
+ "text": "It is dog eat dog, and you ate them up.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0356.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT IS DOG EAT DOG AND YOU EAT THEM UP",
+ "raw": {
+ "normalized_reference": "it is dog eat dog, and you ate them up.",
+ "normalized_hypothesis": "it is dog eat dog and you eat them up",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ },
+ "semantic": {
+ "normalized_reference": "it is dog eat dog and you ate them up",
+ "normalized_hypothesis": "it is dog eat dog and you eat them up",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ }
+ },
+ {
+ "id": "release500_0357",
+ "category": "phonetic_balance",
+ "text": "You can take a vacation on pay.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0357.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "YOU CAN TAKE A VACATION ON PAY",
+ "raw": {
+ "normalized_reference": "you can take a vacation on pay.",
+ "normalized_hypothesis": "you can take a vacation on pay",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "you can take a vacation on pay",
+ "normalized_hypothesis": "you can take a vacation on pay",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0358",
+ "category": "phonetic_balance",
+ "text": "I have seen myself that one man contemplated by Pascal's philosophic eye.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0358.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I HAVE SEEN MYSELF THAT ONE MAN CONTEMPLATED BY PASCELL'S PHILOSOPHIC EYE",
+ "raw": {
+ "normalized_reference": "i have seen myself that one man contemplated by pascal's philosophic eye.",
+ "normalized_hypothesis": "i have seen myself that one man contemplated by pascell's philosophic eye",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "i have seen myself that one man contemplated by pascal is philosophic eye",
+ "normalized_hypothesis": "i have seen myself that one man contemplated by pascell is philosophic eye",
+ "errors": 1,
+ "words": 13,
+ "wer": 0.07692307692307693
+ }
+ },
+ {
+ "id": "release500_0359",
+ "category": "phonetic_balance",
+ "text": "Ah, I had forgotten, he exclaimed.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0359.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "AUH I HAD FORGOTTEN HE EXCLAIMED",
+ "raw": {
+ "normalized_reference": "ah, i had forgotten, he exclaimed.",
+ "normalized_hypothesis": "auh i had forgotten he exclaimed",
+ "errors": 1,
+ "words": 6,
+ "wer": 0.16666666666666666
+ },
+ "semantic": {
+ "normalized_reference": "ah i had forgotten he exclaimed",
+ "normalized_hypothesis": "auh i had forgotten he exclaimed",
+ "errors": 1,
+ "words": 6,
+ "wer": 0.16666666666666666
+ }
+ },
+ {
+ "id": "release500_0360",
+ "category": "phonetic_balance",
+ "text": "Providence had delivered him through the maelstrom.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0360.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "PROVIDENCE HAD DELIVERED HIM THROUGH THE MAILSTRAM",
+ "raw": {
+ "normalized_reference": "providence had delivered him through the maelstrom.",
+ "normalized_hypothesis": "providence had delivered him through the mailstram",
+ "errors": 1,
+ "words": 7,
+ "wer": 0.14285714285714285
+ },
+ "semantic": {
+ "normalized_reference": "providence had delivered him through the maelstrom",
+ "normalized_hypothesis": "providence had delivered him through the mailstram",
+ "errors": 1,
+ "words": 7,
+ "wer": 0.14285714285714285
+ }
+ },
+ {
+ "id": "release500_0361",
+ "category": "phonetic_balance",
+ "text": "She was his now, forever.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0361.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SHE WAS HIS NOW FOREVER",
+ "raw": {
+ "normalized_reference": "she was his now, forever.",
+ "normalized_hypothesis": "she was his now forever",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "she was his now forever",
+ "normalized_hypothesis": "she was his now forever",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0362",
+ "category": "phonetic_balance",
+ "text": "It was unobtrusive, yet it was there.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0362.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT WAS UNOBTRUSIVE YET IT WAS THERE",
+ "raw": {
+ "normalized_reference": "it was unobtrusive, yet it was there.",
+ "normalized_hypothesis": "it was unobtrusive yet it was there",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it was unobtrusive yet it was there",
+ "normalized_hypothesis": "it was unobtrusive yet it was there",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0363",
+ "category": "phonetic_balance",
+ "text": "Men like Joe Goose dated existence from drunk to drunk.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0363.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "MEN LIKE JOE GOOSE DATED EXISTENCE FROM DRUNK TO DRUNK",
+ "raw": {
+ "normalized_reference": "men like joe goose dated existence from drunk to drunk.",
+ "normalized_hypothesis": "men like joe goose dated existence from drunk to drunk",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "men like joe goose dated existence from drunk to drunk",
+ "normalized_hypothesis": "men like joe goose dated existence from drunk to drunk",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0364",
+ "category": "phonetic_balance",
+ "text": "I was completely lost in my work.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0364.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I WAS COMPLETELY LOST IN MY WORK",
+ "raw": {
+ "normalized_reference": "i was completely lost in my work.",
+ "normalized_hypothesis": "i was completely lost in my work",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i was completely lost in my work",
+ "normalized_hypothesis": "i was completely lost in my work",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0365",
+ "category": "phonetic_balance",
+ "text": "The very thing, Ernest agreed.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0365.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE VERY THING ERNEST AGREED",
+ "raw": {
+ "normalized_reference": "the very thing, ernest agreed.",
+ "normalized_hypothesis": "the very thing ernest agreed",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the very thing ernest agreed",
+ "normalized_hypothesis": "the very thing ernest agreed",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0366",
+ "category": "phonetic_balance",
+ "text": "He was the leader, and Tudor was his lieutenant.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0366.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE WAS THE LEADER AND TUDOR WAS HIS LIEUTENANT",
+ "raw": {
+ "normalized_reference": "he was the leader, and tudor was his lieutenant.",
+ "normalized_hypothesis": "he was the leader and tudor was his lieutenant",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he was the leader and tudor was his lieutenant",
+ "normalized_hypothesis": "he was the leader and tudor was his lieutenant",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0367",
+ "category": "phonetic_balance",
+ "text": "The Fire-Men wore animal skins around their waists and across their shoulders.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0367.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE FIREMEN WORE ANIMAL SKINS AROUND THEIR WAISTS AND ACROSS THEIR SHOULDERS",
+ "raw": {
+ "normalized_reference": "the fire-men wore animal skins around their waists and across their shoulders.",
+ "normalized_hypothesis": "the firemen wore animal skins around their waists and across their shoulders",
+ "errors": 2,
+ "words": 13,
+ "wer": 0.15384615384615385
+ },
+ "semantic": {
+ "normalized_reference": "the fire men wore animal skins around their waists and across their shoulders",
+ "normalized_hypothesis": "the firemen wore animal skins around their waists and across their shoulders",
+ "errors": 2,
+ "words": 13,
+ "wer": 0.15384615384615385
+ }
+ },
+ {
+ "id": "release500_0368",
+ "category": "phonetic_balance",
+ "text": "We could throw stones with our feet.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0368.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WE COULD THROW STONES WITH OUR FEET",
+ "raw": {
+ "normalized_reference": "we could throw stones with our feet.",
+ "normalized_hypothesis": "we could throw stones with our feet",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "we could throw stones with our feet",
+ "normalized_hypothesis": "we could throw stones with our feet",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0369",
+ "category": "phonetic_balance",
+ "text": "He turned sharply, and faced Gregson across the table.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0369.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE TURNED SHARPLY AND FACED GREGSON ACROSS THE TABLE",
+ "raw": {
+ "normalized_reference": "he turned sharply, and faced gregson across the table.",
+ "normalized_hypothesis": "he turned sharply and faced gregson across the table",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he turned sharply and faced gregson across the table",
+ "normalized_hypothesis": "he turned sharply and faced gregson across the table",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0370",
+ "category": "phonetic_balance",
+ "text": "They are to attack your camp tomorrow night.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0370.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THEY ARE TO ATTACK YOUR CAMP TO MORROW NIGHT",
+ "raw": {
+ "normalized_reference": "they are to attack your camp tomorrow night.",
+ "normalized_hypothesis": "they are to attack your camp to morrow night",
+ "errors": 2,
+ "words": 8,
+ "wer": 0.25
+ },
+ "semantic": {
+ "normalized_reference": "they are to attack your camp tomorrow night",
+ "normalized_hypothesis": "they are to attack your camp to morrow night",
+ "errors": 2,
+ "words": 8,
+ "wer": 0.25
+ }
+ },
+ {
+ "id": "release500_0371",
+ "category": "phonetic_balance",
+ "text": "And watch out for wet feet, was his parting advice.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0371.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "AND WATCH OUT FOR WET FEET WAS HIS PARTING ADVICE",
+ "raw": {
+ "normalized_reference": "and watch out for wet feet, was his parting advice.",
+ "normalized_hypothesis": "and watch out for wet feet was his parting advice",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "and watch out for wet feet was his parting advice",
+ "normalized_hypothesis": "and watch out for wet feet was his parting advice",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0372",
+ "category": "phonetic_balance",
+ "text": "Two weeks passed, and in that time Thorpe left camp three times.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0372.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "TWO WEEKS PASSED AND IN THAT TIME THORPE LEFT CAMP THREE TIMES",
+ "raw": {
+ "normalized_reference": "two weeks passed, and in that time thorpe left camp three times.",
+ "normalized_hypothesis": "two weeks passed and in that time thorpe left camp three times",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "2 weeks passed and in that time thorpe left camp 3 times",
+ "normalized_hypothesis": "2 weeks passed and in that time thorpe left camp 3 times",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0373",
+ "category": "phonetic_balance",
+ "text": "But ever his gaze returned to that Crouched Venus on the piano.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0373.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "BUT EVER HIS GAZE RETURNED TO THAT CROUCHED VENUS ON THE PIANO",
+ "raw": {
+ "normalized_reference": "but ever his gaze returned to that crouched venus on the piano.",
+ "normalized_hypothesis": "but ever his gaze returned to that crouched venus on the piano",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "but ever his gaze returned to that crouched venus on the piano",
+ "normalized_hypothesis": "but ever his gaze returned to that crouched venus on the piano",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0374",
+ "category": "phonetic_balance",
+ "text": "But she had become an automaton.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0374.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "BUT SHE HAD BECOME AN AUTOMATON",
+ "raw": {
+ "normalized_reference": "but she had become an automaton.",
+ "normalized_hypothesis": "but she had become an automaton",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "but she had become an automaton",
+ "normalized_hypothesis": "but she had become an automaton",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0375",
+ "category": "phonetic_balance",
+ "text": "I do not blame you for anything; remember that.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0375.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I DO NOT BLAME YOU FOR ANYTHING REMEMBER THAT",
+ "raw": {
+ "normalized_reference": "i do not blame you for anything; remember that.",
+ "normalized_hypothesis": "i do not blame you for anything remember that",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i do not blame you for anything remember that",
+ "normalized_hypothesis": "i do not blame you for anything remember that",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0376",
+ "category": "phonetic_balance",
+ "text": "The next thing to watch out for is bed sores.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0376.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE NEXT THING TO WATCH OUT FOR IS BED SORES",
+ "raw": {
+ "normalized_reference": "the next thing to watch out for is bed sores.",
+ "normalized_hypothesis": "the next thing to watch out for is bed sores",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the next thing to watch out for is bed sores",
+ "normalized_hypothesis": "the next thing to watch out for is bed sores",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0377",
+ "category": "phonetic_balance",
+ "text": "He will follow us soon.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0377.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE WILL FOLLOW US SOON",
+ "raw": {
+ "normalized_reference": "he will follow us soon.",
+ "normalized_hypothesis": "he will follow us soon",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he will follow us soon",
+ "normalized_hypothesis": "he will follow us soon",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0378",
+ "category": "phonetic_balance",
+ "text": "His beady black eyes saw bargains where other men saw bankruptcy.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0378.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HIS BEADY BLACK EYES SAW BARGAINS WHERE OTHER MEN SAW BANKRUPTCY",
+ "raw": {
+ "normalized_reference": "his beady black eyes saw bargains where other men saw bankruptcy.",
+ "normalized_hypothesis": "his beady black eyes saw bargains where other men saw bankruptcy",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "his beady black eyes saw bargains where other men saw bankruptcy",
+ "normalized_hypothesis": "his beady black eyes saw bargains where other men saw bankruptcy",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0379",
+ "category": "phonetic_balance",
+ "text": "They robbed me a few years later.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0379.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THEY ROBBED ME A FEW YEARS LATER",
+ "raw": {
+ "normalized_reference": "they robbed me a few years later.",
+ "normalized_hypothesis": "they robbed me a few years later",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "they robbed me a few years later",
+ "normalized_hypothesis": "they robbed me a few years later",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0380",
+ "category": "phonetic_balance",
+ "text": "Gregson held a lighted match until it burnt his fingertips.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0380.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "GREGSEN HELD A LIGHTED MATCH UNTIL IT BURNT HIS FINGER TIPS",
+ "raw": {
+ "normalized_reference": "gregson held a lighted match until it burnt his fingertips.",
+ "normalized_hypothesis": "gregsen held a lighted match until it burnt his finger tips",
+ "errors": 3,
+ "words": 10,
+ "wer": 0.3
+ },
+ "semantic": {
+ "normalized_reference": "gregson held a lighted match until it burnt his fingertips",
+ "normalized_hypothesis": "gregsen held a lighted match until it burnt his finger tips",
+ "errors": 3,
+ "words": 10,
+ "wer": 0.3
+ }
+ },
+ {
+ "id": "release500_0381",
+ "category": "phonetic_balance",
+ "text": "A burst of laughter was his reward.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0381.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "A BURST OF LAUGHTER WAS HIS REWARD",
+ "raw": {
+ "normalized_reference": "a burst of laughter was his reward.",
+ "normalized_hypothesis": "a burst of laughter was his reward",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "a burst of laughter was his reward",
+ "normalized_hypothesis": "a burst of laughter was his reward",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0382",
+ "category": "phonetic_balance",
+ "text": "He cried, and swung the club wildly.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0382.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE CRIED AND SWUNG THE CLUB WILDLY",
+ "raw": {
+ "normalized_reference": "he cried, and swung the club wildly.",
+ "normalized_hypothesis": "he cried and swung the club wildly",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he cried and swung the club wildly",
+ "normalized_hypothesis": "he cried and swung the club wildly",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0383",
+ "category": "phonetic_balance",
+ "text": "Sandel would never become a world champion.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0383.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SANDALL WOULD NEVER BECOME A WORLD CHAMPION",
+ "raw": {
+ "normalized_reference": "sandel would never become a world champion.",
+ "normalized_hypothesis": "sandall would never become a world champion",
+ "errors": 1,
+ "words": 7,
+ "wer": 0.14285714285714285
+ },
+ "semantic": {
+ "normalized_reference": "sandel would never become a world champion",
+ "normalized_hypothesis": "sandall would never become a world champion",
+ "errors": 1,
+ "words": 7,
+ "wer": 0.14285714285714285
+ }
+ },
+ {
+ "id": "release500_0384",
+ "category": "phonetic_balance",
+ "text": "And after the bath a shave would not be bad.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0384.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "AND AFTER THE BATH A SHAVE WOULD NOT BE BAD",
+ "raw": {
+ "normalized_reference": "and after the bath a shave would not be bad.",
+ "normalized_hypothesis": "and after the bath a shave would not be bad",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "and after the bath a shave would not be bad",
+ "normalized_hypothesis": "and after the bath a shave would not be bad",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0385",
+ "category": "phonetic_balance",
+ "text": "They saw each other for the first time in Boston.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0385.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THEY SAW EACH OTHER FOR THE FIRST TIME IN BOSTON",
+ "raw": {
+ "normalized_reference": "they saw each other for the first time in boston.",
+ "normalized_hypothesis": "they saw each other for the first time in boston",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "they saw each other for the 1st time in boston",
+ "normalized_hypothesis": "they saw each other for the 1st time in boston",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0386",
+ "category": "phonetic_balance",
+ "text": "I had been born with no organic, chemical predisposition toward alcohol.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0386.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I HAD BEEN BORN WITH NO ORGANIC CHEMICAL PREDISPOSITION TOWARD ALCOHOL",
+ "raw": {
+ "normalized_reference": "i had been born with no organic, chemical predisposition toward alcohol.",
+ "normalized_hypothesis": "i had been born with no organic chemical predisposition toward alcohol",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i had been born with no organic chemical predisposition toward alcohol",
+ "normalized_hypothesis": "i had been born with no organic chemical predisposition toward alcohol",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0387",
+ "category": "phonetic_balance",
+ "text": "They had no fixed values, to be altered by adjectives and adverbs.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0387.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THEY HAD NO FIXED VALUES TO BE ALTERED BY ADJECTIVES AND ADVERBS",
+ "raw": {
+ "normalized_reference": "they had no fixed values, to be altered by adjectives and adverbs.",
+ "normalized_hypothesis": "they had no fixed values to be altered by adjectives and adverbs",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "they had no fixed values to be altered by adjectives and adverbs",
+ "normalized_hypothesis": "they had no fixed values to be altered by adjectives and adverbs",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0388",
+ "category": "phonetic_balance",
+ "text": "Nor was Elam Harnish an exception.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0388.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "NOR WAS A LAMB HARNISH AN EXCEPTION",
+ "raw": {
+ "normalized_reference": "nor was elam harnish an exception.",
+ "normalized_hypothesis": "nor was a lamb harnish an exception",
+ "errors": 2,
+ "words": 6,
+ "wer": 0.3333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "nor was elam harnish an exception",
+ "normalized_hypothesis": "nor was a lamb harnish an exception",
+ "errors": 2,
+ "words": 6,
+ "wer": 0.3333333333333333
+ }
+ },
+ {
+ "id": "release500_0389",
+ "category": "phonetic_balance",
+ "text": "He was a wise hyena.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0389.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE WAS A WISE HYENA",
+ "raw": {
+ "normalized_reference": "he was a wise hyena.",
+ "normalized_hypothesis": "he was a wise hyena",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he was a wise hyena",
+ "normalized_hypothesis": "he was a wise hyena",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0390",
+ "category": "phonetic_balance",
+ "text": "And now behold the perversity of things.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0390.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "AND NOW BEHOLD THE PERVERSITY OF THINGS",
+ "raw": {
+ "normalized_reference": "and now behold the perversity of things.",
+ "normalized_hypothesis": "and now behold the perversity of things",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "and now behold the perversity of things",
+ "normalized_hypothesis": "and now behold the perversity of things",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0391",
+ "category": "phonetic_balance",
+ "text": "The Fire People, like ourselves, lived in caves.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0391.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE FIRE PEOPLE LIKE OURSELVES LIVED IN CAVES",
+ "raw": {
+ "normalized_reference": "the fire people, like ourselves, lived in caves.",
+ "normalized_hypothesis": "the fire people like ourselves lived in caves",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the fire people like ourselves lived in caves",
+ "normalized_hypothesis": "the fire people like ourselves lived in caves",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0392",
+ "category": "phonetic_balance",
+ "text": "To her the bridge was tambo, which is the native for taboo.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0392.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "TO HER THE BRIDGE WAS TAMBO WHICH IS THE NATIVE FOR TABOO",
+ "raw": {
+ "normalized_reference": "to her the bridge was tambo, which is the native for taboo.",
+ "normalized_hypothesis": "to her the bridge was tambo which is the native for taboo",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "to her the bridge was tambo which is the native for taboo",
+ "normalized_hypothesis": "to her the bridge was tambo which is the native for taboo",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0393",
+ "category": "phonetic_balance",
+ "text": "The voyage was our idea of a good time.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0393.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE VOYAGE WAS OUR IDEA OF A GOOD TIME",
+ "raw": {
+ "normalized_reference": "the voyage was our idea of a good time.",
+ "normalized_hypothesis": "the voyage was our idea of a good time",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the voyage was our idea of a good time",
+ "normalized_hypothesis": "the voyage was our idea of a good time",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0394",
+ "category": "phonetic_balance",
+ "text": "Of course much grumbling went on, and little outbursts were continually occurring.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0394.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "OF COURSE MUCH GRUMBLING WENT ON AND LITTLE OUTBURSTS WERE CONTINUALLY OCCURRING",
+ "raw": {
+ "normalized_reference": "of course much grumbling went on, and little outbursts were continually occurring.",
+ "normalized_hypothesis": "of course much grumbling went on and little outbursts were continually occurring",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "of course much grumbling went on and little outbursts were continually occurring",
+ "normalized_hypothesis": "of course much grumbling went on and little outbursts were continually occurring",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0395",
+ "category": "phonetic_balance",
+ "text": "Get down and dig in.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0395.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "GET DOWN AND DIG IN",
+ "raw": {
+ "normalized_reference": "get down and dig in.",
+ "normalized_hypothesis": "get down and dig in",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "get down and dig in",
+ "normalized_hypothesis": "get down and dig in",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0396",
+ "category": "phonetic_balance",
+ "text": "It was put together in a casual, helter-skelter sort of way.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0396.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT WAS PUT TOGETHER IN A CASUAL HELTER SKELTER SORT OF WAY",
+ "raw": {
+ "normalized_reference": "it was put together in a casual, helter-skelter sort of way.",
+ "normalized_hypothesis": "it was put together in a casual helter skelter sort of way",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it was put together in a casual helter skelter sort of way",
+ "normalized_hypothesis": "it was put together in a casual helter skelter sort of way",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0397",
+ "category": "phonetic_balance",
+ "text": "He deluged me, overwhelmed me with argument.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0397.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE DELUGED ME OVERWHELMED ME WITH ARGUMENT",
+ "raw": {
+ "normalized_reference": "he deluged me, overwhelmed me with argument.",
+ "normalized_hypothesis": "he deluged me overwhelmed me with argument",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he deluged me overwhelmed me with argument",
+ "normalized_hypothesis": "he deluged me overwhelmed me with argument",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0398",
+ "category": "phonetic_balance",
+ "text": "The girl faced him, her eyes shining with sudden fear.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0398.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE GIRL FACED HIM HER EYES SHINING WITH SUDDEN FEAR",
+ "raw": {
+ "normalized_reference": "the girl faced him, her eyes shining with sudden fear.",
+ "normalized_hypothesis": "the girl faced him her eyes shining with sudden fear",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the girl faced him her eyes shining with sudden fear",
+ "normalized_hypothesis": "the girl faced him her eyes shining with sudden fear",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0399",
+ "category": "phonetic_balance",
+ "text": "It will break our hearts and our backs to hoist anchor by hand.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0399.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT WILL BREAK OUR HEARTS AND OUR BACKS TO HOIST ANCHOR BY HAND",
+ "raw": {
+ "normalized_reference": "it will break our hearts and our backs to hoist anchor by hand.",
+ "normalized_hypothesis": "it will break our hearts and our backs to hoist anchor by hand",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it will break our hearts and our backs to hoist anchor by hand",
+ "normalized_hypothesis": "it will break our hearts and our backs to hoist anchor by hand",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0400",
+ "category": "phonetic_balance",
+ "text": "He made no reply as he waited for Whittemore to continue.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0400.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE MADE NO REPLY AS HE WAITED FOR WHITTAMORE TO CONTINUE",
+ "raw": {
+ "normalized_reference": "he made no reply as he waited for whittemore to continue.",
+ "normalized_hypothesis": "he made no reply as he waited for whittamore to continue",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ },
+ "semantic": {
+ "normalized_reference": "he made no reply as he waited for whittemore to continue",
+ "normalized_hypothesis": "he made no reply as he waited for whittamore to continue",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ }
+ },
+ {
+ "id": "release500_0401",
+ "category": "phonetic_balance",
+ "text": "Who the devil gave it to you to be judge and jury.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0401.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WHO THE DEVIL GAVE IT TO YOU TO BE JUDGE AND JURY",
+ "raw": {
+ "normalized_reference": "who the devil gave it to you to be judge and jury.",
+ "normalized_hypothesis": "who the devil gave it to you to be judge and jury",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "who the devil gave it to you to be judge and jury",
+ "normalized_hypothesis": "who the devil gave it to you to be judge and jury",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0402",
+ "category": "phonetic_balance",
+ "text": "Daylight found himself charmed and made curious by the little man.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0402.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "DAYLIGHT FOUND HIMSELF CHARMED AND MADE CURIOUS BY THE LITTLE MAN",
+ "raw": {
+ "normalized_reference": "daylight found himself charmed and made curious by the little man.",
+ "normalized_hypothesis": "daylight found himself charmed and made curious by the little man",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "daylight found himself charmed and made curious by the little man",
+ "normalized_hypothesis": "daylight found himself charmed and made curious by the little man",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0403",
+ "category": "phonetic_balance",
+ "text": "It's a Yankee, Joan cried.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0403.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT'S A YANKEE JOAN CRIED",
+ "raw": {
+ "normalized_reference": "it's a yankee, joan cried.",
+ "normalized_hypothesis": "it's a yankee joan cried",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it is a yankee joan cried",
+ "normalized_hypothesis": "it is a yankee joan cried",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0404",
+ "category": "phonetic_balance",
+ "text": "But what they want with your toothbrush is more than I can imagine.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0404.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "BUT WHAT THEY WANT WITH YOUR TOOTH BRUSH IS MORE THAN I CAN IMAGINE",
+ "raw": {
+ "normalized_reference": "but what they want with your toothbrush is more than i can imagine.",
+ "normalized_hypothesis": "but what they want with your tooth brush is more than i can imagine",
+ "errors": 2,
+ "words": 13,
+ "wer": 0.15384615384615385
+ },
+ "semantic": {
+ "normalized_reference": "but what they want with your toothbrush is more than i can imagine",
+ "normalized_hypothesis": "but what they want with your tooth brush is more than i can imagine",
+ "errors": 2,
+ "words": 13,
+ "wer": 0.15384615384615385
+ }
+ },
+ {
+ "id": "release500_0405",
+ "category": "phonetic_balance",
+ "text": "They were babbling and chattering all together.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0405.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THEY WERE BABBLING AND CHATTERING ALTOGETHER",
+ "raw": {
+ "normalized_reference": "they were babbling and chattering all together.",
+ "normalized_hypothesis": "they were babbling and chattering altogether",
+ "errors": 2,
+ "words": 7,
+ "wer": 0.2857142857142857
+ },
+ "semantic": {
+ "normalized_reference": "they were babbling and chattering all together",
+ "normalized_hypothesis": "they were babbling and chattering altogether",
+ "errors": 2,
+ "words": 7,
+ "wer": 0.2857142857142857
+ }
+ },
+ {
+ "id": "release500_0406",
+ "category": "phonetic_balance",
+ "text": "Men of Selden's stamp don't stop at women and children.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0406.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "MEN OF SELDON'S STAMP DON'T STOP AT WOMEN AND CHILDREN",
+ "raw": {
+ "normalized_reference": "men of selden's stamp don't stop at women and children.",
+ "normalized_hypothesis": "men of seldon's stamp don't stop at women and children",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ },
+ "semantic": {
+ "normalized_reference": "men of selden is stamp do not stop at women and children",
+ "normalized_hypothesis": "men of seldon is stamp do not stop at women and children",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ }
+ },
+ {
+ "id": "release500_0407",
+ "category": "phonetic_balance",
+ "text": "Cameron looked at his hands with their long, sinewy fingers.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0407.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "CAMERON LOOKED AT HIS HANDS WITH THEIR LONG SINEWY FINGERS",
+ "raw": {
+ "normalized_reference": "cameron looked at his hands with their long, sinewy fingers.",
+ "normalized_hypothesis": "cameron looked at his hands with their long sinewy fingers",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "cameron looked at his hands with their long sinewy fingers",
+ "normalized_hypothesis": "cameron looked at his hands with their long sinewy fingers",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0408",
+ "category": "phonetic_balance",
+ "text": "Saxon waited, for she knew a fresh idea had struck Billy.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0408.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SAXON WAITED FOR SHE KNEW A FRESH IDEA HAD STRUCK BILLY",
+ "raw": {
+ "normalized_reference": "saxon waited, for she knew a fresh idea had struck billy.",
+ "normalized_hypothesis": "saxon waited for she knew a fresh idea had struck billy",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "saxon waited for she knew a fresh idea had struck billy",
+ "normalized_hypothesis": "saxon waited for she knew a fresh idea had struck billy",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0409",
+ "category": "phonetic_balance",
+ "text": "He saw only the effect in a general, sketchy way.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0409.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE SAW ONLY THE EFFECT IN A GENERAL SKETCHY WAY",
+ "raw": {
+ "normalized_reference": "he saw only the effect in a general, sketchy way.",
+ "normalized_hypothesis": "he saw only the effect in a general sketchy way",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he saw only the effect in a general sketchy way",
+ "normalized_hypothesis": "he saw only the effect in a general sketchy way",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0410",
+ "category": "phonetic_balance",
+ "text": "He began to follow the footprints of the dog.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0410.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE BEGAN TO FOLLOW THE FOOTPRINTS OF THE DOG",
+ "raw": {
+ "normalized_reference": "he began to follow the footprints of the dog.",
+ "normalized_hypothesis": "he began to follow the footprints of the dog",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he began to follow the footprints of the dog",
+ "normalized_hypothesis": "he began to follow the footprints of the dog",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0411",
+ "category": "phonetic_balance",
+ "text": "Do you value your hide.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0411.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "DO YOU VALUE YOUR HIDE",
+ "raw": {
+ "normalized_reference": "do you value your hide.",
+ "normalized_hypothesis": "do you value your hide",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "do you value your hide",
+ "normalized_hypothesis": "do you value your hide",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0412",
+ "category": "phonetic_balance",
+ "text": "A rising tide of fat had submerged them.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0412.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "A RISING TIDE OF FAT HAD SUBMERGED THEM",
+ "raw": {
+ "normalized_reference": "a rising tide of fat had submerged them.",
+ "normalized_hypothesis": "a rising tide of fat had submerged them",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "a rising tide of fat had submerged them",
+ "normalized_hypothesis": "a rising tide of fat had submerged them",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0413",
+ "category": "phonetic_balance",
+ "text": "There was nothing on the rock.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0413.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THERE WAS NOTHING ON THE ROCK",
+ "raw": {
+ "normalized_reference": "there was nothing on the rock.",
+ "normalized_hypothesis": "there was nothing on the rock",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "there was nothing on the rock",
+ "normalized_hypothesis": "there was nothing on the rock",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0414",
+ "category": "phonetic_balance",
+ "text": "Whatever he guessed he locked away in the taboo room of Naomi.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0414.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WHATEVER HE GUESSED HE LOCKED AWAY IN THE TABOO ROOM OF NAOMI",
+ "raw": {
+ "normalized_reference": "whatever he guessed he locked away in the taboo room of naomi.",
+ "normalized_hypothesis": "whatever he guessed he locked away in the taboo room of naomi",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "whatever he guessed he locked away in the taboo room of naomi",
+ "normalized_hypothesis": "whatever he guessed he locked away in the taboo room of naomi",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0415",
+ "category": "phonetic_balance",
+ "text": "Will we ever forget it.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0415.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WILL WE EVER FORGET IT",
+ "raw": {
+ "normalized_reference": "will we ever forget it.",
+ "normalized_hypothesis": "will we ever forget it",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "will we ever forget it",
+ "normalized_hypothesis": "will we ever forget it",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0416",
+ "category": "phonetic_balance",
+ "text": "Jeanne and Pierre both gazed toward the great rock.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0416.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "JEAN AND PIERRE BOTH GAZED TOWARD THE GREAT ROCK",
+ "raw": {
+ "normalized_reference": "jeanne and pierre both gazed toward the great rock.",
+ "normalized_hypothesis": "jean and pierre both gazed toward the great rock",
+ "errors": 1,
+ "words": 9,
+ "wer": 0.1111111111111111
+ },
+ "semantic": {
+ "normalized_reference": "jeanne and pierre both gazed toward the great rock",
+ "normalized_hypothesis": "jean and pierre both gazed toward the great rock",
+ "errors": 1,
+ "words": 9,
+ "wer": 0.1111111111111111
+ }
+ },
+ {
+ "id": "release500_0417",
+ "category": "phonetic_balance",
+ "text": "Does the old boy often go off at half-cock that way.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0417.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "DOES THE OLD BOY OFTEN GO OFF AT HALF COCK THAT WAY",
+ "raw": {
+ "normalized_reference": "does the old boy often go off at half-cock that way.",
+ "normalized_hypothesis": "does the old boy often go off at half cock that way",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "does the old boy often go off at half cock that way",
+ "normalized_hypothesis": "does the old boy often go off at half cock that way",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0418",
+ "category": "phonetic_balance",
+ "text": "The temperature dropped to fifty below zero and remained there the whole trip.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0418.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE TEMPERATURE DROPPED TO FIFTY BELOW ZERO AND REMAINED THERE THE WHOLE TRIP",
+ "raw": {
+ "normalized_reference": "the temperature dropped to fifty below zero and remained there the whole trip.",
+ "normalized_hypothesis": "the temperature dropped to fifty below zero and remained there the whole trip",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the temperature dropped to 50 below 0 and remained there the whole trip",
+ "normalized_hypothesis": "the temperature dropped to 50 below 0 and remained there the whole trip",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0419",
+ "category": "phonetic_balance",
+ "text": "It's worth eight dollars.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0419.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT'S WORTH EIGHT DOLLARS",
+ "raw": {
+ "normalized_reference": "it's worth eight dollars.",
+ "normalized_hypothesis": "it's worth eight dollars",
+ "errors": 0,
+ "words": 4,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it is worth $8",
+ "normalized_hypothesis": "it is worth $8",
+ "errors": 0,
+ "words": 4,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0420",
+ "category": "phonetic_balance",
+ "text": "But I did not enjoy it long.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0420.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "BUT I DID NOT ENJOY IT LONG",
+ "raw": {
+ "normalized_reference": "but i did not enjoy it long.",
+ "normalized_hypothesis": "but i did not enjoy it long",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "but i did not enjoy it long",
+ "normalized_hypothesis": "but i did not enjoy it long",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0421",
+ "category": "phonetic_balance",
+ "text": "Shorty turned to their employers.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0421.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SHORTY TURNED TO THEIR EMPLOYERS",
+ "raw": {
+ "normalized_reference": "shorty turned to their employers.",
+ "normalized_hypothesis": "shorty turned to their employers",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "shorty turned to their employers",
+ "normalized_hypothesis": "shorty turned to their employers",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0422",
+ "category": "phonetic_balance",
+ "text": "It was the third or fourth time that Philip had heard MacDougall swear.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0422.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT WAS THE THIRD OR FOURTH TIME THAT PHILIP HAD HEARD MACDUGAL SWEAR",
+ "raw": {
+ "normalized_reference": "it was the third or fourth time that philip had heard macdougall swear.",
+ "normalized_hypothesis": "it was the third or fourth time that philip had heard macdugal swear",
+ "errors": 1,
+ "words": 13,
+ "wer": 0.07692307692307693
+ },
+ "semantic": {
+ "normalized_reference": "it was the 3rd or 4th time that philip had heard macdougall swear",
+ "normalized_hypothesis": "it was the 3rd or 4th time that philip had heard macdugal swear",
+ "errors": 1,
+ "words": 13,
+ "wer": 0.07692307692307693
+ }
+ },
+ {
+ "id": "release500_0423",
+ "category": "phonetic_balance",
+ "text": "Ernest saw in the affair the most sinister import.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0423.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "ERNEST SAW IN THE AFFAIR THE MOST SINISTER IMPORT",
+ "raw": {
+ "normalized_reference": "ernest saw in the affair the most sinister import.",
+ "normalized_hypothesis": "ernest saw in the affair the most sinister import",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "ernest saw in the affair the most sinister import",
+ "normalized_hypothesis": "ernest saw in the affair the most sinister import",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0424",
+ "category": "phonetic_balance",
+ "text": "Also, at regular intervals, he would mutter.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0424.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "ALSO AT REGULAR INTERVALS HE WOULD MUTTER",
+ "raw": {
+ "normalized_reference": "also, at regular intervals, he would mutter.",
+ "normalized_hypothesis": "also at regular intervals he would mutter",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "also at regular intervals he would mutter",
+ "normalized_hypothesis": "also at regular intervals he would mutter",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0425",
+ "category": "phonetic_balance",
+ "text": "He may anticipate the day of his death.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0425.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE MAY ANTICIPATE THE DAY OF HIS DEATH",
+ "raw": {
+ "normalized_reference": "he may anticipate the day of his death.",
+ "normalized_hypothesis": "he may anticipate the day of his death",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he may anticipate the day of his death",
+ "normalized_hypothesis": "he may anticipate the day of his death",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0426",
+ "category": "phonetic_balance",
+ "text": "I defy any man to get a Solomon Island sore in California.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0426.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I DEFY ANY MAN TO GET A SOLOMON ISLAND SORE IN CALIFORNIA",
+ "raw": {
+ "normalized_reference": "i defy any man to get a solomon island sore in california.",
+ "normalized_hypothesis": "i defy any man to get a solomon island sore in california",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i defy any man to get a solomon island sore in california",
+ "normalized_hypothesis": "i defy any man to get a solomon island sore in california",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0427",
+ "category": "phonetic_balance",
+ "text": "Now our figuring was all right.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0427.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "NOW OUR FIGURING WAS ALL RIGHT",
+ "raw": {
+ "normalized_reference": "now our figuring was all right.",
+ "normalized_hypothesis": "now our figuring was all right",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "now our figuring was all right",
+ "normalized_hypothesis": "now our figuring was all right",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0428",
+ "category": "phonetic_balance",
+ "text": "MacDougall tapped his forehead suspiciously with a stubby forefinger.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0428.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "MACDUGAL TAPPED HIS FOREHEAD SUSPICIOUSLY WITH A STUBBY FOREFINGER",
+ "raw": {
+ "normalized_reference": "macdougall tapped his forehead suspiciously with a stubby forefinger.",
+ "normalized_hypothesis": "macdugal tapped his forehead suspiciously with a stubby forefinger",
+ "errors": 1,
+ "words": 9,
+ "wer": 0.1111111111111111
+ },
+ "semantic": {
+ "normalized_reference": "macdougall tapped his forehead suspiciously with a stubby forefinger",
+ "normalized_hypothesis": "macdugal tapped his forehead suspiciously with a stubby forefinger",
+ "errors": 1,
+ "words": 9,
+ "wer": 0.1111111111111111
+ }
+ },
+ {
+ "id": "release500_0429",
+ "category": "phonetic_balance",
+ "text": "O'Brien took off his coat and bared his right arm.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0429.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "O'BRIAN TOOK OFF HIS COAT AND BARED HIS RIGHT ARM",
+ "raw": {
+ "normalized_reference": "o'brien took off his coat and bared his right arm.",
+ "normalized_hypothesis": "o'brian took off his coat and bared his right arm",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ },
+ "semantic": {
+ "normalized_reference": "0 brien took off his coat and bared his right arm",
+ "normalized_hypothesis": "0 brian took off his coat and bared his right arm",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ }
+ },
+ {
+ "id": "release500_0430",
+ "category": "phonetic_balance",
+ "text": "But it won't continue, she said with easy confidence.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0430.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "BUT IT WON'T CONTINUE SHE SAID WITH EASY CONFIDENCE",
+ "raw": {
+ "normalized_reference": "but it won't continue, she said with easy confidence.",
+ "normalized_hypothesis": "but it won't continue she said with easy confidence",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "but it will not continue she said with easy confidence",
+ "normalized_hypothesis": "but it will not continue she said with easy confidence",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0431",
+ "category": "phonetic_balance",
+ "text": "I was still weak from my prolonged immersion.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0431.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I WAS STILL WEAK FROM MY PROLONGED IMMERSION",
+ "raw": {
+ "normalized_reference": "i was still weak from my prolonged immersion.",
+ "normalized_hypothesis": "i was still weak from my prolonged immersion",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i was still weak from my prolonged immersion",
+ "normalized_hypothesis": "i was still weak from my prolonged immersion",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0432",
+ "category": "phonetic_balance",
+ "text": "Oolong was two hundred and fifty miles from the nearest land.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0432.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "ULHLONG WAS TWO HUNDRED AND FIFTY MILES FROM THE NEAREST LAND",
+ "raw": {
+ "normalized_reference": "oolong was two hundred and fifty miles from the nearest land.",
+ "normalized_hypothesis": "ulhlong was two hundred and fifty miles from the nearest land",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ },
+ "semantic": {
+ "normalized_reference": "oolong was 250 miles from the nearest land",
+ "normalized_hypothesis": "ulhlong was 250 miles from the nearest land",
+ "errors": 1,
+ "words": 8,
+ "wer": 0.125
+ }
+ },
+ {
+ "id": "release500_0433",
+ "category": "phonetic_balance",
+ "text": "I just do appreciate it without being able to express my feelings.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0433.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I JUST DO APPRECIATE IT WITHOUT BEING ABLE TO EXPRESS MY FEELINGS",
+ "raw": {
+ "normalized_reference": "i just do appreciate it without being able to express my feelings.",
+ "normalized_hypothesis": "i just do appreciate it without being able to express my feelings",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i just do appreciate it without being able to express my feelings",
+ "normalized_hypothesis": "i just do appreciate it without being able to express my feelings",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0434",
+ "category": "phonetic_balance",
+ "text": "Now just what do you want to know.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0434.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "NOW JUST WHAT DO YOU WANT TO KNOW",
+ "raw": {
+ "normalized_reference": "now just what do you want to know.",
+ "normalized_hypothesis": "now just what do you want to know",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "now just what do you want to know",
+ "normalized_hypothesis": "now just what do you want to know",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0435",
+ "category": "phonetic_balance",
+ "text": "But she swung obediently on her heel into the wind.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0435.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "BUT SHE SWUNG OBEDIENTLY ON HER HEEL INTO THE WIND",
+ "raw": {
+ "normalized_reference": "but she swung obediently on her heel into the wind.",
+ "normalized_hypothesis": "but she swung obediently on her heel into the wind",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "but she swung obediently on her heel into the wind",
+ "normalized_hypothesis": "but she swung obediently on her heel into the wind",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0436",
+ "category": "phonetic_balance",
+ "text": "Ask him, I laughed, then turned to Pasquini.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0436.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "ASK HIM I LAUGHED THEN TURNED TO PASQUINI",
+ "raw": {
+ "normalized_reference": "ask him, i laughed, then turned to pasquini.",
+ "normalized_hypothesis": "ask him i laughed then turned to pasquini",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "ask him i laughed then turned to pasquini",
+ "normalized_hypothesis": "ask him i laughed then turned to pasquini",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0437",
+ "category": "phonetic_balance",
+ "text": "He did not rush in.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0437.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE DID NOT RUSH IN",
+ "raw": {
+ "normalized_reference": "he did not rush in.",
+ "normalized_hypothesis": "he did not rush in",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he did not rush in",
+ "normalized_hypothesis": "he did not rush in",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0438",
+ "category": "phonetic_balance",
+ "text": "They obeyed him, and went here and there at his commands.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0438.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THEY OBEYED HIM AND WENT HERE AND THERE AT HIS COMMANDS",
+ "raw": {
+ "normalized_reference": "they obeyed him, and went here and there at his commands.",
+ "normalized_hypothesis": "they obeyed him and went here and there at his commands",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "they obeyed him and went here and there at his commands",
+ "normalized_hypothesis": "they obeyed him and went here and there at his commands",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0439",
+ "category": "phonetic_balance",
+ "text": "Tom Spink has a harpoon.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0439.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "TOM SPINK HAS A HARPOON",
+ "raw": {
+ "normalized_reference": "tom spink has a harpoon.",
+ "normalized_hypothesis": "tom spink has a harpoon",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "tom spink has a harpoon",
+ "normalized_hypothesis": "tom spink has a harpoon",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0440",
+ "category": "phonetic_balance",
+ "text": "It was my reports from the north which chiefly induced people to buy.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0440.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT WAS MY REPORTS FROM THE NORTH WHICH CHIEFLY INDUCED PEOPLE TO BUY",
+ "raw": {
+ "normalized_reference": "it was my reports from the north which chiefly induced people to buy.",
+ "normalized_hypothesis": "it was my reports from the north which chiefly induced people to buy",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it was my reports from the north which chiefly induced people to buy",
+ "normalized_hypothesis": "it was my reports from the north which chiefly induced people to buy",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0441",
+ "category": "phonetic_balance",
+ "text": "Cherokee identified himself with his instinct.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0441.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "CHEROKEY IDENTIFIED HIMSELF WITH HIS INSTINCT",
+ "raw": {
+ "normalized_reference": "cherokee identified himself with his instinct.",
+ "normalized_hypothesis": "cherokey identified himself with his instinct",
+ "errors": 1,
+ "words": 6,
+ "wer": 0.16666666666666666
+ },
+ "semantic": {
+ "normalized_reference": "cherokee identified himself with his instinct",
+ "normalized_hypothesis": "cherokey identified himself with his instinct",
+ "errors": 1,
+ "words": 6,
+ "wer": 0.16666666666666666
+ }
+ },
+ {
+ "id": "release500_0442",
+ "category": "phonetic_balance",
+ "text": "I graduated last of my class.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0442.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I GRADUATED THE LAST OF MY CLASS",
+ "raw": {
+ "normalized_reference": "i graduated last of my class.",
+ "normalized_hypothesis": "i graduated the last of my class",
+ "errors": 1,
+ "words": 6,
+ "wer": 0.16666666666666666
+ },
+ "semantic": {
+ "normalized_reference": "i graduated last of my class",
+ "normalized_hypothesis": "i graduated the last of my class",
+ "errors": 1,
+ "words": 6,
+ "wer": 0.16666666666666666
+ }
+ },
+ {
+ "id": "release500_0443",
+ "category": "phonetic_balance",
+ "text": "It was a gigantic inadequacy.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0443.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT WAS A GIGANTIC INADEQUACY",
+ "raw": {
+ "normalized_reference": "it was a gigantic inadequacy.",
+ "normalized_hypothesis": "it was a gigantic inadequacy",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it was a gigantic inadequacy",
+ "normalized_hypothesis": "it was a gigantic inadequacy",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0444",
+ "category": "phonetic_balance",
+ "text": "Did I possess too much vitality.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0444.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "DID I POSSESS TOO MUCH VITALITY",
+ "raw": {
+ "normalized_reference": "did i possess too much vitality.",
+ "normalized_hypothesis": "did i possess too much vitality",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "did i possess too much vitality",
+ "normalized_hypothesis": "did i possess too much vitality",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0445",
+ "category": "phonetic_balance",
+ "text": "Therefore, hurrah for the game.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0445.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THEREFORE HURRAH FOR THE GAME",
+ "raw": {
+ "normalized_reference": "therefore, hurrah for the game.",
+ "normalized_hypothesis": "therefore hurrah for the game",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "therefore hurrah for the game",
+ "normalized_hypothesis": "therefore hurrah for the game",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0446",
+ "category": "phonetic_balance",
+ "text": "He had fulfilled his duty and paid properly.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0446.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE HAD FULFILLED HIS DUTY AND PAID PROPERLY",
+ "raw": {
+ "normalized_reference": "he had fulfilled his duty and paid properly.",
+ "normalized_hypothesis": "he had fulfilled his duty and paid properly",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he had fulfilled his duty and paid properly",
+ "normalized_hypothesis": "he had fulfilled his duty and paid properly",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0447",
+ "category": "phonetic_balance",
+ "text": "How can you manage all alone, Mr Young.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0447.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HOW CAN YOU MANAGE ALL ALONE MISTER YOUNG",
+ "raw": {
+ "normalized_reference": "how can you manage all alone, mr young.",
+ "normalized_hypothesis": "how can you manage all alone mister young",
+ "errors": 1,
+ "words": 8,
+ "wer": 0.125
+ },
+ "semantic": {
+ "normalized_reference": "how can you manage all alone mister young",
+ "normalized_hypothesis": "how can you manage all alone mister young",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0448",
+ "category": "phonetic_balance",
+ "text": "Red-Eye never committed a more outrageous deed.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0448.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "RED EYE NEVER COMMITTED A MORE OUTRAGEOUS DEED",
+ "raw": {
+ "normalized_reference": "red-eye never committed a more outrageous deed.",
+ "normalized_hypothesis": "red eye never committed a more outrageous deed",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "red eye never committed a more outrageous deed",
+ "normalized_hypothesis": "red eye never committed a more outrageous deed",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0449",
+ "category": "phonetic_balance",
+ "text": "He would keep his faith with Gregson for the promised day or two.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0449.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE WOULD KEEP HIS FAITH WITH GREGSON FOR THE PROMISED DAY OR TWO",
+ "raw": {
+ "normalized_reference": "he would keep his faith with gregson for the promised day or two.",
+ "normalized_hypothesis": "he would keep his faith with gregson for the promised day or two",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he would keep his faith with gregson for the promised day or 2",
+ "normalized_hypothesis": "he would keep his faith with gregson for the promised day or 2",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0450",
+ "category": "phonetic_balance",
+ "text": "There was one chance, and only one, of saving Jeanne.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0450.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THERE WAS ONE CHANCE AND ONLY ONE OF SAVING JEAN",
+ "raw": {
+ "normalized_reference": "there was one chance, and only one, of saving jeanne.",
+ "normalized_hypothesis": "there was one chance and only one of saving jean",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ },
+ "semantic": {
+ "normalized_reference": "there was one chance and only one of saving jeanne",
+ "normalized_hypothesis": "there was one chance and only one of saving jean",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ }
+ },
+ {
+ "id": "release500_0451",
+ "category": "phonetic_balance",
+ "text": "I was beneath the water, suffocating and drowning.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0451.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I WAS BENEATH THE WATER SUFFOCATING AND DROWNING",
+ "raw": {
+ "normalized_reference": "i was beneath the water, suffocating and drowning.",
+ "normalized_hypothesis": "i was beneath the water suffocating and drowning",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i was beneath the water suffocating and drowning",
+ "normalized_hypothesis": "i was beneath the water suffocating and drowning",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0452",
+ "category": "phonetic_balance",
+ "text": "Give them their choice between a fine or an official whipping.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0452.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "GIVE THEM THEIR CHOICE BETWEEN A FINE OR AN OFFICIAL WHIPPING",
+ "raw": {
+ "normalized_reference": "give them their choice between a fine or an official whipping.",
+ "normalized_hypothesis": "give them their choice between a fine or an official whipping",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "give them their choice between a fine or an official whipping",
+ "normalized_hypothesis": "give them their choice between a fine or an official whipping",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0453",
+ "category": "phonetic_balance",
+ "text": "They must have been swept away by the chaotic currents.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0453.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THEY MUST HAVE BEEN SWEPT AWAY BY THE CHAOTIC CURRENTS",
+ "raw": {
+ "normalized_reference": "they must have been swept away by the chaotic currents.",
+ "normalized_hypothesis": "they must have been swept away by the chaotic currents",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "they must have been swept away by the chaotic currents",
+ "normalized_hypothesis": "they must have been swept away by the chaotic currents",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0454",
+ "category": "phonetic_balance",
+ "text": "You yellow giant thing of the frost.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0454.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "YOU YELLOW GIANT THING OF THE FROST",
+ "raw": {
+ "normalized_reference": "you yellow giant thing of the frost.",
+ "normalized_hypothesis": "you yellow giant thing of the frost",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "you yellow giant thing of the frost",
+ "normalized_hypothesis": "you yellow giant thing of the frost",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0455",
+ "category": "phonetic_balance",
+ "text": "He had peeled off his shirt and was wildly waving it.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0455.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE HAD PEELED OFF HIS SHIRT AND WAS WILDLY WAVING IT",
+ "raw": {
+ "normalized_reference": "he had peeled off his shirt and was wildly waving it.",
+ "normalized_hypothesis": "he had peeled off his shirt and was wildly waving it",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he had peeled off his shirt and was wildly waving it",
+ "normalized_hypothesis": "he had peeled off his shirt and was wildly waving it",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0456",
+ "category": "phonetic_balance",
+ "text": "To these he gave castor oil.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0456.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "TO THESE HE GAVE CASTER OIL",
+ "raw": {
+ "normalized_reference": "to these he gave castor oil.",
+ "normalized_hypothesis": "to these he gave caster oil",
+ "errors": 1,
+ "words": 6,
+ "wer": 0.16666666666666666
+ },
+ "semantic": {
+ "normalized_reference": "to these he gave castor oil",
+ "normalized_hypothesis": "to these he gave caster oil",
+ "errors": 1,
+ "words": 6,
+ "wer": 0.16666666666666666
+ }
+ },
+ {
+ "id": "release500_0457",
+ "category": "phonetic_balance",
+ "text": "Tudor surveyed him with withering disgust.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0457.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "TUDOR SURVEYED HIM WITH WITHERING DISGUST",
+ "raw": {
+ "normalized_reference": "tudor surveyed him with withering disgust.",
+ "normalized_hypothesis": "tudor surveyed him with withering disgust",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "tudor surveyed him with withering disgust",
+ "normalized_hypothesis": "tudor surveyed him with withering disgust",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0458",
+ "category": "phonetic_balance",
+ "text": "After all, it was simply a mistake in judgment.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0458.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "AFTER ALL IT WAS SIMPLY A MISTAKE IN JUDGMENT",
+ "raw": {
+ "normalized_reference": "after all, it was simply a mistake in judgment.",
+ "normalized_hypothesis": "after all it was simply a mistake in judgment",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "after all it was simply a mistake in judgment",
+ "normalized_hypothesis": "after all it was simply a mistake in judgment",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0459",
+ "category": "phonetic_balance",
+ "text": "That is the strange part of it.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0459.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THAT IS THE STRANGE PART OF IT",
+ "raw": {
+ "normalized_reference": "that is the strange part of it.",
+ "normalized_hypothesis": "that is the strange part of it",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "that is the strange part of it",
+ "normalized_hypothesis": "that is the strange part of it",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0460",
+ "category": "phonetic_balance",
+ "text": "Not a wheel moved in his empire.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0460.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "NOT A WHEEL MOVED IN HIS EMPIRE",
+ "raw": {
+ "normalized_reference": "not a wheel moved in his empire.",
+ "normalized_hypothesis": "not a wheel moved in his empire",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "not a wheel moved in his empire",
+ "normalized_hypothesis": "not a wheel moved in his empire",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0461",
+ "category": "phonetic_balance",
+ "text": "Its diameter was not more than two hundred yards.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0461.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "ITS DIAMETER WAS NOT MORE THAN TWO HUNDRED YARDS",
+ "raw": {
+ "normalized_reference": "its diameter was not more than two hundred yards.",
+ "normalized_hypothesis": "its diameter was not more than two hundred yards",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "its diameter was not more than 200 yards",
+ "normalized_hypothesis": "its diameter was not more than 200 yards",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0462",
+ "category": "phonetic_balance",
+ "text": "Nimrod replied, with a slight manifestation of sensitiveness.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0462.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "NIMROD REPLIED WITH A SLIGHT MANIFESTATION OF SENSITIVENESS",
+ "raw": {
+ "normalized_reference": "nimrod replied, with a slight manifestation of sensitiveness.",
+ "normalized_hypothesis": "nimrod replied with a slight manifestation of sensitiveness",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "nimrod replied with a slight manifestation of sensitiveness",
+ "normalized_hypothesis": "nimrod replied with a slight manifestation of sensitiveness",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0463",
+ "category": "phonetic_balance",
+ "text": "I tried to read George Moore last night, and was dreadfully bored.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0463.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I TRIED TO READ GEORGE MOORE LAST NIGHT AND WAS DREADFULLY BORED",
+ "raw": {
+ "normalized_reference": "i tried to read george moore last night, and was dreadfully bored.",
+ "normalized_hypothesis": "i tried to read george moore last night and was dreadfully bored",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i tried to read george moore last night and was dreadfully bored",
+ "normalized_hypothesis": "i tried to read george moore last night and was dreadfully bored",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0464",
+ "category": "phonetic_balance",
+ "text": "Yes, it was a man who asked, a stranger.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0464.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "YES IT WAS A MAN WHO ASKED A STRANGER",
+ "raw": {
+ "normalized_reference": "yes, it was a man who asked, a stranger.",
+ "normalized_hypothesis": "yes it was a man who asked a stranger",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "yes it was a man who asked a stranger",
+ "normalized_hypothesis": "yes it was a man who asked a stranger",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0465",
+ "category": "phonetic_balance",
+ "text": "We were now good friends.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0465.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WE WERE NOW GOOD FRIENDS",
+ "raw": {
+ "normalized_reference": "we were now good friends.",
+ "normalized_hypothesis": "we were now good friends",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "we were now good friends",
+ "normalized_hypothesis": "we were now good friends",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0466",
+ "category": "phonetic_balance",
+ "text": "Yet, in accordance with Ernest's test of truth, it worked.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0466.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "YET IN ACCORDANCE WITH ERNEST'S TEST OF TRUTH IT WORKED",
+ "raw": {
+ "normalized_reference": "yet, in accordance with ernest's test of truth, it worked.",
+ "normalized_hypothesis": "yet in accordance with ernest's test of truth it worked",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "yet in accordance with ernest is test of truth it worked",
+ "normalized_hypothesis": "yet in accordance with ernest is test of truth it worked",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0467",
+ "category": "phonetic_balance",
+ "text": "It's the strap hangers that'll keep us from going under.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0467.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT'S THE STRAP HANGERS THAT'LL KEEP US FROM GOING UNDER",
+ "raw": {
+ "normalized_reference": "it's the strap hangers that'll keep us from going under.",
+ "normalized_hypothesis": "it's the strap hangers that'll keep us from going under",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it is the strap hangers that will keep us from going under",
+ "normalized_hypothesis": "it is the strap hangers that will keep us from going under",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0468",
+ "category": "phonetic_balance",
+ "text": "I never allow what can't be changed to annoy me.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0468.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I NEVER ALLOW WHAT CAN'T BE CHANGED TO ANNOY ME",
+ "raw": {
+ "normalized_reference": "i never allow what can't be changed to annoy me.",
+ "normalized_hypothesis": "i never allow what can't be changed to annoy me",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i never allow what can not be changed to annoy me",
+ "normalized_hypothesis": "i never allow what can not be changed to annoy me",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0469",
+ "category": "phonetic_balance",
+ "text": "After all superfluous flesh is gone what is left is stringy and resistant.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0469.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "AFTER ALL SUPERFLUOUS FLESH IS GONE WHAT IS LEFT IS STRINGY AND RESISTANT",
+ "raw": {
+ "normalized_reference": "after all superfluous flesh is gone what is left is stringy and resistant.",
+ "normalized_hypothesis": "after all superfluous flesh is gone what is left is stringy and resistant",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "after all superfluous flesh is gone what is left is stringy and resistant",
+ "normalized_hypothesis": "after all superfluous flesh is gone what is left is stringy and resistant",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0470",
+ "category": "phonetic_balance",
+ "text": "Then he shouted, Shut up.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0470.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THEN HE SHOUTED SHUT UP",
+ "raw": {
+ "normalized_reference": "then he shouted, shut up.",
+ "normalized_hypothesis": "then he shouted shut up",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "then he shouted shut up",
+ "normalized_hypothesis": "then he shouted shut up",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0471",
+ "category": "phonetic_balance",
+ "text": "He had worshiped her, as Dante might have worshiped Beatrice.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0471.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE HAD WORSHIPPED HER AS DANTES MIGHT HAVE WORSHIPPED BEATRICE",
+ "raw": {
+ "normalized_reference": "he had worshiped her, as dante might have worshiped beatrice.",
+ "normalized_hypothesis": "he had worshipped her as dantes might have worshipped beatrice",
+ "errors": 3,
+ "words": 10,
+ "wer": 0.3
+ },
+ "semantic": {
+ "normalized_reference": "he had worshiped her as dante might have worshiped beatrice",
+ "normalized_hypothesis": "he had worshipped her as dantes might have worshipped beatrice",
+ "errors": 3,
+ "words": 10,
+ "wer": 0.3
+ }
+ },
+ {
+ "id": "release500_0472",
+ "category": "phonetic_balance",
+ "text": "He looked like one who had passed through an uncomfortable hour or two.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0472.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE LOOKED LIKE ONE WHO HAD PASSED THROUGH AN UNCOMFORTABLE HOUR OR TWO",
+ "raw": {
+ "normalized_reference": "he looked like one who had passed through an uncomfortable hour or two.",
+ "normalized_hypothesis": "he looked like one who had passed through an uncomfortable hour or two",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he looked like one who had passed through an uncomfortable hour or 2",
+ "normalized_hypothesis": "he looked like one who had passed through an uncomfortable hour or 2",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0473",
+ "category": "phonetic_balance",
+ "text": "The history of our westward-faring race is written in it.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0473.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE HISTORY OF OUR WESTWARD FARING RACE IS WRITTEN IN IT",
+ "raw": {
+ "normalized_reference": "the history of our westward-faring race is written in it.",
+ "normalized_hypothesis": "the history of our westward faring race is written in it",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the history of our westward faring race is written in it",
+ "normalized_hypothesis": "the history of our westward faring race is written in it",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0474",
+ "category": "phonetic_balance",
+ "text": "He may desire to escape pain, or to enjoy pleasure.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0474.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE MAY DESIRE TO ESCAPE PAIN OR TO ENJOY PLEASURE",
+ "raw": {
+ "normalized_reference": "he may desire to escape pain, or to enjoy pleasure.",
+ "normalized_hypothesis": "he may desire to escape pain or to enjoy pleasure",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he may desire to escape pain or to enjoy pleasure",
+ "normalized_hypothesis": "he may desire to escape pain or to enjoy pleasure",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0475",
+ "category": "phonetic_balance",
+ "text": "And the air was growing chilly.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0475.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "AND THE AIR WAS GROWING CHILLY",
+ "raw": {
+ "normalized_reference": "and the air was growing chilly.",
+ "normalized_hypothesis": "and the air was growing chilly",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "and the air was growing chilly",
+ "normalized_hypothesis": "and the air was growing chilly",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0476",
+ "category": "phonetic_balance",
+ "text": "Dig in; you're sure good, was Daylight's answer.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0476.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "DIGGIN YOU'RE SURE GOOD WAS DAYLIGHT'S ANSWER",
+ "raw": {
+ "normalized_reference": "dig in; you're sure good, was daylight's answer.",
+ "normalized_hypothesis": "diggin you're sure good was daylight's answer",
+ "errors": 2,
+ "words": 8,
+ "wer": 0.25
+ },
+ "semantic": {
+ "normalized_reference": "dig in you are sure good was daylight is answer",
+ "normalized_hypothesis": "diggin you are sure good was daylight is answer",
+ "errors": 2,
+ "words": 10,
+ "wer": 0.2
+ }
+ },
+ {
+ "id": "release500_0477",
+ "category": "phonetic_balance",
+ "text": "My idea was that he would have more influence over the natives.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0477.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "MY IDEA WAS THAT HE WOULD HAVE MORE INFLUENCE OVER THE NATIVES",
+ "raw": {
+ "normalized_reference": "my idea was that he would have more influence over the natives.",
+ "normalized_hypothesis": "my idea was that he would have more influence over the natives",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "my idea was that he would have more influence over the natives",
+ "normalized_hypothesis": "my idea was that he would have more influence over the natives",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0478",
+ "category": "phonetic_balance",
+ "text": "Now run along, and tell them to hurry.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0478.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "NOW RUN ALONG AND TELL THEM TO HURRY",
+ "raw": {
+ "normalized_reference": "now run along, and tell them to hurry.",
+ "normalized_hypothesis": "now run along and tell them to hurry",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "now run along and tell them to hurry",
+ "normalized_hypothesis": "now run along and tell them to hurry",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0479",
+ "category": "phonetic_balance",
+ "text": "I have no idea, replied Philip.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0479.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I HAVE NO IDEA REPLIED PHILIP",
+ "raw": {
+ "normalized_reference": "i have no idea, replied philip.",
+ "normalized_hypothesis": "i have no idea replied philip",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i have no idea replied philip",
+ "normalized_hypothesis": "i have no idea replied philip",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0480",
+ "category": "phonetic_balance",
+ "text": "For the rest, he was a mere automaton.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0480.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "FOR THE REST HE WAS A MERE AUTOMATON",
+ "raw": {
+ "normalized_reference": "for the rest, he was a mere automaton.",
+ "normalized_hypothesis": "for the rest he was a mere automaton",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "for the rest he was a mere automaton",
+ "normalized_hypothesis": "for the rest he was a mere automaton",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0481",
+ "category": "phonetic_balance",
+ "text": "Why, the average review is more nauseating than cod liver oil.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0481.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WHY THE AVERAGE REVIEW IS MORE NAUSEATING THAN COD LIVER OIL",
+ "raw": {
+ "normalized_reference": "why, the average review is more nauseating than cod liver oil.",
+ "normalized_hypothesis": "why the average review is more nauseating than cod liver oil",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "why the average review is more nauseating than cod liver oil",
+ "normalized_hypothesis": "why the average review is more nauseating than cod liver oil",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0482",
+ "category": "phonetic_balance",
+ "text": "I have been robbed, sir, I amended.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0482.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I HAVE BEEN ROBBED SIR I AMENDED",
+ "raw": {
+ "normalized_reference": "i have been robbed, sir, i amended.",
+ "normalized_hypothesis": "i have been robbed sir i amended",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i have been robbed sir i amended",
+ "normalized_hypothesis": "i have been robbed sir i amended",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0483",
+ "category": "phonetic_balance",
+ "text": "He moved away as quietly as he had come.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0483.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE MOVED AWAY AS QUIETLY AS HE HAD COME",
+ "raw": {
+ "normalized_reference": "he moved away as quietly as he had come.",
+ "normalized_hypothesis": "he moved away as quietly as he had come",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he moved away as quietly as he had come",
+ "normalized_hypothesis": "he moved away as quietly as he had come",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0484",
+ "category": "phonetic_balance",
+ "text": "A trickle of fresh blood ran over his face.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0484.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "A TRICKLE OF FRESH BLOOD RAN OVER HIS FACE",
+ "raw": {
+ "normalized_reference": "a trickle of fresh blood ran over his face.",
+ "normalized_hypothesis": "a trickle of fresh blood ran over his face",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "a trickle of fresh blood ran over his face",
+ "normalized_hypothesis": "a trickle of fresh blood ran over his face",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0485",
+ "category": "phonetic_balance",
+ "text": "Suppose you saw me at work through the window.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0485.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SUPPOSE YOU SAW ME AT WORK THROUGH THE WINDOW",
+ "raw": {
+ "normalized_reference": "suppose you saw me at work through the window.",
+ "normalized_hypothesis": "suppose you saw me at work through the window",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "suppose you saw me at work through the window",
+ "normalized_hypothesis": "suppose you saw me at work through the window",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0486",
+ "category": "phonetic_balance",
+ "text": "But who was Eileen's double.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0486.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "BUT WHO WAS ILEEN'S DOUBLE",
+ "raw": {
+ "normalized_reference": "but who was eileen's double.",
+ "normalized_hypothesis": "but who was ileen's double",
+ "errors": 1,
+ "words": 5,
+ "wer": 0.2
+ },
+ "semantic": {
+ "normalized_reference": "but who was eileen is double",
+ "normalized_hypothesis": "but who was ileen is double",
+ "errors": 1,
+ "words": 6,
+ "wer": 0.16666666666666666
+ }
+ },
+ {
+ "id": "release500_0487",
+ "category": "phonetic_balance",
+ "text": "Darkness hid him from Jeanne.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0487.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "DARKNESS HID HIM FROM JEAN",
+ "raw": {
+ "normalized_reference": "darkness hid him from jeanne.",
+ "normalized_hypothesis": "darkness hid him from jean",
+ "errors": 1,
+ "words": 5,
+ "wer": 0.2
+ },
+ "semantic": {
+ "normalized_reference": "darkness hid him from jeanne",
+ "normalized_hypothesis": "darkness hid him from jean",
+ "errors": 1,
+ "words": 5,
+ "wer": 0.2
+ }
+ },
+ {
+ "id": "release500_0488",
+ "category": "phonetic_balance",
+ "text": "By this time Charley was as enraged as the Greek.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0488.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "BY THIS TIME CHARLEY WAS AS ENRAGED AS THE GREEK",
+ "raw": {
+ "normalized_reference": "by this time charley was as enraged as the greek.",
+ "normalized_hypothesis": "by this time charley was as enraged as the greek",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "by this time charley was as enraged as the greek",
+ "normalized_hypothesis": "by this time charley was as enraged as the greek",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0489",
+ "category": "phonetic_balance",
+ "text": "I am writing these lines in Honolulu, Hawaii.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0489.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I AM WRITING THESE LINES IN HONOLULU HAWAII",
+ "raw": {
+ "normalized_reference": "i am writing these lines in honolulu, hawaii.",
+ "normalized_hypothesis": "i am writing these lines in honolulu hawaii",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i am writing these lines in honolulu hawaii",
+ "normalized_hypothesis": "i am writing these lines in honolulu hawaii",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0490",
+ "category": "phonetic_balance",
+ "text": "Nobody knows how the natives got them.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0490.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "NOBODY KNOWS HOW THE NATIVES GOT THEM",
+ "raw": {
+ "normalized_reference": "nobody knows how the natives got them.",
+ "normalized_hypothesis": "nobody knows how the natives got them",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "nobody knows how the natives got them",
+ "normalized_hypothesis": "nobody knows how the natives got them",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0491",
+ "category": "phonetic_balance",
+ "text": "And how would we ever find ourselves.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0491.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "AND HOW WOULD WE EVER FIND OURSELVES",
+ "raw": {
+ "normalized_reference": "and how would we ever find ourselves.",
+ "normalized_hypothesis": "and how would we ever find ourselves",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "and how would we ever find ourselves",
+ "normalized_hypothesis": "and how would we ever find ourselves",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0492",
+ "category": "phonetic_balance",
+ "text": "They are coming ashore, whoever they are.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0492.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THEY ARE COMING ASHORE WHOEVER THEY ARE",
+ "raw": {
+ "normalized_reference": "they are coming ashore, whoever they are.",
+ "normalized_hypothesis": "they are coming ashore whoever they are",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "they are coming ashore whoever they are",
+ "normalized_hypothesis": "they are coming ashore whoever they are",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0493",
+ "category": "phonetic_balance",
+ "text": "This is my fifth voyage.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0493.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THIS IS MY FIFTH VOYAGE",
+ "raw": {
+ "normalized_reference": "this is my fifth voyage.",
+ "normalized_hypothesis": "this is my fifth voyage",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "this is my 5th voyage",
+ "normalized_hypothesis": "this is my 5th voyage",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0494",
+ "category": "phonetic_balance",
+ "text": "And yet if she came he had no words to say.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0494.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "AND YET IF SHE CAME HE HAD NO WORDS TO SAY",
+ "raw": {
+ "normalized_reference": "and yet if she came he had no words to say.",
+ "normalized_hypothesis": "and yet if she came he had no words to say",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "and yet if she came he had no words to say",
+ "normalized_hypothesis": "and yet if she came he had no words to say",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0495",
+ "category": "phonetic_balance",
+ "text": "You must sleep, he urged.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0495.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "YOU MUST SLEEP HE URGED",
+ "raw": {
+ "normalized_reference": "you must sleep, he urged.",
+ "normalized_hypothesis": "you must sleep he urged",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "you must sleep he urged",
+ "normalized_hypothesis": "you must sleep he urged",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0496",
+ "category": "phonetic_balance",
+ "text": "Could the incident have anything to do with Jeanne and Pierre.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0496.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "COULD THE INCIDENT HAVE ANYTHING TO DO WITH JEAN AND PIERRE",
+ "raw": {
+ "normalized_reference": "could the incident have anything to do with jeanne and pierre.",
+ "normalized_hypothesis": "could the incident have anything to do with jean and pierre",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ },
+ "semantic": {
+ "normalized_reference": "could the incident have anything to do with jeanne and pierre",
+ "normalized_hypothesis": "could the incident have anything to do with jean and pierre",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ }
+ },
+ {
+ "id": "release500_0497",
+ "category": "phonetic_balance",
+ "text": "Each day she became a more vital part of him.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0497.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "EACH DAY SHE BECAME A MORE VITAL PART OF HIM",
+ "raw": {
+ "normalized_reference": "each day she became a more vital part of him.",
+ "normalized_hypothesis": "each day she became a more vital part of him",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "each day she became a more vital part of him",
+ "normalized_hypothesis": "each day she became a more vital part of him",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0498",
+ "category": "phonetic_balance",
+ "text": "But why continue the tirade, for tirade it was.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0498.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "BUT WHY CONTINUE THE TIRADE FOR TIRADE IT WAS",
+ "raw": {
+ "normalized_reference": "but why continue the tirade, for tirade it was.",
+ "normalized_hypothesis": "but why continue the tirade for tirade it was",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "but why continue the tirade for tirade it was",
+ "normalized_hypothesis": "but why continue the tirade for tirade it was",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0499",
+ "category": "phonetic_balance",
+ "text": "That is what distinguishes all of us from the lower animals.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0499.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THAT IS WHAT DISTINGUISHES ALL OF US FROM THE LOWER ANIMALS",
+ "raw": {
+ "normalized_reference": "that is what distinguishes all of us from the lower animals.",
+ "normalized_hypothesis": "that is what distinguishes all of us from the lower animals",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "that is what distinguishes all of us from the lower animals",
+ "normalized_hypothesis": "that is what distinguishes all of us from the lower animals",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0500",
+ "category": "phonetic_balance",
+ "text": "There was something pathetic in the girl's attitude now.",
+ "model": "kitten-nano-hugo",
+ "audio": "benchmark/audio/kitten-nano-hugo/release500_0500.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THERE WAS SOMETHING PATHETIC IN THE GIRL'S ATTITUDE NOW",
+ "raw": {
+ "normalized_reference": "there was something pathetic in the girl's attitude now.",
+ "normalized_hypothesis": "there was something pathetic in the girl's attitude now",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "there was something pathetic in the girl is attitude now",
+ "normalized_hypothesis": "there was something pathetic in the girl is attitude now",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0001",
+ "category": "conversational",
+ "text": "It wasn't until later that I realized what had actually happened.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0001.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT WASN'T UNTIL LATER THAT I REALIZED WHAT HAD ACTUALLY HAPPENED",
+ "raw": {
+ "normalized_reference": "it wasn't until later that i realized what had actually happened.",
+ "normalized_hypothesis": "it wasn't until later that i realized what had actually happened",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it was not until later that i realized what had actually happened",
+ "normalized_hypothesis": "it was not until later that i realized what had actually happened",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0002",
+ "category": "descriptive",
+ "text": "The light at that time of afternoon is perfect for about twenty minutes.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0002.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "E LIGHT AT THAT TIME OF AFTERNOON IS PERFECT FOR ABOUT TWENTY MINUTES",
+ "raw": {
+ "normalized_reference": "the light at that time of afternoon is perfect for about twenty minutes.",
+ "normalized_hypothesis": "e light at that time of afternoon is perfect for about twenty minutes",
+ "errors": 1,
+ "words": 13,
+ "wer": 0.07692307692307693
+ },
+ "semantic": {
+ "normalized_reference": "the light at that time of afternoon is perfect for about 20 minutes",
+ "normalized_hypothesis": "e light at that time of afternoon is perfect for about 20 minutes",
+ "errors": 1,
+ "words": 13,
+ "wer": 0.07692307692307693
+ }
+ },
+ {
+ "id": "release500_0003",
+ "category": "emotional",
+ "text": "That was the scariest thing I've ever done, and also maybe the best.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0003.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THAT WAS THE SCARIEST THING I'VE EVER DONE AND ALSO MAYBE THE BEST",
+ "raw": {
+ "normalized_reference": "that was the scariest thing i've ever done, and also maybe the best.",
+ "normalized_hypothesis": "that was the scariest thing i've ever done and also maybe the best",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "that was the scariest thing i have ever done and also maybe the best",
+ "normalized_hypothesis": "that was the scariest thing i have ever done and also maybe the best",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0004",
+ "category": "homographs",
+ "text": "After I record the record, please present the present to the bass player.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0004.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "AFTER I RECORD THE RECORD PLEASE PRESENT THE PRESENT TO THE BASE PLAYER",
+ "raw": {
+ "normalized_reference": "after i record the record, please present the present to the bass player.",
+ "normalized_hypothesis": "after i record the record please present the present to the base player",
+ "errors": 1,
+ "words": 13,
+ "wer": 0.07692307692307693
+ },
+ "semantic": {
+ "normalized_reference": "after i record the record please present the present to the bass player",
+ "normalized_hypothesis": "after i record the record please present the present to the base player",
+ "errors": 1,
+ "words": 13,
+ "wer": 0.07692307692307693
+ }
+ },
+ {
+ "id": "release500_0005",
+ "category": "long",
+ "text": "If I'm being completely honest, I had no idea what I was doing for most of it, and I think that might be the most normal thing about the whole experience.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0005.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IF I'M BEING COMPLETELY HONEST I HAD NO IDEA WHAT I WAS DOING FOR MOST OF IT AND I THINK THAT MIGHT BE THE MOST NORMAL THING ABOUT THE WHOLE EXPERIENCE",
+ "raw": {
+ "normalized_reference": "if i'm being completely honest, i had no idea what i was doing for most of it, and i think that might be the most normal thing about the whole experience.",
+ "normalized_hypothesis": "if i'm being completely honest i had no idea what i was doing for most of it and i think that might be the most normal thing about the whole experience",
+ "errors": 0,
+ "words": 31,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "if i am being completely honest i had no idea what i was doing for most of it and i think that might be the most normal thing about the whole experience",
+ "normalized_hypothesis": "if i am being completely honest i had no idea what i was doing for most of it and i think that might be the most normal thing about the whole experience",
+ "errors": 0,
+ "words": 32,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0006",
+ "category": "long_range",
+ "text": "The committee postponed its decision after discovering that three pages of the report had been printed in the wrong order.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0006.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "COMMITTEE POSTPONED ITS DECISION AFTER DISCOVERING THAT THREE PAGES OF THE REPORT HAD BEEN PRINTED IN THE WRONG ORDER",
+ "raw": {
+ "normalized_reference": "the committee postponed its decision after discovering that three pages of the report had been printed in the wrong order.",
+ "normalized_hypothesis": "committee postponed its decision after discovering that three pages of the report had been printed in the wrong order",
+ "errors": 1,
+ "words": 20,
+ "wer": 0.05
+ },
+ "semantic": {
+ "normalized_reference": "the committee postponed its decision after discovering that 3 pages of the report had been printed in the wrong order",
+ "normalized_hypothesis": "committee postponed its decision after discovering that 3 pages of the report had been printed in the wrong order",
+ "errors": 1,
+ "words": 20,
+ "wer": 0.05
+ }
+ },
+ {
+ "id": "release500_0007",
+ "category": "names_places",
+ "text": "Gwendolyn photographed the eucalyptus trees outside Ljubljana.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0007.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "GWENDOLEN PHOTOGRAPHED THE EUCALYPTUS TREES OUTSIDE THE UBLIANA",
+ "raw": {
+ "normalized_reference": "gwendolyn photographed the eucalyptus trees outside ljubljana.",
+ "normalized_hypothesis": "gwendolen photographed the eucalyptus trees outside the ubliana",
+ "errors": 3,
+ "words": 7,
+ "wer": 0.42857142857142855
+ },
+ "semantic": {
+ "normalized_reference": "gwendolyn photographed the eucalyptus trees outside ljubljana",
+ "normalized_hypothesis": "gwendolen photographed the eucalyptus trees outside the ubliana",
+ "errors": 3,
+ "words": 7,
+ "wer": 0.42857142857142855
+ }
+ },
+ {
+ "id": "release500_0008",
+ "category": "narrative",
+ "text": "He had this theory, and the more time passes, the more I think he was right.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0008.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE HAD THIS THEORY AND THE MORE TIME PASSES THE MORE I THINK HE WAS RIGHT",
+ "raw": {
+ "normalized_reference": "he had this theory, and the more time passes, the more i think he was right.",
+ "normalized_hypothesis": "he had this theory and the more time passes the more i think he was right",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he had this theory and the more time passes the more i think he was right",
+ "normalized_hypothesis": "he had this theory and the more time passes the more i think he was right",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0009",
+ "category": "numbers",
+ "text": "The package weighs twelve point six kilograms and arrived on July twenty-first.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0009.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE PACKAGE WEIGHS TWELVE POINT SIX KILOGRAMS AND ARRIVED ON JULY TWENTY FIRST",
+ "raw": {
+ "normalized_reference": "the package weighs twelve point six kilograms and arrived on july twenty-first.",
+ "normalized_hypothesis": "the package weighs twelve point six kilograms and arrived on july twenty first",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the package weighs 12.6 kilograms and arrived on july 21st",
+ "normalized_hypothesis": "the package weighs 12.6 kilograms and arrived on july 21st",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0010",
+ "category": "punctuation",
+ "text": "First, close the window; second, turn off the lamp; finally, lock the door.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0010.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "FIRST CLOSED THE WINDOW SECOND TURN OFF THE LAMP FINALLY LOCK THE DOOR",
+ "raw": {
+ "normalized_reference": "first, close the window; second, turn off the lamp; finally, lock the door.",
+ "normalized_hypothesis": "first closed the window second turn off the lamp finally lock the door",
+ "errors": 1,
+ "words": 13,
+ "wer": 0.07692307692307693
+ },
+ "semantic": {
+ "normalized_reference": "1st close the window 2nd turn off the lamp finally lock the door",
+ "normalized_hypothesis": "1st closed the window 2nd turn off the lamp finally lock the door",
+ "errors": 1,
+ "words": 13,
+ "wer": 0.07692307692307693
+ }
+ },
+ {
+ "id": "release500_0011",
+ "category": "question",
+ "text": "When did you first realize something was wrong?",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0011.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WHEN DID YOU FIRST REALIZE SOMETHING WAS WRONG",
+ "raw": {
+ "normalized_reference": "when did you first realize something was wrong?",
+ "normalized_hypothesis": "when did you first realize something was wrong",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "when did you 1st realize something was wrong",
+ "normalized_hypothesis": "when did you 1st realize something was wrong",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0012",
+ "category": "rare_words",
+ "text": "The archaeologist catalogued a Byzantine reliquary and several anachronistic fragments.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0012.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE ARCHAEOLOGIST CATALOGUED E BIZANTINE RULIQUARY AND SEVERAL ANACERNISTIC FRAGMENTS",
+ "raw": {
+ "normalized_reference": "the archaeologist catalogued a byzantine reliquary and several anachronistic fragments.",
+ "normalized_hypothesis": "the archaeologist catalogued e bizantine ruliquary and several anacernistic fragments",
+ "errors": 4,
+ "words": 10,
+ "wer": 0.4
+ },
+ "semantic": {
+ "normalized_reference": "the archaeologist catalogued a byzantine reliquary and several anachronistic fragments",
+ "normalized_hypothesis": "the archaeologist catalogued e bizantine ruliquary and several anacernistic fragments",
+ "errors": 4,
+ "words": 10,
+ "wer": 0.4
+ }
+ },
+ {
+ "id": "release500_0013",
+ "category": "short",
+ "text": "I have no idea.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0013.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I HAVE NO IDEA",
+ "raw": {
+ "normalized_reference": "i have no idea.",
+ "normalized_hypothesis": "i have no idea",
+ "errors": 0,
+ "words": 4,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i have no idea",
+ "normalized_hypothesis": "i have no idea",
+ "errors": 0,
+ "words": 4,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0014",
+ "category": "technical",
+ "text": "The system runs on three core components that all have to stay in sync.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0014.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE SYSTEM RUNS ON THREE CORPS COMPONENTS THAT ALL HAVE TO STAY AND SINK",
+ "raw": {
+ "normalized_reference": "the system runs on three core components that all have to stay in sync.",
+ "normalized_hypothesis": "the system runs on three corps components that all have to stay and sink",
+ "errors": 3,
+ "words": 14,
+ "wer": 0.21428571428571427
+ },
+ "semantic": {
+ "normalized_reference": "the system runs on 3 core components that all have to stay in sync",
+ "normalized_hypothesis": "the system runs on 3 corps components that all have to stay and sink",
+ "errors": 3,
+ "words": 14,
+ "wer": 0.21428571428571427
+ }
+ },
+ {
+ "id": "release500_0015",
+ "category": "conversational",
+ "text": "We improvised the entire second half and somehow nobody noticed.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0015.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WE IMPROVISED THE ENTIRE SECOND HALF AND SOMEHOW NOBODY NOTICED",
+ "raw": {
+ "normalized_reference": "we improvised the entire second half and somehow nobody noticed.",
+ "normalized_hypothesis": "we improvised the entire second half and somehow nobody noticed",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "we improvised the entire 2nd half and somehow nobody noticed",
+ "normalized_hypothesis": "we improvised the entire 2nd half and somehow nobody noticed",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0016",
+ "category": "descriptive",
+ "text": "There's a smell in old buildings that's impossible to describe but immediately familiar.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0016.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THERE'S A SMELL IN OLD BUILDINGS THAT'S IMPOSSIBLE TO DESCRIBE BUT IMMEDIATELY FAMILIAR",
+ "raw": {
+ "normalized_reference": "there's a smell in old buildings that's impossible to describe but immediately familiar.",
+ "normalized_hypothesis": "there's a smell in old buildings that's impossible to describe but immediately familiar",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "there is a smell in old buildings that is impossible to describe but immediately familiar",
+ "normalized_hypothesis": "there is a smell in old buildings that is impossible to describe but immediately familiar",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0017",
+ "category": "emotional",
+ "text": "I specifically asked for one thing. One thing. And it still didn't happen.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0017.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I SPECIFICALLY ASKED FOR ONE THING ONE THING AND IT STILL DIDN'T HAPPEN",
+ "raw": {
+ "normalized_reference": "i specifically asked for one thing. one thing. and it still didn't happen.",
+ "normalized_hypothesis": "i specifically asked for one thing one thing and it still didn't happen",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i specifically asked for one thing one thing and it still did not happen",
+ "normalized_hypothesis": "i specifically asked for one thing one thing and it still did not happen",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0018",
+ "category": "homographs",
+ "text": "They refuse to place the refuse beside the produce aisle.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0018.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THEY REFUSED TO PLACE THE REFUSE BESIDE THE PRODUCE ISLE",
+ "raw": {
+ "normalized_reference": "they refuse to place the refuse beside the produce aisle.",
+ "normalized_hypothesis": "they refused to place the refuse beside the produce isle",
+ "errors": 2,
+ "words": 10,
+ "wer": 0.2
+ },
+ "semantic": {
+ "normalized_reference": "they refuse to place the refuse beside the produce aisle",
+ "normalized_hypothesis": "they refused to place the refuse beside the produce isle",
+ "errors": 2,
+ "words": 10,
+ "wer": 0.2
+ }
+ },
+ {
+ "id": "release500_0019",
+ "category": "long",
+ "text": "There were days last year when I genuinely didn't know how I was going to get through it, and somehow I did, and I still don't fully understand how.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0019.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HERE WERE DAYS LAST YEAR WHEN I GENUINELY DIDN'T KNOW HOW I WAS GOING TO GET THROUGH IT AND SOMEHOW I DID AND I STILL DON'T FULLY UNDERSTAND HOW",
+ "raw": {
+ "normalized_reference": "there were days last year when i genuinely didn't know how i was going to get through it, and somehow i did, and i still don't fully understand how.",
+ "normalized_hypothesis": "here were days last year when i genuinely didn't know how i was going to get through it and somehow i did and i still don't fully understand how",
+ "errors": 1,
+ "words": 29,
+ "wer": 0.034482758620689655
+ },
+ "semantic": {
+ "normalized_reference": "there were days last year when i genuinely did not know how i was going to get through it and somehow i did and i still do not fully understand how",
+ "normalized_hypothesis": "here were days last year when i genuinely did not know how i was going to get through it and somehow i did and i still do not fully understand how",
+ "errors": 1,
+ "words": 31,
+ "wer": 0.03225806451612903
+ }
+ },
+ {
+ "id": "release500_0020",
+ "category": "long_range",
+ "text": "Before announcing the winner, the judge thanked the volunteers who had worked through the storm to repair the damaged stage.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0020.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "EFORE ANNOUNCING THE WINNER THE JUDGE THANKED THE VOLUNTEERS WHO HAD WORKED THROUGH THE STORM TO REPAIR THE DAMAGE STAGE",
+ "raw": {
+ "normalized_reference": "before announcing the winner, the judge thanked the volunteers who had worked through the storm to repair the damaged stage.",
+ "normalized_hypothesis": "efore announcing the winner the judge thanked the volunteers who had worked through the storm to repair the damage stage",
+ "errors": 2,
+ "words": 20,
+ "wer": 0.1
+ },
+ "semantic": {
+ "normalized_reference": "before announcing the winner the judge thanked the volunteers who had worked through the storm to repair the damaged stage",
+ "normalized_hypothesis": "efore announcing the winner the judge thanked the volunteers who had worked through the storm to repair the damage stage",
+ "errors": 2,
+ "words": 20,
+ "wer": 0.1
+ }
+ },
+ {
+ "id": "release500_0021",
+ "category": "names_places",
+ "text": "Professor Nguyen mailed the fluorescent poster to Reykjavik on Thursday.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0021.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "PROFESSOR NIA MAILED THE FLORESCENT POSTER TO RAKEVICK ON THURSDAY",
+ "raw": {
+ "normalized_reference": "professor nguyen mailed the fluorescent poster to reykjavik on thursday.",
+ "normalized_hypothesis": "professor nia mailed the florescent poster to rakevick on thursday",
+ "errors": 3,
+ "words": 10,
+ "wer": 0.3
+ },
+ "semantic": {
+ "normalized_reference": "professor nguyen mailed the fluorescent poster to reykjavik on thursday",
+ "normalized_hypothesis": "professor nia mailed the florescent poster to rakevick on thursday",
+ "errors": 3,
+ "words": 10,
+ "wer": 0.3
+ }
+ },
+ {
+ "id": "release500_0022",
+ "category": "narrative",
+ "text": "The plan was simple. The execution was anything but.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0022.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE PLAN WAS SIMPLE THE EXECUTION WAS ANYTHING BUT",
+ "raw": {
+ "normalized_reference": "the plan was simple. the execution was anything but.",
+ "normalized_hypothesis": "the plan was simple the execution was anything but",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the plan was simple the execution was anything but",
+ "normalized_hypothesis": "the plan was simple the execution was anything but",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0023",
+ "category": "numbers",
+ "text": "Her score improved from eighty-two percent to ninety-seven percent in two weeks.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0023.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HER SCORE IMPROVED FROM EIGHTY TWO PER CENT TO NINETY SEVEN PER CENT IN TWO WEEKS",
+ "raw": {
+ "normalized_reference": "her score improved from eighty-two percent to ninety-seven percent in two weeks.",
+ "normalized_hypothesis": "her score improved from eighty two per cent to ninety seven per cent in two weeks",
+ "errors": 4,
+ "words": 14,
+ "wer": 0.2857142857142857
+ },
+ "semantic": {
+ "normalized_reference": "her score improved from 82% to 97% in 2 weeks",
+ "normalized_hypothesis": "her score improved from 82% to 97% in 2 weeks",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0024",
+ "category": "punctuation",
+ "text": "Why would anyone pack spoons, batteries, and a violin for one short trip?",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0024.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WHY WOULD ANYONE PACK SPOONS BATTERIES AND A VIOLIN FOR ONE SHORT TRIP",
+ "raw": {
+ "normalized_reference": "why would anyone pack spoons, batteries, and a violin for one short trip?",
+ "normalized_hypothesis": "why would anyone pack spoons batteries and a violin for one short trip",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "why would anyone pack spoons batteries and a violin for one short trip",
+ "normalized_hypothesis": "why would anyone pack spoons batteries and a violin for one short trip",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0025",
+ "category": "question",
+ "text": "How long have you been sitting on this?",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0025.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HOWLONG HAVE YOU BEEN SITTING ON THIS",
+ "raw": {
+ "normalized_reference": "how long have you been sitting on this?",
+ "normalized_hypothesis": "howlong have you been sitting on this",
+ "errors": 2,
+ "words": 8,
+ "wer": 0.25
+ },
+ "semantic": {
+ "normalized_reference": "how long have you been sitting on this",
+ "normalized_hypothesis": "howlong have you been sitting on this",
+ "errors": 2,
+ "words": 8,
+ "wer": 0.25
+ }
+ },
+ {
+ "id": "release500_0026",
+ "category": "rare_words",
+ "text": "A mischievous chameleon rested beside the chrysanthemum and the bougainvillea.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0026.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "A MISCHIEVOUS CAMELION RESTED BESIDE THE CHRYSANTHEMUM AND THE BAUGAN VILLIA",
+ "raw": {
+ "normalized_reference": "a mischievous chameleon rested beside the chrysanthemum and the bougainvillea.",
+ "normalized_hypothesis": "a mischievous camelion rested beside the chrysanthemum and the baugan villia",
+ "errors": 3,
+ "words": 10,
+ "wer": 0.3
+ },
+ "semantic": {
+ "normalized_reference": "a mischievous chameleon rested beside the chrysanthemum and the bougainvillea",
+ "normalized_hypothesis": "a mischievous camelion rested beside the chrysanthemum and the baugan villia",
+ "errors": 3,
+ "words": 10,
+ "wer": 0.3
+ }
+ },
+ {
+ "id": "release500_0027",
+ "category": "short",
+ "text": "We don't talk about that.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0027.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WE DON'T TALK ABOUT THAT",
+ "raw": {
+ "normalized_reference": "we don't talk about that.",
+ "normalized_hypothesis": "we don't talk about that",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "we do not talk about that",
+ "normalized_hypothesis": "we do not talk about that",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0028",
+ "category": "technical",
+ "text": "Hash the UTF eight payload with S H A two fifty-six before opening the checkpoint.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0028.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HASH THE U T F EIGHT PALOAD WITH S H A TWO FIFTY SIX BEFORE OPENING THE CHECK POINT",
+ "raw": {
+ "normalized_reference": "hash the utf eight payload with s h a two fifty-six before opening the checkpoint.",
+ "normalized_hypothesis": "hash the u t f eight paload with s h a two fifty six before opening the check point",
+ "errors": 6,
+ "words": 16,
+ "wer": 0.375
+ },
+ "semantic": {
+ "normalized_reference": "hash the utf 8 payload with s h a 256 before opening the checkpoint",
+ "normalized_hypothesis": "hash the u t f 8 paload with s h a 256 before opening the check point",
+ "errors": 6,
+ "words": 14,
+ "wer": 0.42857142857142855
+ }
+ },
+ {
+ "id": "release500_0029",
+ "category": "conversational",
+ "text": "I thought I had it all figured out, and then this happened.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0029.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I THOUGHT I HAD IT ALL FIGURED OUT AND THEN THIS HAPPENED",
+ "raw": {
+ "normalized_reference": "i thought i had it all figured out, and then this happened.",
+ "normalized_hypothesis": "i thought i had it all figured out and then this happened",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i thought i had it all figured out and then this happened",
+ "normalized_hypothesis": "i thought i had it all figured out and then this happened",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0030",
+ "category": "descriptive",
+ "text": "The kind of quiet you only get at three in the morning in a city.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0030.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE KIND OF QUIET YOU ONLY GET AT THREE IN THE MORNING IN A CITY",
+ "raw": {
+ "normalized_reference": "the kind of quiet you only get at three in the morning in a city.",
+ "normalized_hypothesis": "the kind of quiet you only get at three in the morning in a city",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the kind of quiet you only get at 3 in the morning in a city",
+ "normalized_hypothesis": "the kind of quiet you only get at 3 in the morning in a city",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0031",
+ "category": "emotional",
+ "text": "The joy of it was almost overwhelming, like too much good all at once.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0031.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "E JOY OF IT WAS ALMOST OVERWHELMING LIKE TOO MUCH GOOD ALL AT ONCE",
+ "raw": {
+ "normalized_reference": "the joy of it was almost overwhelming, like too much good all at once.",
+ "normalized_hypothesis": "e joy of it was almost overwhelming like too much good all at once",
+ "errors": 1,
+ "words": 14,
+ "wer": 0.07142857142857142
+ },
+ "semantic": {
+ "normalized_reference": "the joy of it was almost overwhelming like too much good all at once",
+ "normalized_hypothesis": "e joy of it was almost overwhelming like too much good all at once",
+ "errors": 1,
+ "words": 14,
+ "wer": 0.07142857142857142
+ }
+ },
+ {
+ "id": "release500_0032",
+ "category": "homographs",
+ "text": "Please close the door while I stand close to the microphone.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0032.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "PLEASE CLOSE THE DOOR WHILE I STAND CLOSE TO THE MICROPHONE",
+ "raw": {
+ "normalized_reference": "please close the door while i stand close to the microphone.",
+ "normalized_hypothesis": "please close the door while i stand close to the microphone",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "please close the door while i stand close to the microphone",
+ "normalized_hypothesis": "please close the door while i stand close to the microphone",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0033",
+ "category": "long",
+ "text": "I think the most underrated skill in the world is knowing when to stop explaining yourself and just let people draw their own conclusions.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0033.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I THINK THE MOST UNDERRATED SKILL IN THE WORLD IS KNOWING ONES STOP EXPLAINING YOURSELF AND JUST LET PEOPLE DRAW THEIR OWN CONCLUSIONS",
+ "raw": {
+ "normalized_reference": "i think the most underrated skill in the world is knowing when to stop explaining yourself and just let people draw their own conclusions.",
+ "normalized_hypothesis": "i think the most underrated skill in the world is knowing ones stop explaining yourself and just let people draw their own conclusions",
+ "errors": 2,
+ "words": 24,
+ "wer": 0.08333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "i think the most underrated skill in the world is knowing when to stop explaining yourself and just let people draw their own conclusions",
+ "normalized_hypothesis": "i think the most underrated skill in the world is knowing ones stop explaining yourself and just let people draw their own conclusions",
+ "errors": 2,
+ "words": 24,
+ "wer": 0.08333333333333333
+ }
+ },
+ {
+ "id": "release500_0034",
+ "category": "long_range",
+ "text": "When the final bus disappeared around the corner, Nadia remembered that her umbrella, notebook, and lunch were still inside the quiet office.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0034.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WHEN THE FINAL BUST DISAPPEARED AROUND THE CORNER NOJA REMEMBERED THAT HER UMBRELLA NOTE BOOK AND LUNCH WERE STILL INSIDE THE QUIET OFFICE",
+ "raw": {
+ "normalized_reference": "when the final bus disappeared around the corner, nadia remembered that her umbrella, notebook, and lunch were still inside the quiet office.",
+ "normalized_hypothesis": "when the final bust disappeared around the corner noja remembered that her umbrella note book and lunch were still inside the quiet office",
+ "errors": 4,
+ "words": 22,
+ "wer": 0.18181818181818182
+ },
+ "semantic": {
+ "normalized_reference": "when the final bus disappeared around the corner nadia remembered that her umbrella notebook and lunch were still inside the quiet office",
+ "normalized_hypothesis": "when the final bust disappeared around the corner noja remembered that her umbrella note book and lunch were still inside the quiet office",
+ "errors": 4,
+ "words": 22,
+ "wer": 0.18181818181818182
+ }
+ },
+ {
+ "id": "release500_0035",
+ "category": "names_places",
+ "text": "Aurelia and Joaquin crossed the Schuylkill River before noon.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0035.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "ORELIAH AND JOKWIN CROSSED THE SHIOLKOLE RIVER BEFORE NOON",
+ "raw": {
+ "normalized_reference": "aurelia and joaquin crossed the schuylkill river before noon.",
+ "normalized_hypothesis": "oreliah and jokwin crossed the shiolkole river before noon",
+ "errors": 3,
+ "words": 9,
+ "wer": 0.3333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "aurelia and joaquin crossed the schuylkill river before noon",
+ "normalized_hypothesis": "oreliah and jokwin crossed the shiolkole river before noon",
+ "errors": 3,
+ "words": 9,
+ "wer": 0.3333333333333333
+ }
+ },
+ {
+ "id": "release500_0036",
+ "category": "narrative",
+ "text": "The village had a name, but nobody used it. Everyone just called it home.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0036.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE VILLAGE HAD A NAME BUT NOBODY USED IT EVERYONE JUST CALLED IT HOME",
+ "raw": {
+ "normalized_reference": "the village had a name, but nobody used it. everyone just called it home.",
+ "normalized_hypothesis": "the village had a name but nobody used it everyone just called it home",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the village had a name but nobody used it everyone just called it home",
+ "normalized_hypothesis": "the village had a name but nobody used it everyone just called it home",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0037",
+ "category": "numbers",
+ "text": "Increase the gain by three decibels, wait five seconds, then reduce it by one.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0037.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "INCREASE THE GAIN BY THREE DESIBALS WAIGHT FIVE SECONDS THEN REDUCE IT BY ONE",
+ "raw": {
+ "normalized_reference": "increase the gain by three decibels, wait five seconds, then reduce it by one.",
+ "normalized_hypothesis": "increase the gain by three desibals waight five seconds then reduce it by one",
+ "errors": 2,
+ "words": 14,
+ "wer": 0.14285714285714285
+ },
+ "semantic": {
+ "normalized_reference": "increase the gain by 3 decibels wait 5 seconds then reduce it by one",
+ "normalized_hypothesis": "increase the gain by 3 desibals waight 5 seconds then reduce it by one",
+ "errors": 2,
+ "words": 14,
+ "wer": 0.14285714285714285
+ }
+ },
+ {
+ "id": "release500_0038",
+ "category": "punctuation",
+ "text": "The result was unexpected: clear, stable, and surprisingly natural.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0038.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE RESULT WAS UNEXPECTED CLEAR STABLE AND SURPRISINGLY NATURAL",
+ "raw": {
+ "normalized_reference": "the result was unexpected: clear, stable, and surprisingly natural.",
+ "normalized_hypothesis": "the result was unexpected clear stable and surprisingly natural",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the result was unexpected clear stable and surprisingly natural",
+ "normalized_hypothesis": "the result was unexpected clear stable and surprisingly natural",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0039",
+ "category": "question",
+ "text": "Who told you that was the right way to do it?",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0039.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WHOTOLD YOU THAT WAS THE RIGHT WAY TO DO IT",
+ "raw": {
+ "normalized_reference": "who told you that was the right way to do it?",
+ "normalized_hypothesis": "whotold you that was the right way to do it",
+ "errors": 2,
+ "words": 11,
+ "wer": 0.18181818181818182
+ },
+ "semantic": {
+ "normalized_reference": "who told you that was the right way to do it",
+ "normalized_hypothesis": "whotold you that was the right way to do it",
+ "errors": 2,
+ "words": 11,
+ "wer": 0.18181818181818182
+ }
+ },
+ {
+ "id": "release500_0040",
+ "category": "rare_words",
+ "text": "The meticulous cartographer examined an iridescent archipelago through binoculars.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0040.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HEMETICULOUS CHROTOGRAPHER EXAMINED AN IRIDESCENT ARCHIPELAGO THROUGH BONOCULARS",
+ "raw": {
+ "normalized_reference": "the meticulous cartographer examined an iridescent archipelago through binoculars.",
+ "normalized_hypothesis": "hemeticulous chrotographer examined an iridescent archipelago through bonoculars",
+ "errors": 4,
+ "words": 9,
+ "wer": 0.4444444444444444
+ },
+ "semantic": {
+ "normalized_reference": "the meticulous cartographer examined an iridescent archipelago through binoculars",
+ "normalized_hypothesis": "hemeticulous chrotographer examined an iridescent archipelago through bonoculars",
+ "errors": 4,
+ "words": 9,
+ "wer": 0.4444444444444444
+ }
+ },
+ {
+ "id": "release500_0041",
+ "category": "short",
+ "text": "I mean. Yeah. Kind of.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0041.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I MEAN NOW KIND O",
+ "raw": {
+ "normalized_reference": "i mean. yeah. kind of.",
+ "normalized_hypothesis": "i mean now kind o",
+ "errors": 2,
+ "words": 5,
+ "wer": 0.4
+ },
+ "semantic": {
+ "normalized_reference": "i mean yeah kind of",
+ "normalized_hypothesis": "i mean now kind 0",
+ "errors": 2,
+ "words": 5,
+ "wer": 0.4
+ }
+ },
+ {
+ "id": "release500_0042",
+ "category": "technical",
+ "text": "The logs are your best friend when something behaves unexpectedly.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0042.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE LOGS ARE YOUR BEST FRIEND WHEN SOMETHING BEHAVES UNEXPECTEDLY",
+ "raw": {
+ "normalized_reference": "the logs are your best friend when something behaves unexpectedly.",
+ "normalized_hypothesis": "he logs are your best friend when something behaves unexpectedly",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ },
+ "semantic": {
+ "normalized_reference": "the logs are your best friend when something behaves unexpectedly",
+ "normalized_hypothesis": "he logs are your best friend when something behaves unexpectedly",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ }
+ },
+ {
+ "id": "release500_0043",
+ "category": "conversational",
+ "text": "He kept talking and I kept nodding and neither of us was listening.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0043.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE KEPT TALKING AND I KEPT NODDING AND NEITHER OF US WAS LISTENING",
+ "raw": {
+ "normalized_reference": "he kept talking and i kept nodding and neither of us was listening.",
+ "normalized_hypothesis": "he kept talking and i kept nodding and neither of us was listening",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he kept talking and i kept nodding and neither of us was listening",
+ "normalized_hypothesis": "he kept talking and i kept nodding and neither of us was listening",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0044",
+ "category": "descriptive",
+ "text": "It's one of those places where everyone seems to know each other already.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0044.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT'S ONE OF THOSE PLACES WHERE EVERYONE SEEMS TO KNOW EACH OTHER ALREADY",
+ "raw": {
+ "normalized_reference": "it's one of those places where everyone seems to know each other already.",
+ "normalized_hypothesis": "it's one of those places where everyone seems to know each other already",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it is one of those places where everyone seems to know each other already",
+ "normalized_hypothesis": "it is one of those places where everyone seems to know each other already",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0045",
+ "category": "emotional",
+ "text": "He said he was proud of me and nobody had ever said that to me before.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0045.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE SAID HE WAS PROUD OF ME AND NOBODY HAD EVER SAID THAT TO ME BEFORE",
+ "raw": {
+ "normalized_reference": "he said he was proud of me and nobody had ever said that to me before.",
+ "normalized_hypothesis": "he said he was proud of me and nobody had ever said that to me before",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he said he was proud of me and nobody had ever said that to me before",
+ "normalized_hypothesis": "he said he was proud of me and nobody had ever said that to me before",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0046",
+ "category": "homographs",
+ "text": "The minute detail took a minute to explain clearly.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0046.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "MINUTE DETAIL TOOK A MINUTE TO EXPLAIN CLEARLY",
+ "raw": {
+ "normalized_reference": "the minute detail took a minute to explain clearly.",
+ "normalized_hypothesis": "minute detail took a minute to explain clearly",
+ "errors": 1,
+ "words": 9,
+ "wer": 0.1111111111111111
+ },
+ "semantic": {
+ "normalized_reference": "the minute detail took a minute to explain clearly",
+ "normalized_hypothesis": "minute detail took a minute to explain clearly",
+ "errors": 1,
+ "words": 9,
+ "wer": 0.1111111111111111
+ }
+ },
+ {
+ "id": "release500_0047",
+ "category": "long",
+ "text": "The strange thing about really wanting something is that getting it never quite feels the way you imagined it would, and not getting it never quite feels as bad.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0047.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE STRANGE THING ABOUT REALLY WANTING SOMETHING IS THAT GETTING IT NEVER QUITE FEELS THE WAY YOU IMAGINED IT WOULD AND NOT GETTING IT NEVER QUITE FEELS AS BAD",
+ "raw": {
+ "normalized_reference": "the strange thing about really wanting something is that getting it never quite feels the way you imagined it would, and not getting it never quite feels as bad.",
+ "normalized_hypothesis": "the strange thing about really wanting something is that getting it never quite feels the way you imagined it would and not getting it never quite feels as bad",
+ "errors": 0,
+ "words": 29,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the strange thing about really wanting something is that getting it never quite feels the way you imagined it would and not getting it never quite feels as bad",
+ "normalized_hypothesis": "the strange thing about really wanting something is that getting it never quite feels the way you imagined it would and not getting it never quite feels as bad",
+ "errors": 0,
+ "words": 29,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0048",
+ "category": "long_range",
+ "text": "Even though the recipe looked simple at first, preparing the sauce required patience, careful timing, and a much larger pan.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0048.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "EVEN THOUGH THE RECIPE LOOKED SIMPLE AT FIRST PREPARING THIS SAUCE REQUIRED PATIENCE CAREFUL TIMING AND A MUCH LARGER PAN",
+ "raw": {
+ "normalized_reference": "even though the recipe looked simple at first, preparing the sauce required patience, careful timing, and a much larger pan.",
+ "normalized_hypothesis": "even though the recipe looked simple at first preparing this sauce required patience careful timing and a much larger pan",
+ "errors": 1,
+ "words": 20,
+ "wer": 0.05
+ },
+ "semantic": {
+ "normalized_reference": "even though the recipe looked simple at 1st preparing the sauce required patience careful timing and a much larger pan",
+ "normalized_hypothesis": "even though the recipe looked simple at 1st preparing this sauce required patience careful timing and a much larger pan",
+ "errors": 1,
+ "words": 20,
+ "wer": 0.05
+ }
+ },
+ {
+ "id": "release500_0049",
+ "category": "names_places",
+ "text": "Xiomara met Siobhan near the old library in Worcestershire.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0049.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SIIMARA MET CHAUVON NEAR THE OLD LIBRARY IN WORCESTERSHIRE",
+ "raw": {
+ "normalized_reference": "xiomara met siobhan near the old library in worcestershire.",
+ "normalized_hypothesis": "siimara met chauvon near the old library in worcestershire",
+ "errors": 2,
+ "words": 9,
+ "wer": 0.2222222222222222
+ },
+ "semantic": {
+ "normalized_reference": "xiomara met siobhan near the old library in worcestershire",
+ "normalized_hypothesis": "siimara met chauvon near the old library in worcestershire",
+ "errors": 2,
+ "words": 9,
+ "wer": 0.2222222222222222
+ }
+ },
+ {
+ "id": "release500_0050",
+ "category": "narrative",
+ "text": "She walked into the room and everyone went quiet, not because she was loud, but because she wasn't.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0050.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SHE WALKED INTO THE ROOM AND EVERYONE WENT QUIET NOT BECAUSE SHE WAS ALOWD BUT BECAUSE SHE WASN'T",
+ "raw": {
+ "normalized_reference": "she walked into the room and everyone went quiet, not because she was loud, but because she wasn't.",
+ "normalized_hypothesis": "she walked into the room and everyone went quiet not because she was alowd but because she wasn't",
+ "errors": 1,
+ "words": 18,
+ "wer": 0.05555555555555555
+ },
+ "semantic": {
+ "normalized_reference": "she walked into the room and everyone went quiet not because she was loud but because she was not",
+ "normalized_hypothesis": "she walked into the room and everyone went quiet not because she was alowd but because she was not",
+ "errors": 1,
+ "words": 19,
+ "wer": 0.05263157894736842
+ }
+ },
+ {
+ "id": "release500_0051",
+ "category": "numbers",
+ "text": "At exactly seven forty-five, Morgan placed eighteen dollars and fifty cents beside the notebook.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0051.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "AT EXACTLY SEVEN FOUR TO FIVE MORGAN PLACED EIGHTEEN DOLLARS AND FIFTY CENTS BESIDE THE NOTEBOOK",
+ "raw": {
+ "normalized_reference": "at exactly seven forty-five, morgan placed eighteen dollars and fifty cents beside the notebook.",
+ "normalized_hypothesis": "at exactly seven four to five morgan placed eighteen dollars and fifty cents beside the notebook",
+ "errors": 2,
+ "words": 15,
+ "wer": 0.13333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "at exactly 745 morgan placed $18.50 beside the notebook",
+ "normalized_hypothesis": "at exactly 74 to 5 morgan placed $18.50 beside the notebook",
+ "errors": 3,
+ "words": 10,
+ "wer": 0.3
+ }
+ },
+ {
+ "id": "release500_0052",
+ "category": "punctuation",
+ "text": "Does punctuation work? Yes: commas, semicolons, and periods should create sensible boundaries.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0052.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IS PUNCTUATION WORK YES COMMAS SEMICOLANS AND PERIODS SHOULD CREATE SENSIBLE BOUNDARIES",
+ "raw": {
+ "normalized_reference": "does punctuation work? yes: commas, semicolons, and periods should create sensible boundaries.",
+ "normalized_hypothesis": "is punctuation work yes commas semicolans and periods should create sensible boundaries",
+ "errors": 2,
+ "words": 12,
+ "wer": 0.16666666666666666
+ },
+ "semantic": {
+ "normalized_reference": "does punctuation work yes commas semicolons and periods should create sensible boundaries",
+ "normalized_hypothesis": "is punctuation work yes commas semicolans and periods should create sensible boundaries",
+ "errors": 2,
+ "words": 12,
+ "wer": 0.16666666666666666
+ }
+ },
+ {
+ "id": "release500_0053",
+ "category": "question",
+ "text": "Have you ever just stopped and thought about how strange all of this is?",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0053.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HAVE YOU EVER JUST STOPPED AND THOUGHT ABOUT HOW STRANGE ALL OF THIS IS",
+ "raw": {
+ "normalized_reference": "have you ever just stopped and thought about how strange all of this is?",
+ "normalized_hypothesis": "have you ever just stopped and thought about how strange all of this is",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "have you ever just stopped and thought about how strange all of this is",
+ "normalized_hypothesis": "have you ever just stopped and thought about how strange all of this is",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0054",
+ "category": "rare_words",
+ "text": "Her idiosyncratic explanation sounded plausible but remained fundamentally ambiguous.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0054.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HER IDIOSYNCRATIC EXPLANATIONS SOUNDED PLAUSIBLE BUT REMAINED FUNDAMENTALLY AMBIGUOUS",
+ "raw": {
+ "normalized_reference": "her idiosyncratic explanation sounded plausible but remained fundamentally ambiguous.",
+ "normalized_hypothesis": "her idiosyncratic explanations sounded plausible but remained fundamentally ambiguous",
+ "errors": 1,
+ "words": 9,
+ "wer": 0.1111111111111111
+ },
+ "semantic": {
+ "normalized_reference": "her idiosyncratic explanation sounded plausible but remained fundamentally ambiguous",
+ "normalized_hypothesis": "her idiosyncratic explanations sounded plausible but remained fundamentally ambiguous",
+ "errors": 1,
+ "words": 9,
+ "wer": 0.1111111111111111
+ }
+ },
+ {
+ "id": "release500_0055",
+ "category": "short",
+ "text": "That tracks.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0055.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THAT DRACKS",
+ "raw": {
+ "normalized_reference": "that tracks.",
+ "normalized_hypothesis": "that dracks",
+ "errors": 1,
+ "words": 2,
+ "wer": 0.5
+ },
+ "semantic": {
+ "normalized_reference": "that tracks",
+ "normalized_hypothesis": "that dracks",
+ "errors": 1,
+ "words": 2,
+ "wer": 0.5
+ }
+ },
+ {
+ "id": "release500_0056",
+ "category": "technical",
+ "text": "The key thing to remember is that order matters here more than you'd think.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0056.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE KEY THING TO REMEMBER IS THAT ORDER MATTERS HERE MORE THAN YOU THINK",
+ "raw": {
+ "normalized_reference": "the key thing to remember is that order matters here more than you'd think.",
+ "normalized_hypothesis": "the key thing to remember is that order matters here more than you think",
+ "errors": 1,
+ "words": 14,
+ "wer": 0.07142857142857142
+ },
+ "semantic": {
+ "normalized_reference": "the key thing to remember is that order matters here more than you would think",
+ "normalized_hypothesis": "the key thing to remember is that order matters here more than you think",
+ "errors": 1,
+ "words": 15,
+ "wer": 0.06666666666666667
+ }
+ },
+ {
+ "id": "release500_0057",
+ "category": "conversational",
+ "text": "I would not say that again in this room if I were you.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0057.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WOULD NOT SAY THAT AGAIN IN THIS ROOM IF I WERE YOU",
+ "raw": {
+ "normalized_reference": "i would not say that again in this room if i were you.",
+ "normalized_hypothesis": "would not say that again in this room if i were you",
+ "errors": 1,
+ "words": 13,
+ "wer": 0.07692307692307693
+ },
+ "semantic": {
+ "normalized_reference": "i would not say that again in this room if i were you",
+ "normalized_hypothesis": "would not say that again in this room if i were you",
+ "errors": 1,
+ "words": 13,
+ "wer": 0.07692307692307693
+ }
+ },
+ {
+ "id": "release500_0058",
+ "category": "descriptive",
+ "text": "The room was filled with the quiet that means something is about to change.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0058.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE ROOM WAS FILLED WITH THE QUIET THAT MEANS SOMETHING IS ABOUT TO CHANGE",
+ "raw": {
+ "normalized_reference": "the room was filled with the quiet that means something is about to change.",
+ "normalized_hypothesis": "the room was filled with the quiet that means something is about to change",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the room was filled with the quiet that means something is about to change",
+ "normalized_hypothesis": "the room was filled with the quiet that means something is about to change",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0059",
+ "category": "emotional",
+ "text": "I didn't expect to feel that way about it, but I did.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0059.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I DIDN'T EXPECT TO FEEL THAT WAY ABOUT IT BUT I DID",
+ "raw": {
+ "normalized_reference": "i didn't expect to feel that way about it, but i did.",
+ "normalized_hypothesis": "i didn't expect to feel that way about it but i did",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i did not expect to feel that way about it but i did",
+ "normalized_hypothesis": "i did not expect to feel that way about it but i did",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0060",
+ "category": "homographs",
+ "text": "I object to the strange object that was left near the entrance.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0060.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I OBJECT TO THE STRANGE OBJECT THAT WAS LEFT NEAR TE ENTRANCE",
+ "raw": {
+ "normalized_reference": "i object to the strange object that was left near the entrance.",
+ "normalized_hypothesis": "i object to the strange object that was left near te entrance",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "i object to the strange object that was left near the entrance",
+ "normalized_hypothesis": "i object to the strange object that was left near te entrance",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ }
+ },
+ {
+ "id": "release500_0061",
+ "category": "long",
+ "text": "I've learned more from the things that went wrong than from the things that worked, which is a frustrating way to have to learn things but apparently the only way that stuck.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0061.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I'VE LEARNED MORE FROM THE THINGS THAT WENT WRONG THAN FROM THE THINGS THAT WORKED WHICH IS A FRUSTRATING WAY TO HAVE TO LEARN THINGS BUT APPARENTLY THE ONLY WAY THAT STUCK",
+ "raw": {
+ "normalized_reference": "i've learned more from the things that went wrong than from the things that worked, which is a frustrating way to have to learn things but apparently the only way that stuck.",
+ "normalized_hypothesis": "i've learned more from the things that went wrong than from the things that worked which is a frustrating way to have to learn things but apparently the only way that stuck",
+ "errors": 0,
+ "words": 32,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i have learned more from the things that went wrong than from the things that worked which is a frustrating way to have to learn things but apparently the only way that stuck",
+ "normalized_hypothesis": "i have learned more from the things that went wrong than from the things that worked which is a frustrating way to have to learn things but apparently the only way that stuck",
+ "errors": 0,
+ "words": 33,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0062",
+ "category": "long_range",
+ "text": "Because every nearby hotel was full, the exhausted hikers followed the narrow road until they found a farmhouse with one empty room.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0062.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "BECAUSE EVERY NEAR BY HOTEL WAS FULL THE EXHAUSTED HIKERS FOLLOWED THE NARROW ROAD UNTIL THEY FOUND A FARMHOUSE WITH ONE EMPTY ROOM",
+ "raw": {
+ "normalized_reference": "because every nearby hotel was full, the exhausted hikers followed the narrow road until they found a farmhouse with one empty room.",
+ "normalized_hypothesis": "because every near by hotel was full the exhausted hikers followed the narrow road until they found a farmhouse with one empty room",
+ "errors": 2,
+ "words": 22,
+ "wer": 0.09090909090909091
+ },
+ "semantic": {
+ "normalized_reference": "because every nearby hotel was full the exhausted hikers followed the narrow road until they found a farmhouse with one empty room",
+ "normalized_hypothesis": "because every near by hotel was full the exhausted hikers followed the narrow road until they found a farmhouse with one empty room",
+ "errors": 2,
+ "words": 22,
+ "wer": 0.09090909090909091
+ }
+ },
+ {
+ "id": "release500_0063",
+ "category": "names_places",
+ "text": "The itinerary sends Beatrice from Albuquerque to Des Moines and then to Boise.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0063.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE ITINERARY SENDS BEATRICE FROM ALBAKIRKY TO DEMOGN AND THEN TO BOISY",
+ "raw": {
+ "normalized_reference": "the itinerary sends beatrice from albuquerque to des moines and then to boise.",
+ "normalized_hypothesis": "the itinerary sends beatrice from albakirky to demogn and then to boisy",
+ "errors": 4,
+ "words": 13,
+ "wer": 0.3076923076923077
+ },
+ "semantic": {
+ "normalized_reference": "the itinerary sends beatrice from albuquerque to des moines and then to boise",
+ "normalized_hypothesis": "the itinerary sends beatrice from albakirky to demogn and then to boisy",
+ "errors": 4,
+ "words": 13,
+ "wer": 0.3076923076923077
+ }
+ },
+ {
+ "id": "release500_0064",
+ "category": "narrative",
+ "text": "We had been driving for six hours when we finally saw the sign for the town.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0064.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WE HAD BEEN DRIVING FOR SIX HOURS WHEN WE FINALLY SAW THE SIGN FOR THE TOWN",
+ "raw": {
+ "normalized_reference": "we had been driving for six hours when we finally saw the sign for the town.",
+ "normalized_hypothesis": "we had been driving for six hours when we finally saw the sign for the town",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "we had been driving for 6 hours when we finally saw the sign for the town",
+ "normalized_hypothesis": "we had been driving for 6 hours when we finally saw the sign for the town",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0065",
+ "category": "numbers",
+ "text": "Call extension zero four nine after eight thirty in the morning.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0065.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "CALL EXTENSION ZERO FOR NINE AFTER EIGHT THIRTY IN THE MORNING",
+ "raw": {
+ "normalized_reference": "call extension zero four nine after eight thirty in the morning.",
+ "normalized_hypothesis": "call extension zero for nine after eight thirty in the morning",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ },
+ "semantic": {
+ "normalized_reference": "call extension 049 after 830 in the morning",
+ "normalized_hypothesis": "call extension 0 for 9 after 830 in the morning",
+ "errors": 3,
+ "words": 8,
+ "wer": 0.375
+ }
+ },
+ {
+ "id": "release500_0066",
+ "category": "punctuation",
+ "text": "Stop - listen carefully - then answer the question without guessing.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0066.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "STOP LISTEN CAREFULLY THEN ANSWER THE QUESTION WITHOUT GUESSING",
+ "raw": {
+ "normalized_reference": "stop - listen carefully - then answer the question without guessing.",
+ "normalized_hypothesis": "stop listen carefully then answer the question without guessing",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "stop listen carefully then answer the question without guessing",
+ "normalized_hypothesis": "stop listen carefully then answer the question without guessing",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0067",
+ "category": "question",
+ "text": "Does it matter at this point, honestly?",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0067.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "DOES IT MATTER AT THIS POINT HONESTLY",
+ "raw": {
+ "normalized_reference": "does it matter at this point, honestly?",
+ "normalized_hypothesis": "does it matter at this point honestly",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "does it matter at this point honestly",
+ "normalized_hypothesis": "does it matter at this point honestly",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0068",
+ "category": "rare_words",
+ "text": "An entrepreneurial meteorologist described the unprecedented atmospheric phenomenon.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0068.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "AN ANTRIPONERIAL MEDEOROLOGIST DESCRIBED THE UNPRECEDENTED ATMOSPHERIC PHENOMENON",
+ "raw": {
+ "normalized_reference": "an entrepreneurial meteorologist described the unprecedented atmospheric phenomenon.",
+ "normalized_hypothesis": "an antriponerial medeorologist described the unprecedented atmospheric phenomenon",
+ "errors": 2,
+ "words": 8,
+ "wer": 0.25
+ },
+ "semantic": {
+ "normalized_reference": "an entrepreneurial meteorologist described the unprecedented atmospheric phenomenon",
+ "normalized_hypothesis": "an antriponerial medeorologist described the unprecedented atmospheric phenomenon",
+ "errors": 2,
+ "words": 8,
+ "wer": 0.25
+ }
+ },
+ {
+ "id": "release500_0069",
+ "category": "short",
+ "text": "That's not what I expected.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0069.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THAT'S NOT WHAT I EXPECTED",
+ "raw": {
+ "normalized_reference": "that's not what i expected.",
+ "normalized_hypothesis": "that's not what i expected",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "that is not what i expected",
+ "normalized_hypothesis": "that is not what i expected",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0070",
+ "category": "technical",
+ "text": "There's a known issue with that version that causes failures under load.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0070.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HERE'S A KNOWN ISSUE WITH THAT VERSION THAT CAUSES FAILURES UNDER LOAD",
+ "raw": {
+ "normalized_reference": "there's a known issue with that version that causes failures under load.",
+ "normalized_hypothesis": "here's a known issue with that version that causes failures under load",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "there is a known issue with that version that causes failures under load",
+ "normalized_hypothesis": "here is a known issue with that version that causes failures under load",
+ "errors": 1,
+ "words": 13,
+ "wer": 0.07692307692307693
+ }
+ },
+ {
+ "id": "release500_0071",
+ "category": "conversational",
+ "text": "Don't look now but I think that's him over there.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0071.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "ON'T LOOK NOW BUT I THINK THAT'S HIM OVER THERE",
+ "raw": {
+ "normalized_reference": "don't look now but i think that's him over there.",
+ "normalized_hypothesis": "on't look now but i think that's him over there",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ },
+ "semantic": {
+ "normalized_reference": "do not look now but i think that is him over there",
+ "normalized_hypothesis": "0 not look now but i think that is him over there",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ }
+ },
+ {
+ "id": "release500_0072",
+ "category": "descriptive",
+ "text": "The city at night looks completely different from how it feels during the day.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0072.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE CITY AT NIGHT LOOKS COMPLETELY DIFFERENT FROM HOW IT FEELS DURING THE DAY",
+ "raw": {
+ "normalized_reference": "the city at night looks completely different from how it feels during the day.",
+ "normalized_hypothesis": "the city at night looks completely different from how it feels during the day",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the city at night looks completely different from how it feels during the day",
+ "normalized_hypothesis": "the city at night looks completely different from how it feels during the day",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0073",
+ "category": "emotional",
+ "text": "It's the kind of happy that makes you nervous because you know it can't last.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0073.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT'S THE KIND OF HAPPY THAT MAKES YOU NERVOUS BECAUSE YOU KNOW IT CAN'T LAST",
+ "raw": {
+ "normalized_reference": "it's the kind of happy that makes you nervous because you know it can't last.",
+ "normalized_hypothesis": "it's the kind of happy that makes you nervous because you know it can't last",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it is the kind of happy that makes you nervous because you know it can not last",
+ "normalized_hypothesis": "it is the kind of happy that makes you nervous because you know it can not last",
+ "errors": 0,
+ "words": 17,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0074",
+ "category": "homographs",
+ "text": "The dove dove beneath the branch while the wind began to wind the loose cable.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0074.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE DOVE DOVED BENEATH THE BRANCH WHILE THE WIND BEGAN TO WIND THE LOOSE CABLE",
+ "raw": {
+ "normalized_reference": "the dove dove beneath the branch while the wind began to wind the loose cable.",
+ "normalized_hypothesis": "he dove doved beneath the branch while the wind began to wind the loose cable",
+ "errors": 2,
+ "words": 15,
+ "wer": 0.13333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "the dove dove beneath the branch while the wind began to wind the loose cable",
+ "normalized_hypothesis": "he dove doved beneath the branch while the wind began to wind the loose cable",
+ "errors": 2,
+ "words": 15,
+ "wer": 0.13333333333333333
+ }
+ },
+ {
+ "id": "release500_0075",
+ "category": "long",
+ "text": "I've thought a lot about what I would do differently if I could go back, and the honest answer is probably not as much as I'd like to think.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0075.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I'VE THOUGHT A LOT ABOUT WHAT I WOULD DO DIFFERENTLY IF I COULD GO BACK AND THE HONEST ANSWER IS PROBABLY NOT AS MUCH AS I'D LIKE TO THINK",
+ "raw": {
+ "normalized_reference": "i've thought a lot about what i would do differently if i could go back, and the honest answer is probably not as much as i'd like to think.",
+ "normalized_hypothesis": "i've thought a lot about what i would do differently if i could go back and the honest answer is probably not as much as i'd like to think",
+ "errors": 0,
+ "words": 29,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i have thought a lot about what i would do differently if i could go back and the honest answer is probably not as much as i would like to think",
+ "normalized_hypothesis": "i have thought a lot about what i would do differently if i could go back and the honest answer is probably not as much as i would like to think",
+ "errors": 0,
+ "words": 31,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0076",
+ "category": "long_range",
+ "text": "Although the weather changed without warning, the musicians finished packing before anyone opened the enormous glass door.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0076.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "ALTHOUGH THE WEATHER CHANGED WITHOUT WARNING THE MUSICIANS FINISHED PACKING BEFORE ANYONE OPENED THE ENORMOUS GLASS DOOR",
+ "raw": {
+ "normalized_reference": "although the weather changed without warning, the musicians finished packing before anyone opened the enormous glass door.",
+ "normalized_hypothesis": "although the weather changed without warning the musicians finished packing before anyone opened the enormous glass door",
+ "errors": 0,
+ "words": 17,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "although the weather changed without warning the musicians finished packing before anyone opened the enormous glass door",
+ "normalized_hypothesis": "although the weather changed without warning the musicians finished packing before anyone opened the enormous glass door",
+ "errors": 0,
+ "words": 17,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0077",
+ "category": "names_places",
+ "text": "Saskatchewan looked unusually bright beneath the pale February sunrise.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0077.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SAS KETUIN LOOKED UNUSUALLY BRIGHT BENEATH THE PALE FEBRUARY SUNRISE",
+ "raw": {
+ "normalized_reference": "saskatchewan looked unusually bright beneath the pale february sunrise.",
+ "normalized_hypothesis": "sas ketuin looked unusually bright beneath the pale february sunrise",
+ "errors": 2,
+ "words": 9,
+ "wer": 0.2222222222222222
+ },
+ "semantic": {
+ "normalized_reference": "saskatchewan looked unusually bright beneath the pale february sunrise",
+ "normalized_hypothesis": "sas ketuin looked unusually bright beneath the pale february sunrise",
+ "errors": 2,
+ "words": 9,
+ "wer": 0.2222222222222222
+ }
+ },
+ {
+ "id": "release500_0078",
+ "category": "narrative",
+ "text": "She handled it with a grace I genuinely did not expect.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0078.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SHE HANDLED IT WITH A GRACE I GENUINELY DID NOT EXPECT",
+ "raw": {
+ "normalized_reference": "she handled it with a grace i genuinely did not expect.",
+ "normalized_hypothesis": "she handled it with a grace i genuinely did not expect",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "she handled it with a grace i genuinely did not expect",
+ "normalized_hypothesis": "she handled it with a grace i genuinely did not expect",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0079",
+ "category": "numbers",
+ "text": "Apartment four hundred seven is across from room twenty-nine on the third floor.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0079.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "APARTMENT FOUR HUNDRED SEVEN IS ACROSS FROM ROOM TWENTY NINE ON THE THIRD FLOOR",
+ "raw": {
+ "normalized_reference": "apartment four hundred seven is across from room twenty-nine on the third floor.",
+ "normalized_hypothesis": "apartment four hundred seven is across from room twenty nine on the third floor",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "apartment 407 is across from room 29 on the 3rd floor",
+ "normalized_hypothesis": "apartment 407 is across from room 29 on the 3rd floor",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0080",
+ "category": "punctuation",
+ "text": "\"Not yet,\" Rowan whispered. \"Give me another minute... please.\"",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0080.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "NOT YET ROWEN WHISPERED GIVE ME ANOTHER MINUTE PLEASE",
+ "raw": {
+ "normalized_reference": "\"not yet,\" rowan whispered. \"give me another minute... please.\"",
+ "normalized_hypothesis": "not yet rowen whispered give me another minute please",
+ "errors": 1,
+ "words": 9,
+ "wer": 0.1111111111111111
+ },
+ "semantic": {
+ "normalized_reference": "not yet rowan whispered give me another minute please",
+ "normalized_hypothesis": "not yet rowen whispered give me another minute please",
+ "errors": 1,
+ "words": 9,
+ "wer": 0.1111111111111111
+ }
+ },
+ {
+ "id": "release500_0081",
+ "category": "question",
+ "text": "What are we actually trying to accomplish here?",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0081.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WHAT ARE WE ACTUALLY TRYING TO ACCOMPLISH HERE",
+ "raw": {
+ "normalized_reference": "what are we actually trying to accomplish here?",
+ "normalized_hypothesis": "what are we actually trying to accomplish here",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "what are we actually trying to accomplish here",
+ "normalized_hypothesis": "what are we actually trying to accomplish here",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0082",
+ "category": "rare_words",
+ "text": "The physician recommended a noninvasive procedure for the recurring neurological symptoms.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0082.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE PHYSICIAN RECOMMENDED A NON EVASIVE PROCEDURE FOR THE RECURRING NEUROLOGICAL SYMPTOMS",
+ "raw": {
+ "normalized_reference": "the physician recommended a noninvasive procedure for the recurring neurological symptoms.",
+ "normalized_hypothesis": "the physician recommended a non evasive procedure for the recurring neurological symptoms",
+ "errors": 2,
+ "words": 11,
+ "wer": 0.18181818181818182
+ },
+ "semantic": {
+ "normalized_reference": "the physician recommended a noninvasive procedure for the recurring neurological symptoms",
+ "normalized_hypothesis": "the physician recommended a non evasive procedure for the recurring neurological symptoms",
+ "errors": 2,
+ "words": 11,
+ "wer": 0.18181818181818182
+ }
+ },
+ {
+ "id": "release500_0083",
+ "category": "short",
+ "text": "Are you serious right now?",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0083.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "AR YOUSERIOUS RIGHT NOW",
+ "raw": {
+ "normalized_reference": "are you serious right now?",
+ "normalized_hypothesis": "ar youserious right now",
+ "errors": 3,
+ "words": 5,
+ "wer": 0.6
+ },
+ "semantic": {
+ "normalized_reference": "are you serious right now",
+ "normalized_hypothesis": "ar youserious right now",
+ "errors": 3,
+ "words": 5,
+ "wer": 0.6
+ }
+ },
+ {
+ "id": "release500_0084",
+ "category": "technical",
+ "text": "The performance improvement is real, but it comes with tradeoffs.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0084.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "PERFORMANCE IMPROVEMENT IS REAL BUT IT COMES WITH TRADOS",
+ "raw": {
+ "normalized_reference": "the performance improvement is real, but it comes with tradeoffs.",
+ "normalized_hypothesis": "performance improvement is real but it comes with trados",
+ "errors": 2,
+ "words": 10,
+ "wer": 0.2
+ },
+ "semantic": {
+ "normalized_reference": "the performance improvement is real but it comes with tradeoffs",
+ "normalized_hypothesis": "performance improvement is real but it comes with trados",
+ "errors": 2,
+ "words": 10,
+ "wer": 0.2
+ }
+ },
+ {
+ "id": "release500_0085",
+ "category": "conversational",
+ "text": "I don't know. It's just been one of those weeks, you know?",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0085.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IDON'T KNOW IT'S JUST BEEN ONE OF THOSE WEEKS YOU KNOW",
+ "raw": {
+ "normalized_reference": "i don't know. it's just been one of those weeks, you know?",
+ "normalized_hypothesis": "idon't know it's just been one of those weeks you know",
+ "errors": 2,
+ "words": 12,
+ "wer": 0.16666666666666666
+ },
+ "semantic": {
+ "normalized_reference": "i do not know it is just been one of those weeks you know",
+ "normalized_hypothesis": "ido not know it is just been one of those weeks you know",
+ "errors": 2,
+ "words": 14,
+ "wer": 0.14285714285714285
+ }
+ },
+ {
+ "id": "release500_0086",
+ "category": "descriptive",
+ "text": "The moment you stop expecting perfection, things get a lot more interesting.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0086.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE MOMENT YOU STOP EXPECTING PERFECTION THINGS GET A LOT MORE INTERESTING",
+ "raw": {
+ "normalized_reference": "the moment you stop expecting perfection, things get a lot more interesting.",
+ "normalized_hypothesis": "the moment you stop expecting perfection things get a lot more interesting",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the moment you stop expecting perfection things get a lot more interesting",
+ "normalized_hypothesis": "the moment you stop expecting perfection things get a lot more interesting",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0087",
+ "category": "emotional",
+ "text": "It's the kind of thing that stays with you even years later.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0087.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT'S THE KIND OF THING THAT STAYS WITH YOU EVEN YEARS LATER",
+ "raw": {
+ "normalized_reference": "it's the kind of thing that stays with you even years later.",
+ "normalized_hypothesis": "it's the kind of thing that stays with you even years later",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it is the kind of thing that stays with you even years later",
+ "normalized_hypothesis": "it is the kind of thing that stays with you even years later",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0088",
+ "category": "long",
+ "text": "She told me once that the most important skill she'd developed was learning to disagree with someone without making them feel they were wrong.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0088.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SHE TOLD ME ONCE THAT THE MOST IMPORTANT SKILL SHE DEVELOPED WAS LEARNING TO DISAGREE WITH SOMEONE WITHOUT MAKING THEM FEEL THEY WERE WRONG",
+ "raw": {
+ "normalized_reference": "she told me once that the most important skill she'd developed was learning to disagree with someone without making them feel they were wrong.",
+ "normalized_hypothesis": "she told me once that the most important skill she developed was learning to disagree with someone without making them feel they were wrong",
+ "errors": 1,
+ "words": 24,
+ "wer": 0.041666666666666664
+ },
+ "semantic": {
+ "normalized_reference": "she told me once that the most important skill she would developed was learning to disagree with someone without making them feel they were wrong",
+ "normalized_hypothesis": "she told me once that the most important skill she developed was learning to disagree with someone without making them feel they were wrong",
+ "errors": 1,
+ "words": 25,
+ "wer": 0.04
+ }
+ },
+ {
+ "id": "release500_0089",
+ "category": "narrative",
+ "text": "He stood at the edge of the cliff and wondered, not for the first time, how he'd gotten here.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0089.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE STOOD AT THE EDGE OF THE CLIFF AND WONDERED NOT FOR THE FIRST TIME HOW HE'D GOTTEN HERE",
+ "raw": {
+ "normalized_reference": "he stood at the edge of the cliff and wondered, not for the first time, how he'd gotten here.",
+ "normalized_hypothesis": "he stood at the edge of the cliff and wondered not for the first time how he'd gotten here",
+ "errors": 0,
+ "words": 19,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he stood at the edge of the cliff and wondered not for the 1st time how he would gotten here",
+ "normalized_hypothesis": "he stood at the edge of the cliff and wondered not for the 1st time how he would gotten here",
+ "errors": 0,
+ "words": 20,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0090",
+ "category": "question",
+ "text": "Why does asking for help feel so hard when you actually need it?",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0090.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WHY DOES ASKING FOR HELP FEEL SO HARD WHEN YOU ACTUALLY NEED IT",
+ "raw": {
+ "normalized_reference": "why does asking for help feel so hard when you actually need it?",
+ "normalized_hypothesis": "why does asking for help feel so hard when you actually need it",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "why does asking for help feel so hard when you actually need it",
+ "normalized_hypothesis": "why does asking for help feel so hard when you actually need it",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0091",
+ "category": "short",
+ "text": "That's fair.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0091.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "DES FAR",
+ "raw": {
+ "normalized_reference": "that's fair.",
+ "normalized_hypothesis": "des far",
+ "errors": 2,
+ "words": 2,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "that is fair",
+ "normalized_hypothesis": "des far",
+ "errors": 3,
+ "words": 3,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0092",
+ "category": "technical",
+ "text": "If you run into that error, the fix is usually just restarting the service.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0092.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IF YOU RUN INTO THAT ERROR THE FIX IS USUALLY JUST RESTORTING THE SERVICE",
+ "raw": {
+ "normalized_reference": "if you run into that error, the fix is usually just restarting the service.",
+ "normalized_hypothesis": "if you run into that error the fix is usually just restorting the service",
+ "errors": 1,
+ "words": 14,
+ "wer": 0.07142857142857142
+ },
+ "semantic": {
+ "normalized_reference": "if you run into that error the fix is usually just restarting the service",
+ "normalized_hypothesis": "if you run into that error the fix is usually just restorting the service",
+ "errors": 1,
+ "words": 14,
+ "wer": 0.07142857142857142
+ }
+ },
+ {
+ "id": "release500_0093",
+ "category": "conversational",
+ "text": "It was a disaster in the best possible way.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0093.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT WAS A DISASTER IN THE BEST POSSIBLE WAY",
+ "raw": {
+ "normalized_reference": "it was a disaster in the best possible way.",
+ "normalized_hypothesis": "it was a disaster in the best possible way",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it was a disaster in the best possible way",
+ "normalized_hypothesis": "it was a disaster in the best possible way",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0094",
+ "category": "descriptive",
+ "text": "It had the look of a place that used to mean something to someone.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0094.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT HAD THE LOOK OF A PLACE THAT USED TO MEAN SOMETHING TO SOME ONE",
+ "raw": {
+ "normalized_reference": "it had the look of a place that used to mean something to someone.",
+ "normalized_hypothesis": "it had the look of a place that used to mean something to some one",
+ "errors": 2,
+ "words": 14,
+ "wer": 0.14285714285714285
+ },
+ "semantic": {
+ "normalized_reference": "it had the look of a place that used to mean something to someone",
+ "normalized_hypothesis": "it had the look of a place that used to mean something to some one",
+ "errors": 2,
+ "words": 14,
+ "wer": 0.14285714285714285
+ }
+ },
+ {
+ "id": "release500_0095",
+ "category": "emotional",
+ "text": "There's this feeling you get when something finally works after weeks of failure.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0095.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THERE'S THIS FEELING YOU GET WHEN SOMETHING FINALLY WORKS AFTER WEEKS OF FAILURE",
+ "raw": {
+ "normalized_reference": "there's this feeling you get when something finally works after weeks of failure.",
+ "normalized_hypothesis": "there's this feeling you get when something finally works after weeks of failure",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "there is this feeling you get when something finally works after weeks of failure",
+ "normalized_hypothesis": "there is this feeling you get when something finally works after weeks of failure",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0096",
+ "category": "long",
+ "text": "He had a theory that most arguments aren't actually about what they seem to be about, and the longer I live the more I think he was onto something.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0096.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE HAD A THEORY THAT MOST ARGUMENTS AREN'T ACTUALLY ABOUT WHAT THEY SEEM TO BE ABOUT AND THE LONGER I LIVE THE MORE I THINK HE WAS ON TO SOMETHING",
+ "raw": {
+ "normalized_reference": "he had a theory that most arguments aren't actually about what they seem to be about, and the longer i live the more i think he was onto something.",
+ "normalized_hypothesis": "he had a theory that most arguments aren't actually about what they seem to be about and the longer i live the more i think he was on to something",
+ "errors": 2,
+ "words": 29,
+ "wer": 0.06896551724137931
+ },
+ "semantic": {
+ "normalized_reference": "he had a theory that most arguments are not actually about what they seem to be about and the longer i live the more i think he was onto something",
+ "normalized_hypothesis": "he had a theory that most arguments are not actually about what they seem to be about and the longer i live the more i think he was on to something",
+ "errors": 2,
+ "words": 30,
+ "wer": 0.06666666666666667
+ }
+ },
+ {
+ "id": "release500_0097",
+ "category": "narrative",
+ "text": "The thing nobody tells you about starting over is how ordinary most of it feels.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0097.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE THING NOBODY TELLS YOU ABOUT STARTING OVER IS HOW ORDINARY MOST OF IT FEELS",
+ "raw": {
+ "normalized_reference": "the thing nobody tells you about starting over is how ordinary most of it feels.",
+ "normalized_hypothesis": "the thing nobody tells you about starting over is how ordinary most of it feels",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the thing nobody tells you about starting over is how ordinary most of it feels",
+ "normalized_hypothesis": "the thing nobody tells you about starting over is how ordinary most of it feels",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0098",
+ "category": "question",
+ "text": "What exactly are we waiting for at this point?",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0098.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WHAT EXACTLY ARE WE WAITING FOR AT THIS POINT",
+ "raw": {
+ "normalized_reference": "what exactly are we waiting for at this point?",
+ "normalized_hypothesis": "what exactly are we waiting for at this point",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "what exactly are we waiting for at this point",
+ "normalized_hypothesis": "what exactly are we waiting for at this point",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0099",
+ "category": "short",
+ "text": "I stand by what I said.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0099.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I STAND BY WHAT I SAY",
+ "raw": {
+ "normalized_reference": "i stand by what i said.",
+ "normalized_hypothesis": "i stand by what i say",
+ "errors": 1,
+ "words": 6,
+ "wer": 0.16666666666666666
+ },
+ "semantic": {
+ "normalized_reference": "i stand by what i said",
+ "normalized_hypothesis": "i stand by what i say",
+ "errors": 1,
+ "words": 6,
+ "wer": 0.16666666666666666
+ }
+ },
+ {
+ "id": "release500_0100",
+ "category": "technical",
+ "text": "Double-check that value before you push anything to production.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0100.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "DOUBLE CHECK THAIT VALUE BEFORE YOU PUSH ANYTHING TO PRODUCTION",
+ "raw": {
+ "normalized_reference": "double-check that value before you push anything to production.",
+ "normalized_hypothesis": "double check thait value before you push anything to production",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ },
+ "semantic": {
+ "normalized_reference": "double check that value before you push anything to production",
+ "normalized_hypothesis": "double check thait value before you push anything to production",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ }
+ },
+ {
+ "id": "release500_0101",
+ "category": "conversational",
+ "text": "I understand your point; I just do not agree with the conclusion.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0101.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "UNDERSTAND YOUR POINT I JUST DO NOT AGREE WITH THE CONCLUSION",
+ "raw": {
+ "normalized_reference": "i understand your point; i just do not agree with the conclusion.",
+ "normalized_hypothesis": "understand your point i just do not agree with the conclusion",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "i understand your point i just do not agree with the conclusion",
+ "normalized_hypothesis": "understand your point i just do not agree with the conclusion",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ }
+ },
+ {
+ "id": "release500_0102",
+ "category": "descriptive",
+ "text": "The sky that day was a strange shade of orange I'd never seen before.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0102.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SKY THA DAY WAS A STRANGE SHADE OF ORANGE I'D NEVER SEEN BEFORE",
+ "raw": {
+ "normalized_reference": "the sky that day was a strange shade of orange i'd never seen before.",
+ "normalized_hypothesis": "sky tha day was a strange shade of orange i'd never seen before",
+ "errors": 2,
+ "words": 14,
+ "wer": 0.14285714285714285
+ },
+ "semantic": {
+ "normalized_reference": "the sky that day was a strange shade of orange i would never seen before",
+ "normalized_hypothesis": "sky tha day was a strange shade of orange i would never seen before",
+ "errors": 2,
+ "words": 15,
+ "wer": 0.13333333333333333
+ }
+ },
+ {
+ "id": "release500_0103",
+ "category": "emotional",
+ "text": "It hurt more than I expected, and I thought I was prepared.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0103.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT HURT MORE THAN I EXPECTED AND I THOUGHT I WAS PREPARED",
+ "raw": {
+ "normalized_reference": "it hurt more than i expected, and i thought i was prepared.",
+ "normalized_hypothesis": "it hurt more than i expected and i thought i was prepared",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it hurt more than i expected and i thought i was prepared",
+ "normalized_hypothesis": "it hurt more than i expected and i thought i was prepared",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0104",
+ "category": "long",
+ "text": "The hardest part wasn't making the decision, it was sitting with the uncertainty of not knowing whether it was the right one for months afterward.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0104.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE HARDEST PART WAS IN MAKING THE DECISION IT WAS SITTING WITH THE UNCERTAINTY OF NOT KNOWING WHETHER IT WAS THE RIGHT ONE FOR MONTHS AFTERWARD",
+ "raw": {
+ "normalized_reference": "the hardest part wasn't making the decision, it was sitting with the uncertainty of not knowing whether it was the right one for months afterward.",
+ "normalized_hypothesis": "the hardest part was in making the decision it was sitting with the uncertainty of not knowing whether it was the right one for months afterward",
+ "errors": 2,
+ "words": 25,
+ "wer": 0.08
+ },
+ "semantic": {
+ "normalized_reference": "the hardest part was not making the decision it was sitting with the uncertainty of not knowing whether it was the right one for months afterward",
+ "normalized_hypothesis": "the hardest part was in making the decision it was sitting with the uncertainty of not knowing whether it was the right one for months afterward",
+ "errors": 1,
+ "words": 26,
+ "wer": 0.038461538461538464
+ }
+ },
+ {
+ "id": "release500_0105",
+ "category": "narrative",
+ "text": "There was this one day last summer when everything just clicked into place.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0105.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THERE WAS THIS ONE DAY LAST SUMMER WHEN EVERYTHING JUST CLICKED INTO PLACE",
+ "raw": {
+ "normalized_reference": "there was this one day last summer when everything just clicked into place.",
+ "normalized_hypothesis": "there was this one day last summer when everything just clicked into place",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "there was this one day last summer when everything just clicked into place",
+ "normalized_hypothesis": "there was this one day last summer when everything just clicked into place",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0106",
+ "category": "question",
+ "text": "Have you ever thought about why we keep doing things the same way when they stop working?",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0106.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HAVE YOU EVER THOUGHT ABOUT WHY WE KEEP DOING THINGS THE SAME WAY WHEN THEY STOP WORKING",
+ "raw": {
+ "normalized_reference": "have you ever thought about why we keep doing things the same way when they stop working?",
+ "normalized_hypothesis": "have you ever thought about why we keep doing things the same way when they stop working",
+ "errors": 0,
+ "words": 17,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "have you ever thought about why we keep doing things the same way when they stop working",
+ "normalized_hypothesis": "have you ever thought about why we keep doing things the same way when they stop working",
+ "errors": 0,
+ "words": 17,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0107",
+ "category": "short",
+ "text": "I can't believe you just said that.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0107.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I CAN'T BELIEVE YOU JUST SAID THAT",
+ "raw": {
+ "normalized_reference": "i can't believe you just said that.",
+ "normalized_hypothesis": "i can't believe you just said that",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i can not believe you just said that",
+ "normalized_hypothesis": "i can not believe you just said that",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0108",
+ "category": "technical",
+ "text": "You'll want to isolate that variable before drawing any conclusions.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0108.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "YOUILL WANT TO ISOLATE THAT VARIABLE BEFORE DRAWING ANY CONCLUSIONS",
+ "raw": {
+ "normalized_reference": "you'll want to isolate that variable before drawing any conclusions.",
+ "normalized_hypothesis": "youill want to isolate that variable before drawing any conclusions",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ },
+ "semantic": {
+ "normalized_reference": "you will want to isolate that variable before drawing any conclusions",
+ "normalized_hypothesis": "youill want to isolate that variable before drawing any conclusions",
+ "errors": 2,
+ "words": 11,
+ "wer": 0.18181818181818182
+ }
+ },
+ {
+ "id": "release500_0109",
+ "category": "conversational",
+ "text": "The meeting that could have been an email lasted two hours.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0109.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE MEETING THAT COULD HAVE BEEN AN ENEL LASTED TWO HOURS",
+ "raw": {
+ "normalized_reference": "the meeting that could have been an email lasted two hours.",
+ "normalized_hypothesis": "he meeting that could have been an enel lasted two hours",
+ "errors": 2,
+ "words": 11,
+ "wer": 0.18181818181818182
+ },
+ "semantic": {
+ "normalized_reference": "the meeting that could have been an email lasted 2 hours",
+ "normalized_hypothesis": "he meeting that could have been an enel lasted 2 hours",
+ "errors": 2,
+ "words": 11,
+ "wer": 0.18181818181818182
+ }
+ },
+ {
+ "id": "release500_0110",
+ "category": "descriptive",
+ "text": "There's a specific kind of tired that sleep doesn't fix.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0110.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HERE'S A SPECIFIC KIND OF TIRED THAT SLEEP DOESN'T FIX",
+ "raw": {
+ "normalized_reference": "there's a specific kind of tired that sleep doesn't fix.",
+ "normalized_hypothesis": "here's a specific kind of tired that sleep doesn't fix",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ },
+ "semantic": {
+ "normalized_reference": "there is a specific kind of tired that sleep does not fix",
+ "normalized_hypothesis": "here is a specific kind of tired that sleep does not fix",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ }
+ },
+ {
+ "id": "release500_0111",
+ "category": "emotional",
+ "text": "Something about that moment broke something open in me that I hadn't realized was closed.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0111.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SOMETHING ABOUT THAT MOMENT BROOKES SOMETHING OPEN IN ME THAT I HADN'T REALIZED WAS CLOSE",
+ "raw": {
+ "normalized_reference": "something about that moment broke something open in me that i hadn't realized was closed.",
+ "normalized_hypothesis": "something about that moment brookes something open in me that i hadn't realized was close",
+ "errors": 2,
+ "words": 15,
+ "wer": 0.13333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "something about that moment broke something open in me that i had not realized was closed",
+ "normalized_hypothesis": "something about that moment brookes something open in me that i had not realized was close",
+ "errors": 2,
+ "words": 16,
+ "wer": 0.125
+ }
+ },
+ {
+ "id": "release500_0112",
+ "category": "long",
+ "text": "The thing about making something good is that you don't usually know it's good while you're making it. You just know you're not done yet.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0112.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE THING ABOUT MAKING SOMETHING GOOD IS THAT YOU DON'T USUALLY KNOW IT'S GOOD WHINE YOU'RE MAKING IT YOU JUST KNOW YOU'RE NOT DONE YET",
+ "raw": {
+ "normalized_reference": "the thing about making something good is that you don't usually know it's good while you're making it. you just know you're not done yet.",
+ "normalized_hypothesis": "the thing about making something good is that you don't usually know it's good whine you're making it you just know you're not done yet",
+ "errors": 1,
+ "words": 25,
+ "wer": 0.04
+ },
+ "semantic": {
+ "normalized_reference": "the thing about making something good is that you do not usually know it is good while you are making it you just know you are not done yet",
+ "normalized_hypothesis": "the thing about making something good is that you do not usually know it is good whine you are making it you just know you are not done yet",
+ "errors": 1,
+ "words": 29,
+ "wer": 0.034482758620689655
+ }
+ },
+ {
+ "id": "release500_0113",
+ "category": "narrative",
+ "text": "She opened the door slowly, as if she expected to find something she wasn't ready for.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0113.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SHE OPENED THE DOOR SLOWLY AS IF SHE EXPECTED TO FIND SOMETHING SHE WASN'T READY FOR",
+ "raw": {
+ "normalized_reference": "she opened the door slowly, as if she expected to find something she wasn't ready for.",
+ "normalized_hypothesis": "she opened the door slowly as if she expected to find something she wasn't ready for",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "she opened the door slowly as if she expected to find something she was not ready for",
+ "normalized_hypothesis": "she opened the door slowly as if she expected to find something she was not ready for",
+ "errors": 0,
+ "words": 17,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0114",
+ "category": "question",
+ "text": "What would you do differently if you could go back?",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0114.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WHAT WOULD YOU DO DIFFERENTLY IF YOU COULD GO BACK",
+ "raw": {
+ "normalized_reference": "what would you do differently if you could go back?",
+ "normalized_hypothesis": "what would you do differently if you could go back",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "what would you do differently if you could go back",
+ "normalized_hypothesis": "what would you do differently if you could go back",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0115",
+ "category": "short",
+ "text": "Okay. Okay okay okay.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0115.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "OK O K O K OK",
+ "raw": {
+ "normalized_reference": "okay. okay okay okay.",
+ "normalized_hypothesis": "ok o k o k ok",
+ "errors": 6,
+ "words": 4,
+ "wer": 1.5
+ },
+ "semantic": {
+ "normalized_reference": "okay okay okay okay",
+ "normalized_hypothesis": "ok 0 k 0 k ok",
+ "errors": 6,
+ "words": 4,
+ "wer": 1.5
+ }
+ },
+ {
+ "id": "release500_0116",
+ "category": "technical",
+ "text": "The latency goes up significantly at scale, which is expected.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0116.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "ELATENCY GOES UP SIGNIFICANTLY AT SCALE WHICH IS EXPECTED",
+ "raw": {
+ "normalized_reference": "the latency goes up significantly at scale, which is expected.",
+ "normalized_hypothesis": "elatency goes up significantly at scale which is expected",
+ "errors": 2,
+ "words": 10,
+ "wer": 0.2
+ },
+ "semantic": {
+ "normalized_reference": "the latency goes up significantly at scale which is expected",
+ "normalized_hypothesis": "elatency goes up significantly at scale which is expected",
+ "errors": 2,
+ "words": 10,
+ "wer": 0.2
+ }
+ },
+ {
+ "id": "release500_0117",
+ "category": "conversational",
+ "text": "We ended up staying until two in the morning talking about nothing.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0117.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WE ENDED UP STAYING UNTIL TWO IN THE MORNING TALKING ABOUT NOTHING",
+ "raw": {
+ "normalized_reference": "we ended up staying until two in the morning talking about nothing.",
+ "normalized_hypothesis": "we ended up staying until two in the morning talking about nothing",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "we ended up staying until 2 in the morning talking about nothing",
+ "normalized_hypothesis": "we ended up staying until 2 in the morning talking about nothing",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0118",
+ "category": "descriptive",
+ "text": "The rain started slowly, and then all at once, the way it usually does.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0118.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE RAIN STARTED SLOWLY AND THEN ALL AT ONCE THE WAY IT USUALLY DOES",
+ "raw": {
+ "normalized_reference": "the rain started slowly, and then all at once, the way it usually does.",
+ "normalized_hypothesis": "the rain started slowly and then all at once the way it usually does",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the rain started slowly and then all at once the way it usually does",
+ "normalized_hypothesis": "the rain started slowly and then all at once the way it usually does",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0119",
+ "category": "emotional",
+ "text": "I honestly did not think I was going to make it through that year.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0119.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I HONESTLY DID NOT THINK I WAS GOING TO MAKE IT THROUGH THAT YEAR",
+ "raw": {
+ "normalized_reference": "i honestly did not think i was going to make it through that year.",
+ "normalized_hypothesis": "i honestly did not think i was going to make it through that year",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i honestly did not think i was going to make it through that year",
+ "normalized_hypothesis": "i honestly did not think i was going to make it through that year",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0120",
+ "category": "long",
+ "text": "There's an assumption that people who seem confident all the time actually feel confident, and in my experience that's almost never true.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0120.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HERE'S AN ASSUMPTION THAT PEOPLE WHO SEEM CONFIDENT ALL THE TIME ACTUALLY FEEL CONFIDENT AND IN MY EXPERIENCE THAT'S ALMOST NEVER TRUE",
+ "raw": {
+ "normalized_reference": "there's an assumption that people who seem confident all the time actually feel confident, and in my experience that's almost never true.",
+ "normalized_hypothesis": "here's an assumption that people who seem confident all the time actually feel confident and in my experience that's almost never true",
+ "errors": 1,
+ "words": 22,
+ "wer": 0.045454545454545456
+ },
+ "semantic": {
+ "normalized_reference": "there is an assumption that people who seem confident all the time actually feel confident and in my experience that is almost never true",
+ "normalized_hypothesis": "here is an assumption that people who seem confident all the time actually feel confident and in my experience that is almost never true",
+ "errors": 1,
+ "words": 24,
+ "wer": 0.041666666666666664
+ }
+ },
+ {
+ "id": "release500_0121",
+ "category": "narrative",
+ "text": "The moment I opened the door I knew something was different, even if I couldn't say what.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0121.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE MOMENT I OPENED THE DOOR I KNEW SOMETHING WAS DIFFERENT EVEN IF I COULDN'T SAY WHAT",
+ "raw": {
+ "normalized_reference": "the moment i opened the door i knew something was different, even if i couldn't say what.",
+ "normalized_hypothesis": "the moment i opened the door i knew something was different even if i couldn't say what",
+ "errors": 0,
+ "words": 17,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the moment i opened the door i knew something was different even if i could not say what",
+ "normalized_hypothesis": "the moment i opened the door i knew something was different even if i could not say what",
+ "errors": 0,
+ "words": 18,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0122",
+ "category": "question",
+ "text": "How do you explain something that doesn't make sense even to you?",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0122.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "DO YOU EXPLAIN SOMETHING THAT DOESN'T MAKE SENSE EVEN TO YOU",
+ "raw": {
+ "normalized_reference": "how do you explain something that doesn't make sense even to you?",
+ "normalized_hypothesis": "do you explain something that doesn't make sense even to you",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "how do you explain something that does not make sense even to you",
+ "normalized_hypothesis": "do you explain something that does not make sense even to you",
+ "errors": 1,
+ "words": 13,
+ "wer": 0.07692307692307693
+ }
+ },
+ {
+ "id": "release500_0123",
+ "category": "short",
+ "text": "Hold on a second.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0123.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HOLD ON A SECOND",
+ "raw": {
+ "normalized_reference": "hold on a second.",
+ "normalized_hypothesis": "hold on a second",
+ "errors": 0,
+ "words": 4,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "hold on a 2nd",
+ "normalized_hypothesis": "hold on a 2nd",
+ "errors": 0,
+ "words": 4,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0124",
+ "category": "technical",
+ "text": "A deterministic split prevents duplicate normalized text from leaking into evaluation.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0124.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "A DETERMINISTIC SPLIT PREVENTS DUPLICATE NORMALIZED TEXT FROM LEAKING INTO EVALUATION",
+ "raw": {
+ "normalized_reference": "a deterministic split prevents duplicate normalized text from leaking into evaluation.",
+ "normalized_hypothesis": "a deterministic split prevents duplicate normalized text from leaking into evaluation",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "a deterministic split prevents duplicate normalized text from leaking into evaluation",
+ "normalized_hypothesis": "a deterministic split prevents duplicate normalized text from leaking into evaluation",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0125",
+ "category": "conversational",
+ "text": "Honestly, that was much better than I expected, but the ending still felt rushed.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0125.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HONESTLY THAT WAS MUCH BETTER THAN I EXPECTED BUT THE ENDING STILL FELT RUSHED",
+ "raw": {
+ "normalized_reference": "honestly, that was much better than i expected, but the ending still felt rushed.",
+ "normalized_hypothesis": "honestly that was much better than i expected but the ending still felt rushed",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "honestly that was much better than i expected but the ending still felt rushed",
+ "normalized_hypothesis": "honestly that was much better than i expected but the ending still felt rushed",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0126",
+ "category": "descriptive",
+ "text": "She had a habit of tilting her head slightly when thinking hard about something.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0126.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SHE HAD A HABIT OF TILTING HER HEAD SLIGHTLY WHEN THINKING HARD ABOUT SOMETHING",
+ "raw": {
+ "normalized_reference": "she had a habit of tilting her head slightly when thinking hard about something.",
+ "normalized_hypothesis": "she had a habit of tilting her head slightly when thinking hard about something",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "she had a habit of tilting her head slightly when thinking hard about something",
+ "normalized_hypothesis": "she had a habit of tilting her head slightly when thinking hard about something",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0127",
+ "category": "emotional",
+ "text": "I cannot stress enough how much I did not want to be right about this.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0127.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I CANNOT STRESS ENOUGH HOW MUCH I DID NOT WANT TO BE RIGHT ABOUT THIS",
+ "raw": {
+ "normalized_reference": "i cannot stress enough how much i did not want to be right about this.",
+ "normalized_hypothesis": "i cannot stress enough how much i did not want to be right about this",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i cannot stress enough how much i did not want to be right about this",
+ "normalized_hypothesis": "i cannot stress enough how much i did not want to be right about this",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0128",
+ "category": "long",
+ "text": "I used to think the goal was to get to a point where nothing was hard anymore, and now I think that point doesn't exist and I'm not sure I'd want it to.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0128.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I USED TO THINK THE GOAL WAS TO GET TO A POINT WHERE NOTHING WAS HARD ANY MORE AND NOW I THINK THAT POINT DOESN'T EXIST AND I'M NOT SURE I'D WANT IT TO",
+ "raw": {
+ "normalized_reference": "i used to think the goal was to get to a point where nothing was hard anymore, and now i think that point doesn't exist and i'm not sure i'd want it to.",
+ "normalized_hypothesis": "i used to think the goal was to get to a point where nothing was hard any more and now i think that point doesn't exist and i'm not sure i'd want it to",
+ "errors": 2,
+ "words": 33,
+ "wer": 0.06060606060606061
+ },
+ "semantic": {
+ "normalized_reference": "i used to think the goal was to get to a point where nothing was hard anymore and now i think that point does not exist and i am not sure i would want it to",
+ "normalized_hypothesis": "i used to think the goal was to get to a point where nothing was hard any more and now i think that point does not exist and i am not sure i would want it to",
+ "errors": 2,
+ "words": 36,
+ "wer": 0.05555555555555555
+ }
+ },
+ {
+ "id": "release500_0129",
+ "category": "narrative",
+ "text": "So this is the part of the story where everything goes completely sideways.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0129.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SO THIS IS THE PART OF THE STORY WHERE EVERYTHING GOES COMPLETELY SIDEWAYS",
+ "raw": {
+ "normalized_reference": "so this is the part of the story where everything goes completely sideways.",
+ "normalized_hypothesis": "so this is the part of the story where everything goes completely sideways",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "so this is the part of the story where everything goes completely sideways",
+ "normalized_hypothesis": "so this is the part of the story where everything goes completely sideways",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0130",
+ "category": "question",
+ "text": "Is there a version of this where it actually works out okay?",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0130.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IS THERE A VERSION OF THIS WHERE IT ACTUALLY WORKS OUT BOUQUET",
+ "raw": {
+ "normalized_reference": "is there a version of this where it actually works out okay?",
+ "normalized_hypothesis": "is there a version of this where it actually works out bouquet",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "is there a version of this where it actually works out okay",
+ "normalized_hypothesis": "is there a version of this where it actually works out bouquet",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ }
+ },
+ {
+ "id": "release500_0131",
+ "category": "short",
+ "text": "Oh no. Oh no no no.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0131.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "OH NO OH NO NO NO",
+ "raw": {
+ "normalized_reference": "oh no. oh no no no.",
+ "normalized_hypothesis": "oh no oh no no no",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "0 no 0 no no no",
+ "normalized_hypothesis": "0 no 0 no no no",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0132",
+ "category": "technical",
+ "text": "Always test on a copy before modifying the original.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0132.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "ALWAYS TESTS ON A COPY BEFORE MODIFYING THE ORIGINAL",
+ "raw": {
+ "normalized_reference": "always test on a copy before modifying the original.",
+ "normalized_hypothesis": "always tests on a copy before modifying the original",
+ "errors": 1,
+ "words": 9,
+ "wer": 0.1111111111111111
+ },
+ "semantic": {
+ "normalized_reference": "always test on a copy before modifying the original",
+ "normalized_hypothesis": "always tests on a copy before modifying the original",
+ "errors": 1,
+ "words": 9,
+ "wer": 0.1111111111111111
+ }
+ },
+ {
+ "id": "release500_0133",
+ "category": "conversational",
+ "text": "I wasn't even supposed to be there. That's the wild part.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0133.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I WASN'T EVEN SUPPOSED TO BE THERE THAT'S THE WILD PART",
+ "raw": {
+ "normalized_reference": "i wasn't even supposed to be there. that's the wild part.",
+ "normalized_hypothesis": "i wasn't even supposed to be there that's the wild part",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i was not even supposed to be there that is the wild part",
+ "normalized_hypothesis": "i was not even supposed to be there that is the wild part",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0134",
+ "category": "descriptive",
+ "text": "It's a small town. Not the kind you leave. The kind you leave and come back to.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0134.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT'S A SMALL TOWN NOT THE KIND YOU LEAVE THE KIND YOU LEAVE AND COME BACK TO",
+ "raw": {
+ "normalized_reference": "it's a small town. not the kind you leave. the kind you leave and come back to.",
+ "normalized_hypothesis": "it's a small town not the kind you leave the kind you leave and come back to",
+ "errors": 0,
+ "words": 17,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it is a small town not the kind you leave the kind you leave and come back to",
+ "normalized_hypothesis": "it is a small town not the kind you leave the kind you leave and come back to",
+ "errors": 0,
+ "words": 18,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0135",
+ "category": "emotional",
+ "text": "She cried. I cried. We all just sort of stood there crying.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0135.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SHE CRIED I CRIE WE ALL JUST SORT OF STOOD THERE CRYING",
+ "raw": {
+ "normalized_reference": "she cried. i cried. we all just sort of stood there crying.",
+ "normalized_hypothesis": "she cried i crie we all just sort of stood there crying",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "she cried i cried we all just sort of stood there crying",
+ "normalized_hypothesis": "she cried i crie we all just sort of stood there crying",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ }
+ },
+ {
+ "id": "release500_0136",
+ "category": "long",
+ "text": "There's a version of being busy that's actually just a very elaborate way of avoiding the one thing you know you should be doing.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0136.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THERE'S A VERSION OF BEING BUSY THAT'S ACTUALLY JUST A VERY ELABORATE WAY OF AVOIDING THE ONE THING YOU KNOW YOU SHOULD BE DOIN",
+ "raw": {
+ "normalized_reference": "there's a version of being busy that's actually just a very elaborate way of avoiding the one thing you know you should be doing.",
+ "normalized_hypothesis": "there's a version of being busy that's actually just a very elaborate way of avoiding the one thing you know you should be doin",
+ "errors": 1,
+ "words": 24,
+ "wer": 0.041666666666666664
+ },
+ "semantic": {
+ "normalized_reference": "there is a version of being busy that is actually just a very elaborate way of avoiding the one thing you know you should be doing",
+ "normalized_hypothesis": "there is a version of being busy that is actually just a very elaborate way of avoiding the one thing you know you should be doin",
+ "errors": 1,
+ "words": 26,
+ "wer": 0.038461538461538464
+ }
+ },
+ {
+ "id": "release500_0137",
+ "category": "narrative",
+ "text": "The letter arrived on a Tuesday, which felt appropriate somehow.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0137.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE LETTER ARRIVED ON A TUESDAY WHICH FELT APPROPRIATE SOMEHOW",
+ "raw": {
+ "normalized_reference": "the letter arrived on a tuesday, which felt appropriate somehow.",
+ "normalized_hypothesis": "the letter arrived on a tuesday which felt appropriate somehow",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the letter arrived on a tuesday which felt appropriate somehow",
+ "normalized_hypothesis": "the letter arrived on a tuesday which felt appropriate somehow",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0138",
+ "category": "question",
+ "text": "What exactly did she say when you told her?",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0138.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WHAT EXACTLY DID SHE SAY WHEN YOU TOLD HER",
+ "raw": {
+ "normalized_reference": "what exactly did she say when you told her?",
+ "normalized_hypothesis": "what exactly did she say when you told her",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "what exactly did she say when you told her",
+ "normalized_hypothesis": "what exactly did she say when you told her",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0139",
+ "category": "short",
+ "text": "No way. Absolutely not.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0139.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "NO WAY ABSOLUTELY NOT",
+ "raw": {
+ "normalized_reference": "no way. absolutely not.",
+ "normalized_hypothesis": "no way absolutely not",
+ "errors": 0,
+ "words": 4,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "no way absolutely not",
+ "normalized_hypothesis": "no way absolutely not",
+ "errors": 0,
+ "words": 4,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0140",
+ "category": "technical",
+ "text": "The first thing you want to do is make sure everything is backed up.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0140.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE FIRST THING YOU WANT TO DO IS MAKE SURE EVERYTHING IS BACKED UP",
+ "raw": {
+ "normalized_reference": "the first thing you want to do is make sure everything is backed up.",
+ "normalized_hypothesis": "the first thing you want to do is make sure everything is backed up",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the 1st thing you want to do is make sure everything is backed up",
+ "normalized_hypothesis": "the 1st thing you want to do is make sure everything is backed up",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0141",
+ "category": "conversational",
+ "text": "I have the most insane idea and you need to just hear me out.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0141.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I HAVE THE MOST INSANE IDEA AND YOU NEED TO JUST HEAR ME OUT",
+ "raw": {
+ "normalized_reference": "i have the most insane idea and you need to just hear me out.",
+ "normalized_hypothesis": "i have the most insane idea and you need to just hear me out",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i have the most insane idea and you need to just hear me out",
+ "normalized_hypothesis": "i have the most insane idea and you need to just hear me out",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0142",
+ "category": "descriptive",
+ "text": "There's a coffee shop on the corner that's been there longer than anyone can remember.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0142.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HERE'S A COFFEE SHOP ON THE CORNER THAT'S BEEN THERE LONGER THAN ANYONE CAN REMEMBER",
+ "raw": {
+ "normalized_reference": "there's a coffee shop on the corner that's been there longer than anyone can remember.",
+ "normalized_hypothesis": "here's a coffee shop on the corner that's been there longer than anyone can remember",
+ "errors": 1,
+ "words": 15,
+ "wer": 0.06666666666666667
+ },
+ "semantic": {
+ "normalized_reference": "there is a coffee shop on the corner that has been there longer than anyone can remember",
+ "normalized_hypothesis": "here is a coffee shop on the corner that has been there longer than anyone can remember",
+ "errors": 1,
+ "words": 17,
+ "wer": 0.058823529411764705
+ }
+ },
+ {
+ "id": "release500_0143",
+ "category": "emotional",
+ "text": "I kept it together until I got to the car, and then I completely fell apart.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0143.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I KEPT IT TOGETHER UNTIL I GOT TO THE CAR AND THEN I COMPLETELY FELL APART",
+ "raw": {
+ "normalized_reference": "i kept it together until i got to the car, and then i completely fell apart.",
+ "normalized_hypothesis": "i kept it together until i got to the car and then i completely fell apart",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i kept it together until i got to the car and then i completely fell apart",
+ "normalized_hypothesis": "i kept it together until i got to the car and then i completely fell apart",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0144",
+ "category": "long",
+ "text": "The older I get, the more I think that most of what we call wisdom is just pattern recognition built from having been wrong enough times to start noticing.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0144.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE OLDER I GET THE MORE I THINK THAT MOST OF WHAT WE CALL WISDOM IS JUST PATERN RECOGNITION BUILT FROM HAVING BEEN WRONG ENOUGH TIMES TO START NOTICING",
+ "raw": {
+ "normalized_reference": "the older i get, the more i think that most of what we call wisdom is just pattern recognition built from having been wrong enough times to start noticing.",
+ "normalized_hypothesis": "the older i get the more i think that most of what we call wisdom is just patern recognition built from having been wrong enough times to start noticing",
+ "errors": 1,
+ "words": 29,
+ "wer": 0.034482758620689655
+ },
+ "semantic": {
+ "normalized_reference": "the older i get the more i think that most of what we call wisdom is just pattern recognition built from having been wrong enough times to start noticing",
+ "normalized_hypothesis": "the older i get the more i think that most of what we call wisdom is just patern recognition built from having been wrong enough times to start noticing",
+ "errors": 1,
+ "words": 29,
+ "wer": 0.034482758620689655
+ }
+ },
+ {
+ "id": "release500_0145",
+ "category": "narrative",
+ "text": "I remember the exact moment I realized I had made a huge mistake.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0145.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I REMEMBERED THE EXACT MOMENT I REALIZED I HAD MADE A HUGE MISTAKE",
+ "raw": {
+ "normalized_reference": "i remember the exact moment i realized i had made a huge mistake.",
+ "normalized_hypothesis": "i remembered the exact moment i realized i had made a huge mistake",
+ "errors": 1,
+ "words": 13,
+ "wer": 0.07692307692307693
+ },
+ "semantic": {
+ "normalized_reference": "i remember the exact moment i realized i had made a huge mistake",
+ "normalized_hypothesis": "i remembered the exact moment i realized i had made a huge mistake",
+ "errors": 1,
+ "words": 13,
+ "wer": 0.07692307692307693
+ }
+ },
+ {
+ "id": "release500_0146",
+ "category": "question",
+ "text": "Do you think it would have gone differently if we'd left earlier?",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0146.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "DO YOU THINK IT WOULD HAVE GONE DIFFERENTLY IF WE'D LEFT EARLIER",
+ "raw": {
+ "normalized_reference": "do you think it would have gone differently if we'd left earlier?",
+ "normalized_hypothesis": "do you think it would have gone differently if we'd left earlier",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "do you think it would have gone differently if we would left earlier",
+ "normalized_hypothesis": "do you think it would have gone differently if we would left earlier",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0147",
+ "category": "short",
+ "text": "You have to be kidding me.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0147.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "YOU HAVE TO BE KITTING ME",
+ "raw": {
+ "normalized_reference": "you have to be kidding me.",
+ "normalized_hypothesis": "you have to be kitting me",
+ "errors": 1,
+ "words": 6,
+ "wer": 0.16666666666666666
+ },
+ "semantic": {
+ "normalized_reference": "you have to be kidding me",
+ "normalized_hypothesis": "you have to be kitting me",
+ "errors": 1,
+ "words": 6,
+ "wer": 0.16666666666666666
+ }
+ },
+ {
+ "id": "release500_0148",
+ "category": "technical",
+ "text": "If the checksum fails, restart the worker process without deleting the cached tensors.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0148.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IF THE CHEXUM FAILS RE START THE WORKER PROCESS WITHOUT DELEADING THE CASH TENSERS",
+ "raw": {
+ "normalized_reference": "if the checksum fails, restart the worker process without deleting the cached tensors.",
+ "normalized_hypothesis": "if the chexum fails re start the worker process without deleading the cash tensers",
+ "errors": 6,
+ "words": 13,
+ "wer": 0.46153846153846156
+ },
+ "semantic": {
+ "normalized_reference": "if the checksum fails restart the worker process without deleting the cached tensors",
+ "normalized_hypothesis": "if the chexum fails re start the worker process without deleading the cash tensers",
+ "errors": 6,
+ "words": 13,
+ "wer": 0.46153846153846156
+ }
+ },
+ {
+ "id": "release500_0149",
+ "category": "conversational",
+ "text": "I need you to understand how close that was to not working.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0149.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I NEED YOU TO UNDERSTAND HOW CLOSE THAT WAS TO NOT WORKING",
+ "raw": {
+ "normalized_reference": "i need you to understand how close that was to not working.",
+ "normalized_hypothesis": "i need you to understand how close that was to not working",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i need you to understand how close that was to not working",
+ "normalized_hypothesis": "i need you to understand how close that was to not working",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0150",
+ "category": "descriptive",
+ "text": "It was the sort of morning where the whole world looks rinsed clean.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0150.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT WAS THE SORT OF MORNING WHERE THE WHOLE WORLD LOOKS RENCED CLEAN",
+ "raw": {
+ "normalized_reference": "it was the sort of morning where the whole world looks rinsed clean.",
+ "normalized_hypothesis": "it was the sort of morning where the whole world looks renced clean",
+ "errors": 1,
+ "words": 13,
+ "wer": 0.07692307692307693
+ },
+ "semantic": {
+ "normalized_reference": "it was the sort of morning where the whole world looks rinsed clean",
+ "normalized_hypothesis": "it was the sort of morning where the whole world looks renced clean",
+ "errors": 1,
+ "words": 13,
+ "wer": 0.07692307692307693
+ }
+ },
+ {
+ "id": "release500_0151",
+ "category": "emotional",
+ "text": "She started laughing and I started laughing and we couldn't stop for a full minute.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0151.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SHE STARTED LAUGHING AND I STARTED LAUGHING AND WE COULDN'T STOP FOR A FULL MINUTE",
+ "raw": {
+ "normalized_reference": "she started laughing and i started laughing and we couldn't stop for a full minute.",
+ "normalized_hypothesis": "she started laughing and i started laughing and we couldn't stop for a full minute",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "she started laughing and i started laughing and we could not stop for a full minute",
+ "normalized_hypothesis": "she started laughing and i started laughing and we could not stop for a full minute",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0152",
+ "category": "long",
+ "text": "We spent three years building something we were genuinely proud of, and then the market shifted and none of it mattered the way we thought it would.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0152.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WE SPENT THREE YEARS BUILDING SOMETHING WE WERE GENUINELY PROUD OF AND THEN THE MARKET SHIFTED AND NONE OF IT MATTERED THE WAY WE THOUGHT IT WERE",
+ "raw": {
+ "normalized_reference": "we spent three years building something we were genuinely proud of, and then the market shifted and none of it mattered the way we thought it would.",
+ "normalized_hypothesis": "we spent three years building something we were genuinely proud of and then the market shifted and none of it mattered the way we thought it were",
+ "errors": 1,
+ "words": 27,
+ "wer": 0.037037037037037035
+ },
+ "semantic": {
+ "normalized_reference": "we spent 3 years building something we were genuinely proud of and then the market shifted and none of it mattered the way we thought it would",
+ "normalized_hypothesis": "we spent 3 years building something we were genuinely proud of and then the market shifted and none of it mattered the way we thought it were",
+ "errors": 1,
+ "words": 27,
+ "wer": 0.037037037037037035
+ }
+ },
+ {
+ "id": "release500_0153",
+ "category": "narrative",
+ "text": "And somehow, despite all of that, it actually turned out okay.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0153.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "AND SOMEHOW DESPITE ALL OF THAT IT ACTUALLY TURNED OUT O K",
+ "raw": {
+ "normalized_reference": "and somehow, despite all of that, it actually turned out okay.",
+ "normalized_hypothesis": "and somehow despite all of that it actually turned out o k",
+ "errors": 2,
+ "words": 11,
+ "wer": 0.18181818181818182
+ },
+ "semantic": {
+ "normalized_reference": "and somehow despite all of that it actually turned out okay",
+ "normalized_hypothesis": "and somehow despite all of that it actually turned out 0 k",
+ "errors": 2,
+ "words": 11,
+ "wer": 0.18181818181818182
+ }
+ },
+ {
+ "id": "release500_0154",
+ "category": "question",
+ "text": "Is that something you actually want, or something you think you should want?",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0154.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IS THAT SOMETHING YOU ACTUALLY WANT OR SOMETHING YOU THINK YOU SHOULD WANT",
+ "raw": {
+ "normalized_reference": "is that something you actually want, or something you think you should want?",
+ "normalized_hypothesis": "is that something you actually want or something you think you should want",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "is that something you actually want or something you think you should want",
+ "normalized_hypothesis": "is that something you actually want or something you think you should want",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0155",
+ "category": "short",
+ "text": "This is fine.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0155.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THIS IS FINE",
+ "raw": {
+ "normalized_reference": "this is fine.",
+ "normalized_hypothesis": "this is fine",
+ "errors": 0,
+ "words": 3,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "this is fine",
+ "normalized_hypothesis": "this is fine",
+ "errors": 0,
+ "words": 3,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0156",
+ "category": "technical",
+ "text": "The tricky part isn't the setup, it's making sure all dependencies line up.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0156.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE TRICKY PART ISN'T THE SET UP IT'S MAKING SURE ALL DEPENDENCIES LINE UP",
+ "raw": {
+ "normalized_reference": "the tricky part isn't the setup, it's making sure all dependencies line up.",
+ "normalized_hypothesis": "he tricky part isn't the set up it's making sure all dependencies line up",
+ "errors": 3,
+ "words": 13,
+ "wer": 0.23076923076923078
+ },
+ "semantic": {
+ "normalized_reference": "the tricky part is not the setup it is making sure all dependencies line up",
+ "normalized_hypothesis": "he tricky part is not the set up it is making sure all dependencies line up",
+ "errors": 3,
+ "words": 15,
+ "wer": 0.2
+ }
+ },
+ {
+ "id": "release500_0157",
+ "category": "conversational",
+ "text": "He called it a minor setback. I called it a disaster.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0157.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE CALLED IT A MINOR SETBACK I CALLED IT A DISASTER",
+ "raw": {
+ "normalized_reference": "he called it a minor setback. i called it a disaster.",
+ "normalized_hypothesis": "he called it a minor setback i called it a disaster",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he called it a minor setback i called it a disaster",
+ "normalized_hypothesis": "he called it a minor setback i called it a disaster",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0158",
+ "category": "descriptive",
+ "text": "The whole place smelled like pine and old wood and cold air.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0158.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE WHOLE PLACE SMELLED LIKE PINE AND OLD WOOD AND COLD AIR",
+ "raw": {
+ "normalized_reference": "the whole place smelled like pine and old wood and cold air.",
+ "normalized_hypothesis": "the whole place smelled like pine and old wood and cold air",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the whole place smelled like pine and old wood and cold air",
+ "normalized_hypothesis": "the whole place smelled like pine and old wood and cold air",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0159",
+ "category": "emotional",
+ "text": "I've never been so relieved in my entire life.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0159.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I'VE NEVER BEEN SO RELIEVED IN MY ENTIRE LIFE",
+ "raw": {
+ "normalized_reference": "i've never been so relieved in my entire life.",
+ "normalized_hypothesis": "i've never been so relieved in my entire life",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i have never been so relieved in my entire life",
+ "normalized_hypothesis": "i have never been so relieved in my entire life",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0160",
+ "category": "narrative",
+ "text": "I wrote it all down that night so I wouldn't forget, and I still haven't looked at it.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0160.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I WRODE IT ALL DOWN THAT NIGHT SO I WOULDN'T FORGET AND I STILL HAVEN'T LOOKED AT",
+ "raw": {
+ "normalized_reference": "i wrote it all down that night so i wouldn't forget, and i still haven't looked at it.",
+ "normalized_hypothesis": "i wrode it all down that night so i wouldn't forget and i still haven't looked at",
+ "errors": 2,
+ "words": 18,
+ "wer": 0.1111111111111111
+ },
+ "semantic": {
+ "normalized_reference": "i wrote it all down that night so i would not forget and i still have not looked at it",
+ "normalized_hypothesis": "i wrode it all down that night so i would not forget and i still have not looked at",
+ "errors": 2,
+ "words": 20,
+ "wer": 0.1
+ }
+ },
+ {
+ "id": "release500_0161",
+ "category": "short",
+ "text": "That's kind of wild actually.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0161.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "TS KIND OF WILD ACTUALLY",
+ "raw": {
+ "normalized_reference": "that's kind of wild actually.",
+ "normalized_hypothesis": "ts kind of wild actually",
+ "errors": 1,
+ "words": 5,
+ "wer": 0.2
+ },
+ "semantic": {
+ "normalized_reference": "that is kind of wild actually",
+ "normalized_hypothesis": "ts kind of wild actually",
+ "errors": 2,
+ "words": 6,
+ "wer": 0.3333333333333333
+ }
+ },
+ {
+ "id": "release500_0162",
+ "category": "technical",
+ "text": "Version pinning will save you a lot of pain later down the line.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0162.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "VERSION PINNING WILL SAVE YOU A LOT OF PAIN LATER DOWN THE LINE",
+ "raw": {
+ "normalized_reference": "version pinning will save you a lot of pain later down the line.",
+ "normalized_hypothesis": "version pinning will save you a lot of pain later down the line",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "version pinning will save you a lot of pain later down the line",
+ "normalized_hypothesis": "version pinning will save you a lot of pain later down the line",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0163",
+ "category": "conversational",
+ "text": "At some point you just have to commit and see what happens.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0163.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "AT SOME POINT YOU JUST HAVE TO COMMIT AND SEE WHAT HAPPENS",
+ "raw": {
+ "normalized_reference": "at some point you just have to commit and see what happens.",
+ "normalized_hypothesis": "at some point you just have to commit and see what happens",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "at some point you just have to commit and see what happens",
+ "normalized_hypothesis": "at some point you just have to commit and see what happens",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0164",
+ "category": "emotional",
+ "text": "Nobody told me it was going to be this hard, and honestly, I'm grateful.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0164.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "NOBODY TOLD ME IT WAS GOING TO BE THIS HARD AND HONESTLY I'M GRATEFUL",
+ "raw": {
+ "normalized_reference": "nobody told me it was going to be this hard, and honestly, i'm grateful.",
+ "normalized_hypothesis": "nobody told me it was going to be this hard and honestly i'm grateful",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "nobody told me it was going to be this hard and honestly i am grateful",
+ "normalized_hypothesis": "nobody told me it was going to be this hard and honestly i am grateful",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0165",
+ "category": "narrative",
+ "text": "Chapter seven. In which everything that could go wrong did, and a few things that couldn't.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0165.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "CHAPTER SEVEN IN WHICH EVERYTHING THAT COULD GO WRONG DID AND A FEW THINGS THAT COULDN'T",
+ "raw": {
+ "normalized_reference": "chapter seven. in which everything that could go wrong did, and a few things that couldn't.",
+ "normalized_hypothesis": "chapter seven in which everything that could go wrong did and a few things that couldn't",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "chapter 7 in which everything that could go wrong did and a few things that could not",
+ "normalized_hypothesis": "chapter 7 in which everything that could go wrong did and a few things that could not",
+ "errors": 0,
+ "words": 17,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0166",
+ "category": "short",
+ "text": "Wait, what?",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0166.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WAKE WHAT",
+ "raw": {
+ "normalized_reference": "wait, what?",
+ "normalized_hypothesis": "wake what",
+ "errors": 1,
+ "words": 2,
+ "wer": 0.5
+ },
+ "semantic": {
+ "normalized_reference": "wait what",
+ "normalized_hypothesis": "wake what",
+ "errors": 1,
+ "words": 2,
+ "wer": 0.5
+ }
+ },
+ {
+ "id": "release500_0167",
+ "category": "technical",
+ "text": "The graphics processor remained idle because the data loader exhausted every central processing core.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0167.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE GRAFHICS PROCESS REMAINED IDLE BECAUSE THE DATA LODER EXHAUSTED EVERY CENTRAL PROCESS IN CORE",
+ "raw": {
+ "normalized_reference": "the graphics processor remained idle because the data loader exhausted every central processing core.",
+ "normalized_hypothesis": "the grafhics process remained idle because the data loder exhausted every central process in core",
+ "errors": 5,
+ "words": 14,
+ "wer": 0.35714285714285715
+ },
+ "semantic": {
+ "normalized_reference": "the graphics processor remained idle because the data loader exhausted every central processing core",
+ "normalized_hypothesis": "the grafhics process remained idle because the data loder exhausted every central process in core",
+ "errors": 5,
+ "words": 14,
+ "wer": 0.35714285714285715
+ }
+ },
+ {
+ "id": "release500_0168",
+ "category": "conversational",
+ "text": "Could you speak a little more slowly? I am trying to write every detail down.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0168.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "COULD YOU SPEAK A LITTLE MORE SLOWLY I AM TRYING TO WRITE EVERY DETAIL DOWN",
+ "raw": {
+ "normalized_reference": "could you speak a little more slowly? i am trying to write every detail down.",
+ "normalized_hypothesis": "could you speak a little more slowly i am trying to write every detail down",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "could you speak a little more slowly i am trying to write every detail down",
+ "normalized_hypothesis": "could you speak a little more slowly i am trying to write every detail down",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0169",
+ "category": "emotional",
+ "text": "I was so excited I could barely form a complete sentence.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0169.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WAS SO EXCITED I COULD BARELY FORM A COMPLETE SENTENCE",
+ "raw": {
+ "normalized_reference": "i was so excited i could barely form a complete sentence.",
+ "normalized_hypothesis": "was so excited i could barely form a complete sentence",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ },
+ "semantic": {
+ "normalized_reference": "i was so excited i could barely form a complete sentence",
+ "normalized_hypothesis": "was so excited i could barely form a complete sentence",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ }
+ },
+ {
+ "id": "release500_0170",
+ "category": "narrative",
+ "text": "He looked at me like I had said something in a language he didn't speak.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0170.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE LOOKED AT ME LIKE I HAD SAID SOMETHING IN A LANGUAGE HE DIDN'T SPEAK",
+ "raw": {
+ "normalized_reference": "he looked at me like i had said something in a language he didn't speak.",
+ "normalized_hypothesis": "he looked at me like i had said something in a language he didn't speak",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he looked at me like i had said something in a language he did not speak",
+ "normalized_hypothesis": "he looked at me like i had said something in a language he did not speak",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0171",
+ "category": "short",
+ "text": "That's actually hilarious.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0171.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THAT'S ACTUALLY HILARIOUS",
+ "raw": {
+ "normalized_reference": "that's actually hilarious.",
+ "normalized_hypothesis": "that's actually hilarious",
+ "errors": 0,
+ "words": 3,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "that is actually hilarious",
+ "normalized_hypothesis": "that is actually hilarious",
+ "errors": 0,
+ "words": 4,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0172",
+ "category": "technical",
+ "text": "The encoder maps phonemes, stress, and phrase boundaries into a shared hidden representation.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0172.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE INCOTER MAPS PHONIUMS STRESS AND PHRASE BOUNDARIES INTO A SHARED HIDIN REPRESENTATION",
+ "raw": {
+ "normalized_reference": "the encoder maps phonemes, stress, and phrase boundaries into a shared hidden representation.",
+ "normalized_hypothesis": "the incoter maps phoniums stress and phrase boundaries into a shared hidin representation",
+ "errors": 3,
+ "words": 13,
+ "wer": 0.23076923076923078
+ },
+ "semantic": {
+ "normalized_reference": "the encoder maps phonemes stress and phrase boundaries into a shared hidden representation",
+ "normalized_hypothesis": "the incoter maps phoniums stress and phrase boundaries into a shared hidin representation",
+ "errors": 3,
+ "words": 13,
+ "wer": 0.23076923076923078
+ }
+ },
+ {
+ "id": "release500_0173",
+ "category": "conversational",
+ "text": "She showed up three hours late and acted like nothing had happened.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0173.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SHE SHOWED UP THREE HOURS LATE AND ACTED LIKE NOTHING HAD HAPPENED",
+ "raw": {
+ "normalized_reference": "she showed up three hours late and acted like nothing had happened.",
+ "normalized_hypothesis": "she showed up three hours late and acted like nothing had happened",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "she showed up 3 hours late and acted like nothing had happened",
+ "normalized_hypothesis": "she showed up 3 hours late and acted like nothing had happened",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0174",
+ "category": "emotional",
+ "text": "Oh my god. This is the best thing I've ever tasted.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0174.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "O MY GOD THIS IS THE BEST THING IHAVE EVER TASTED",
+ "raw": {
+ "normalized_reference": "oh my god. this is the best thing i've ever tasted.",
+ "normalized_hypothesis": "o my god this is the best thing ihave ever tasted",
+ "errors": 2,
+ "words": 11,
+ "wer": 0.18181818181818182
+ },
+ "semantic": {
+ "normalized_reference": "0 my god this is the best thing i have ever tasted",
+ "normalized_hypothesis": "0 my god this is the best thing ihave ever tasted",
+ "errors": 2,
+ "words": 12,
+ "wer": 0.16666666666666666
+ }
+ },
+ {
+ "id": "release500_0175",
+ "category": "narrative",
+ "text": "The note was short, which was somehow more frightening than a long one would have been.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0175.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE NOTE WAS SHORT WHICH WAS SOMEHOW MORE FRIGHTENING THAN A LONG ONE WOULD HAVE BEEN",
+ "raw": {
+ "normalized_reference": "the note was short, which was somehow more frightening than a long one would have been.",
+ "normalized_hypothesis": "the note was short which was somehow more frightening than a long one would have been",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the note was short which was somehow more frightening than a long one would have been",
+ "normalized_hypothesis": "the note was short which was somehow more frightening than a long one would have been",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0176",
+ "category": "short",
+ "text": "Noted.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0176.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "NODDED",
+ "raw": {
+ "normalized_reference": "noted.",
+ "normalized_hypothesis": "nodded",
+ "errors": 1,
+ "words": 1,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "noted",
+ "normalized_hypothesis": "nodded",
+ "errors": 1,
+ "words": 1,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0177",
+ "category": "technical",
+ "text": "It's a simple process once you understand the underlying structure.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0177.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT'S A SIMPLE PROCESS ONCE YOU UNDERSTAND THE UNDERLYING STRUCTURE",
+ "raw": {
+ "normalized_reference": "it's a simple process once you understand the underlying structure.",
+ "normalized_hypothesis": "it's a simple process once you understand the underlying structure",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it is a simple process once you understand the underlying structure",
+ "normalized_hypothesis": "it is a simple process once you understand the underlying structure",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0178",
+ "category": "conversational",
+ "text": "You have to hear this story because it's almost too good to be true.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0178.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "YOU HAVE TO HEAR THIS STORY BECAUSE IT'S ALMOST TOO GOOD TO BE TRUE",
+ "raw": {
+ "normalized_reference": "you have to hear this story because it's almost too good to be true.",
+ "normalized_hypothesis": "you have to hear this story because it's almost too good to be true",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "you have to hear this story because it is almost too good to be true",
+ "normalized_hypothesis": "you have to hear this story because it is almost too good to be true",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0179",
+ "category": "emotional",
+ "text": "I have never felt more seen in my entire life than in that moment.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0179.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I HAVE NEVER FELT MORE SCEEN IN MY ENTIRE LIFE THAN IN THAT MOMENT",
+ "raw": {
+ "normalized_reference": "i have never felt more seen in my entire life than in that moment.",
+ "normalized_hypothesis": "i have never felt more sceen in my entire life than in that moment",
+ "errors": 1,
+ "words": 14,
+ "wer": 0.07142857142857142
+ },
+ "semantic": {
+ "normalized_reference": "i have never felt more seen in my entire life than in that moment",
+ "normalized_hypothesis": "i have never felt more sceen in my entire life than in that moment",
+ "errors": 1,
+ "words": 14,
+ "wer": 0.07142857142857142
+ }
+ },
+ {
+ "id": "release500_0180",
+ "category": "narrative",
+ "text": "She knew. She had always known. She just waited for the right moment.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0180.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SHE KNEW SHE HAD ALWAYS KNOWN SHE JUST WAITED FOR THE RIGHT MOMENT",
+ "raw": {
+ "normalized_reference": "she knew. she had always known. she just waited for the right moment.",
+ "normalized_hypothesis": "she knew she had always known she just waited for the right moment",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "she knew she had always known she just waited for the right moment",
+ "normalized_hypothesis": "she knew she had always known she just waited for the right moment",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0181",
+ "category": "short",
+ "text": "Yeah, no. Hard pass.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0181.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "YELED NO HARD PASS",
+ "raw": {
+ "normalized_reference": "yeah, no. hard pass.",
+ "normalized_hypothesis": "yeled no hard pass",
+ "errors": 1,
+ "words": 4,
+ "wer": 0.25
+ },
+ "semantic": {
+ "normalized_reference": "yeah no hard pass",
+ "normalized_hypothesis": "yeled no hard pass",
+ "errors": 1,
+ "words": 4,
+ "wer": 0.25
+ }
+ },
+ {
+ "id": "release500_0182",
+ "category": "technical",
+ "text": "Please verify the sample rate, hop length, and mel filter configuration before inference.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0182.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "PLEASE VERIFY THE SAMPLE RATE HOP LENGTH AND MELL FILTER CONFIGURATION BEFORE INFERENCE",
+ "raw": {
+ "normalized_reference": "please verify the sample rate, hop length, and mel filter configuration before inference.",
+ "normalized_hypothesis": "please verify the sample rate hop length and mell filter configuration before inference",
+ "errors": 1,
+ "words": 13,
+ "wer": 0.07692307692307693
+ },
+ "semantic": {
+ "normalized_reference": "please verify the sample rate hop length and mel filter configuration before inference",
+ "normalized_hypothesis": "please verify the sample rate hop length and mell filter configuration before inference",
+ "errors": 1,
+ "words": 13,
+ "wer": 0.07692307692307693
+ }
+ },
+ {
+ "id": "release500_0183",
+ "category": "conversational",
+ "text": "That's the worst idea I've ever heard and I'm completely in.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0183.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THAT'S THE WORST IDEA I'VE EVER HEARD AND I'M COMPLETELY IN",
+ "raw": {
+ "normalized_reference": "that's the worst idea i've ever heard and i'm completely in.",
+ "normalized_hypothesis": "that's the worst idea i've ever heard and i'm completely in",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "that is the worst idea i have ever heard and i am completely in",
+ "normalized_hypothesis": "that is the worst idea i have ever heard and i am completely in",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0184",
+ "category": "emotional",
+ "text": "I remember thinking: this is it. This is the thing I've been waiting for.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0184.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I REMEMBER THINKING THIS IS IT THIS IS THE THING I'VE BEEN WAITING FOR",
+ "raw": {
+ "normalized_reference": "i remember thinking: this is it. this is the thing i've been waiting for.",
+ "normalized_hypothesis": "i remember thinking this is it this is the thing i've been waiting for",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i remember thinking this is it this is the thing i have been waiting for",
+ "normalized_hypothesis": "i remember thinking this is it this is the thing i have been waiting for",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0185",
+ "category": "narrative",
+ "text": "We never talked about it again, which I think was the right call.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0185.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WE NEVER TALKED ABOUT IT AGAIN WHICH I THINK WAS THE RIGHT CALL",
+ "raw": {
+ "normalized_reference": "we never talked about it again, which i think was the right call.",
+ "normalized_hypothesis": "we never talked about it again which i think was the right call",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "we never talked about it again which i think was the right call",
+ "normalized_hypothesis": "we never talked about it again which i think was the right call",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0186",
+ "category": "short",
+ "text": "In hindsight, obvious.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0186.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IN HINDSIGHT OBVIOUS",
+ "raw": {
+ "normalized_reference": "in hindsight, obvious.",
+ "normalized_hypothesis": "in hindsight obvious",
+ "errors": 0,
+ "words": 3,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "in hindsight obvious",
+ "normalized_hypothesis": "in hindsight obvious",
+ "errors": 0,
+ "words": 3,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0187",
+ "category": "technical",
+ "text": "Once it's configured correctly, it basically runs itself.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0187.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "ONCE IT'S CONFIGURED CORRECTLY IT BASELY RUNS ITSELF",
+ "raw": {
+ "normalized_reference": "once it's configured correctly, it basically runs itself.",
+ "normalized_hypothesis": "once it's configured correctly it basely runs itself",
+ "errors": 1,
+ "words": 8,
+ "wer": 0.125
+ },
+ "semantic": {
+ "normalized_reference": "once it is configured correctly it basically runs itself",
+ "normalized_hypothesis": "once it is configured correctly it basely runs itself",
+ "errors": 1,
+ "words": 9,
+ "wer": 0.1111111111111111
+ }
+ },
+ {
+ "id": "release500_0188",
+ "category": "conversational",
+ "text": "You got engaged? When? How? Tell me everything right now.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0188.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "YOU GOT ENGAGED WE OH TELL ME EVERYTHING RIGHT NOW",
+ "raw": {
+ "normalized_reference": "you got engaged? when? how? tell me everything right now.",
+ "normalized_hypothesis": "you got engaged we oh tell me everything right now",
+ "errors": 2,
+ "words": 10,
+ "wer": 0.2
+ },
+ "semantic": {
+ "normalized_reference": "you got engaged when how tell me everything right now",
+ "normalized_hypothesis": "you got engaged we 0 tell me everything right now",
+ "errors": 2,
+ "words": 10,
+ "wer": 0.2
+ }
+ },
+ {
+ "id": "release500_0189",
+ "category": "emotional",
+ "text": "It's hard to explain how much that moment meant to me.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0189.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT'S HARD TO EXPLAIN HOW MUCH THAT MOMENT MEANT TO ME",
+ "raw": {
+ "normalized_reference": "it's hard to explain how much that moment meant to me.",
+ "normalized_hypothesis": "it's hard to explain how much that moment meant to me",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it is hard to explain how much that moment meant to me",
+ "normalized_hypothesis": "it is hard to explain how much that moment meant to me",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0190",
+ "category": "narrative",
+ "text": "He told the story three different times that night and somehow it got better each time.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0190.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HETOLD THE STORY THREE DIFFERENT TIMES THAT NIGHT AND SOMEHOW IT GOT BETTER EACH TIME",
+ "raw": {
+ "normalized_reference": "he told the story three different times that night and somehow it got better each time.",
+ "normalized_hypothesis": "hetold the story three different times that night and somehow it got better each time",
+ "errors": 2,
+ "words": 16,
+ "wer": 0.125
+ },
+ "semantic": {
+ "normalized_reference": "he told the story 3 different times that night and somehow it got better each time",
+ "normalized_hypothesis": "hetold the story 3 different times that night and somehow it got better each time",
+ "errors": 2,
+ "words": 16,
+ "wer": 0.125
+ }
+ },
+ {
+ "id": "release500_0191",
+ "category": "short",
+ "text": "Right. Moving on.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0191.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "RIVED MOVING ON",
+ "raw": {
+ "normalized_reference": "right. moving on.",
+ "normalized_hypothesis": "rived moving on",
+ "errors": 1,
+ "words": 3,
+ "wer": 0.3333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "right moving on",
+ "normalized_hypothesis": "rived moving on",
+ "errors": 1,
+ "words": 3,
+ "wer": 0.3333333333333333
+ }
+ },
+ {
+ "id": "release500_0192",
+ "category": "conversational",
+ "text": "No, I did not move your keys; check beside the small ceramic bowl.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0192.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "NO I DID NOT MOVE YOUR KEYS CHECKED ASIDE THE SMALL SERANNIC BOWL",
+ "raw": {
+ "normalized_reference": "no, i did not move your keys; check beside the small ceramic bowl.",
+ "normalized_hypothesis": "no i did not move your keys checked aside the small serannic bowl",
+ "errors": 3,
+ "words": 13,
+ "wer": 0.23076923076923078
+ },
+ "semantic": {
+ "normalized_reference": "no i did not move your keys check beside the small ceramic bowl",
+ "normalized_hypothesis": "no i did not move your keys checked aside the small serannic bowl",
+ "errors": 3,
+ "words": 13,
+ "wer": 0.23076923076923078
+ }
+ },
+ {
+ "id": "release500_0193",
+ "category": "emotional",
+ "text": "I was devastated. I'm not going to pretend otherwise.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0193.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I WAS DEVASTATED I'M NOT GOING TO PRETEND OTHERWISE",
+ "raw": {
+ "normalized_reference": "i was devastated. i'm not going to pretend otherwise.",
+ "normalized_hypothesis": "i was devastated i'm not going to pretend otherwise",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i was devastated i am not going to pretend otherwise",
+ "normalized_hypothesis": "i was devastated i am not going to pretend otherwise",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0194",
+ "category": "narrative",
+ "text": "That was the last summer before everything changed, though we didn't know it yet.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0194.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THAT WAS THE LAST SUMMER BEFORE EVERYTHING CHANGED THOUGH WE DIDN'T KNOW IT YET",
+ "raw": {
+ "normalized_reference": "that was the last summer before everything changed, though we didn't know it yet.",
+ "normalized_hypothesis": "that was the last summer before everything changed though we didn't know it yet",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "that was the last summer before everything changed though we did not know it yet",
+ "normalized_hypothesis": "that was the last summer before everything changed though we did not know it yet",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0195",
+ "category": "short",
+ "text": "Bold of you to assume.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0195.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "BOLD OF YOU TO ASSUME",
+ "raw": {
+ "normalized_reference": "bold of you to assume.",
+ "normalized_hypothesis": "bold of you to assume",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "bold of you to assume",
+ "normalized_hypothesis": "bold of you to assume",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0196",
+ "category": "conversational",
+ "text": "It's one of those decisions that makes total sense in the moment.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0196.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT'S ONE OF THOSE DECISIONS THAT MAKES TOTAL SENSE IN THE MOMENT",
+ "raw": {
+ "normalized_reference": "it's one of those decisions that makes total sense in the moment.",
+ "normalized_hypothesis": "it's one of those decisions that makes total sense in the moment",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it is one of those decisions that makes total sense in the moment",
+ "normalized_hypothesis": "it is one of those decisions that makes total sense in the moment",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0197",
+ "category": "emotional",
+ "text": "I was furious. Not the loud kind. The quiet kind that lasts for days.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0197.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I WAS FURIOUS NOT THE LOUD KIND THE QUIET KIND THAT LASTS FOR DAYS",
+ "raw": {
+ "normalized_reference": "i was furious. not the loud kind. the quiet kind that lasts for days.",
+ "normalized_hypothesis": "i was furious not the loud kind the quiet kind that lasts for days",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i was furious not the loud kind the quiet kind that lasts for days",
+ "normalized_hypothesis": "i was furious not the loud kind the quiet kind that lasts for days",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0198",
+ "category": "narrative",
+ "text": "That's the part of the story I usually leave out.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0198.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THAT'S THE PART OF THE STORY I USUALLY LEAVE OUT",
+ "raw": {
+ "normalized_reference": "that's the part of the story i usually leave out.",
+ "normalized_hypothesis": "that's the part of the story i usually leave out",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "that is the part of the story i usually leave out",
+ "normalized_hypothesis": "that is the part of the story i usually leave out",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0199",
+ "category": "short",
+ "text": "I've said too much.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0199.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I SAID TOO MUCH",
+ "raw": {
+ "normalized_reference": "i've said too much.",
+ "normalized_hypothesis": "i said too much",
+ "errors": 1,
+ "words": 4,
+ "wer": 0.25
+ },
+ "semantic": {
+ "normalized_reference": "i have said too much",
+ "normalized_hypothesis": "i said too much",
+ "errors": 1,
+ "words": 5,
+ "wer": 0.2
+ }
+ },
+ {
+ "id": "release500_0200",
+ "category": "conversational",
+ "text": "I've been faking that I understood this the entire time.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0200.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I'VE BEEN FAKING THAT I UNDERSTOOD THIS THE ENTIRE TIME",
+ "raw": {
+ "normalized_reference": "i've been faking that i understood this the entire time.",
+ "normalized_hypothesis": "i've been faking that i understood this the entire time",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i have been faking that i understood this the entire time",
+ "normalized_hypothesis": "i have been faking that i understood this the entire time",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0201",
+ "category": "phonetic_balance",
+ "text": "Peace be unto you and grace before the Lord.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0201.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "PEACE BE UNTO YOU AND GRACE BEFORE THE LORD",
+ "raw": {
+ "normalized_reference": "peace be unto you and grace before the lord.",
+ "normalized_hypothesis": "peace be unto you and grace before the lord",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "peace be unto you and grace before the lord",
+ "normalized_hypothesis": "peace be unto you and grace before the lord",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0202",
+ "category": "phonetic_balance",
+ "text": "From now on we're pals.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0202.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "FROM NOW ON WE ARE PALES",
+ "raw": {
+ "normalized_reference": "from now on we're pals.",
+ "normalized_hypothesis": "from now on we are pales",
+ "errors": 3,
+ "words": 5,
+ "wer": 0.6
+ },
+ "semantic": {
+ "normalized_reference": "from now on we are pals",
+ "normalized_hypothesis": "from now on we are pales",
+ "errors": 1,
+ "words": 6,
+ "wer": 0.16666666666666666
+ }
+ },
+ {
+ "id": "release500_0203",
+ "category": "phonetic_balance",
+ "text": "Here we allow our solicitors to look after our legal work.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0203.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HERE WE ALLOW OUR SOLICITORS TO LOOK AFTER OUR ILLEGAL WORK",
+ "raw": {
+ "normalized_reference": "here we allow our solicitors to look after our legal work.",
+ "normalized_hypothesis": "here we allow our solicitors to look after our illegal work",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ },
+ "semantic": {
+ "normalized_reference": "here we allow our solicitors to look after our legal work",
+ "normalized_hypothesis": "here we allow our solicitors to look after our illegal work",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ }
+ },
+ {
+ "id": "release500_0204",
+ "category": "phonetic_balance",
+ "text": "Her words sent a strange chill through Philip.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0204.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HER WORDS SENT A STRANGE CHILL THROUGH PHILIP",
+ "raw": {
+ "normalized_reference": "her words sent a strange chill through philip.",
+ "normalized_hypothesis": "her words sent a strange chill through philip",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "her words sent a strange chill through philip",
+ "normalized_hypothesis": "her words sent a strange chill through philip",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0205",
+ "category": "phonetic_balance",
+ "text": "It was introduced by Representative Dick of Ohio.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0205.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT WAS INTRODUCED BY REPRESENTATIVE DICK OF OHIO",
+ "raw": {
+ "normalized_reference": "it was introduced by representative dick of ohio.",
+ "normalized_hypothesis": "it was introduced by representative dick of ohio",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it was introduced by representative dick of ohio",
+ "normalized_hypothesis": "it was introduced by representative dick of ohio",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0206",
+ "category": "phonetic_balance",
+ "text": "It's the aurora borealis.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0206.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT'S THE AURORA BOREALIS",
+ "raw": {
+ "normalized_reference": "it's the aurora borealis.",
+ "normalized_hypothesis": "it's the aurora borealis",
+ "errors": 0,
+ "words": 4,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it is the aurora borealis",
+ "normalized_hypothesis": "it is the aurora borealis",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0207",
+ "category": "phonetic_balance",
+ "text": "The boy grew and prospered.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0207.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HEBOY GREW AND PROSPERED",
+ "raw": {
+ "normalized_reference": "the boy grew and prospered.",
+ "normalized_hypothesis": "heboy grew and prospered",
+ "errors": 2,
+ "words": 5,
+ "wer": 0.4
+ },
+ "semantic": {
+ "normalized_reference": "the boy grew and prospered",
+ "normalized_hypothesis": "heboy grew and prospered",
+ "errors": 2,
+ "words": 5,
+ "wer": 0.4
+ }
+ },
+ {
+ "id": "release500_0208",
+ "category": "phonetic_balance",
+ "text": "His voice was passionately rebellious.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0208.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HIS VOICE WAS PASSIONATELY REBELLIOUS",
+ "raw": {
+ "normalized_reference": "his voice was passionately rebellious.",
+ "normalized_hypothesis": "his voice was passionately rebellious",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "his voice was passionately rebellious",
+ "normalized_hypothesis": "his voice was passionately rebellious",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0209",
+ "category": "phonetic_balance",
+ "text": "That's where they cut off the Scottish Chiefs and killed all hands.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0209.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THAT'S WHERE THEY CUT OFF THE SCOTTISH CHIEFS AND KILLED ALL HANDS",
+ "raw": {
+ "normalized_reference": "that's where they cut off the scottish chiefs and killed all hands.",
+ "normalized_hypothesis": "that's where they cut off the scottish chiefs and killed all hands",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "that is where they cut off the scottish chiefs and killed all hands",
+ "normalized_hypothesis": "that is where they cut off the scottish chiefs and killed all hands",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0210",
+ "category": "phonetic_balance",
+ "text": "And Tom King patiently endured.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0210.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "AND TOM KING PATIENTLY ENDURED",
+ "raw": {
+ "normalized_reference": "and tom king patiently endured.",
+ "normalized_hypothesis": "and tom king patiently endured",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "and tom king patiently endured",
+ "normalized_hypothesis": "and tom king patiently endured",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0211",
+ "category": "phonetic_balance",
+ "text": "From that moment his friendship for Belize turns to hatred and jealousy.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0211.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "FROM THAT MOMENT HIS FRIENDSHIP FOR BILIS TURNS TO HATRED AND JEALOUSY",
+ "raw": {
+ "normalized_reference": "from that moment his friendship for belize turns to hatred and jealousy.",
+ "normalized_hypothesis": "from that moment his friendship for bilis turns to hatred and jealousy",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "from that moment his friendship for belize turns to hatred and jealousy",
+ "normalized_hypothesis": "from that moment his friendship for bilis turns to hatred and jealousy",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ }
+ },
+ {
+ "id": "release500_0212",
+ "category": "phonetic_balance",
+ "text": "Then you don't believe in altruism.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0212.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HAN YOU DON'T BELIEVE IN ALTRUISM",
+ "raw": {
+ "normalized_reference": "then you don't believe in altruism.",
+ "normalized_hypothesis": "han you don't believe in altruism",
+ "errors": 1,
+ "words": 6,
+ "wer": 0.16666666666666666
+ },
+ "semantic": {
+ "normalized_reference": "then you do not believe in altruism",
+ "normalized_hypothesis": "han you do not believe in altruism",
+ "errors": 1,
+ "words": 7,
+ "wer": 0.14285714285714285
+ }
+ },
+ {
+ "id": "release500_0213",
+ "category": "phonetic_balance",
+ "text": "What Pascal glimpsed with the vision of a seer, I have lived.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0213.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WHAT PASCO GLIMPSED WITH THE VISION OF A SEER I HAVE LIVED",
+ "raw": {
+ "normalized_reference": "what pascal glimpsed with the vision of a seer, i have lived.",
+ "normalized_hypothesis": "what pasco glimpsed with the vision of a seer i have lived",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "what pascal glimpsed with the vision of a seer i have lived",
+ "normalized_hypothesis": "what pasco glimpsed with the vision of a seer i have lived",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ }
+ },
+ {
+ "id": "release500_0214",
+ "category": "phonetic_balance",
+ "text": "You were making them talk shop, Ruth charged him.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0214.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "YOU WERE MAKING THEM TALK SHOP RUTH CHARGED HIM",
+ "raw": {
+ "normalized_reference": "you were making them talk shop, ruth charged him.",
+ "normalized_hypothesis": "you were making them talk shop ruth charged him",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "you were making them talk shop ruth charged him",
+ "normalized_hypothesis": "you were making them talk shop ruth charged him",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0215",
+ "category": "phonetic_balance",
+ "text": "Philip snatched at the letter which Gregson held out to him.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0215.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "PHILIP SNATCHED AT THE LETTER WHICH GREGSON HELD OUT TO HIM",
+ "raw": {
+ "normalized_reference": "philip snatched at the letter which gregson held out to him.",
+ "normalized_hypothesis": "philip snatched at the letter which gregson held out to him",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "philip snatched at the letter which gregson held out to him",
+ "normalized_hypothesis": "philip snatched at the letter which gregson held out to him",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0216",
+ "category": "phonetic_balance",
+ "text": "For a time the exciting thrill of his adventure was gone.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0216.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "FOR A TIME THE EXCITING DRILL OF HIS ADVENTURE WAS GONE",
+ "raw": {
+ "normalized_reference": "for a time the exciting thrill of his adventure was gone.",
+ "normalized_hypothesis": "for a time the exciting drill of his adventure was gone",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ },
+ "semantic": {
+ "normalized_reference": "for a time the exciting thrill of his adventure was gone",
+ "normalized_hypothesis": "for a time the exciting drill of his adventure was gone",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ }
+ },
+ {
+ "id": "release500_0217",
+ "category": "phonetic_balance",
+ "text": "Obviously, it was a disease that could be contracted by contact.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0217.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "OBVIOUSLY IT WAS A DISEASE THAT COULD BE CONTRACTED BY CONTACT",
+ "raw": {
+ "normalized_reference": "obviously, it was a disease that could be contracted by contact.",
+ "normalized_hypothesis": "obviously it was a disease that could be contracted by contact",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "obviously it was a disease that could be contracted by contact",
+ "normalized_hypothesis": "obviously it was a disease that could be contracted by contact",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0218",
+ "category": "phonetic_balance",
+ "text": "Blind with rage, he darted in.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0218.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "BLIND WITH RAGE HE DARTED IN",
+ "raw": {
+ "normalized_reference": "blind with rage, he darted in.",
+ "normalized_hypothesis": "blind with rage he darted in",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "blind with rage he darted in",
+ "normalized_hypothesis": "blind with rage he darted in",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0219",
+ "category": "phonetic_balance",
+ "text": "You live on an income which your father earned.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0219.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "YOU LIVE ON AN INCOME WHICH YOUR FATHER EARNED",
+ "raw": {
+ "normalized_reference": "you live on an income which your father earned.",
+ "normalized_hypothesis": "you live on an income which your father earned",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "you live on an income which your father earned",
+ "normalized_hypothesis": "you live on an income which your father earned",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0220",
+ "category": "phonetic_balance",
+ "text": "Something about it seemed to fascinate him, to challenge his presence.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0220.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SOMETHING ABOUT IT SEEMED TO FASCINATE HIM TO CHALLENGE HIS PRESENCE",
+ "raw": {
+ "normalized_reference": "something about it seemed to fascinate him, to challenge his presence.",
+ "normalized_hypothesis": "something about it seemed to fascinate him to challenge his presence",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "something about it seemed to fascinate him to challenge his presence",
+ "normalized_hypothesis": "something about it seemed to fascinate him to challenge his presence",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0221",
+ "category": "phonetic_balance",
+ "text": "Her beautiful hair was done up in shining coils.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0221.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HER BEAUTIFUL HAIR WAS DONE UP IN SHINING COILS",
+ "raw": {
+ "normalized_reference": "her beautiful hair was done up in shining coils.",
+ "normalized_hypothesis": "her beautiful hair was done up in shining coils",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "her beautiful hair was done up in shining coils",
+ "normalized_hypothesis": "her beautiful hair was done up in shining coils",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0222",
+ "category": "phonetic_balance",
+ "text": "It does, was her audacious answer.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0222.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT DOES WAS HER AUDACIOUS ANSWER",
+ "raw": {
+ "normalized_reference": "it does, was her audacious answer.",
+ "normalized_hypothesis": "it does was her audacious answer",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it does was her audacious answer",
+ "normalized_hypothesis": "it does was her audacious answer",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0223",
+ "category": "phonetic_balance",
+ "text": "He caught himself with a jerk.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0223.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE CAUGHT HIMSELF WITH A JERK",
+ "raw": {
+ "normalized_reference": "he caught himself with a jerk.",
+ "normalized_hypothesis": "he caught himself with a jerk",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he caught himself with a jerk",
+ "normalized_hypothesis": "he caught himself with a jerk",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0224",
+ "category": "phonetic_balance",
+ "text": "He did not believe in the burning of daylight for such a luxury.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0224.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE DID NOT BELIEVE IN THE BURNING OF DAYLIGHT FOR SUCH A LUXURY",
+ "raw": {
+ "normalized_reference": "he did not believe in the burning of daylight for such a luxury.",
+ "normalized_hypothesis": "he did not believe in the burning of daylight for such a luxury",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he did not believe in the burning of daylight for such a luxury",
+ "normalized_hypothesis": "he did not believe in the burning of daylight for such a luxury",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0225",
+ "category": "phonetic_balance",
+ "text": "He had comparatively no advantages at first.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0225.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE HAD COMPARATIVELY NO ADVANTAGES AT FIRST",
+ "raw": {
+ "normalized_reference": "he had comparatively no advantages at first.",
+ "normalized_hypothesis": "he had comparatively no advantages at first",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he had comparatively no advantages at 1st",
+ "normalized_hypothesis": "he had comparatively no advantages at 1st",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0226",
+ "category": "phonetic_balance",
+ "text": "Such men believe, when they come together.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0226.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SUCH MEN BELIEVE WHEN THEY COME TOGETHER",
+ "raw": {
+ "normalized_reference": "such men believe, when they come together.",
+ "normalized_hypothesis": "such men believe when they come together",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "such men believe when they come together",
+ "normalized_hypothesis": "such men believe when they come together",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0227",
+ "category": "phonetic_balance",
+ "text": "The ship should be in within a week or ten days.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0227.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE SHIP SHOULD BE IN WITHIN A WEEK OR TEN DAYS",
+ "raw": {
+ "normalized_reference": "the ship should be in within a week or ten days.",
+ "normalized_hypothesis": "the ship should be in within a week or ten days",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the ship should be in within a week or 10 days",
+ "normalized_hypothesis": "the ship should be in within a week or 10 days",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0228",
+ "category": "phonetic_balance",
+ "text": "Such a dog the wise driver kills, or turns loose.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0228.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SUCH A DOG THE WISE DRIVER KILLS OR TURNS LOOSE",
+ "raw": {
+ "normalized_reference": "such a dog the wise driver kills, or turns loose.",
+ "normalized_hypothesis": "such a dog the wise driver kills or turns loose",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "such a dog the wise driver kills or turns loose",
+ "normalized_hypothesis": "such a dog the wise driver kills or turns loose",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0229",
+ "category": "phonetic_balance",
+ "text": "You have heard always how he was the lover of the Princess Naomi.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0229.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "YOU HAVE HEARD ALWAYS HOW HE WAS THE LOVER OF THE PRINCESS NAOMI",
+ "raw": {
+ "normalized_reference": "you have heard always how he was the lover of the princess naomi.",
+ "normalized_hypothesis": "you have heard always how he was the lover of the princess naomi",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "you have heard always how he was the lover of the princess naomi",
+ "normalized_hypothesis": "you have heard always how he was the lover of the princess naomi",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0230",
+ "category": "phonetic_balance",
+ "text": "I want to know how all this is possible.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0230.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I WANT TO KNOW HOW ALL THIS IS POSSIBLE",
+ "raw": {
+ "normalized_reference": "i want to know how all this is possible.",
+ "normalized_hypothesis": "i want to know how all this is possible",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i want to know how all this is possible",
+ "normalized_hypothesis": "i want to know how all this is possible",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0231",
+ "category": "phonetic_balance",
+ "text": "It was my idea to a tee.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0231.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT WAS MY IDEA TOTEE",
+ "raw": {
+ "normalized_reference": "it was my idea to a tee.",
+ "normalized_hypothesis": "it was my idea totee",
+ "errors": 3,
+ "words": 7,
+ "wer": 0.42857142857142855
+ },
+ "semantic": {
+ "normalized_reference": "it was my idea to a tee",
+ "normalized_hypothesis": "it was my idea totee",
+ "errors": 3,
+ "words": 7,
+ "wer": 0.42857142857142855
+ }
+ },
+ {
+ "id": "release500_0232",
+ "category": "phonetic_balance",
+ "text": "They likewise are disinclined to being eaten.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0232.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THEY LIKEWISE ARE DISINCLINED TO BEING EATEN",
+ "raw": {
+ "normalized_reference": "they likewise are disinclined to being eaten.",
+ "normalized_hypothesis": "they likewise are disinclined to being eaten",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "they likewise are disinclined to being eaten",
+ "normalized_hypothesis": "they likewise are disinclined to being eaten",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0233",
+ "category": "phonetic_balance",
+ "text": "This state of mind comes of an undue prominence of the ego.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0233.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THIS STATE OF MIND COMES OF AN UNDUE PROMINENCE OF THE EGO",
+ "raw": {
+ "normalized_reference": "this state of mind comes of an undue prominence of the ego.",
+ "normalized_hypothesis": "this state of mind comes of an undue prominence of the ego",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "this state of mind comes of an undue prominence of the ego",
+ "normalized_hypothesis": "this state of mind comes of an undue prominence of the ego",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0234",
+ "category": "phonetic_balance",
+ "text": "Then it is as I said, Womble announced with finality.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0234.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THEN IT IS AS I SAID WAMBLE ANNOUNCED WITH FINALITY",
+ "raw": {
+ "normalized_reference": "then it is as i said, womble announced with finality.",
+ "normalized_hypothesis": "then it is as i said wamble announced with finality",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ },
+ "semantic": {
+ "normalized_reference": "then it is as i said womble announced with finality",
+ "normalized_hypothesis": "then it is as i said wamble announced with finality",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ }
+ },
+ {
+ "id": "release500_0235",
+ "category": "phonetic_balance",
+ "text": "I can see that knife now.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0235.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I CAN SEE THAT KNIFE NOW",
+ "raw": {
+ "normalized_reference": "i can see that knife now.",
+ "normalized_hypothesis": "i can see that knife now",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i can see that knife now",
+ "normalized_hypothesis": "i can see that knife now",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0236",
+ "category": "phonetic_balance",
+ "text": "If I meet her again I shall apologize, said Eileen.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0236.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IF I MEET HER AGAIN I SHALL APOLOGIZE SAID ILEIN",
+ "raw": {
+ "normalized_reference": "if i meet her again i shall apologize, said eileen.",
+ "normalized_hypothesis": "if i meet her again i shall apologize said ilein",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ },
+ "semantic": {
+ "normalized_reference": "if i meet her again i shall apologize said eileen",
+ "normalized_hypothesis": "if i meet her again i shall apologize said ilein",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ }
+ },
+ {
+ "id": "release500_0237",
+ "category": "phonetic_balance",
+ "text": "A wildly exciting time was his during the week preceding Thursday the eighteenth.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0237.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "A WILDLY EXCITING TIME WAS HIS DURING THE WEEK PRECEDING THURSDAY THE EIGHTEENTH",
+ "raw": {
+ "normalized_reference": "a wildly exciting time was his during the week preceding thursday the eighteenth.",
+ "normalized_hypothesis": "a wildly exciting time was his during the week preceding thursday the eighteenth",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "a wildly exciting time was his during the week preceding thursday the 18th",
+ "normalized_hypothesis": "a wildly exciting time was his during the week preceding thursday the 18th",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0238",
+ "category": "phonetic_balance",
+ "text": "In what bucolic school of fence he had been taught was beyond imagining.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0238.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IN WHAT BEAUCALIC SCHOOL OF FENCE HE HAD BEEN TAUGHT WAS BEYOND IMAGINING",
+ "raw": {
+ "normalized_reference": "in what bucolic school of fence he had been taught was beyond imagining.",
+ "normalized_hypothesis": "in what beaucalic school of fence he had been taught was beyond imagining",
+ "errors": 1,
+ "words": 13,
+ "wer": 0.07692307692307693
+ },
+ "semantic": {
+ "normalized_reference": "in what bucolic school of fence he had been taught was beyond imagining",
+ "normalized_hypothesis": "in what beaucalic school of fence he had been taught was beyond imagining",
+ "errors": 1,
+ "words": 13,
+ "wer": 0.07692307692307693
+ }
+ },
+ {
+ "id": "release500_0239",
+ "category": "phonetic_balance",
+ "text": "So she said, the irate skipper dashed on.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0239.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SO SHE SAID THE IRATE SKIPPER DASHED ON",
+ "raw": {
+ "normalized_reference": "so she said, the irate skipper dashed on.",
+ "normalized_hypothesis": "so she said the irate skipper dashed on",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "so she said the irate skipper dashed on",
+ "normalized_hypothesis": "so she said the irate skipper dashed on",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0240",
+ "category": "phonetic_balance",
+ "text": "It is not an attempt to smash the market.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0240.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT IS NOT AN ATTEMPT TO SMASH THE MARKET",
+ "raw": {
+ "normalized_reference": "it is not an attempt to smash the market.",
+ "normalized_hypothesis": "it is not an attempt to smash the market",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it is not an attempt to smash the market",
+ "normalized_hypothesis": "it is not an attempt to smash the market",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0241",
+ "category": "phonetic_balance",
+ "text": "He destroyed everything that had belonged to the woman.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0241.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE DESTROYED EVERYTHING THAT HAD BELONGED TO THE WOMEN",
+ "raw": {
+ "normalized_reference": "he destroyed everything that had belonged to the woman.",
+ "normalized_hypothesis": "he destroyed everything that had belonged to the women",
+ "errors": 1,
+ "words": 9,
+ "wer": 0.1111111111111111
+ },
+ "semantic": {
+ "normalized_reference": "he destroyed everything that had belonged to the woman",
+ "normalized_hypothesis": "he destroyed everything that had belonged to the women",
+ "errors": 1,
+ "words": 9,
+ "wer": 0.1111111111111111
+ }
+ },
+ {
+ "id": "release500_0242",
+ "category": "phonetic_balance",
+ "text": "Did Thorpe go to see any one in Churchill.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0242.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "DID THORK GO TO SEE ANY ONE IN CHURCHILL",
+ "raw": {
+ "normalized_reference": "did thorpe go to see any one in churchill.",
+ "normalized_hypothesis": "did thork go to see any one in churchill",
+ "errors": 1,
+ "words": 9,
+ "wer": 0.1111111111111111
+ },
+ "semantic": {
+ "normalized_reference": "did thorpe go to see any one in churchill",
+ "normalized_hypothesis": "did thork go to see any one in churchill",
+ "errors": 1,
+ "words": 9,
+ "wer": 0.1111111111111111
+ }
+ },
+ {
+ "id": "release500_0243",
+ "category": "phonetic_balance",
+ "text": "Many other similar disconcerting omissions will be noticed in the Manuscript.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0243.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "MANY OTHER SIMILAR DISCONCERTING OMISSIONS WILL BE NOTICED IN THE MANUSCRIPT",
+ "raw": {
+ "normalized_reference": "many other similar disconcerting omissions will be noticed in the manuscript.",
+ "normalized_hypothesis": "many other similar disconcerting omissions will be noticed in the manuscript",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "many other similar disconcerting omissions will be noticed in the manuscript",
+ "normalized_hypothesis": "many other similar disconcerting omissions will be noticed in the manuscript",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0244",
+ "category": "phonetic_balance",
+ "text": "The ringing of the big bell aroused him.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0244.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE RINGING OF THE BIG BELL AROUSED HIM",
+ "raw": {
+ "normalized_reference": "the ringing of the big bell aroused him.",
+ "normalized_hypothesis": "the ringing of the big bell aroused him",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the ringing of the big bell aroused him",
+ "normalized_hypothesis": "the ringing of the big bell aroused him",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0245",
+ "category": "phonetic_balance",
+ "text": "A sinewy hand, dripping with water, was clutching the rail.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0245.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "A SINEWY HAND DRIPPING WITH WATER WAS CLUTCHING THE RAIL",
+ "raw": {
+ "normalized_reference": "a sinewy hand, dripping with water, was clutching the rail.",
+ "normalized_hypothesis": "a sinewy hand dripping with water was clutching the rail",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "a sinewy hand dripping with water was clutching the rail",
+ "normalized_hypothesis": "a sinewy hand dripping with water was clutching the rail",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0246",
+ "category": "phonetic_balance",
+ "text": "Enters now the psychology of the situation.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0246.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "ENTERS NOW THE PSYCHOLOGY OF THE SITUATION",
+ "raw": {
+ "normalized_reference": "enters now the psychology of the situation.",
+ "normalized_hypothesis": "enters now the psychology of the situation",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "enters now the psychology of the situation",
+ "normalized_hypothesis": "enters now the psychology of the situation",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0247",
+ "category": "phonetic_balance",
+ "text": "Wash your hands of me.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0247.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WASH YOUR HANDS OF ME",
+ "raw": {
+ "normalized_reference": "wash your hands of me.",
+ "normalized_hypothesis": "wash your hands of me",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "wash your hands of me",
+ "normalized_hypothesis": "wash your hands of me",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0248",
+ "category": "phonetic_balance",
+ "text": "He confessed that the sketch had startled him.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0248.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE CONFESSED THAT THE SKETCH HAD STARTLED HIM",
+ "raw": {
+ "normalized_reference": "he confessed that the sketch had startled him.",
+ "normalized_hypothesis": "he confessed that the sketch had startled him",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he confessed that the sketch had startled him",
+ "normalized_hypothesis": "he confessed that the sketch had startled him",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0249",
+ "category": "phonetic_balance",
+ "text": "Her face was against his breast.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0249.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HER FACE WAS AGAINST HIS BREAST",
+ "raw": {
+ "normalized_reference": "her face was against his breast.",
+ "normalized_hypothesis": "her face was against his breast",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "her face was against his breast",
+ "normalized_hypothesis": "her face was against his breast",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0250",
+ "category": "phonetic_balance",
+ "text": "He was a pariah; a wanderer without a friend or a home.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0250.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE WAS A PARIAH A WANDERER WITHOUT A FRIEND OR A HOME",
+ "raw": {
+ "normalized_reference": "he was a pariah; a wanderer without a friend or a home.",
+ "normalized_hypothesis": "he was a pariah a wanderer without a friend or a home",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he was a pariah a wanderer without a friend or a home",
+ "normalized_hypothesis": "he was a pariah a wanderer without a friend or a home",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0251",
+ "category": "phonetic_balance",
+ "text": "A cry of joy burst from Philip's lips.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0251.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "A CRY OF JOY BURST FROM PHILIP'S LIPS",
+ "raw": {
+ "normalized_reference": "a cry of joy burst from philip's lips.",
+ "normalized_hypothesis": "a cry of joy burst from philip's lips",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "a cry of joy burst from philip is lips",
+ "normalized_hypothesis": "a cry of joy burst from philip is lips",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0252",
+ "category": "phonetic_balance",
+ "text": "And the big Persian knew of his existence before he did of hers.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0252.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "AND THE BIG PERSIAN KNEW OF HIS EXISTENCE BEFORE HE DID OF HERS",
+ "raw": {
+ "normalized_reference": "and the big persian knew of his existence before he did of hers.",
+ "normalized_hypothesis": "and the big persian knew of his existence before he did of hers",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "and the big persian knew of his existence before he did of hers",
+ "normalized_hypothesis": "and the big persian knew of his existence before he did of hers",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0253",
+ "category": "phonetic_balance",
+ "text": "The men stared into each other's face.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0253.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE MEN STARED INTO EACH OTHER'S FACE",
+ "raw": {
+ "normalized_reference": "the men stared into each other's face.",
+ "normalized_hypothesis": "the men stared into each other's face",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the men stared into each other is face",
+ "normalized_hypothesis": "the men stared into each other is face",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0254",
+ "category": "phonetic_balance",
+ "text": "It is the fire, partly, she said.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0254.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IS THE FIRE PARTLY SHE SAID",
+ "raw": {
+ "normalized_reference": "it is the fire, partly, she said.",
+ "normalized_hypothesis": "is the fire partly she said",
+ "errors": 1,
+ "words": 7,
+ "wer": 0.14285714285714285
+ },
+ "semantic": {
+ "normalized_reference": "it is the fire partly she said",
+ "normalized_hypothesis": "is the fire partly she said",
+ "errors": 1,
+ "words": 7,
+ "wer": 0.14285714285714285
+ }
+ },
+ {
+ "id": "release500_0255",
+ "category": "phonetic_balance",
+ "text": "It seemed nearer to him since he had seen and talked with Gregson.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0255.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SEEMED NEARER TO HIM SINCE HE HAD SEEN AND TALKED WITH GRIGSON",
+ "raw": {
+ "normalized_reference": "it seemed nearer to him since he had seen and talked with gregson.",
+ "normalized_hypothesis": "seemed nearer to him since he had seen and talked with grigson",
+ "errors": 2,
+ "words": 13,
+ "wer": 0.15384615384615385
+ },
+ "semantic": {
+ "normalized_reference": "it seemed nearer to him since he had seen and talked with gregson",
+ "normalized_hypothesis": "seemed nearer to him since he had seen and talked with grigson",
+ "errors": 2,
+ "words": 13,
+ "wer": 0.15384615384615385
+ }
+ },
+ {
+ "id": "release500_0256",
+ "category": "phonetic_balance",
+ "text": "Among my minor afflictions, I may mention a new and mysterious one.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0256.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "AMONG MY MINOR AFFLICTIONS I MAY MENTION A NEW AND MYSTERIOUS ONE",
+ "raw": {
+ "normalized_reference": "among my minor afflictions, i may mention a new and mysterious one.",
+ "normalized_hypothesis": "among my minor afflictions i may mention a new and mysterious one",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "among my minor afflictions i may mention a new and mysterious one",
+ "normalized_hypothesis": "among my minor afflictions i may mention a new and mysterious one",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0257",
+ "category": "phonetic_balance",
+ "text": "Between him and all domestic animals there must be no hostilities.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0257.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "BETWEEN HIM AND ALL DOMESTIC ANIMALS THERE MUST BE NO HOSTILITIES",
+ "raw": {
+ "normalized_reference": "between him and all domestic animals there must be no hostilities.",
+ "normalized_hypothesis": "between him and all domestic animals there must be no hostilities",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "between him and all domestic animals there must be no hostilities",
+ "normalized_hypothesis": "between him and all domestic animals there must be no hostilities",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0258",
+ "category": "phonetic_balance",
+ "text": "With these arguments he convinced himself that he should go on alone.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0258.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WITH THESE ARGUMENTS HE CONVINCED HIMSELF THAT HE SHOULD GO ON ALONE",
+ "raw": {
+ "normalized_reference": "with these arguments he convinced himself that he should go on alone.",
+ "normalized_hypothesis": "with these arguments he convinced himself that he should go on alone",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "with these arguments he convinced himself that he should go on alone",
+ "normalized_hypothesis": "with these arguments he convinced himself that he should go on alone",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0259",
+ "category": "phonetic_balance",
+ "text": "We are consumed in our own flesh-pots.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0259.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WE ARE CONSUMED IN OUR OWN FLESH POTS",
+ "raw": {
+ "normalized_reference": "we are consumed in our own flesh-pots.",
+ "normalized_hypothesis": "we are consumed in our own flesh pots",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "we are consumed in our own flesh pots",
+ "normalized_hypothesis": "we are consumed in our own flesh pots",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0260",
+ "category": "phonetic_balance",
+ "text": "Mercedes screamed, cried, laughed, and manifested the chaotic abandonment of hysteria.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0260.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "MERCEDES SCREEN CRIED LAUGHED AND MANIFESTED THE CHAOTIC ABANDONMENT OF HYSTERIA",
+ "raw": {
+ "normalized_reference": "mercedes screamed, cried, laughed, and manifested the chaotic abandonment of hysteria.",
+ "normalized_hypothesis": "mercedes screen cried laughed and manifested the chaotic abandonment of hysteria",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ },
+ "semantic": {
+ "normalized_reference": "mercedes screamed cried laughed and manifested the chaotic abandonment of hysteria",
+ "normalized_hypothesis": "mercedes screen cried laughed and manifested the chaotic abandonment of hysteria",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ }
+ },
+ {
+ "id": "release500_0261",
+ "category": "phonetic_balance",
+ "text": "Philip made no effort to follow.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0261.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "PHILIP MADE NO EFFORT TO FOLLOW",
+ "raw": {
+ "normalized_reference": "philip made no effort to follow.",
+ "normalized_hypothesis": "philip made no effort to follow",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "philip made no effort to follow",
+ "normalized_hypothesis": "philip made no effort to follow",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0262",
+ "category": "phonetic_balance",
+ "text": "She said with chattering teeth.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0262.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SHE SAID WITH CHATTERING TEETH",
+ "raw": {
+ "normalized_reference": "she said with chattering teeth.",
+ "normalized_hypothesis": "she said with chattering teeth",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "she said with chattering teeth",
+ "normalized_hypothesis": "she said with chattering teeth",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0263",
+ "category": "phonetic_balance",
+ "text": "Halfway around the track one donkey got into an argument with its rider.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0263.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HALF WAY ROUND THE TRACK ONE DONKEY GOT INTO AN ARGUMENT WITH ITS RIDER",
+ "raw": {
+ "normalized_reference": "halfway around the track one donkey got into an argument with its rider.",
+ "normalized_hypothesis": "half way round the track one donkey got into an argument with its rider",
+ "errors": 3,
+ "words": 13,
+ "wer": 0.23076923076923078
+ },
+ "semantic": {
+ "normalized_reference": "halfway around the track one donkey got into an argument with its rider",
+ "normalized_hypothesis": "half way round the track one donkey got into an argument with its rider",
+ "errors": 3,
+ "words": 13,
+ "wer": 0.23076923076923078
+ }
+ },
+ {
+ "id": "release500_0264",
+ "category": "phonetic_balance",
+ "text": "Earth and gravel seemed to fill the pan.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0264.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "EARTH AND GRAVEL SEEMED TO FILL THE PAN",
+ "raw": {
+ "normalized_reference": "earth and gravel seemed to fill the pan.",
+ "normalized_hypothesis": "earth and gravel seemed to fill the pan",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "earth and gravel seemed to fill the pan",
+ "normalized_hypothesis": "earth and gravel seemed to fill the pan",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0265",
+ "category": "phonetic_balance",
+ "text": "Her achievements with cocoanuts were a revelation.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0265.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HER ACHIEVEMENTS WITH COCOANUTS WERE A REVELATION",
+ "raw": {
+ "normalized_reference": "her achievements with cocoanuts were a revelation.",
+ "normalized_hypothesis": "her achievements with cocoanuts were a revelation",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "her achievements with cocoanuts were a revelation",
+ "normalized_hypothesis": "her achievements with cocoanuts were a revelation",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0266",
+ "category": "phonetic_balance",
+ "text": "For a full minute he crouched and listened.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0266.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "FOR A FULL MINUTE HE CROUCHED AND LISTENED",
+ "raw": {
+ "normalized_reference": "for a full minute he crouched and listened.",
+ "normalized_hypothesis": "for a full minute he crouched and listened",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "for a full minute he crouched and listened",
+ "normalized_hypothesis": "for a full minute he crouched and listened",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0267",
+ "category": "phonetic_balance",
+ "text": "It was edged with ice.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0267.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT WAS EDGED WITH ICE",
+ "raw": {
+ "normalized_reference": "it was edged with ice.",
+ "normalized_hypothesis": "it was edged with ice",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it was edged with ice",
+ "normalized_hypothesis": "it was edged with ice",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0268",
+ "category": "phonetic_balance",
+ "text": "The Portuguese boy passed the Hawaiian.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0268.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE PORTUGUESE BOY PASSED THE HAWAIAN",
+ "raw": {
+ "normalized_reference": "the portuguese boy passed the hawaiian.",
+ "normalized_hypothesis": "the portuguese boy passed the hawaian",
+ "errors": 1,
+ "words": 6,
+ "wer": 0.16666666666666666
+ },
+ "semantic": {
+ "normalized_reference": "the portuguese boy passed the hawaiian",
+ "normalized_hypothesis": "the portuguese boy passed the hawaian",
+ "errors": 1,
+ "words": 6,
+ "wer": 0.16666666666666666
+ }
+ },
+ {
+ "id": "release500_0269",
+ "category": "phonetic_balance",
+ "text": "But he did not broach it, preferring to mature it carefully.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0269.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "BUT HE DID NOT BROACH IT PREFERRING TO MATURE IT CAREFULLY",
+ "raw": {
+ "normalized_reference": "but he did not broach it, preferring to mature it carefully.",
+ "normalized_hypothesis": "but he did not broach it preferring to mature it carefully",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "but he did not broach it preferring to mature it carefully",
+ "normalized_hypothesis": "but he did not broach it preferring to mature it carefully",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0270",
+ "category": "phonetic_balance",
+ "text": "Come on, Del Mar challenged.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0270.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "COME ON DELMAR CHALLENGE",
+ "raw": {
+ "normalized_reference": "come on, del mar challenged.",
+ "normalized_hypothesis": "come on delmar challenge",
+ "errors": 3,
+ "words": 5,
+ "wer": 0.6
+ },
+ "semantic": {
+ "normalized_reference": "come on del mar challenged",
+ "normalized_hypothesis": "come on delmar challenge",
+ "errors": 3,
+ "words": 5,
+ "wer": 0.6
+ }
+ },
+ {
+ "id": "release500_0271",
+ "category": "phonetic_balance",
+ "text": "For a full half minute Jeanne looked at him without speaking.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0271.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "FOR A FULL HALF MINUTE JEAN LOOKED AT HIM WITHOUT SPEAKING",
+ "raw": {
+ "normalized_reference": "for a full half minute jeanne looked at him without speaking.",
+ "normalized_hypothesis": "for a full half minute jean looked at him without speaking",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ },
+ "semantic": {
+ "normalized_reference": "for a full half minute jeanne looked at him without speaking",
+ "normalized_hypothesis": "for a full half minute jean looked at him without speaking",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ }
+ },
+ {
+ "id": "release500_0272",
+ "category": "phonetic_balance",
+ "text": "He bore no grudges and had few enemies.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0272.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE BORE NO GRUDGES AND HAD FEW ENEMIES",
+ "raw": {
+ "normalized_reference": "he bore no grudges and had few enemies.",
+ "normalized_hypothesis": "he bore no grudges and had few enemies",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he bore no grudges and had few enemies",
+ "normalized_hypothesis": "he bore no grudges and had few enemies",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0273",
+ "category": "phonetic_balance",
+ "text": "I'll go over tomorrow afternoon.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0273.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I'LL GO OVER TO MORROW AFTERNOON",
+ "raw": {
+ "normalized_reference": "i'll go over tomorrow afternoon.",
+ "normalized_hypothesis": "i'll go over to morrow afternoon",
+ "errors": 2,
+ "words": 5,
+ "wer": 0.4
+ },
+ "semantic": {
+ "normalized_reference": "i will go over tomorrow afternoon",
+ "normalized_hypothesis": "i will go over to morrow afternoon",
+ "errors": 2,
+ "words": 6,
+ "wer": 0.3333333333333333
+ }
+ },
+ {
+ "id": "release500_0274",
+ "category": "phonetic_balance",
+ "text": "Well, there are better men in Hawaii, that's all.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0274.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WELL THERE ARE BETTER MEN IN HAWAI'I THAT'S ALL",
+ "raw": {
+ "normalized_reference": "well, there are better men in hawaii, that's all.",
+ "normalized_hypothesis": "well there are better men in hawai'i that's all",
+ "errors": 1,
+ "words": 9,
+ "wer": 0.1111111111111111
+ },
+ "semantic": {
+ "normalized_reference": "well there are better men in hawaii that is all",
+ "normalized_hypothesis": "well there are better men in hawai i that is all",
+ "errors": 2,
+ "words": 10,
+ "wer": 0.2
+ }
+ },
+ {
+ "id": "release500_0275",
+ "category": "phonetic_balance",
+ "text": "He was an athlete and a giant.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0275.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE WAS AN ATHLETE AND A GIANT",
+ "raw": {
+ "normalized_reference": "he was an athlete and a giant.",
+ "normalized_hypothesis": "he was an athlete and a giant",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he was an athlete and a giant",
+ "normalized_hypothesis": "he was an athlete and a giant",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0276",
+ "category": "phonetic_balance",
+ "text": "McVeigh when he returns from a trip to Honolulu.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0276.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "UKVAY WHEN HE RETURNS FROM A TRIP TO HONNOLULU",
+ "raw": {
+ "normalized_reference": "mcveigh when he returns from a trip to honolulu.",
+ "normalized_hypothesis": "ukvay when he returns from a trip to honnolulu",
+ "errors": 2,
+ "words": 9,
+ "wer": 0.2222222222222222
+ },
+ "semantic": {
+ "normalized_reference": "mcveigh when he returns from a trip to honolulu",
+ "normalized_hypothesis": "ukvay when he returns from a trip to honnolulu",
+ "errors": 2,
+ "words": 9,
+ "wer": 0.2222222222222222
+ }
+ },
+ {
+ "id": "release500_0277",
+ "category": "phonetic_balance",
+ "text": "You're a devil for fighting, and will surely win.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0277.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "YOU ARE A DEVIL FOR FIGHTING AND WILL SURELY WIN",
+ "raw": {
+ "normalized_reference": "you're a devil for fighting, and will surely win.",
+ "normalized_hypothesis": "you are a devil for fighting and will surely win",
+ "errors": 2,
+ "words": 9,
+ "wer": 0.2222222222222222
+ },
+ "semantic": {
+ "normalized_reference": "you are a devil for fighting and will surely win",
+ "normalized_hypothesis": "you are a devil for fighting and will surely win",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0278",
+ "category": "phonetic_balance",
+ "text": "All an appearance can know is mirage.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0278.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "ALL AN APPEARANCE CAN KNOW IS MUROCHE",
+ "raw": {
+ "normalized_reference": "all an appearance can know is mirage.",
+ "normalized_hypothesis": "all an appearance can know is muroche",
+ "errors": 1,
+ "words": 7,
+ "wer": 0.14285714285714285
+ },
+ "semantic": {
+ "normalized_reference": "all an appearance can know is mirage",
+ "normalized_hypothesis": "all an appearance can know is muroche",
+ "errors": 1,
+ "words": 7,
+ "wer": 0.14285714285714285
+ }
+ },
+ {
+ "id": "release500_0279",
+ "category": "phonetic_balance",
+ "text": "You fired me out of your house, in short.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0279.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "YOU FIRED ME OUT OF YOUR HOUSE IN SHORT",
+ "raw": {
+ "normalized_reference": "you fired me out of your house, in short.",
+ "normalized_hypothesis": "you fired me out of your house in short",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "you fired me out of your house in short",
+ "normalized_hypothesis": "you fired me out of your house in short",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0280",
+ "category": "phonetic_balance",
+ "text": "He made sure that the magazine was loaded, and resumed his paddling.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0280.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE MADE SURE THAT THE MAGAZINE WAS LOADED AND RESUMED HIS PADDLING",
+ "raw": {
+ "normalized_reference": "he made sure that the magazine was loaded, and resumed his paddling.",
+ "normalized_hypothesis": "he made sure that the magazine was loaded and resumed his paddling",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he made sure that the magazine was loaded and resumed his paddling",
+ "normalized_hypothesis": "he made sure that the magazine was loaded and resumed his paddling",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0281",
+ "category": "phonetic_balance",
+ "text": "Change chairs, Daylight commanded.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0281.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "CHANGED CHAIRS DAYLIGHT COMMANDED",
+ "raw": {
+ "normalized_reference": "change chairs, daylight commanded.",
+ "normalized_hypothesis": "changed chairs daylight commanded",
+ "errors": 1,
+ "words": 4,
+ "wer": 0.25
+ },
+ "semantic": {
+ "normalized_reference": "change chairs daylight commanded",
+ "normalized_hypothesis": "changed chairs daylight commanded",
+ "errors": 1,
+ "words": 4,
+ "wer": 0.25
+ }
+ },
+ {
+ "id": "release500_0282",
+ "category": "phonetic_balance",
+ "text": "By that answer my professional medical prestige stood or fell.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0282.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "Y THAT ANSWER MY PROFESSIONAL MEDICAL PRESTIGE STOOD OR FELL",
+ "raw": {
+ "normalized_reference": "by that answer my professional medical prestige stood or fell.",
+ "normalized_hypothesis": "y that answer my professional medical prestige stood or fell",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ },
+ "semantic": {
+ "normalized_reference": "by that answer my professional medical prestige stood or fell",
+ "normalized_hypothesis": "y that answer my professional medical prestige stood or fell",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ }
+ },
+ {
+ "id": "release500_0283",
+ "category": "phonetic_balance",
+ "text": "The moon had already begun its westward decline.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0283.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE MOON HAD ALREADY BEGUN ITS WESTWARD DECLINE",
+ "raw": {
+ "normalized_reference": "the moon had already begun its westward decline.",
+ "normalized_hypothesis": "the moon had already begun its westward decline",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the moon had already begun its westward decline",
+ "normalized_hypothesis": "the moon had already begun its westward decline",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0284",
+ "category": "phonetic_balance",
+ "text": "At sea, Monday, March 16, 1908.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0284.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "AT CE MONDAY MARCH SIXTEENTH NINETEEN HUNDRED EIGHT",
+ "raw": {
+ "normalized_reference": "at sea, monday, march 16, 1908.",
+ "normalized_hypothesis": "at ce monday march sixteenth nineteen hundred eight",
+ "errors": 5,
+ "words": 6,
+ "wer": 0.8333333333333334
+ },
+ "semantic": {
+ "normalized_reference": "at sea monday march 16 1908",
+ "normalized_hypothesis": "at ce monday march 16th 1908",
+ "errors": 2,
+ "words": 6,
+ "wer": 0.3333333333333333
+ }
+ },
+ {
+ "id": "release500_0285",
+ "category": "phonetic_balance",
+ "text": "And wherever I ranged, the way lay along alcohol-drenched roads.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0285.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "AND WHEREVER I RANGED THE WAY LAY ALONG ALCOHOL DRENCHED ROADS",
+ "raw": {
+ "normalized_reference": "and wherever i ranged, the way lay along alcohol-drenched roads.",
+ "normalized_hypothesis": "and wherever i ranged the way lay along alcohol drenched roads",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "and wherever i ranged the way lay along alcohol drenched roads",
+ "normalized_hypothesis": "and wherever i ranged the way lay along alcohol drenched roads",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0286",
+ "category": "phonetic_balance",
+ "text": "The date was nearly eighteen years old.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0286.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE DATE WAS NEARLY EIGHTEEN YEARS OLD",
+ "raw": {
+ "normalized_reference": "the date was nearly eighteen years old.",
+ "normalized_hypothesis": "the date was nearly eighteen years old",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the date was nearly 18 years old",
+ "normalized_hypothesis": "the date was nearly 18 years old",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0287",
+ "category": "phonetic_balance",
+ "text": "Very early in my life, I separated from my mother.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0287.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "VERY EARLY IN MY LIFE I SEPARATED FROM MY MOTHER",
+ "raw": {
+ "normalized_reference": "very early in my life, i separated from my mother.",
+ "normalized_hypothesis": "very early in my life i separated from my mother",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "very early in my life i separated from my mother",
+ "normalized_hypothesis": "very early in my life i separated from my mother",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0288",
+ "category": "phonetic_balance",
+ "text": "There was pride and strength, the ring of triumph in his voice.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0288.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THERE WAS PRIDE AND STRENGTH THE RING OF TRIUMPH IN HIS VOICE",
+ "raw": {
+ "normalized_reference": "there was pride and strength, the ring of triumph in his voice.",
+ "normalized_hypothesis": "there was pride and strength the ring of triumph in his voice",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "there was pride and strength the ring of triumph in his voice",
+ "normalized_hypothesis": "there was pride and strength the ring of triumph in his voice",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0289",
+ "category": "phonetic_balance",
+ "text": "From my earliest recollection my sleep was a period of terror.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0289.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "FROM MY EARLIEST RECOLLECTION MY SLEEP WAS A PERIOD OF TERROR",
+ "raw": {
+ "normalized_reference": "from my earliest recollection my sleep was a period of terror.",
+ "normalized_hypothesis": "from my earliest recollection my sleep was a period of terror",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "from my earliest recollection my sleep was a period of terror",
+ "normalized_hypothesis": "from my earliest recollection my sleep was a period of terror",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0290",
+ "category": "phonetic_balance",
+ "text": "We'll have to watch our chances.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0290.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WE'LL HAVE TO WATCH OUR CHANCES",
+ "raw": {
+ "normalized_reference": "we'll have to watch our chances.",
+ "normalized_hypothesis": "we'll have to watch our chances",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "we will have to watch our chances",
+ "normalized_hypothesis": "we will have to watch our chances",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0291",
+ "category": "phonetic_balance",
+ "text": "Close beside him gleamed the white fangs of the wolf-dog.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0291.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "CLOSE BESIDE HIM GLEAMED THE WHITE FANGS OF THE WOLF DOG",
+ "raw": {
+ "normalized_reference": "close beside him gleamed the white fangs of the wolf-dog.",
+ "normalized_hypothesis": "close beside him gleamed the white fangs of the wolf dog",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "close beside him gleamed the white fangs of the wolf dog",
+ "normalized_hypothesis": "close beside him gleamed the white fangs of the wolf dog",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0292",
+ "category": "phonetic_balance",
+ "text": "Would you be satisfied with that one hundredth part of me.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0292.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WOULD YOU BE SATISFIED WITH THAT ONE HUNDREDTH PART OF ME",
+ "raw": {
+ "normalized_reference": "would you be satisfied with that one hundredth part of me.",
+ "normalized_hypothesis": "would you be satisfied with that one hundredth part of me",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "would you be satisfied with that 100th part of me",
+ "normalized_hypothesis": "would you be satisfied with that 100th part of me",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0293",
+ "category": "phonetic_balance",
+ "text": "Two gigantic owls were tearing at the carcass.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0293.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "TWO GIGANTIC OWLS WERE TEARING AT THE CARCASE",
+ "raw": {
+ "normalized_reference": "two gigantic owls were tearing at the carcass.",
+ "normalized_hypothesis": "two gigantic owls were tearing at the carcase",
+ "errors": 1,
+ "words": 8,
+ "wer": 0.125
+ },
+ "semantic": {
+ "normalized_reference": "2 gigantic owls were tearing at the carcass",
+ "normalized_hypothesis": "2 gigantic owls were tearing at the carcase",
+ "errors": 1,
+ "words": 8,
+ "wer": 0.125
+ }
+ },
+ {
+ "id": "release500_0294",
+ "category": "phonetic_balance",
+ "text": "The lush vegetation of that sheltered spot make a natural shield.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0294.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "ELUSH VEGETATION OF THAT SHELTERED SPOT MAKE AND NATURAL SHIELD",
+ "raw": {
+ "normalized_reference": "the lush vegetation of that sheltered spot make a natural shield.",
+ "normalized_hypothesis": "elush vegetation of that sheltered spot make and natural shield",
+ "errors": 3,
+ "words": 11,
+ "wer": 0.2727272727272727
+ },
+ "semantic": {
+ "normalized_reference": "the lush vegetation of that sheltered spot make a natural shield",
+ "normalized_hypothesis": "elush vegetation of that sheltered spot make and natural shield",
+ "errors": 3,
+ "words": 11,
+ "wer": 0.2727272727272727
+ }
+ },
+ {
+ "id": "release500_0295",
+ "category": "phonetic_balance",
+ "text": "Below him the shadow was broken into a pool of rippling starlight.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0295.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "BELOW HIM THE SHADOW WAS BROKEN INTO A POOL OF RIPPLING STARLIGHT",
+ "raw": {
+ "normalized_reference": "below him the shadow was broken into a pool of rippling starlight.",
+ "normalized_hypothesis": "below him the shadow was broken into a pool of rippling starlight",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "below him the shadow was broken into a pool of rippling starlight",
+ "normalized_hypothesis": "below him the shadow was broken into a pool of rippling starlight",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0296",
+ "category": "phonetic_balance",
+ "text": "Lots of men take women buggy riding.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0296.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "LOTS OF MEN TAKE WOMEN BUGGY RIDING",
+ "raw": {
+ "normalized_reference": "lots of men take women buggy riding.",
+ "normalized_hypothesis": "lots of men take women buggy riding",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "lots of men take women buggy riding",
+ "normalized_hypothesis": "lots of men take women buggy riding",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0297",
+ "category": "phonetic_balance",
+ "text": "But it contributed to the smash.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0297.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "BUT IT CONTRIBUTED TO THE SMASH",
+ "raw": {
+ "normalized_reference": "but it contributed to the smash.",
+ "normalized_hypothesis": "but it contributed to the smash",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "but it contributed to the smash",
+ "normalized_hypothesis": "but it contributed to the smash",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0298",
+ "category": "phonetic_balance",
+ "text": "No man ate of the seal meat or the oil.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0298.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "NO MEN ATE OF THE SEAL MEAT OR THE OIL",
+ "raw": {
+ "normalized_reference": "no man ate of the seal meat or the oil.",
+ "normalized_hypothesis": "no men ate of the seal meat or the oil",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ },
+ "semantic": {
+ "normalized_reference": "no man ate of the seal meat or the oil",
+ "normalized_hypothesis": "no men ate of the seal meat or the oil",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ }
+ },
+ {
+ "id": "release500_0299",
+ "category": "phonetic_balance",
+ "text": "You're going in for grab sharing.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0299.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "OU RE GOING IN FOR GRABS SHARING",
+ "raw": {
+ "normalized_reference": "you're going in for grab sharing.",
+ "normalized_hypothesis": "ou re going in for grabs sharing",
+ "errors": 3,
+ "words": 6,
+ "wer": 0.5
+ },
+ "semantic": {
+ "normalized_reference": "you are going in for grab sharing",
+ "normalized_hypothesis": "ou re going in for grabs sharing",
+ "errors": 3,
+ "words": 7,
+ "wer": 0.42857142857142855
+ }
+ },
+ {
+ "id": "release500_0300",
+ "category": "phonetic_balance",
+ "text": "He it was that lived to found the family of the Patino.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0300.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT WAS THAT LIVED TO FOUND THE FAMILY OF THE PETINO",
+ "raw": {
+ "normalized_reference": "he it was that lived to found the family of the patino.",
+ "normalized_hypothesis": "it was that lived to found the family of the petino",
+ "errors": 2,
+ "words": 12,
+ "wer": 0.16666666666666666
+ },
+ "semantic": {
+ "normalized_reference": "he it was that lived to found the family of the patino",
+ "normalized_hypothesis": "it was that lived to found the family of the petino",
+ "errors": 2,
+ "words": 12,
+ "wer": 0.16666666666666666
+ }
+ },
+ {
+ "id": "release500_0301",
+ "category": "phonetic_balance",
+ "text": "For such countries nothing remained but reorganization.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0301.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "FOR SUCH COUNTRIES NOTHING REMAINED BUT REORGANIZATION",
+ "raw": {
+ "normalized_reference": "for such countries nothing remained but reorganization.",
+ "normalized_hypothesis": "for such countries nothing remained but reorganization",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "for such countries nothing remained but reorganization",
+ "normalized_hypothesis": "for such countries nothing remained but reorganization",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0302",
+ "category": "phonetic_balance",
+ "text": "Stand off butcher and baker and all the rest.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0302.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "STAND OFF BUTCHER AND BAKER AND ALL THE REST",
+ "raw": {
+ "normalized_reference": "stand off butcher and baker and all the rest.",
+ "normalized_hypothesis": "stand off butcher and baker and all the rest",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "stand off butcher and baker and all the rest",
+ "normalized_hypothesis": "stand off butcher and baker and all the rest",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0303",
+ "category": "phonetic_balance",
+ "text": "All eyes, however, were staring at him in certitude of expectancy.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0303.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "ALL EYES HOWEVER WERE STARING AT HIM IN CERTITUDE OF EXPECTANCY",
+ "raw": {
+ "normalized_reference": "all eyes, however, were staring at him in certitude of expectancy.",
+ "normalized_hypothesis": "all eyes however were staring at him in certitude of expectancy",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "all eyes however were staring at him in certitude of expectancy",
+ "normalized_hypothesis": "all eyes however were staring at him in certitude of expectancy",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0304",
+ "category": "phonetic_balance",
+ "text": "Two days ago Jeanne learned where her father's men were hiding.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0304.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "TWO DAYS AGO JEAN LEARNED WHERE HER FATHER'S MEN WERE HIDING",
+ "raw": {
+ "normalized_reference": "two days ago jeanne learned where her father's men were hiding.",
+ "normalized_hypothesis": "two days ago jean learned where her father's men were hiding",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ },
+ "semantic": {
+ "normalized_reference": "2 days ago jeanne learned where her father is men were hiding",
+ "normalized_hypothesis": "2 days ago jean learned where her father is men were hiding",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ }
+ },
+ {
+ "id": "release500_0305",
+ "category": "phonetic_balance",
+ "text": "His previous wives had never lived long enough to bear him children.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0305.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HIS PREVIOUS WIVES HAD NEVER LIVED LONG ENOUGH TO BEAR HIM CHILDREN",
+ "raw": {
+ "normalized_reference": "his previous wives had never lived long enough to bear him children.",
+ "normalized_hypothesis": "his previous wives had never lived long enough to bear him children",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "his previous wives had never lived long enough to bear him children",
+ "normalized_hypothesis": "his previous wives had never lived long enough to bear him children",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0306",
+ "category": "phonetic_balance",
+ "text": "But this little defect did not worry him.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0306.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "BUT THIS LITTLE DEFECT DID NOT WORRY HIM",
+ "raw": {
+ "normalized_reference": "but this little defect did not worry him.",
+ "normalized_hypothesis": "but this little defect did not worry him",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "but this little defect did not worry him",
+ "normalized_hypothesis": "but this little defect did not worry him",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0307",
+ "category": "phonetic_balance",
+ "text": "Eli Harding asked, as Shunk started to follow.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0307.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "ILY HARDING ASKED AS SHUNK STARTED TO FOLLOW",
+ "raw": {
+ "normalized_reference": "eli harding asked, as shunk started to follow.",
+ "normalized_hypothesis": "ily harding asked as shunk started to follow",
+ "errors": 1,
+ "words": 8,
+ "wer": 0.125
+ },
+ "semantic": {
+ "normalized_reference": "eli harding asked as shunk started to follow",
+ "normalized_hypothesis": "ily harding asked as shunk started to follow",
+ "errors": 1,
+ "words": 8,
+ "wer": 0.125
+ }
+ },
+ {
+ "id": "release500_0308",
+ "category": "phonetic_balance",
+ "text": "And this was their sole conversation throughout the meal.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0308.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "AND THIS WAS THEIR SOLE CONVERSATION THROUGHOUT THE MEAL",
+ "raw": {
+ "normalized_reference": "and this was their sole conversation throughout the meal.",
+ "normalized_hypothesis": "and this was their sole conversation throughout the meal",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "and this was their sole conversation throughout the meal",
+ "normalized_hypothesis": "and this was their sole conversation throughout the meal",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0309",
+ "category": "phonetic_balance",
+ "text": "Then, again, Tudor had such an irritating way about him.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0309.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THEN AGAIN TUTOR HAD SUCH AN IRRITATING WAY ABOUT HIM",
+ "raw": {
+ "normalized_reference": "then, again, tudor had such an irritating way about him.",
+ "normalized_hypothesis": "then again tutor had such an irritating way about him",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ },
+ "semantic": {
+ "normalized_reference": "then again tudor had such an irritating way about him",
+ "normalized_hypothesis": "then again tutor had such an irritating way about him",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ }
+ },
+ {
+ "id": "release500_0310",
+ "category": "phonetic_balance",
+ "text": "In it there was something that was almost tragedy.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0310.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IN IT THERE WAS SOMETHING THAT WAS ALMOST TRAGEDY",
+ "raw": {
+ "normalized_reference": "in it there was something that was almost tragedy.",
+ "normalized_hypothesis": "in it there was something that was almost tragedy",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "in it there was something that was almost tragedy",
+ "normalized_hypothesis": "in it there was something that was almost tragedy",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0311",
+ "category": "phonetic_balance",
+ "text": "Thirty pounds, said the captain with finality.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0311.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THIRTY POUNDS SAID THE CAPTAIN WITH FINALITY",
+ "raw": {
+ "normalized_reference": "thirty pounds, said the captain with finality.",
+ "normalized_hypothesis": "thirty pounds said the captain with finality",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "\u00a330 said the captain with finality",
+ "normalized_hypothesis": "\u00a330 said the captain with finality",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0312",
+ "category": "phonetic_balance",
+ "text": "It was over when he made his way through the ring of spectators.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0312.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT WAS OVER WHEN HE MADE HIS WAY THROUGH THE RING OF SPECTATORS",
+ "raw": {
+ "normalized_reference": "it was over when he made his way through the ring of spectators.",
+ "normalized_hypothesis": "it was over when he made his way through the ring of spectators",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it was over when he made his way through the ring of spectators",
+ "normalized_hypothesis": "it was over when he made his way through the ring of spectators",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0313",
+ "category": "phonetic_balance",
+ "text": "She had died from cold and starvation.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0313.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SHE HAD DIED FROM COLD AND STARVATION",
+ "raw": {
+ "normalized_reference": "she had died from cold and starvation.",
+ "normalized_hypothesis": "she had died from cold and starvation",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "she had died from cold and starvation",
+ "normalized_hypothesis": "she had died from cold and starvation",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0314",
+ "category": "phonetic_balance",
+ "text": "We fished sharks on Niihau together.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0314.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WE FISHED SHARKS ON KNEE HOW TOGETHER",
+ "raw": {
+ "normalized_reference": "we fished sharks on niihau together.",
+ "normalized_hypothesis": "we fished sharks on knee how together",
+ "errors": 2,
+ "words": 6,
+ "wer": 0.3333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "we fished sharks on niihau together",
+ "normalized_hypothesis": "we fished sharks on knee how together",
+ "errors": 2,
+ "words": 6,
+ "wer": 0.3333333333333333
+ }
+ },
+ {
+ "id": "release500_0315",
+ "category": "phonetic_balance",
+ "text": "His slim hands gripped the edges of the table.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0315.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HIS SLIM HANDS GRIPPED THE EDGES OF THE TABLE",
+ "raw": {
+ "normalized_reference": "his slim hands gripped the edges of the table.",
+ "normalized_hypothesis": "his slim hands gripped the edges of the table",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "his slim hands gripped the edges of the table",
+ "normalized_hypothesis": "his slim hands gripped the edges of the table",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0316",
+ "category": "phonetic_balance",
+ "text": "We had been chased by them ourselves, more than once.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0316.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WE HAD BEEN CHASED BY THEM OURSELVES MORE THAN ONCE",
+ "raw": {
+ "normalized_reference": "we had been chased by them ourselves, more than once.",
+ "normalized_hypothesis": "we had been chased by them ourselves more than once",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "we had been chased by them ourselves more than once",
+ "normalized_hypothesis": "we had been chased by them ourselves more than once",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0317",
+ "category": "phonetic_balance",
+ "text": "Or have they already devised one.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0317.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "OR HAVE THEY ALREADY DEVISED ONE",
+ "raw": {
+ "normalized_reference": "or have they already devised one.",
+ "normalized_hypothesis": "or have they already devised one",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "or have they already devised one",
+ "normalized_hypothesis": "or have they already devised one",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0318",
+ "category": "phonetic_balance",
+ "text": "I cannot follow you, she said.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0318.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I CANNOT FOLLOW YOU SHE SAID",
+ "raw": {
+ "normalized_reference": "i cannot follow you, she said.",
+ "normalized_hypothesis": "i cannot follow you she said",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i cannot follow you she said",
+ "normalized_hypothesis": "i cannot follow you she said",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0319",
+ "category": "phonetic_balance",
+ "text": "He went down in midstream, searching the shadows of both shores.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0319.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE WENT DOWN IN MID STREAM SEARCHING THE SHADOWS OF BOTH SHORES",
+ "raw": {
+ "normalized_reference": "he went down in midstream, searching the shadows of both shores.",
+ "normalized_hypothesis": "he went down in mid stream searching the shadows of both shores",
+ "errors": 2,
+ "words": 11,
+ "wer": 0.18181818181818182
+ },
+ "semantic": {
+ "normalized_reference": "he went down in midstream searching the shadows of both shores",
+ "normalized_hypothesis": "he went down in mid stream searching the shadows of both shores",
+ "errors": 2,
+ "words": 11,
+ "wer": 0.18181818181818182
+ }
+ },
+ {
+ "id": "release500_0320",
+ "category": "phonetic_balance",
+ "text": "I was sick once -- typhoid.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0320.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I WAS SICK ONCE TYFO",
+ "raw": {
+ "normalized_reference": "i was sick once -- typhoid.",
+ "normalized_hypothesis": "i was sick once tyfo",
+ "errors": 1,
+ "words": 5,
+ "wer": 0.2
+ },
+ "semantic": {
+ "normalized_reference": "i was sick once typhoid",
+ "normalized_hypothesis": "i was sick once tyfo",
+ "errors": 1,
+ "words": 5,
+ "wer": 0.2
+ }
+ },
+ {
+ "id": "release500_0321",
+ "category": "phonetic_balance",
+ "text": "Why not like any railroad station or ferry depot.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0321.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WHY NOT LIKE ANY RAILROAD STATION OR FERRY DEPOT",
+ "raw": {
+ "normalized_reference": "why not like any railroad station or ferry depot.",
+ "normalized_hypothesis": "why not like any railroad station or ferry depot",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "why not like any railroad station or ferry depot",
+ "normalized_hypothesis": "why not like any railroad station or ferry depot",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0322",
+ "category": "phonetic_balance",
+ "text": "Red-Eye swung back and forth on the branch farther down.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0322.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "RED EYE SWUNG BACK AND FORTH ON THE BRANCH FARTHER DOWN",
+ "raw": {
+ "normalized_reference": "red-eye swung back and forth on the branch farther down.",
+ "normalized_hypothesis": "red eye swung back and forth on the branch farther down",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "red eye swung back and forth on the branch farther down",
+ "normalized_hypothesis": "red eye swung back and forth on the branch farther down",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0323",
+ "category": "phonetic_balance",
+ "text": "Within himself he called it no longer his own.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0323.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WITHIN HIMSELF HE CALLED IT NO LONGER HIS OWN",
+ "raw": {
+ "normalized_reference": "within himself he called it no longer his own.",
+ "normalized_hypothesis": "within himself he called it no longer his own",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "within himself he called it no longer his own",
+ "normalized_hypothesis": "within himself he called it no longer his own",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0324",
+ "category": "phonetic_balance",
+ "text": "It lasted as a deterrent for two days.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0324.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT LASTED AS A DETERRANT FOR TWO DAYS",
+ "raw": {
+ "normalized_reference": "it lasted as a deterrent for two days.",
+ "normalized_hypothesis": "it lasted as a deterrant for two days",
+ "errors": 1,
+ "words": 8,
+ "wer": 0.125
+ },
+ "semantic": {
+ "normalized_reference": "it lasted as a deterrent for 2 days",
+ "normalized_hypothesis": "it lasted as a deterrant for 2 days",
+ "errors": 1,
+ "words": 8,
+ "wer": 0.125
+ }
+ },
+ {
+ "id": "release500_0325",
+ "category": "phonetic_balance",
+ "text": "He loved to play Chinese lottery.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0325.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE LOVED TO PLAY CHINESE LOTTERY",
+ "raw": {
+ "normalized_reference": "he loved to play chinese lottery.",
+ "normalized_hypothesis": "he loved to play chinese lottery",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he loved to play chinese lottery",
+ "normalized_hypothesis": "he loved to play chinese lottery",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0326",
+ "category": "phonetic_balance",
+ "text": "Lord, but I'm glad to see you again, Phil.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0326.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "LORD BUT I'M GLAD TO SEE YOU AGAIN PHIL",
+ "raw": {
+ "normalized_reference": "lord, but i'm glad to see you again, phil.",
+ "normalized_hypothesis": "lord but i'm glad to see you again phil",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "lord but i am glad to see you again phil",
+ "normalized_hypothesis": "lord but i am glad to see you again phil",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0327",
+ "category": "phonetic_balance",
+ "text": "There is not an iota of truth in it, certainly not.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0327.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THERE IS NOT AN IODA OF TRUTH IN IT CERTAINLY NOT",
+ "raw": {
+ "normalized_reference": "there is not an iota of truth in it, certainly not.",
+ "normalized_hypothesis": "there is not an ioda of truth in it certainly not",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ },
+ "semantic": {
+ "normalized_reference": "there is not an iota of truth in it certainly not",
+ "normalized_hypothesis": "there is not an ioda of truth in it certainly not",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ }
+ },
+ {
+ "id": "release500_0328",
+ "category": "phonetic_balance",
+ "text": "She added, with genuine sympathy in her face and voice.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0328.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SHE ADDED WITH GENUINE SYMPATHY IN HER FACE AND VOICE",
+ "raw": {
+ "normalized_reference": "she added, with genuine sympathy in her face and voice.",
+ "normalized_hypothesis": "she added with genuine sympathy in her face and voice",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "she added with genuine sympathy in her face and voice",
+ "normalized_hypothesis": "she added with genuine sympathy in her face and voice",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0329",
+ "category": "phonetic_balance",
+ "text": "We have plenty of capital ourselves, and yet we want more.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0329.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WE HAVE PLENTY OF CAPITAL OURSELVES AND YET WE WANT MORE",
+ "raw": {
+ "normalized_reference": "we have plenty of capital ourselves, and yet we want more.",
+ "normalized_hypothesis": "we have plenty of capital ourselves and yet we want more",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "we have plenty of capital ourselves and yet we want more",
+ "normalized_hypothesis": "we have plenty of capital ourselves and yet we want more",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0330",
+ "category": "phonetic_balance",
+ "text": "And now put yourself in my place for a moment.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0330.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "AND NOW PUT YOURSELF IN MY PLACE FOR A MOMENT",
+ "raw": {
+ "normalized_reference": "and now put yourself in my place for a moment.",
+ "normalized_hypothesis": "and now put yourself in my place for a moment",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "and now put yourself in my place for a moment",
+ "normalized_hypothesis": "and now put yourself in my place for a moment",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0331",
+ "category": "phonetic_balance",
+ "text": "Fast, but endure.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0331.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "PAST BUT IN DOOR",
+ "raw": {
+ "normalized_reference": "fast, but endure.",
+ "normalized_hypothesis": "past but in door",
+ "errors": 3,
+ "words": 3,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "fast but endure",
+ "normalized_hypothesis": "past but in door",
+ "errors": 3,
+ "words": 3,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0332",
+ "category": "phonetic_balance",
+ "text": "You were looking squeamish this afternoon, he began.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0332.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "YOU ARE LOOKING SQUEAMISH THIS AFTERNOON HE BEGAN",
+ "raw": {
+ "normalized_reference": "you were looking squeamish this afternoon, he began.",
+ "normalized_hypothesis": "you are looking squeamish this afternoon he began",
+ "errors": 1,
+ "words": 8,
+ "wer": 0.125
+ },
+ "semantic": {
+ "normalized_reference": "you were looking squeamish this afternoon he began",
+ "normalized_hypothesis": "you are looking squeamish this afternoon he began",
+ "errors": 1,
+ "words": 8,
+ "wer": 0.125
+ }
+ },
+ {
+ "id": "release500_0333",
+ "category": "phonetic_balance",
+ "text": "It is the nearest refuge.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0333.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT IS THE NEAREST REFUGE",
+ "raw": {
+ "normalized_reference": "it is the nearest refuge.",
+ "normalized_hypothesis": "it is the nearest refuge",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it is the nearest refuge",
+ "normalized_hypothesis": "it is the nearest refuge",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0334",
+ "category": "phonetic_balance",
+ "text": "I want to die in it.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0334.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I WANT TO DIE IN IT",
+ "raw": {
+ "normalized_reference": "i want to die in it.",
+ "normalized_hypothesis": "i want to die in it",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i want to die in it",
+ "normalized_hypothesis": "i want to die in it",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0335",
+ "category": "phonetic_balance",
+ "text": "He stopped, and Philip nodded at the horrified question in his eyes.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0335.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE STOPPED AND PHILIP NODDED AT THE HORRIFIED QUESTION IN HIS EYES",
+ "raw": {
+ "normalized_reference": "he stopped, and philip nodded at the horrified question in his eyes.",
+ "normalized_hypothesis": "he stopped and philip nodded at the horrified question in his eyes",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he stopped and philip nodded at the horrified question in his eyes",
+ "normalized_hypothesis": "he stopped and philip nodded at the horrified question in his eyes",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0336",
+ "category": "phonetic_balance",
+ "text": "She turned, fearing that Jacques might see what was in her face.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0336.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SHE TURNED FEARING THAT JACK MIGHT SEE WHAT WAS IN HER FACE",
+ "raw": {
+ "normalized_reference": "she turned, fearing that jacques might see what was in her face.",
+ "normalized_hypothesis": "she turned fearing that jack might see what was in her face",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "she turned fearing that jacques might see what was in her face",
+ "normalized_hypothesis": "she turned fearing that jack might see what was in her face",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ }
+ },
+ {
+ "id": "release500_0337",
+ "category": "phonetic_balance",
+ "text": "It was a curious coincidence.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0337.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT WAS A CURIOUS COINCIDENCE",
+ "raw": {
+ "normalized_reference": "it was a curious coincidence.",
+ "normalized_hypothesis": "it was a curious coincidence",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it was a curious coincidence",
+ "normalized_hypothesis": "it was a curious coincidence",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0338",
+ "category": "phonetic_balance",
+ "text": "Daylight made no answer, and the door closed behind him.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0338.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "DAYLIGHT MADE NO ANSWER AND THE DOOR CLOSED BEHIND HIM",
+ "raw": {
+ "normalized_reference": "daylight made no answer, and the door closed behind him.",
+ "normalized_hypothesis": "daylight made no answer and the door closed behind him",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "daylight made no answer and the door closed behind him",
+ "normalized_hypothesis": "daylight made no answer and the door closed behind him",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0339",
+ "category": "phonetic_balance",
+ "text": "And he did hurt my arm.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0339.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "AND HE DID HURT MY YARN",
+ "raw": {
+ "normalized_reference": "and he did hurt my arm.",
+ "normalized_hypothesis": "and he did hurt my yarn",
+ "errors": 1,
+ "words": 6,
+ "wer": 0.16666666666666666
+ },
+ "semantic": {
+ "normalized_reference": "and he did hurt my arm",
+ "normalized_hypothesis": "and he did hurt my yarn",
+ "errors": 1,
+ "words": 6,
+ "wer": 0.16666666666666666
+ }
+ },
+ {
+ "id": "release500_0340",
+ "category": "phonetic_balance",
+ "text": "Dunham, can your boy go along with Jesse.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0340.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "DUNEM KIN YOUR BOY GO ALONG WITH JESSE",
+ "raw": {
+ "normalized_reference": "dunham, can your boy go along with jesse.",
+ "normalized_hypothesis": "dunem kin your boy go along with jesse",
+ "errors": 2,
+ "words": 8,
+ "wer": 0.25
+ },
+ "semantic": {
+ "normalized_reference": "dunham can your boy go along with jesse",
+ "normalized_hypothesis": "dunem kin your boy go along with jesse",
+ "errors": 2,
+ "words": 8,
+ "wer": 0.25
+ }
+ },
+ {
+ "id": "release500_0341",
+ "category": "phonetic_balance",
+ "text": "The creative joy, I murmured.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0341.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "CREATIVE JOY I MURMUR",
+ "raw": {
+ "normalized_reference": "the creative joy, i murmured.",
+ "normalized_hypothesis": "creative joy i murmur",
+ "errors": 2,
+ "words": 5,
+ "wer": 0.4
+ },
+ "semantic": {
+ "normalized_reference": "the creative joy i murmured",
+ "normalized_hypothesis": "creative joy i murmur",
+ "errors": 2,
+ "words": 5,
+ "wer": 0.4
+ }
+ },
+ {
+ "id": "release500_0342",
+ "category": "phonetic_balance",
+ "text": "He pulled, and the log crashed down to break his back.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0342.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE PULLED AND THE LOG CRASHED DOWN TO BREAK HIS BACK",
+ "raw": {
+ "normalized_reference": "he pulled, and the log crashed down to break his back.",
+ "normalized_hypothesis": "he pulled and the log crashed down to break his back",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he pulled and the log crashed down to break his back",
+ "normalized_hypothesis": "he pulled and the log crashed down to break his back",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0343",
+ "category": "phonetic_balance",
+ "text": "By golly, the boy wins.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0343.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "GOLLY THE BOY WINS",
+ "raw": {
+ "normalized_reference": "by golly, the boy wins.",
+ "normalized_hypothesis": "golly the boy wins",
+ "errors": 1,
+ "words": 5,
+ "wer": 0.2
+ },
+ "semantic": {
+ "normalized_reference": "by golly the boy wins",
+ "normalized_hypothesis": "golly the boy wins",
+ "errors": 1,
+ "words": 5,
+ "wer": 0.2
+ }
+ },
+ {
+ "id": "release500_0344",
+ "category": "phonetic_balance",
+ "text": "The boy threw back his head with pride.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0344.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE BOY THREW BACK HIS HEAD WITH PRIDE",
+ "raw": {
+ "normalized_reference": "the boy threw back his head with pride.",
+ "normalized_hypothesis": "the boy threw back his head with pride",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the boy threw back his head with pride",
+ "normalized_hypothesis": "the boy threw back his head with pride",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0345",
+ "category": "phonetic_balance",
+ "text": "The last one I knew was an overseer.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0345.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE LAST ONE I KNEW WAS AN OVERSEER",
+ "raw": {
+ "normalized_reference": "the last one i knew was an overseer.",
+ "normalized_hypothesis": "the last one i knew was an overseer",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the last one i knew was an overseer",
+ "normalized_hypothesis": "the last one i knew was an overseer",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0346",
+ "category": "phonetic_balance",
+ "text": "Two years ago I gave up civilization for this.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0346.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "TWO YEARS AGO I GAVE UP CIVILIZATION FOR THIS",
+ "raw": {
+ "normalized_reference": "two years ago i gave up civilization for this.",
+ "normalized_hypothesis": "two years ago i gave up civilization for this",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "2 years ago i gave up civilization for this",
+ "normalized_hypothesis": "2 years ago i gave up civilization for this",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0347",
+ "category": "phonetic_balance",
+ "text": "The river bared its bosom, and snorting steamboats challenged the wilderness.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0347.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE RIVER BARED ITS BOSOM AND SNORTING STEAMBOATS CHALLENGED THE WILDERNESS",
+ "raw": {
+ "normalized_reference": "the river bared its bosom, and snorting steamboats challenged the wilderness.",
+ "normalized_hypothesis": "the river bared its bosom and snorting steamboats challenged the wilderness",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the river bared its bosom and snorting steamboats challenged the wilderness",
+ "normalized_hypothesis": "the river bared its bosom and snorting steamboats challenged the wilderness",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0348",
+ "category": "phonetic_balance",
+ "text": "Mab, she said.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0348.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "AV SHE SAID",
+ "raw": {
+ "normalized_reference": "mab, she said.",
+ "normalized_hypothesis": "av she said",
+ "errors": 1,
+ "words": 3,
+ "wer": 0.3333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "mab she said",
+ "normalized_hypothesis": "av she said",
+ "errors": 1,
+ "words": 3,
+ "wer": 0.3333333333333333
+ }
+ },
+ {
+ "id": "release500_0349",
+ "category": "phonetic_balance",
+ "text": "The regret in his voice was provocative of a second burst of laughter.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0349.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "REGRET IN HIS VOICE WAS PROVOCATIVE OF A SECOND BURST OF LAUGHTER",
+ "raw": {
+ "normalized_reference": "the regret in his voice was provocative of a second burst of laughter.",
+ "normalized_hypothesis": "regret in his voice was provocative of a second burst of laughter",
+ "errors": 1,
+ "words": 13,
+ "wer": 0.07692307692307693
+ },
+ "semantic": {
+ "normalized_reference": "the regret in his voice was provocative of a 2nd burst of laughter",
+ "normalized_hypothesis": "regret in his voice was provocative of a 2nd burst of laughter",
+ "errors": 1,
+ "words": 13,
+ "wer": 0.07692307692307693
+ }
+ },
+ {
+ "id": "release500_0350",
+ "category": "phonetic_balance",
+ "text": "And each year something happened, and I did not go.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0350.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "AND EACH YEAR SOMETHING HAPPENED AND I DID NOT GO",
+ "raw": {
+ "normalized_reference": "and each year something happened, and i did not go.",
+ "normalized_hypothesis": "and each year something happened and i did not go",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "and each year something happened and i did not go",
+ "normalized_hypothesis": "and each year something happened and i did not go",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0351",
+ "category": "phonetic_balance",
+ "text": "Gad, your letter came just in time.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0351.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "GAD YOUR LETTER CAME JUST IN TIME",
+ "raw": {
+ "normalized_reference": "gad, your letter came just in time.",
+ "normalized_hypothesis": "gad your letter came just in time",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "gad your letter came just in time",
+ "normalized_hypothesis": "gad your letter came just in time",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0352",
+ "category": "phonetic_balance",
+ "text": "But if Pierre did not return, until tomorrow.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0352.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "BUT IF PIERRE DID NOT RETURN UNTIL TO MORROW",
+ "raw": {
+ "normalized_reference": "but if pierre did not return, until tomorrow.",
+ "normalized_hypothesis": "but if pierre did not return until to morrow",
+ "errors": 2,
+ "words": 8,
+ "wer": 0.25
+ },
+ "semantic": {
+ "normalized_reference": "but if pierre did not return until tomorrow",
+ "normalized_hypothesis": "but if pierre did not return until to morrow",
+ "errors": 2,
+ "words": 8,
+ "wer": 0.25
+ }
+ },
+ {
+ "id": "release500_0353",
+ "category": "phonetic_balance",
+ "text": "Billinger may arrive in time.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0353.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "ILLINGER MAY ARRIVE IN TIME",
+ "raw": {
+ "normalized_reference": "billinger may arrive in time.",
+ "normalized_hypothesis": "illinger may arrive in time",
+ "errors": 1,
+ "words": 5,
+ "wer": 0.2
+ },
+ "semantic": {
+ "normalized_reference": "billinger may arrive in time",
+ "normalized_hypothesis": "illinger may arrive in time",
+ "errors": 1,
+ "words": 5,
+ "wer": 0.2
+ }
+ },
+ {
+ "id": "release500_0354",
+ "category": "phonetic_balance",
+ "text": "In such a tumbling of values was no time to sell.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0354.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IN SUCH A TUMBLING OF VALUES WAS NO TIME TO SELL",
+ "raw": {
+ "normalized_reference": "in such a tumbling of values was no time to sell.",
+ "normalized_hypothesis": "in such a tumbling of values was no time to sell",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "in such a tumbling of values was no time to sell",
+ "normalized_hypothesis": "in such a tumbling of values was no time to sell",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0355",
+ "category": "phonetic_balance",
+ "text": "There's not an iota of truth in it.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0355.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HERE IS NOT AN IODA OF TRUTH IN IT",
+ "raw": {
+ "normalized_reference": "there's not an iota of truth in it.",
+ "normalized_hypothesis": "here is not an ioda of truth in it",
+ "errors": 3,
+ "words": 8,
+ "wer": 0.375
+ },
+ "semantic": {
+ "normalized_reference": "there is not an iota of truth in it",
+ "normalized_hypothesis": "here is not an ioda of truth in it",
+ "errors": 2,
+ "words": 9,
+ "wer": 0.2222222222222222
+ }
+ },
+ {
+ "id": "release500_0356",
+ "category": "phonetic_balance",
+ "text": "It is dog eat dog, and you ate them up.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0356.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT IS DOG EA DOG AND YOU EATE THE",
+ "raw": {
+ "normalized_reference": "it is dog eat dog, and you ate them up.",
+ "normalized_hypothesis": "it is dog ea dog and you eate the",
+ "errors": 4,
+ "words": 10,
+ "wer": 0.4
+ },
+ "semantic": {
+ "normalized_reference": "it is dog eat dog and you ate them up",
+ "normalized_hypothesis": "it is dog ea dog and you eate the",
+ "errors": 4,
+ "words": 10,
+ "wer": 0.4
+ }
+ },
+ {
+ "id": "release500_0357",
+ "category": "phonetic_balance",
+ "text": "You can take a vacation on pay.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0357.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "YOU CAN TAKE A VACATION ON PER",
+ "raw": {
+ "normalized_reference": "you can take a vacation on pay.",
+ "normalized_hypothesis": "you can take a vacation on per",
+ "errors": 1,
+ "words": 7,
+ "wer": 0.14285714285714285
+ },
+ "semantic": {
+ "normalized_reference": "you can take a vacation on pay",
+ "normalized_hypothesis": "you can take a vacation on per",
+ "errors": 1,
+ "words": 7,
+ "wer": 0.14285714285714285
+ }
+ },
+ {
+ "id": "release500_0358",
+ "category": "phonetic_balance",
+ "text": "I have seen myself that one man contemplated by Pascal's philosophic eye.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0358.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I HAVE SEEN MYSELF THAT WHEN MAN CONTEMPLATED BY PASCELL'S PHILOSOPHIC EYE",
+ "raw": {
+ "normalized_reference": "i have seen myself that one man contemplated by pascal's philosophic eye.",
+ "normalized_hypothesis": "i have seen myself that when man contemplated by pascell's philosophic eye",
+ "errors": 2,
+ "words": 12,
+ "wer": 0.16666666666666666
+ },
+ "semantic": {
+ "normalized_reference": "i have seen myself that one man contemplated by pascal is philosophic eye",
+ "normalized_hypothesis": "i have seen myself that when man contemplated by pascell is philosophic eye",
+ "errors": 2,
+ "words": 13,
+ "wer": 0.15384615384615385
+ }
+ },
+ {
+ "id": "release500_0359",
+ "category": "phonetic_balance",
+ "text": "Ah, I had forgotten, he exclaimed.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0359.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "AH I HAD FORGOTTEN HE EXCLAIMED",
+ "raw": {
+ "normalized_reference": "ah, i had forgotten, he exclaimed.",
+ "normalized_hypothesis": "ah i had forgotten he exclaimed",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "ah i had forgotten he exclaimed",
+ "normalized_hypothesis": "ah i had forgotten he exclaimed",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0360",
+ "category": "phonetic_balance",
+ "text": "Providence had delivered him through the maelstrom.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0360.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "PROVIDENCE HAD DELIVERED HIM THROUGH THE MAILSTRONE",
+ "raw": {
+ "normalized_reference": "providence had delivered him through the maelstrom.",
+ "normalized_hypothesis": "providence had delivered him through the mailstrone",
+ "errors": 1,
+ "words": 7,
+ "wer": 0.14285714285714285
+ },
+ "semantic": {
+ "normalized_reference": "providence had delivered him through the maelstrom",
+ "normalized_hypothesis": "providence had delivered him through the mailstrone",
+ "errors": 1,
+ "words": 7,
+ "wer": 0.14285714285714285
+ }
+ },
+ {
+ "id": "release500_0361",
+ "category": "phonetic_balance",
+ "text": "She was his now, forever.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0361.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SHE WAS HIS NOW FOR EVER",
+ "raw": {
+ "normalized_reference": "she was his now, forever.",
+ "normalized_hypothesis": "she was his now for ever",
+ "errors": 2,
+ "words": 5,
+ "wer": 0.4
+ },
+ "semantic": {
+ "normalized_reference": "she was his now forever",
+ "normalized_hypothesis": "she was his now for ever",
+ "errors": 2,
+ "words": 5,
+ "wer": 0.4
+ }
+ },
+ {
+ "id": "release500_0362",
+ "category": "phonetic_balance",
+ "text": "It was unobtrusive, yet it was there.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0362.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT WAS UNOBTRUSIVE YET IT WAS THERE",
+ "raw": {
+ "normalized_reference": "it was unobtrusive, yet it was there.",
+ "normalized_hypothesis": "it was unobtrusive yet it was there",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it was unobtrusive yet it was there",
+ "normalized_hypothesis": "it was unobtrusive yet it was there",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0363",
+ "category": "phonetic_balance",
+ "text": "Men like Joe Goose dated existence from drunk to drunk.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0363.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "MEN LIKE JOE GOOSE DATED EXISTENCE FROM DRUNK TO DRUNK",
+ "raw": {
+ "normalized_reference": "men like joe goose dated existence from drunk to drunk.",
+ "normalized_hypothesis": "men like joe goose dated existence from drunk to drunk",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "men like joe goose dated existence from drunk to drunk",
+ "normalized_hypothesis": "men like joe goose dated existence from drunk to drunk",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0364",
+ "category": "phonetic_balance",
+ "text": "I was completely lost in my work.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0364.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WAS COMPLETELY LOST IN MY WORK",
+ "raw": {
+ "normalized_reference": "i was completely lost in my work.",
+ "normalized_hypothesis": "was completely lost in my work",
+ "errors": 1,
+ "words": 7,
+ "wer": 0.14285714285714285
+ },
+ "semantic": {
+ "normalized_reference": "i was completely lost in my work",
+ "normalized_hypothesis": "was completely lost in my work",
+ "errors": 1,
+ "words": 7,
+ "wer": 0.14285714285714285
+ }
+ },
+ {
+ "id": "release500_0365",
+ "category": "phonetic_balance",
+ "text": "The very thing, Ernest agreed.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0365.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE VERY THING ERNEST AGREED",
+ "raw": {
+ "normalized_reference": "the very thing, ernest agreed.",
+ "normalized_hypothesis": "the very thing ernest agreed",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the very thing ernest agreed",
+ "normalized_hypothesis": "the very thing ernest agreed",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0366",
+ "category": "phonetic_balance",
+ "text": "He was the leader, and Tudor was his lieutenant.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0366.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE WAS THE LEADER AND TUDOR WAS HIS LIEUTENANT",
+ "raw": {
+ "normalized_reference": "he was the leader, and tudor was his lieutenant.",
+ "normalized_hypothesis": "he was the leader and tudor was his lieutenant",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he was the leader and tudor was his lieutenant",
+ "normalized_hypothesis": "he was the leader and tudor was his lieutenant",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0367",
+ "category": "phonetic_balance",
+ "text": "The Fire-Men wore animal skins around their waists and across their shoulders.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0367.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE FIREMEN WORE ANIMAL SKINS AROUND THEIR WAISTS AND ACROSS THEIR SHOULDERS",
+ "raw": {
+ "normalized_reference": "the fire-men wore animal skins around their waists and across their shoulders.",
+ "normalized_hypothesis": "the firemen wore animal skins around their waists and across their shoulders",
+ "errors": 2,
+ "words": 13,
+ "wer": 0.15384615384615385
+ },
+ "semantic": {
+ "normalized_reference": "the fire men wore animal skins around their waists and across their shoulders",
+ "normalized_hypothesis": "the firemen wore animal skins around their waists and across their shoulders",
+ "errors": 2,
+ "words": 13,
+ "wer": 0.15384615384615385
+ }
+ },
+ {
+ "id": "release500_0368",
+ "category": "phonetic_balance",
+ "text": "We could throw stones with our feet.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0368.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WE COULD THROW STONES WITH OUR FEET",
+ "raw": {
+ "normalized_reference": "we could throw stones with our feet.",
+ "normalized_hypothesis": "we could throw stones with our feet",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "we could throw stones with our feet",
+ "normalized_hypothesis": "we could throw stones with our feet",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0369",
+ "category": "phonetic_balance",
+ "text": "He turned sharply, and faced Gregson across the table.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0369.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE TURNED SHARPLY AND FACED GREGSON ACROSS THE TABLE",
+ "raw": {
+ "normalized_reference": "he turned sharply, and faced gregson across the table.",
+ "normalized_hypothesis": "he turned sharply and faced gregson across the table",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he turned sharply and faced gregson across the table",
+ "normalized_hypothesis": "he turned sharply and faced gregson across the table",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0370",
+ "category": "phonetic_balance",
+ "text": "They are to attack your camp tomorrow night.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0370.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THEY ARE TO ATTACK YOUR CAMP TO MORROW NIGHT",
+ "raw": {
+ "normalized_reference": "they are to attack your camp tomorrow night.",
+ "normalized_hypothesis": "they are to attack your camp to morrow night",
+ "errors": 2,
+ "words": 8,
+ "wer": 0.25
+ },
+ "semantic": {
+ "normalized_reference": "they are to attack your camp tomorrow night",
+ "normalized_hypothesis": "they are to attack your camp to morrow night",
+ "errors": 2,
+ "words": 8,
+ "wer": 0.25
+ }
+ },
+ {
+ "id": "release500_0371",
+ "category": "phonetic_balance",
+ "text": "And watch out for wet feet, was his parting advice.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0371.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "AND WATCH OUT FOR WET FEET WAS HIS PARTING ADVICE",
+ "raw": {
+ "normalized_reference": "and watch out for wet feet, was his parting advice.",
+ "normalized_hypothesis": "and watch out for wet feet was his parting advice",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "and watch out for wet feet was his parting advice",
+ "normalized_hypothesis": "and watch out for wet feet was his parting advice",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0372",
+ "category": "phonetic_balance",
+ "text": "Two weeks passed, and in that time Thorpe left camp three times.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0372.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "TWO WEEKS PASSED AND IN THAT TIME THORPE BLUFF CAMPED THREE TIMES",
+ "raw": {
+ "normalized_reference": "two weeks passed, and in that time thorpe left camp three times.",
+ "normalized_hypothesis": "two weeks passed and in that time thorpe bluff camped three times",
+ "errors": 2,
+ "words": 12,
+ "wer": 0.16666666666666666
+ },
+ "semantic": {
+ "normalized_reference": "2 weeks passed and in that time thorpe left camp 3 times",
+ "normalized_hypothesis": "2 weeks passed and in that time thorpe bluff camped 3 times",
+ "errors": 2,
+ "words": 12,
+ "wer": 0.16666666666666666
+ }
+ },
+ {
+ "id": "release500_0373",
+ "category": "phonetic_balance",
+ "text": "But ever his gaze returned to that Crouched Venus on the piano.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0373.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "BUTEVER HIS GAZE RETURNED TO THAT CROUCHED VENUS ON THE PIANO",
+ "raw": {
+ "normalized_reference": "but ever his gaze returned to that crouched venus on the piano.",
+ "normalized_hypothesis": "butever his gaze returned to that crouched venus on the piano",
+ "errors": 2,
+ "words": 12,
+ "wer": 0.16666666666666666
+ },
+ "semantic": {
+ "normalized_reference": "but ever his gaze returned to that crouched venus on the piano",
+ "normalized_hypothesis": "butever his gaze returned to that crouched venus on the piano",
+ "errors": 2,
+ "words": 12,
+ "wer": 0.16666666666666666
+ }
+ },
+ {
+ "id": "release500_0374",
+ "category": "phonetic_balance",
+ "text": "But she had become an automaton.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0374.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SHE HAD BECOME AN AUTOMATON",
+ "raw": {
+ "normalized_reference": "but she had become an automaton.",
+ "normalized_hypothesis": "she had become an automaton",
+ "errors": 1,
+ "words": 6,
+ "wer": 0.16666666666666666
+ },
+ "semantic": {
+ "normalized_reference": "but she had become an automaton",
+ "normalized_hypothesis": "she had become an automaton",
+ "errors": 1,
+ "words": 6,
+ "wer": 0.16666666666666666
+ }
+ },
+ {
+ "id": "release500_0375",
+ "category": "phonetic_balance",
+ "text": "I do not blame you for anything; remember that.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0375.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I DO NOT BLAME YOU FOR ANYTHING REMEMBER THAT",
+ "raw": {
+ "normalized_reference": "i do not blame you for anything; remember that.",
+ "normalized_hypothesis": "i do not blame you for anything remember that",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i do not blame you for anything remember that",
+ "normalized_hypothesis": "i do not blame you for anything remember that",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0376",
+ "category": "phonetic_balance",
+ "text": "The next thing to watch out for is bed sores.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0376.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "NEXT THING TO WATCH OUT FOR HIS BED SORES",
+ "raw": {
+ "normalized_reference": "the next thing to watch out for is bed sores.",
+ "normalized_hypothesis": "next thing to watch out for his bed sores",
+ "errors": 2,
+ "words": 10,
+ "wer": 0.2
+ },
+ "semantic": {
+ "normalized_reference": "the next thing to watch out for is bed sores",
+ "normalized_hypothesis": "next thing to watch out for his bed sores",
+ "errors": 2,
+ "words": 10,
+ "wer": 0.2
+ }
+ },
+ {
+ "id": "release500_0377",
+ "category": "phonetic_balance",
+ "text": "He will follow us soon.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0377.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE WILL FOLLOW US SOON",
+ "raw": {
+ "normalized_reference": "he will follow us soon.",
+ "normalized_hypothesis": "he will follow us soon",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he will follow us soon",
+ "normalized_hypothesis": "he will follow us soon",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0378",
+ "category": "phonetic_balance",
+ "text": "His beady black eyes saw bargains where other men saw bankruptcy.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0378.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HIS BEADY BLACK EYES SAW BARGAINS WHERE OTHER MEN SAW BANKRUPTCY",
+ "raw": {
+ "normalized_reference": "his beady black eyes saw bargains where other men saw bankruptcy.",
+ "normalized_hypothesis": "his beady black eyes saw bargains where other men saw bankruptcy",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "his beady black eyes saw bargains where other men saw bankruptcy",
+ "normalized_hypothesis": "his beady black eyes saw bargains where other men saw bankruptcy",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0379",
+ "category": "phonetic_balance",
+ "text": "They robbed me a few years later.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0379.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THEY ROBBED ME A FEW YEARS LATER",
+ "raw": {
+ "normalized_reference": "they robbed me a few years later.",
+ "normalized_hypothesis": "they robbed me a few years later",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "they robbed me a few years later",
+ "normalized_hypothesis": "they robbed me a few years later",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0380",
+ "category": "phonetic_balance",
+ "text": "Gregson held a lighted match until it burnt his fingertips.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0380.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "GREGEN HELD A LIGHTED MATCH UNTIL IT BURNT HIS FINGER TIPS",
+ "raw": {
+ "normalized_reference": "gregson held a lighted match until it burnt his fingertips.",
+ "normalized_hypothesis": "gregen held a lighted match until it burnt his finger tips",
+ "errors": 3,
+ "words": 10,
+ "wer": 0.3
+ },
+ "semantic": {
+ "normalized_reference": "gregson held a lighted match until it burnt his fingertips",
+ "normalized_hypothesis": "gregen held a lighted match until it burnt his finger tips",
+ "errors": 3,
+ "words": 10,
+ "wer": 0.3
+ }
+ },
+ {
+ "id": "release500_0381",
+ "category": "phonetic_balance",
+ "text": "A burst of laughter was his reward.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0381.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "A BURST OF LAUGHTER WAS HIS REWARD",
+ "raw": {
+ "normalized_reference": "a burst of laughter was his reward.",
+ "normalized_hypothesis": "a burst of laughter was his reward",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "a burst of laughter was his reward",
+ "normalized_hypothesis": "a burst of laughter was his reward",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0382",
+ "category": "phonetic_balance",
+ "text": "He cried, and swung the club wildly.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0382.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE CRYED AND SWUNG THE CLUB WILDLY",
+ "raw": {
+ "normalized_reference": "he cried, and swung the club wildly.",
+ "normalized_hypothesis": "he cryed and swung the club wildly",
+ "errors": 1,
+ "words": 7,
+ "wer": 0.14285714285714285
+ },
+ "semantic": {
+ "normalized_reference": "he cried and swung the club wildly",
+ "normalized_hypothesis": "he cryed and swung the club wildly",
+ "errors": 1,
+ "words": 7,
+ "wer": 0.14285714285714285
+ }
+ },
+ {
+ "id": "release500_0383",
+ "category": "phonetic_balance",
+ "text": "Sandel would never become a world champion.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0383.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SANDAL WOULD NEVER BECOME A WORLD CHAMPION",
+ "raw": {
+ "normalized_reference": "sandel would never become a world champion.",
+ "normalized_hypothesis": "sandal would never become a world champion",
+ "errors": 1,
+ "words": 7,
+ "wer": 0.14285714285714285
+ },
+ "semantic": {
+ "normalized_reference": "sandel would never become a world champion",
+ "normalized_hypothesis": "sandal would never become a world champion",
+ "errors": 1,
+ "words": 7,
+ "wer": 0.14285714285714285
+ }
+ },
+ {
+ "id": "release500_0384",
+ "category": "phonetic_balance",
+ "text": "And after the bath a shave would not be bad.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0384.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "AND AFTER THE BATH A SHAVE WOULD NOT BE BAD",
+ "raw": {
+ "normalized_reference": "and after the bath a shave would not be bad.",
+ "normalized_hypothesis": "and after the bath a shave would not be bad",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "and after the bath a shave would not be bad",
+ "normalized_hypothesis": "and after the bath a shave would not be bad",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0385",
+ "category": "phonetic_balance",
+ "text": "They saw each other for the first time in Boston.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0385.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THEY SAW EACH OTHER FOR THE FIRST TIME IN BOSTON",
+ "raw": {
+ "normalized_reference": "they saw each other for the first time in boston.",
+ "normalized_hypothesis": "they saw each other for the first time in boston",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "they saw each other for the 1st time in boston",
+ "normalized_hypothesis": "they saw each other for the 1st time in boston",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0386",
+ "category": "phonetic_balance",
+ "text": "I had been born with no organic, chemical predisposition toward alcohol.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0386.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I HAD BEEN BORN WITH NO ORGANIC CHEMICAL PREDISPOSITION TOWARD ALCOHOL",
+ "raw": {
+ "normalized_reference": "i had been born with no organic, chemical predisposition toward alcohol.",
+ "normalized_hypothesis": "i had been born with no organic chemical predisposition toward alcohol",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i had been born with no organic chemical predisposition toward alcohol",
+ "normalized_hypothesis": "i had been born with no organic chemical predisposition toward alcohol",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0387",
+ "category": "phonetic_balance",
+ "text": "They had no fixed values, to be altered by adjectives and adverbs.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0387.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THEY HAD NO FIXED VALUES TO BE ALTERED BY ADJECTIVES AND ADVERBS",
+ "raw": {
+ "normalized_reference": "they had no fixed values, to be altered by adjectives and adverbs.",
+ "normalized_hypothesis": "they had no fixed values to be altered by adjectives and adverbs",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "they had no fixed values to be altered by adjectives and adverbs",
+ "normalized_hypothesis": "they had no fixed values to be altered by adjectives and adverbs",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0388",
+ "category": "phonetic_balance",
+ "text": "Nor was Elam Harnish an exception.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0388.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "NOR WAS A LAMB HARNISH AN EXCEPTION",
+ "raw": {
+ "normalized_reference": "nor was elam harnish an exception.",
+ "normalized_hypothesis": "nor was a lamb harnish an exception",
+ "errors": 2,
+ "words": 6,
+ "wer": 0.3333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "nor was elam harnish an exception",
+ "normalized_hypothesis": "nor was a lamb harnish an exception",
+ "errors": 2,
+ "words": 6,
+ "wer": 0.3333333333333333
+ }
+ },
+ {
+ "id": "release500_0389",
+ "category": "phonetic_balance",
+ "text": "He was a wise hyena.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0389.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE WAS A WISE HYENA",
+ "raw": {
+ "normalized_reference": "he was a wise hyena.",
+ "normalized_hypothesis": "he was a wise hyena",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he was a wise hyena",
+ "normalized_hypothesis": "he was a wise hyena",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0390",
+ "category": "phonetic_balance",
+ "text": "And now behold the perversity of things.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0390.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "AND NOW BEHOLD THE PERVERSITY OF THINGS",
+ "raw": {
+ "normalized_reference": "and now behold the perversity of things.",
+ "normalized_hypothesis": "and now behold the perversity of things",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "and now behold the perversity of things",
+ "normalized_hypothesis": "and now behold the perversity of things",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0391",
+ "category": "phonetic_balance",
+ "text": "The Fire People, like ourselves, lived in caves.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0391.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "FIRE PEOPLE LIKE OURSELVES LIVED IN CAVES",
+ "raw": {
+ "normalized_reference": "the fire people, like ourselves, lived in caves.",
+ "normalized_hypothesis": "fire people like ourselves lived in caves",
+ "errors": 1,
+ "words": 8,
+ "wer": 0.125
+ },
+ "semantic": {
+ "normalized_reference": "the fire people like ourselves lived in caves",
+ "normalized_hypothesis": "fire people like ourselves lived in caves",
+ "errors": 1,
+ "words": 8,
+ "wer": 0.125
+ }
+ },
+ {
+ "id": "release500_0392",
+ "category": "phonetic_balance",
+ "text": "To her the bridge was tambo, which is the native for taboo.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0392.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "TO HER THE BRIDGE WAS TAMBO WHICH IS THE NATIVE FOR TABOO",
+ "raw": {
+ "normalized_reference": "to her the bridge was tambo, which is the native for taboo.",
+ "normalized_hypothesis": "to her the bridge was tambo which is the native for taboo",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "to her the bridge was tambo which is the native for taboo",
+ "normalized_hypothesis": "to her the bridge was tambo which is the native for taboo",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0393",
+ "category": "phonetic_balance",
+ "text": "The voyage was our idea of a good time.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0393.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE VOYAGE WAS OUR IDEA OF A GOOD TIME",
+ "raw": {
+ "normalized_reference": "the voyage was our idea of a good time.",
+ "normalized_hypothesis": "the voyage was our idea of a good time",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the voyage was our idea of a good time",
+ "normalized_hypothesis": "the voyage was our idea of a good time",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0394",
+ "category": "phonetic_balance",
+ "text": "Of course much grumbling went on, and little outbursts were continually occurring.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0394.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "OF COURSE MUCH GRUMBLING WENT ON AND LITTLE OUTBURSTS WERE CONTINUALLY OCCURRING",
+ "raw": {
+ "normalized_reference": "of course much grumbling went on, and little outbursts were continually occurring.",
+ "normalized_hypothesis": "of course much grumbling went on and little outbursts were continually occurring",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "of course much grumbling went on and little outbursts were continually occurring",
+ "normalized_hypothesis": "of course much grumbling went on and little outbursts were continually occurring",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0395",
+ "category": "phonetic_balance",
+ "text": "Get down and dig in.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0395.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "T DOWN AND DIG IN",
+ "raw": {
+ "normalized_reference": "get down and dig in.",
+ "normalized_hypothesis": "t down and dig in",
+ "errors": 1,
+ "words": 5,
+ "wer": 0.2
+ },
+ "semantic": {
+ "normalized_reference": "get down and dig in",
+ "normalized_hypothesis": "t down and dig in",
+ "errors": 1,
+ "words": 5,
+ "wer": 0.2
+ }
+ },
+ {
+ "id": "release500_0396",
+ "category": "phonetic_balance",
+ "text": "It was put together in a casual, helter-skelter sort of way.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0396.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT WAS PUT TOGETHER IN A CASUAL HELTER SKELTER SORT OF WAY",
+ "raw": {
+ "normalized_reference": "it was put together in a casual, helter-skelter sort of way.",
+ "normalized_hypothesis": "it was put together in a casual helter skelter sort of way",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it was put together in a casual helter skelter sort of way",
+ "normalized_hypothesis": "it was put together in a casual helter skelter sort of way",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0397",
+ "category": "phonetic_balance",
+ "text": "He deluged me, overwhelmed me with argument.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0397.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE DELUGED ME OVERWHELMED ME WITH ARGUMENT",
+ "raw": {
+ "normalized_reference": "he deluged me, overwhelmed me with argument.",
+ "normalized_hypothesis": "he deluged me overwhelmed me with argument",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he deluged me overwhelmed me with argument",
+ "normalized_hypothesis": "he deluged me overwhelmed me with argument",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0398",
+ "category": "phonetic_balance",
+ "text": "The girl faced him, her eyes shining with sudden fear.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0398.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE GIRL FACED HIM HER EYES SHINING WITH SUDDEN FEAR",
+ "raw": {
+ "normalized_reference": "the girl faced him, her eyes shining with sudden fear.",
+ "normalized_hypothesis": "the girl faced him her eyes shining with sudden fear",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the girl faced him her eyes shining with sudden fear",
+ "normalized_hypothesis": "the girl faced him her eyes shining with sudden fear",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0399",
+ "category": "phonetic_balance",
+ "text": "It will break our hearts and our backs to hoist anchor by hand.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0399.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT WILL BREAK OUR HEARTS AND OUR BACKS TO HOIST ANCHOR BY HAND",
+ "raw": {
+ "normalized_reference": "it will break our hearts and our backs to hoist anchor by hand.",
+ "normalized_hypothesis": "it will break our hearts and our backs to hoist anchor by hand",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it will break our hearts and our backs to hoist anchor by hand",
+ "normalized_hypothesis": "it will break our hearts and our backs to hoist anchor by hand",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0400",
+ "category": "phonetic_balance",
+ "text": "He made no reply as he waited for Whittemore to continue.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0400.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE MADE NO REPLY AS HE WAITED FOR WHITTAMORE TO CONTINUE",
+ "raw": {
+ "normalized_reference": "he made no reply as he waited for whittemore to continue.",
+ "normalized_hypothesis": "he made no reply as he waited for whittamore to continue",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ },
+ "semantic": {
+ "normalized_reference": "he made no reply as he waited for whittemore to continue",
+ "normalized_hypothesis": "he made no reply as he waited for whittamore to continue",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ }
+ },
+ {
+ "id": "release500_0401",
+ "category": "phonetic_balance",
+ "text": "Who the devil gave it to you to be judge and jury.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0401.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HOTHE DEVIL GAVE IT TO YOU TO BE JUDGE AND JURY",
+ "raw": {
+ "normalized_reference": "who the devil gave it to you to be judge and jury.",
+ "normalized_hypothesis": "hothe devil gave it to you to be judge and jury",
+ "errors": 2,
+ "words": 12,
+ "wer": 0.16666666666666666
+ },
+ "semantic": {
+ "normalized_reference": "who the devil gave it to you to be judge and jury",
+ "normalized_hypothesis": "hothe devil gave it to you to be judge and jury",
+ "errors": 2,
+ "words": 12,
+ "wer": 0.16666666666666666
+ }
+ },
+ {
+ "id": "release500_0402",
+ "category": "phonetic_balance",
+ "text": "Daylight found himself charmed and made curious by the little man.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0402.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "DAYLIGHT FOUND HIMSELF CHARMED AND MADE CURIOUS BY THE LITTLE MAN",
+ "raw": {
+ "normalized_reference": "daylight found himself charmed and made curious by the little man.",
+ "normalized_hypothesis": "daylight found himself charmed and made curious by the little man",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "daylight found himself charmed and made curious by the little man",
+ "normalized_hypothesis": "daylight found himself charmed and made curious by the little man",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0403",
+ "category": "phonetic_balance",
+ "text": "It's a Yankee, Joan cried.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0403.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT'S A YANKEE JOAN CRIED",
+ "raw": {
+ "normalized_reference": "it's a yankee, joan cried.",
+ "normalized_hypothesis": "it's a yankee joan cried",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it is a yankee joan cried",
+ "normalized_hypothesis": "it is a yankee joan cried",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0404",
+ "category": "phonetic_balance",
+ "text": "But what they want with your toothbrush is more than I can imagine.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0404.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WHAT THEY WANT WITH YOUR TOOTH BRISH IS MORE THAN I CAN IMAGINE",
+ "raw": {
+ "normalized_reference": "but what they want with your toothbrush is more than i can imagine.",
+ "normalized_hypothesis": "what they want with your tooth brish is more than i can imagine",
+ "errors": 3,
+ "words": 13,
+ "wer": 0.23076923076923078
+ },
+ "semantic": {
+ "normalized_reference": "but what they want with your toothbrush is more than i can imagine",
+ "normalized_hypothesis": "what they want with your tooth brish is more than i can imagine",
+ "errors": 3,
+ "words": 13,
+ "wer": 0.23076923076923078
+ }
+ },
+ {
+ "id": "release500_0405",
+ "category": "phonetic_balance",
+ "text": "They were babbling and chattering all together.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0405.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE WERE BABBLING AND CHATTERING ALL TOGETHER",
+ "raw": {
+ "normalized_reference": "they were babbling and chattering all together.",
+ "normalized_hypothesis": "he were babbling and chattering all together",
+ "errors": 1,
+ "words": 7,
+ "wer": 0.14285714285714285
+ },
+ "semantic": {
+ "normalized_reference": "they were babbling and chattering all together",
+ "normalized_hypothesis": "he were babbling and chattering all together",
+ "errors": 1,
+ "words": 7,
+ "wer": 0.14285714285714285
+ }
+ },
+ {
+ "id": "release500_0406",
+ "category": "phonetic_balance",
+ "text": "Men of Selden's stamp don't stop at women and children.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0406.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "MEN OF SELDOM'S STAMP DON'T STOP AT WOMEN AND CHILDREN",
+ "raw": {
+ "normalized_reference": "men of selden's stamp don't stop at women and children.",
+ "normalized_hypothesis": "men of seldom's stamp don't stop at women and children",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ },
+ "semantic": {
+ "normalized_reference": "men of selden is stamp do not stop at women and children",
+ "normalized_hypothesis": "men of seldom is stamp do not stop at women and children",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ }
+ },
+ {
+ "id": "release500_0407",
+ "category": "phonetic_balance",
+ "text": "Cameron looked at his hands with their long, sinewy fingers.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0407.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "CAMERON LOOKED AT HIS HANDS WITH THEIR LONG SINEWY FINGERS",
+ "raw": {
+ "normalized_reference": "cameron looked at his hands with their long, sinewy fingers.",
+ "normalized_hypothesis": "cameron looked at his hands with their long sinewy fingers",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "cameron looked at his hands with their long sinewy fingers",
+ "normalized_hypothesis": "cameron looked at his hands with their long sinewy fingers",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0408",
+ "category": "phonetic_balance",
+ "text": "Saxon waited, for she knew a fresh idea had struck Billy.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0408.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SAXON WAITED FOR SHE KNEW A FRESH IDEA HAD STRUCK BILLY",
+ "raw": {
+ "normalized_reference": "saxon waited, for she knew a fresh idea had struck billy.",
+ "normalized_hypothesis": "saxon waited for she knew a fresh idea had struck billy",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "saxon waited for she knew a fresh idea had struck billy",
+ "normalized_hypothesis": "saxon waited for she knew a fresh idea had struck billy",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0409",
+ "category": "phonetic_balance",
+ "text": "He saw only the effect in a general, sketchy way.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0409.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE SAW ONLY THE EFFECT IN A GENERAL SKETCHY WAY",
+ "raw": {
+ "normalized_reference": "he saw only the effect in a general, sketchy way.",
+ "normalized_hypothesis": "he saw only the effect in a general sketchy way",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he saw only the effect in a general sketchy way",
+ "normalized_hypothesis": "he saw only the effect in a general sketchy way",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0410",
+ "category": "phonetic_balance",
+ "text": "He began to follow the footprints of the dog.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0410.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE BEGAN TO FOLLOW THE FOOTPRINTS OF THE DOG",
+ "raw": {
+ "normalized_reference": "he began to follow the footprints of the dog.",
+ "normalized_hypothesis": "he began to follow the footprints of the dog",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he began to follow the footprints of the dog",
+ "normalized_hypothesis": "he began to follow the footprints of the dog",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0411",
+ "category": "phonetic_balance",
+ "text": "Do you value your hide.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0411.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "DO YOU VALUE YOUR HIDE",
+ "raw": {
+ "normalized_reference": "do you value your hide.",
+ "normalized_hypothesis": "do you value your hide",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "do you value your hide",
+ "normalized_hypothesis": "do you value your hide",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0412",
+ "category": "phonetic_balance",
+ "text": "A rising tide of fat had submerged them.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0412.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "A RISING TIDE OF FAT HAD SUBMERGED THEM",
+ "raw": {
+ "normalized_reference": "a rising tide of fat had submerged them.",
+ "normalized_hypothesis": "a rising tide of fat had submerged them",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "a rising tide of fat had submerged them",
+ "normalized_hypothesis": "a rising tide of fat had submerged them",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0413",
+ "category": "phonetic_balance",
+ "text": "There was nothing on the rock.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0413.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "ERE WAS NOTHING ON THE ROCK",
+ "raw": {
+ "normalized_reference": "there was nothing on the rock.",
+ "normalized_hypothesis": "ere was nothing on the rock",
+ "errors": 1,
+ "words": 6,
+ "wer": 0.16666666666666666
+ },
+ "semantic": {
+ "normalized_reference": "there was nothing on the rock",
+ "normalized_hypothesis": "ere was nothing on the rock",
+ "errors": 1,
+ "words": 6,
+ "wer": 0.16666666666666666
+ }
+ },
+ {
+ "id": "release500_0414",
+ "category": "phonetic_balance",
+ "text": "Whatever he guessed he locked away in the taboo room of Naomi.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0414.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WHATEVER HE GUESSED HE LOCKED AWAY IN THE TABOO ROOM OF NAOMI",
+ "raw": {
+ "normalized_reference": "whatever he guessed he locked away in the taboo room of naomi.",
+ "normalized_hypothesis": "whatever he guessed he locked away in the taboo room of naomi",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "whatever he guessed he locked away in the taboo room of naomi",
+ "normalized_hypothesis": "whatever he guessed he locked away in the taboo room of naomi",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0415",
+ "category": "phonetic_balance",
+ "text": "Will we ever forget it.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0415.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WILL WE EVER FORGET IT",
+ "raw": {
+ "normalized_reference": "will we ever forget it.",
+ "normalized_hypothesis": "will we ever forget it",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "will we ever forget it",
+ "normalized_hypothesis": "will we ever forget it",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0416",
+ "category": "phonetic_balance",
+ "text": "Jeanne and Pierre both gazed toward the great rock.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0416.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "JEAN AND PIERRE BOTH GAZED TOWARD THE GREAT ROCK",
+ "raw": {
+ "normalized_reference": "jeanne and pierre both gazed toward the great rock.",
+ "normalized_hypothesis": "jean and pierre both gazed toward the great rock",
+ "errors": 1,
+ "words": 9,
+ "wer": 0.1111111111111111
+ },
+ "semantic": {
+ "normalized_reference": "jeanne and pierre both gazed toward the great rock",
+ "normalized_hypothesis": "jean and pierre both gazed toward the great rock",
+ "errors": 1,
+ "words": 9,
+ "wer": 0.1111111111111111
+ }
+ },
+ {
+ "id": "release500_0417",
+ "category": "phonetic_balance",
+ "text": "Does the old boy often go off at half-cock that way.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0417.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "DOES THE OLD BOY OFTEN GO OFF AT HALF COCK THAT WAY",
+ "raw": {
+ "normalized_reference": "does the old boy often go off at half-cock that way.",
+ "normalized_hypothesis": "does the old boy often go off at half cock that way",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "does the old boy often go off at half cock that way",
+ "normalized_hypothesis": "does the old boy often go off at half cock that way",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0418",
+ "category": "phonetic_balance",
+ "text": "The temperature dropped to fifty below zero and remained there the whole trip.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0418.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "TEMPERATURE DROPPED TO FIFTY BELOW ZERO AND REMAINED THERE THE WHOLE TRIP",
+ "raw": {
+ "normalized_reference": "the temperature dropped to fifty below zero and remained there the whole trip.",
+ "normalized_hypothesis": "temperature dropped to fifty below zero and remained there the whole trip",
+ "errors": 1,
+ "words": 13,
+ "wer": 0.07692307692307693
+ },
+ "semantic": {
+ "normalized_reference": "the temperature dropped to 50 below 0 and remained there the whole trip",
+ "normalized_hypothesis": "temperature dropped to 50 below 0 and remained there the whole trip",
+ "errors": 1,
+ "words": 13,
+ "wer": 0.07692307692307693
+ }
+ },
+ {
+ "id": "release500_0419",
+ "category": "phonetic_balance",
+ "text": "It's worth eight dollars.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0419.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT'S WORTH EIGHT DOLLARS",
+ "raw": {
+ "normalized_reference": "it's worth eight dollars.",
+ "normalized_hypothesis": "it's worth eight dollars",
+ "errors": 0,
+ "words": 4,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it is worth $8",
+ "normalized_hypothesis": "it is worth $8",
+ "errors": 0,
+ "words": 4,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0420",
+ "category": "phonetic_balance",
+ "text": "But I did not enjoy it long.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0420.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I DID NOT ENJOY IT LONG",
+ "raw": {
+ "normalized_reference": "but i did not enjoy it long.",
+ "normalized_hypothesis": "i did not enjoy it long",
+ "errors": 1,
+ "words": 7,
+ "wer": 0.14285714285714285
+ },
+ "semantic": {
+ "normalized_reference": "but i did not enjoy it long",
+ "normalized_hypothesis": "i did not enjoy it long",
+ "errors": 1,
+ "words": 7,
+ "wer": 0.14285714285714285
+ }
+ },
+ {
+ "id": "release500_0421",
+ "category": "phonetic_balance",
+ "text": "Shorty turned to their employers.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0421.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SHORTY TURNED TO THEIR EMPLOYERS",
+ "raw": {
+ "normalized_reference": "shorty turned to their employers.",
+ "normalized_hypothesis": "shorty turned to their employers",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "shorty turned to their employers",
+ "normalized_hypothesis": "shorty turned to their employers",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0422",
+ "category": "phonetic_balance",
+ "text": "It was the third or fourth time that Philip had heard MacDougall swear.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0422.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WAS THE THIRD OR FOURTH TIME THAT PHILIP HAD HEARD MAC DUGAL SWEAR",
+ "raw": {
+ "normalized_reference": "it was the third or fourth time that philip had heard macdougall swear.",
+ "normalized_hypothesis": "was the third or fourth time that philip had heard mac dugal swear",
+ "errors": 3,
+ "words": 13,
+ "wer": 0.23076923076923078
+ },
+ "semantic": {
+ "normalized_reference": "it was the 3rd or 4th time that philip had heard macdougall swear",
+ "normalized_hypothesis": "was the 3rd or 4th time that philip had heard mac dugal swear",
+ "errors": 3,
+ "words": 13,
+ "wer": 0.23076923076923078
+ }
+ },
+ {
+ "id": "release500_0423",
+ "category": "phonetic_balance",
+ "text": "Ernest saw in the affair the most sinister import.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0423.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "ERNEST SAW IN THE AFFAIR THE MOST SINISTER IMPORT",
+ "raw": {
+ "normalized_reference": "ernest saw in the affair the most sinister import.",
+ "normalized_hypothesis": "ernest saw in the affair the most sinister import",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "ernest saw in the affair the most sinister import",
+ "normalized_hypothesis": "ernest saw in the affair the most sinister import",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0424",
+ "category": "phonetic_balance",
+ "text": "Also, at regular intervals, he would mutter.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0424.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "ALSO AT REGULAR INTERVALS HE WOULD MUTTER",
+ "raw": {
+ "normalized_reference": "also, at regular intervals, he would mutter.",
+ "normalized_hypothesis": "also at regular intervals he would mutter",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "also at regular intervals he would mutter",
+ "normalized_hypothesis": "also at regular intervals he would mutter",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0425",
+ "category": "phonetic_balance",
+ "text": "He may anticipate the day of his death.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0425.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE MAY ANTICIPATE THE DAY OF HIS DEATH",
+ "raw": {
+ "normalized_reference": "he may anticipate the day of his death.",
+ "normalized_hypothesis": "he may anticipate the day of his death",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he may anticipate the day of his death",
+ "normalized_hypothesis": "he may anticipate the day of his death",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0426",
+ "category": "phonetic_balance",
+ "text": "I defy any man to get a Solomon Island sore in California.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0426.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I DEFY ANY MAN TO GET A SALTLAMMON ISLAND SORE IN CALIFORNIA",
+ "raw": {
+ "normalized_reference": "i defy any man to get a solomon island sore in california.",
+ "normalized_hypothesis": "i defy any man to get a saltlammon island sore in california",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "i defy any man to get a solomon island sore in california",
+ "normalized_hypothesis": "i defy any man to get a saltlammon island sore in california",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ }
+ },
+ {
+ "id": "release500_0427",
+ "category": "phonetic_balance",
+ "text": "Now our figuring was all right.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0427.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "NOW OUR FIGURING WAS ALL RIGHT",
+ "raw": {
+ "normalized_reference": "now our figuring was all right.",
+ "normalized_hypothesis": "now our figuring was all right",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "now our figuring was all right",
+ "normalized_hypothesis": "now our figuring was all right",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0428",
+ "category": "phonetic_balance",
+ "text": "MacDougall tapped his forehead suspiciously with a stubby forefinger.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0428.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "MACDUGLL TAPPED HIS FOREHEAD SUSPICIOUSLY WITH A STUBBY FOREFINGER",
+ "raw": {
+ "normalized_reference": "macdougall tapped his forehead suspiciously with a stubby forefinger.",
+ "normalized_hypothesis": "macdugll tapped his forehead suspiciously with a stubby forefinger",
+ "errors": 1,
+ "words": 9,
+ "wer": 0.1111111111111111
+ },
+ "semantic": {
+ "normalized_reference": "macdougall tapped his forehead suspiciously with a stubby forefinger",
+ "normalized_hypothesis": "macdugll tapped his forehead suspiciously with a stubby forefinger",
+ "errors": 1,
+ "words": 9,
+ "wer": 0.1111111111111111
+ }
+ },
+ {
+ "id": "release500_0429",
+ "category": "phonetic_balance",
+ "text": "O'Brien took off his coat and bared his right arm.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0429.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "O'RIAN TOOK OFF HIS COAT AND BARED HIS RIGHT ARM",
+ "raw": {
+ "normalized_reference": "o'brien took off his coat and bared his right arm.",
+ "normalized_hypothesis": "o'rian took off his coat and bared his right arm",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ },
+ "semantic": {
+ "normalized_reference": "0 brien took off his coat and bared his right arm",
+ "normalized_hypothesis": "0 rian took off his coat and bared his right arm",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ }
+ },
+ {
+ "id": "release500_0430",
+ "category": "phonetic_balance",
+ "text": "But it won't continue, she said with easy confidence.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0430.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "TIT WON'T CONTINUE SHE SAID WITH EASY CONFIDENCE",
+ "raw": {
+ "normalized_reference": "but it won't continue, she said with easy confidence.",
+ "normalized_hypothesis": "tit won't continue she said with easy confidence",
+ "errors": 2,
+ "words": 9,
+ "wer": 0.2222222222222222
+ },
+ "semantic": {
+ "normalized_reference": "but it will not continue she said with easy confidence",
+ "normalized_hypothesis": "tit will not continue she said with easy confidence",
+ "errors": 2,
+ "words": 10,
+ "wer": 0.2
+ }
+ },
+ {
+ "id": "release500_0431",
+ "category": "phonetic_balance",
+ "text": "I was still weak from my prolonged immersion.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0431.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WAS STILL WEAK FROM MY PROLONGED IMMERSION",
+ "raw": {
+ "normalized_reference": "i was still weak from my prolonged immersion.",
+ "normalized_hypothesis": "was still weak from my prolonged immersion",
+ "errors": 1,
+ "words": 8,
+ "wer": 0.125
+ },
+ "semantic": {
+ "normalized_reference": "i was still weak from my prolonged immersion",
+ "normalized_hypothesis": "was still weak from my prolonged immersion",
+ "errors": 1,
+ "words": 8,
+ "wer": 0.125
+ }
+ },
+ {
+ "id": "release500_0432",
+ "category": "phonetic_balance",
+ "text": "Oolong was two hundred and fifty miles from the nearest land.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0432.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "ULAM WAS TWO HUNDRED AND FIFTY MILES FROM THE NEAREST LAND",
+ "raw": {
+ "normalized_reference": "oolong was two hundred and fifty miles from the nearest land.",
+ "normalized_hypothesis": "ulam was two hundred and fifty miles from the nearest land",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ },
+ "semantic": {
+ "normalized_reference": "oolong was 250 miles from the nearest land",
+ "normalized_hypothesis": "ulam was 250 miles from the nearest land",
+ "errors": 1,
+ "words": 8,
+ "wer": 0.125
+ }
+ },
+ {
+ "id": "release500_0433",
+ "category": "phonetic_balance",
+ "text": "I just do appreciate it without being able to express my feelings.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0433.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I JUST DO APPRECIATE IT WITHOUT BEING ABLE TO EXPRESS MY FEELINGS",
+ "raw": {
+ "normalized_reference": "i just do appreciate it without being able to express my feelings.",
+ "normalized_hypothesis": "i just do appreciate it without being able to express my feelings",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i just do appreciate it without being able to express my feelings",
+ "normalized_hypothesis": "i just do appreciate it without being able to express my feelings",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0434",
+ "category": "phonetic_balance",
+ "text": "Now just what do you want to know.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0434.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "NOW JUST WHAT DO YOU WANT TO KNOW",
+ "raw": {
+ "normalized_reference": "now just what do you want to know.",
+ "normalized_hypothesis": "now just what do you want to know",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "now just what do you want to know",
+ "normalized_hypothesis": "now just what do you want to know",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0435",
+ "category": "phonetic_balance",
+ "text": "But she swung obediently on her heel into the wind.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0435.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "BUT SHE SWUNG OBEDIENTLY ON HER HEEL INTO THE WIND",
+ "raw": {
+ "normalized_reference": "but she swung obediently on her heel into the wind.",
+ "normalized_hypothesis": "but she swung obediently on her heel into the wind",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "but she swung obediently on her heel into the wind",
+ "normalized_hypothesis": "but she swung obediently on her heel into the wind",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0436",
+ "category": "phonetic_balance",
+ "text": "Ask him, I laughed, then turned to Pasquini.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0436.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "ASK HIM I LAUGHED THEN TURNED TO PASQUINI",
+ "raw": {
+ "normalized_reference": "ask him, i laughed, then turned to pasquini.",
+ "normalized_hypothesis": "ask him i laughed then turned to pasquini",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "ask him i laughed then turned to pasquini",
+ "normalized_hypothesis": "ask him i laughed then turned to pasquini",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0437",
+ "category": "phonetic_balance",
+ "text": "He did not rush in.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0437.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE DID NOT RUSH IN",
+ "raw": {
+ "normalized_reference": "he did not rush in.",
+ "normalized_hypothesis": "he did not rush in",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he did not rush in",
+ "normalized_hypothesis": "he did not rush in",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0438",
+ "category": "phonetic_balance",
+ "text": "They obeyed him, and went here and there at his commands.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0438.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THEY OBEYED HIM AND WENT HERE AND THERE AT HIS COMMANDS",
+ "raw": {
+ "normalized_reference": "they obeyed him, and went here and there at his commands.",
+ "normalized_hypothesis": "they obeyed him and went here and there at his commands",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "they obeyed him and went here and there at his commands",
+ "normalized_hypothesis": "they obeyed him and went here and there at his commands",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0439",
+ "category": "phonetic_balance",
+ "text": "Tom Spink has a harpoon.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0439.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "TON SPANK HAS A HARPOON",
+ "raw": {
+ "normalized_reference": "tom spink has a harpoon.",
+ "normalized_hypothesis": "ton spank has a harpoon",
+ "errors": 2,
+ "words": 5,
+ "wer": 0.4
+ },
+ "semantic": {
+ "normalized_reference": "tom spink has a harpoon",
+ "normalized_hypothesis": "ton spank has a harpoon",
+ "errors": 2,
+ "words": 5,
+ "wer": 0.4
+ }
+ },
+ {
+ "id": "release500_0440",
+ "category": "phonetic_balance",
+ "text": "It was my reports from the north which chiefly induced people to buy.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0440.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT WAS MY REPORTS FROM THE NORTH WHICH CHIEFLY INDUCED PEOPLE TO BUY",
+ "raw": {
+ "normalized_reference": "it was my reports from the north which chiefly induced people to buy.",
+ "normalized_hypothesis": "it was my reports from the north which chiefly induced people to buy",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it was my reports from the north which chiefly induced people to buy",
+ "normalized_hypothesis": "it was my reports from the north which chiefly induced people to buy",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0441",
+ "category": "phonetic_balance",
+ "text": "Cherokee identified himself with his instinct.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0441.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "CHEROKEY IDENTIFIED HIMSELF WITH HIS INSTINCT",
+ "raw": {
+ "normalized_reference": "cherokee identified himself with his instinct.",
+ "normalized_hypothesis": "cherokey identified himself with his instinct",
+ "errors": 1,
+ "words": 6,
+ "wer": 0.16666666666666666
+ },
+ "semantic": {
+ "normalized_reference": "cherokee identified himself with his instinct",
+ "normalized_hypothesis": "cherokey identified himself with his instinct",
+ "errors": 1,
+ "words": 6,
+ "wer": 0.16666666666666666
+ }
+ },
+ {
+ "id": "release500_0442",
+ "category": "phonetic_balance",
+ "text": "I graduated last of my class.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0442.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I GRADUATED LAST OF MY CLASS",
+ "raw": {
+ "normalized_reference": "i graduated last of my class.",
+ "normalized_hypothesis": "i graduated last of my class",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i graduated last of my class",
+ "normalized_hypothesis": "i graduated last of my class",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0443",
+ "category": "phonetic_balance",
+ "text": "It was a gigantic inadequacy.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0443.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT WAS A GIGANTIC INADEQUACY",
+ "raw": {
+ "normalized_reference": "it was a gigantic inadequacy.",
+ "normalized_hypothesis": "it was a gigantic inadequacy",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it was a gigantic inadequacy",
+ "normalized_hypothesis": "it was a gigantic inadequacy",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0444",
+ "category": "phonetic_balance",
+ "text": "Did I possess too much vitality.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0444.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT I POSSESS TOO MUCH VITALITY",
+ "raw": {
+ "normalized_reference": "did i possess too much vitality.",
+ "normalized_hypothesis": "it i possess too much vitality",
+ "errors": 1,
+ "words": 6,
+ "wer": 0.16666666666666666
+ },
+ "semantic": {
+ "normalized_reference": "did i possess too much vitality",
+ "normalized_hypothesis": "it i possess too much vitality",
+ "errors": 1,
+ "words": 6,
+ "wer": 0.16666666666666666
+ }
+ },
+ {
+ "id": "release500_0445",
+ "category": "phonetic_balance",
+ "text": "Therefore, hurrah for the game.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0445.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THEREFORE HURRAH FOR THE GAME",
+ "raw": {
+ "normalized_reference": "therefore, hurrah for the game.",
+ "normalized_hypothesis": "therefore hurrah for the game",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "therefore hurrah for the game",
+ "normalized_hypothesis": "therefore hurrah for the game",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0446",
+ "category": "phonetic_balance",
+ "text": "He had fulfilled his duty and paid properly.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0446.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HAD FULFILLED HIS DUTY AND PAID PROPERLY",
+ "raw": {
+ "normalized_reference": "he had fulfilled his duty and paid properly.",
+ "normalized_hypothesis": "had fulfilled his duty and paid properly",
+ "errors": 1,
+ "words": 8,
+ "wer": 0.125
+ },
+ "semantic": {
+ "normalized_reference": "he had fulfilled his duty and paid properly",
+ "normalized_hypothesis": "had fulfilled his duty and paid properly",
+ "errors": 1,
+ "words": 8,
+ "wer": 0.125
+ }
+ },
+ {
+ "id": "release500_0447",
+ "category": "phonetic_balance",
+ "text": "How can you manage all alone, Mr Young.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0447.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HOW CAN YOU MANAGE ALL ALONE MISTER YOUNG",
+ "raw": {
+ "normalized_reference": "how can you manage all alone, mr young.",
+ "normalized_hypothesis": "how can you manage all alone mister young",
+ "errors": 1,
+ "words": 8,
+ "wer": 0.125
+ },
+ "semantic": {
+ "normalized_reference": "how can you manage all alone mister young",
+ "normalized_hypothesis": "how can you manage all alone mister young",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0448",
+ "category": "phonetic_balance",
+ "text": "Red-Eye never committed a more outrageous deed.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0448.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "RED EYE NEVER COMMITTED A MORE OUTRAGEOUS DEED",
+ "raw": {
+ "normalized_reference": "red-eye never committed a more outrageous deed.",
+ "normalized_hypothesis": "red eye never committed a more outrageous deed",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "red eye never committed a more outrageous deed",
+ "normalized_hypothesis": "red eye never committed a more outrageous deed",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0449",
+ "category": "phonetic_balance",
+ "text": "He would keep his faith with Gregson for the promised day or two.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0449.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "YOU WOULD KEEP HIS FAITH WITH GREGSEN FOR THE PROMISED DAY OR TWO",
+ "raw": {
+ "normalized_reference": "he would keep his faith with gregson for the promised day or two.",
+ "normalized_hypothesis": "you would keep his faith with gregsen for the promised day or two",
+ "errors": 2,
+ "words": 13,
+ "wer": 0.15384615384615385
+ },
+ "semantic": {
+ "normalized_reference": "he would keep his faith with gregson for the promised day or 2",
+ "normalized_hypothesis": "you would keep his faith with gregsen for the promised day or 2",
+ "errors": 2,
+ "words": 13,
+ "wer": 0.15384615384615385
+ }
+ },
+ {
+ "id": "release500_0450",
+ "category": "phonetic_balance",
+ "text": "There was one chance, and only one, of saving Jeanne.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0450.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THERE WAS ONE CHANCE AND ONLY ONE OF SAVING JEAN",
+ "raw": {
+ "normalized_reference": "there was one chance, and only one, of saving jeanne.",
+ "normalized_hypothesis": "there was one chance and only one of saving jean",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ },
+ "semantic": {
+ "normalized_reference": "there was one chance and only one of saving jeanne",
+ "normalized_hypothesis": "there was one chance and only one of saving jean",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ }
+ },
+ {
+ "id": "release500_0451",
+ "category": "phonetic_balance",
+ "text": "I was beneath the water, suffocating and drowning.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0451.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I WAS BENEATH THE WATER SUFFOCATING AND DROWNING",
+ "raw": {
+ "normalized_reference": "i was beneath the water, suffocating and drowning.",
+ "normalized_hypothesis": "i was beneath the water suffocating and drowning",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i was beneath the water suffocating and drowning",
+ "normalized_hypothesis": "i was beneath the water suffocating and drowning",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0452",
+ "category": "phonetic_balance",
+ "text": "Give them their choice between a fine or an official whipping.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0452.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "GIVE THEM THEIR CHOICE BETWEEN A FINE OR AN OFFICIAL WHIPPING",
+ "raw": {
+ "normalized_reference": "give them their choice between a fine or an official whipping.",
+ "normalized_hypothesis": "give them their choice between a fine or an official whipping",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "give them their choice between a fine or an official whipping",
+ "normalized_hypothesis": "give them their choice between a fine or an official whipping",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0453",
+ "category": "phonetic_balance",
+ "text": "They must have been swept away by the chaotic currents.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0453.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE MUST HAVE BEEN SWEPT AWAY BY THE CHAOTIC OCCURRENTS",
+ "raw": {
+ "normalized_reference": "they must have been swept away by the chaotic currents.",
+ "normalized_hypothesis": "he must have been swept away by the chaotic occurrents",
+ "errors": 2,
+ "words": 10,
+ "wer": 0.2
+ },
+ "semantic": {
+ "normalized_reference": "they must have been swept away by the chaotic currents",
+ "normalized_hypothesis": "he must have been swept away by the chaotic occurrents",
+ "errors": 2,
+ "words": 10,
+ "wer": 0.2
+ }
+ },
+ {
+ "id": "release500_0454",
+ "category": "phonetic_balance",
+ "text": "You yellow giant thing of the frost.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0454.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "YOU YELLOW GIANT THING OF THE FROST",
+ "raw": {
+ "normalized_reference": "you yellow giant thing of the frost.",
+ "normalized_hypothesis": "you yellow giant thing of the frost",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "you yellow giant thing of the frost",
+ "normalized_hypothesis": "you yellow giant thing of the frost",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0455",
+ "category": "phonetic_balance",
+ "text": "He had peeled off his shirt and was wildly waving it.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0455.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE HAD PEELED OFF HIS SHIRT AND WAS WILDLY WAVING IT",
+ "raw": {
+ "normalized_reference": "he had peeled off his shirt and was wildly waving it.",
+ "normalized_hypothesis": "he had peeled off his shirt and was wildly waving it",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he had peeled off his shirt and was wildly waving it",
+ "normalized_hypothesis": "he had peeled off his shirt and was wildly waving it",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0456",
+ "category": "phonetic_balance",
+ "text": "To these he gave castor oil.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0456.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "TO THESE HE GAVE CASTOR OIL",
+ "raw": {
+ "normalized_reference": "to these he gave castor oil.",
+ "normalized_hypothesis": "to these he gave castor oil",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "to these he gave castor oil",
+ "normalized_hypothesis": "to these he gave castor oil",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0457",
+ "category": "phonetic_balance",
+ "text": "Tudor surveyed him with withering disgust.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0457.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "TUTOR SURVEYED HIM WITH WITHERING DISGUST",
+ "raw": {
+ "normalized_reference": "tudor surveyed him with withering disgust.",
+ "normalized_hypothesis": "tutor surveyed him with withering disgust",
+ "errors": 1,
+ "words": 6,
+ "wer": 0.16666666666666666
+ },
+ "semantic": {
+ "normalized_reference": "tudor surveyed him with withering disgust",
+ "normalized_hypothesis": "tutor surveyed him with withering disgust",
+ "errors": 1,
+ "words": 6,
+ "wer": 0.16666666666666666
+ }
+ },
+ {
+ "id": "release500_0458",
+ "category": "phonetic_balance",
+ "text": "After all, it was simply a mistake in judgment.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0458.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "AFTER ALL IT WAS SIMPLY A MISTAKE IN JUDGMENT",
+ "raw": {
+ "normalized_reference": "after all, it was simply a mistake in judgment.",
+ "normalized_hypothesis": "after all it was simply a mistake in judgment",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "after all it was simply a mistake in judgment",
+ "normalized_hypothesis": "after all it was simply a mistake in judgment",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0459",
+ "category": "phonetic_balance",
+ "text": "That is the strange part of it.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0459.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THAT IS THE STRANGE PART OF IT",
+ "raw": {
+ "normalized_reference": "that is the strange part of it.",
+ "normalized_hypothesis": "that is the strange part of it",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "that is the strange part of it",
+ "normalized_hypothesis": "that is the strange part of it",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0460",
+ "category": "phonetic_balance",
+ "text": "Not a wheel moved in his empire.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0460.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "NOT A WHEEL MOVED IN HIS EMPIRE",
+ "raw": {
+ "normalized_reference": "not a wheel moved in his empire.",
+ "normalized_hypothesis": "not a wheel moved in his empire",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "not a wheel moved in his empire",
+ "normalized_hypothesis": "not a wheel moved in his empire",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0461",
+ "category": "phonetic_balance",
+ "text": "Its diameter was not more than two hundred yards.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0461.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "ITS DIAMETER WAS NOT MORE THAN TWO HUNDRED YARDS",
+ "raw": {
+ "normalized_reference": "its diameter was not more than two hundred yards.",
+ "normalized_hypothesis": "its diameter was not more than two hundred yards",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "its diameter was not more than 200 yards",
+ "normalized_hypothesis": "its diameter was not more than 200 yards",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0462",
+ "category": "phonetic_balance",
+ "text": "Nimrod replied, with a slight manifestation of sensitiveness.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0462.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "NIMROD REPLIED WITH A SLIGHT MANIFESTATION OF SENSITIVENESS",
+ "raw": {
+ "normalized_reference": "nimrod replied, with a slight manifestation of sensitiveness.",
+ "normalized_hypothesis": "nimrod replied with a slight manifestation of sensitiveness",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "nimrod replied with a slight manifestation of sensitiveness",
+ "normalized_hypothesis": "nimrod replied with a slight manifestation of sensitiveness",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0463",
+ "category": "phonetic_balance",
+ "text": "I tried to read George Moore last night, and was dreadfully bored.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0463.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I TRIED TO READ GEORGE MOORE LAST NIGHT AND WAS DREADFULLY BORED",
+ "raw": {
+ "normalized_reference": "i tried to read george moore last night, and was dreadfully bored.",
+ "normalized_hypothesis": "i tried to read george moore last night and was dreadfully bored",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i tried to read george moore last night and was dreadfully bored",
+ "normalized_hypothesis": "i tried to read george moore last night and was dreadfully bored",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0464",
+ "category": "phonetic_balance",
+ "text": "Yes, it was a man who asked, a stranger.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0464.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "YES IT WAS A MAN WHO ASKED A STRANGER",
+ "raw": {
+ "normalized_reference": "yes, it was a man who asked, a stranger.",
+ "normalized_hypothesis": "yes it was a man who asked a stranger",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "yes it was a man who asked a stranger",
+ "normalized_hypothesis": "yes it was a man who asked a stranger",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0465",
+ "category": "phonetic_balance",
+ "text": "We were now good friends.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0465.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WE WERE NOW GOOD FRIENDS",
+ "raw": {
+ "normalized_reference": "we were now good friends.",
+ "normalized_hypothesis": "we were now good friends",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "we were now good friends",
+ "normalized_hypothesis": "we were now good friends",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0466",
+ "category": "phonetic_balance",
+ "text": "Yet, in accordance with Ernest's test of truth, it worked.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0466.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "YES IN ACCORDANCE WITH ERNEST'S TEST OF TRUTH A WORK",
+ "raw": {
+ "normalized_reference": "yet, in accordance with ernest's test of truth, it worked.",
+ "normalized_hypothesis": "yes in accordance with ernest's test of truth a work",
+ "errors": 3,
+ "words": 10,
+ "wer": 0.3
+ },
+ "semantic": {
+ "normalized_reference": "yet in accordance with ernest is test of truth it worked",
+ "normalized_hypothesis": "yes in accordance with ernest is test of truth a work",
+ "errors": 3,
+ "words": 11,
+ "wer": 0.2727272727272727
+ }
+ },
+ {
+ "id": "release500_0467",
+ "category": "phonetic_balance",
+ "text": "It's the strap hangers that'll keep us from going under.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0467.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT'S THE STRAP HANGERS THAT'LL KEEP US FROM GOING UNDER",
+ "raw": {
+ "normalized_reference": "it's the strap hangers that'll keep us from going under.",
+ "normalized_hypothesis": "it's the strap hangers that'll keep us from going under",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it is the strap hangers that will keep us from going under",
+ "normalized_hypothesis": "it is the strap hangers that will keep us from going under",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0468",
+ "category": "phonetic_balance",
+ "text": "I never allow what can't be changed to annoy me.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0468.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I NEVER ALLOW WHAT CANT BE CHANGED TO ANNOY ME",
+ "raw": {
+ "normalized_reference": "i never allow what can't be changed to annoy me.",
+ "normalized_hypothesis": "i never allow what cant be changed to annoy me",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ },
+ "semantic": {
+ "normalized_reference": "i never allow what can not be changed to annoy me",
+ "normalized_hypothesis": "i never allow what cant be changed to annoy me",
+ "errors": 2,
+ "words": 11,
+ "wer": 0.18181818181818182
+ }
+ },
+ {
+ "id": "release500_0469",
+ "category": "phonetic_balance",
+ "text": "After all superfluous flesh is gone what is left is stringy and resistant.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0469.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "AFTER ALL SUPERFLUOUS FLESH HIS GONE WHAT IS LEFT IS STRINGY AND RESISTENT",
+ "raw": {
+ "normalized_reference": "after all superfluous flesh is gone what is left is stringy and resistant.",
+ "normalized_hypothesis": "after all superfluous flesh his gone what is left is stringy and resistent",
+ "errors": 2,
+ "words": 13,
+ "wer": 0.15384615384615385
+ },
+ "semantic": {
+ "normalized_reference": "after all superfluous flesh is gone what is left is stringy and resistant",
+ "normalized_hypothesis": "after all superfluous flesh his gone what is left is stringy and resistent",
+ "errors": 2,
+ "words": 13,
+ "wer": 0.15384615384615385
+ }
+ },
+ {
+ "id": "release500_0470",
+ "category": "phonetic_balance",
+ "text": "Then he shouted, Shut up.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0470.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THEN HE SHOUTED SHUT UP",
+ "raw": {
+ "normalized_reference": "then he shouted, shut up.",
+ "normalized_hypothesis": "then he shouted shut up",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "then he shouted shut up",
+ "normalized_hypothesis": "then he shouted shut up",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0471",
+ "category": "phonetic_balance",
+ "text": "He had worshiped her, as Dante might have worshiped Beatrice.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0471.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE HAD WORSHIPPED HER AS DANTES MIGHT HAVE WORSHIPPED BEATRICE",
+ "raw": {
+ "normalized_reference": "he had worshiped her, as dante might have worshiped beatrice.",
+ "normalized_hypothesis": "he had worshipped her as dantes might have worshipped beatrice",
+ "errors": 3,
+ "words": 10,
+ "wer": 0.3
+ },
+ "semantic": {
+ "normalized_reference": "he had worshiped her as dante might have worshiped beatrice",
+ "normalized_hypothesis": "he had worshipped her as dantes might have worshipped beatrice",
+ "errors": 3,
+ "words": 10,
+ "wer": 0.3
+ }
+ },
+ {
+ "id": "release500_0472",
+ "category": "phonetic_balance",
+ "text": "He looked like one who had passed through an uncomfortable hour or two.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0472.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE LOOKED LIKE ONE WHO HAD PASSED THROUGH AN UNCOMFORTABLE HOUR OR TWO",
+ "raw": {
+ "normalized_reference": "he looked like one who had passed through an uncomfortable hour or two.",
+ "normalized_hypothesis": "he looked like one who had passed through an uncomfortable hour or two",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he looked like one who had passed through an uncomfortable hour or 2",
+ "normalized_hypothesis": "he looked like one who had passed through an uncomfortable hour or 2",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0473",
+ "category": "phonetic_balance",
+ "text": "The history of our westward-faring race is written in it.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0473.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HEHISTORY OF OUR WESTWARD FARING RACE IS WRITTEN IN IT",
+ "raw": {
+ "normalized_reference": "the history of our westward-faring race is written in it.",
+ "normalized_hypothesis": "hehistory of our westward faring race is written in it",
+ "errors": 2,
+ "words": 11,
+ "wer": 0.18181818181818182
+ },
+ "semantic": {
+ "normalized_reference": "the history of our westward faring race is written in it",
+ "normalized_hypothesis": "hehistory of our westward faring race is written in it",
+ "errors": 2,
+ "words": 11,
+ "wer": 0.18181818181818182
+ }
+ },
+ {
+ "id": "release500_0474",
+ "category": "phonetic_balance",
+ "text": "He may desire to escape pain, or to enjoy pleasure.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0474.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE MAY DESIRE TO ESCAPE PAIN OR TO ENJOY PLEASURE",
+ "raw": {
+ "normalized_reference": "he may desire to escape pain, or to enjoy pleasure.",
+ "normalized_hypothesis": "he may desire to escape pain or to enjoy pleasure",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he may desire to escape pain or to enjoy pleasure",
+ "normalized_hypothesis": "he may desire to escape pain or to enjoy pleasure",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0475",
+ "category": "phonetic_balance",
+ "text": "And the air was growing chilly.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0475.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "AND THE AIR WAS GROWING CHILLY",
+ "raw": {
+ "normalized_reference": "and the air was growing chilly.",
+ "normalized_hypothesis": "and the air was growing chilly",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "and the air was growing chilly",
+ "normalized_hypothesis": "and the air was growing chilly",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0476",
+ "category": "phonetic_balance",
+ "text": "Dig in; you're sure good, was Daylight's answer.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0476.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "DIG IN YOUR SURE GOOD WAS STALIGHT'S ANSWER",
+ "raw": {
+ "normalized_reference": "dig in; you're sure good, was daylight's answer.",
+ "normalized_hypothesis": "dig in your sure good was stalight's answer",
+ "errors": 2,
+ "words": 8,
+ "wer": 0.25
+ },
+ "semantic": {
+ "normalized_reference": "dig in you are sure good was daylight is answer",
+ "normalized_hypothesis": "dig in your sure good was stalight is answer",
+ "errors": 3,
+ "words": 10,
+ "wer": 0.3
+ }
+ },
+ {
+ "id": "release500_0477",
+ "category": "phonetic_balance",
+ "text": "My idea was that he would have more influence over the natives.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0477.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "MY IDEA WAS THAT HE WOULD HAVE MORE INFLUENCE OVER THE NATIVES",
+ "raw": {
+ "normalized_reference": "my idea was that he would have more influence over the natives.",
+ "normalized_hypothesis": "my idea was that he would have more influence over the natives",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "my idea was that he would have more influence over the natives",
+ "normalized_hypothesis": "my idea was that he would have more influence over the natives",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0478",
+ "category": "phonetic_balance",
+ "text": "Now run along, and tell them to hurry.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0478.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "NOW RUN ALONG AND TELL THEM TO HURRY",
+ "raw": {
+ "normalized_reference": "now run along, and tell them to hurry.",
+ "normalized_hypothesis": "now run along and tell them to hurry",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "now run along and tell them to hurry",
+ "normalized_hypothesis": "now run along and tell them to hurry",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0479",
+ "category": "phonetic_balance",
+ "text": "I have no idea, replied Philip.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0479.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I HAVE NO IDEA REPLIED PHILIP",
+ "raw": {
+ "normalized_reference": "i have no idea, replied philip.",
+ "normalized_hypothesis": "i have no idea replied philip",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i have no idea replied philip",
+ "normalized_hypothesis": "i have no idea replied philip",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0480",
+ "category": "phonetic_balance",
+ "text": "For the rest, he was a mere automaton.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0480.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "FOR THE REST HE WAS A MERE AUTOMATON",
+ "raw": {
+ "normalized_reference": "for the rest, he was a mere automaton.",
+ "normalized_hypothesis": "for the rest he was a mere automaton",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "for the rest he was a mere automaton",
+ "normalized_hypothesis": "for the rest he was a mere automaton",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0481",
+ "category": "phonetic_balance",
+ "text": "Why, the average review is more nauseating than cod liver oil.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0481.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WHY THE AVERAGE REVIEW IS MORE NAUSEATING THAN COD LIVER OIL",
+ "raw": {
+ "normalized_reference": "why, the average review is more nauseating than cod liver oil.",
+ "normalized_hypothesis": "why the average review is more nauseating than cod liver oil",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "why the average review is more nauseating than cod liver oil",
+ "normalized_hypothesis": "why the average review is more nauseating than cod liver oil",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0482",
+ "category": "phonetic_balance",
+ "text": "I have been robbed, sir, I amended.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0482.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I HAVE BEEN ROBBED SIR I AMENDED",
+ "raw": {
+ "normalized_reference": "i have been robbed, sir, i amended.",
+ "normalized_hypothesis": "i have been robbed sir i amended",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i have been robbed sir i amended",
+ "normalized_hypothesis": "i have been robbed sir i amended",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0483",
+ "category": "phonetic_balance",
+ "text": "He moved away as quietly as he had come.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0483.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE MOVED AWAY AS QUIETLY AS HE HAD COME",
+ "raw": {
+ "normalized_reference": "he moved away as quietly as he had come.",
+ "normalized_hypothesis": "he moved away as quietly as he had come",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he moved away as quietly as he had come",
+ "normalized_hypothesis": "he moved away as quietly as he had come",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0484",
+ "category": "phonetic_balance",
+ "text": "A trickle of fresh blood ran over his face.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0484.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "A TRICKLE OF FRESH BLOOD RAN OVER HIS FACE",
+ "raw": {
+ "normalized_reference": "a trickle of fresh blood ran over his face.",
+ "normalized_hypothesis": "a trickle of fresh blood ran over his face",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "a trickle of fresh blood ran over his face",
+ "normalized_hypothesis": "a trickle of fresh blood ran over his face",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0485",
+ "category": "phonetic_balance",
+ "text": "Suppose you saw me at work through the window.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0485.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SUPPOSE YOU SAW ME AT WORK THROUGH THE WINDOW",
+ "raw": {
+ "normalized_reference": "suppose you saw me at work through the window.",
+ "normalized_hypothesis": "suppose you saw me at work through the window",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "suppose you saw me at work through the window",
+ "normalized_hypothesis": "suppose you saw me at work through the window",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0486",
+ "category": "phonetic_balance",
+ "text": "But who was Eileen's double.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0486.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "BUTWHO IS ILEAN'S DUBLE",
+ "raw": {
+ "normalized_reference": "but who was eileen's double.",
+ "normalized_hypothesis": "butwho is ilean's duble",
+ "errors": 5,
+ "words": 5,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "but who was eileen is double",
+ "normalized_hypothesis": "butwho is ilean is duble",
+ "errors": 5,
+ "words": 6,
+ "wer": 0.8333333333333334
+ }
+ },
+ {
+ "id": "release500_0487",
+ "category": "phonetic_balance",
+ "text": "Darkness hid him from Jeanne.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0487.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "DARKESS HID HIM FROM JEAN",
+ "raw": {
+ "normalized_reference": "darkness hid him from jeanne.",
+ "normalized_hypothesis": "darkess hid him from jean",
+ "errors": 2,
+ "words": 5,
+ "wer": 0.4
+ },
+ "semantic": {
+ "normalized_reference": "darkness hid him from jeanne",
+ "normalized_hypothesis": "darkess hid him from jean",
+ "errors": 2,
+ "words": 5,
+ "wer": 0.4
+ }
+ },
+ {
+ "id": "release500_0488",
+ "category": "phonetic_balance",
+ "text": "By this time Charley was as enraged as the Greek.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0488.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "BY THIS TIME CHARLEY WAS AS ENRAGED AS THE GREEK",
+ "raw": {
+ "normalized_reference": "by this time charley was as enraged as the greek.",
+ "normalized_hypothesis": "by this time charley was as enraged as the greek",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "by this time charley was as enraged as the greek",
+ "normalized_hypothesis": "by this time charley was as enraged as the greek",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0489",
+ "category": "phonetic_balance",
+ "text": "I am writing these lines in Honolulu, Hawaii.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0489.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "AM WRITING THESE LINES IN HONNOLULU HAWAI",
+ "raw": {
+ "normalized_reference": "i am writing these lines in honolulu, hawaii.",
+ "normalized_hypothesis": "am writing these lines in honnolulu hawai",
+ "errors": 3,
+ "words": 8,
+ "wer": 0.375
+ },
+ "semantic": {
+ "normalized_reference": "i am writing these lines in honolulu hawaii",
+ "normalized_hypothesis": "am writing these lines in honnolulu hawai",
+ "errors": 3,
+ "words": 8,
+ "wer": 0.375
+ }
+ },
+ {
+ "id": "release500_0490",
+ "category": "phonetic_balance",
+ "text": "Nobody knows how the natives got them.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0490.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "NOBODY KNOWS HOW THE NATIVES GOT THEM",
+ "raw": {
+ "normalized_reference": "nobody knows how the natives got them.",
+ "normalized_hypothesis": "nobody knows how the natives got them",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "nobody knows how the natives got them",
+ "normalized_hypothesis": "nobody knows how the natives got them",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0491",
+ "category": "phonetic_balance",
+ "text": "And how would we ever find ourselves.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0491.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "AND HOW WOULD WE EVER FIND OURSELVES",
+ "raw": {
+ "normalized_reference": "and how would we ever find ourselves.",
+ "normalized_hypothesis": "and how would we ever find ourselves",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "and how would we ever find ourselves",
+ "normalized_hypothesis": "and how would we ever find ourselves",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0492",
+ "category": "phonetic_balance",
+ "text": "They are coming ashore, whoever they are.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0492.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THEY ARE COMING ASHORE WHOEVER THEY ARE",
+ "raw": {
+ "normalized_reference": "they are coming ashore, whoever they are.",
+ "normalized_hypothesis": "they are coming ashore whoever they are",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "they are coming ashore whoever they are",
+ "normalized_hypothesis": "they are coming ashore whoever they are",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0493",
+ "category": "phonetic_balance",
+ "text": "This is my fifth voyage.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0493.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THIS IS MY FIFTH VOYAGE",
+ "raw": {
+ "normalized_reference": "this is my fifth voyage.",
+ "normalized_hypothesis": "this is my fifth voyage",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "this is my 5th voyage",
+ "normalized_hypothesis": "this is my 5th voyage",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0494",
+ "category": "phonetic_balance",
+ "text": "And yet if she came he had no words to say.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0494.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "AND YET IF SHE CAME HE HAD NO WORDS TO SAY",
+ "raw": {
+ "normalized_reference": "and yet if she came he had no words to say.",
+ "normalized_hypothesis": "and yet if she came he had no words to say",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "and yet if she came he had no words to say",
+ "normalized_hypothesis": "and yet if she came he had no words to say",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0495",
+ "category": "phonetic_balance",
+ "text": "You must sleep, he urged.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0495.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "YOU MUST SLEEP HE URGED",
+ "raw": {
+ "normalized_reference": "you must sleep, he urged.",
+ "normalized_hypothesis": "you must sleep he urged",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "you must sleep he urged",
+ "normalized_hypothesis": "you must sleep he urged",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0496",
+ "category": "phonetic_balance",
+ "text": "Could the incident have anything to do with Jeanne and Pierre.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0496.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "OUD THE INCIDENT HAVE ANYTHING TO DO WITH JEAN AND PIERRE",
+ "raw": {
+ "normalized_reference": "could the incident have anything to do with jeanne and pierre.",
+ "normalized_hypothesis": "oud the incident have anything to do with jean and pierre",
+ "errors": 2,
+ "words": 11,
+ "wer": 0.18181818181818182
+ },
+ "semantic": {
+ "normalized_reference": "could the incident have anything to do with jeanne and pierre",
+ "normalized_hypothesis": "oud the incident have anything to do with jean and pierre",
+ "errors": 2,
+ "words": 11,
+ "wer": 0.18181818181818182
+ }
+ },
+ {
+ "id": "release500_0497",
+ "category": "phonetic_balance",
+ "text": "Each day she became a more vital part of him.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0497.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "EACH DAY SHE BECAME A MORE VITAL PART OF HIM",
+ "raw": {
+ "normalized_reference": "each day she became a more vital part of him.",
+ "normalized_hypothesis": "each day she became a more vital part of him",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "each day she became a more vital part of him",
+ "normalized_hypothesis": "each day she became a more vital part of him",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0498",
+ "category": "phonetic_balance",
+ "text": "But why continue the tirade, for tirade it was.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0498.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WHY CONTINUE THE TIRADE FOR TIRADE IT WAS",
+ "raw": {
+ "normalized_reference": "but why continue the tirade, for tirade it was.",
+ "normalized_hypothesis": "why continue the tirade for tirade it was",
+ "errors": 1,
+ "words": 9,
+ "wer": 0.1111111111111111
+ },
+ "semantic": {
+ "normalized_reference": "but why continue the tirade for tirade it was",
+ "normalized_hypothesis": "why continue the tirade for tirade it was",
+ "errors": 1,
+ "words": 9,
+ "wer": 0.1111111111111111
+ }
+ },
+ {
+ "id": "release500_0499",
+ "category": "phonetic_balance",
+ "text": "That is what distinguishes all of us from the lower animals.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0499.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THAT IS WHAT DISTINGUISHES ALL OF US FROM THE LOWER ANIMALS",
+ "raw": {
+ "normalized_reference": "that is what distinguishes all of us from the lower animals.",
+ "normalized_hypothesis": "that is what distinguishes all of us from the lower animals",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "that is what distinguishes all of us from the lower animals",
+ "normalized_hypothesis": "that is what distinguishes all of us from the lower animals",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0500",
+ "category": "phonetic_balance",
+ "text": "There was something pathetic in the girl's attitude now.",
+ "model": "piper-ryan-low",
+ "audio": "benchmark/audio/piper-ryan-low/release500_0500.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THERE WAS SOMETHING PATHETIC IN THE GIRL'S ATTITUDE NOW",
+ "raw": {
+ "normalized_reference": "there was something pathetic in the girl's attitude now.",
+ "normalized_hypothesis": "there was something pathetic in the girl's attitude now",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "there was something pathetic in the girl is attitude now",
+ "normalized_hypothesis": "there was something pathetic in the girl is attitude now",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0001",
+ "category": "conversational",
+ "text": "It wasn't until later that I realized what had actually happened.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0001.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT WASN'T UNTIL LATER THAT I REALIZED WHAT HAD ACTUALLY HAPPENED",
+ "raw": {
+ "normalized_reference": "it wasn't until later that i realized what had actually happened.",
+ "normalized_hypothesis": "it wasn't until later that i realized what had actually happened",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it was not until later that i realized what had actually happened",
+ "normalized_hypothesis": "it was not until later that i realized what had actually happened",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0002",
+ "category": "descriptive",
+ "text": "The light at that time of afternoon is perfect for about twenty minutes.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0002.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE LIGHT AT THAT TIME OF AFTERNOON IS PERFECT FOR ABOUT TWENTY MINUTES",
+ "raw": {
+ "normalized_reference": "the light at that time of afternoon is perfect for about twenty minutes.",
+ "normalized_hypothesis": "the light at that time of afternoon is perfect for about twenty minutes",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the light at that time of afternoon is perfect for about 20 minutes",
+ "normalized_hypothesis": "the light at that time of afternoon is perfect for about 20 minutes",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0003",
+ "category": "emotional",
+ "text": "That was the scariest thing I've ever done, and also maybe the best.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0003.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THAT WAS THE SKARRIEST THING I AVE EVER DONE AND ALSO MEYBE THE BEST",
+ "raw": {
+ "normalized_reference": "that was the scariest thing i've ever done, and also maybe the best.",
+ "normalized_hypothesis": "that was the skarriest thing i ave ever done and also meybe the best",
+ "errors": 4,
+ "words": 13,
+ "wer": 0.3076923076923077
+ },
+ "semantic": {
+ "normalized_reference": "that was the scariest thing i have ever done and also maybe the best",
+ "normalized_hypothesis": "that was the skarriest thing i ave ever done and also meybe the best",
+ "errors": 3,
+ "words": 14,
+ "wer": 0.21428571428571427
+ }
+ },
+ {
+ "id": "release500_0004",
+ "category": "homographs",
+ "text": "After I record the record, please present the present to the bass player.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0004.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "AFTER I RECORD THE RECORD PLEASE PRESENT THE PRESENT TO THE BASE PLAYER",
+ "raw": {
+ "normalized_reference": "after i record the record, please present the present to the bass player.",
+ "normalized_hypothesis": "after i record the record please present the present to the base player",
+ "errors": 1,
+ "words": 13,
+ "wer": 0.07692307692307693
+ },
+ "semantic": {
+ "normalized_reference": "after i record the record please present the present to the bass player",
+ "normalized_hypothesis": "after i record the record please present the present to the base player",
+ "errors": 1,
+ "words": 13,
+ "wer": 0.07692307692307693
+ }
+ },
+ {
+ "id": "release500_0005",
+ "category": "long",
+ "text": "If I'm being completely honest, I had no idea what I was doing for most of it, and I think that might be the most normal thing about the whole experience.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0005.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IF I'M BEING COMPLETELY HONEST I HAD NO IDEA WHAT I WAS DOING FOR MOST OF IT AND I THINK THAT MIGHT BE THE MOST NORMAL THING ABOUT THE WHOLE EXPERIENCE",
+ "raw": {
+ "normalized_reference": "if i'm being completely honest, i had no idea what i was doing for most of it, and i think that might be the most normal thing about the whole experience.",
+ "normalized_hypothesis": "if i'm being completely honest i had no idea what i was doing for most of it and i think that might be the most normal thing about the whole experience",
+ "errors": 0,
+ "words": 31,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "if i am being completely honest i had no idea what i was doing for most of it and i think that might be the most normal thing about the whole experience",
+ "normalized_hypothesis": "if i am being completely honest i had no idea what i was doing for most of it and i think that might be the most normal thing about the whole experience",
+ "errors": 0,
+ "words": 32,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0006",
+ "category": "long_range",
+ "text": "The committee postponed its decision after discovering that three pages of the report had been printed in the wrong order.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0006.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE COMMITTEE POSTPONED ITS DECISION AFTER DISCOVERING THAT THREE PAGES OF THE REPORT HAD BEEN PRINTED IN THE WRONG ORDER",
+ "raw": {
+ "normalized_reference": "the committee postponed its decision after discovering that three pages of the report had been printed in the wrong order.",
+ "normalized_hypothesis": "the committee postponed its decision after discovering that three pages of the report had been printed in the wrong order",
+ "errors": 0,
+ "words": 20,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the committee postponed its decision after discovering that 3 pages of the report had been printed in the wrong order",
+ "normalized_hypothesis": "the committee postponed its decision after discovering that 3 pages of the report had been printed in the wrong order",
+ "errors": 0,
+ "words": 20,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0007",
+ "category": "names_places",
+ "text": "Gwendolyn photographed the eucalyptus trees outside Ljubljana.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0007.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "GWENDOLEN PHOTOGRAPHED THE EUCALYPTUS TREES OUTSIDE LUBLIANA",
+ "raw": {
+ "normalized_reference": "gwendolyn photographed the eucalyptus trees outside ljubljana.",
+ "normalized_hypothesis": "gwendolen photographed the eucalyptus trees outside lubliana",
+ "errors": 2,
+ "words": 7,
+ "wer": 0.2857142857142857
+ },
+ "semantic": {
+ "normalized_reference": "gwendolyn photographed the eucalyptus trees outside ljubljana",
+ "normalized_hypothesis": "gwendolen photographed the eucalyptus trees outside lubliana",
+ "errors": 2,
+ "words": 7,
+ "wer": 0.2857142857142857
+ }
+ },
+ {
+ "id": "release500_0008",
+ "category": "narrative",
+ "text": "He had this theory, and the more time passes, the more I think he was right.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0008.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE HAD THIS THEORY AND THE MORE TIME PASSES THE MORE I THINK HE WAS RIGHT",
+ "raw": {
+ "normalized_reference": "he had this theory, and the more time passes, the more i think he was right.",
+ "normalized_hypothesis": "he had this theory and the more time passes the more i think he was right",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he had this theory and the more time passes the more i think he was right",
+ "normalized_hypothesis": "he had this theory and the more time passes the more i think he was right",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0009",
+ "category": "numbers",
+ "text": "The package weighs twelve point six kilograms and arrived on July twenty-first.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0009.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE PACKAGE WEIGHS TWELVE POINT SIX KILOGRAMS AND ARRIVED ON JULY TWENTY FIRST",
+ "raw": {
+ "normalized_reference": "the package weighs twelve point six kilograms and arrived on july twenty-first.",
+ "normalized_hypothesis": "the package weighs twelve point six kilograms and arrived on july twenty first",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the package weighs 12.6 kilograms and arrived on july 21st",
+ "normalized_hypothesis": "the package weighs 12.6 kilograms and arrived on july 21st",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0010",
+ "category": "punctuation",
+ "text": "First, close the window; second, turn off the lamp; finally, lock the door.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0010.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "FIRST CLOSE THE WINDOW SECOND TURN OFF THE LAMP FINALLY LOCK THE DOOR",
+ "raw": {
+ "normalized_reference": "first, close the window; second, turn off the lamp; finally, lock the door.",
+ "normalized_hypothesis": "first close the window second turn off the lamp finally lock the door",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "1st close the window 2nd turn off the lamp finally lock the door",
+ "normalized_hypothesis": "1st close the window 2nd turn off the lamp finally lock the door",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0011",
+ "category": "question",
+ "text": "When did you first realize something was wrong?",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0011.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WHEN DID YOU FIRST REALIZE SOMETHING WAS WRONG",
+ "raw": {
+ "normalized_reference": "when did you first realize something was wrong?",
+ "normalized_hypothesis": "when did you first realize something was wrong",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "when did you 1st realize something was wrong",
+ "normalized_hypothesis": "when did you 1st realize something was wrong",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0012",
+ "category": "rare_words",
+ "text": "The archaeologist catalogued a Byzantine reliquary and several anachronistic fragments.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0012.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE ARCHAEOLOGIST CATALOGUED A BISANTINE RHILIQUARY AND SEVERAL ANACHRONISTIC FRAGMENTS",
+ "raw": {
+ "normalized_reference": "the archaeologist catalogued a byzantine reliquary and several anachronistic fragments.",
+ "normalized_hypothesis": "the archaeologist catalogued a bisantine rhiliquary and several anachronistic fragments",
+ "errors": 2,
+ "words": 10,
+ "wer": 0.2
+ },
+ "semantic": {
+ "normalized_reference": "the archaeologist catalogued a byzantine reliquary and several anachronistic fragments",
+ "normalized_hypothesis": "the archaeologist catalogued a bisantine rhiliquary and several anachronistic fragments",
+ "errors": 2,
+ "words": 10,
+ "wer": 0.2
+ }
+ },
+ {
+ "id": "release500_0013",
+ "category": "short",
+ "text": "I have no idea.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0013.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I HAVE NO IDEA",
+ "raw": {
+ "normalized_reference": "i have no idea.",
+ "normalized_hypothesis": "i have no idea",
+ "errors": 0,
+ "words": 4,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i have no idea",
+ "normalized_hypothesis": "i have no idea",
+ "errors": 0,
+ "words": 4,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0014",
+ "category": "technical",
+ "text": "The system runs on three core components that all have to stay in sync.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0014.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE SYSTEM RUNS ON THREE COR COMPONENTS THAT ALL HAVE TO STAY IN SINK",
+ "raw": {
+ "normalized_reference": "the system runs on three core components that all have to stay in sync.",
+ "normalized_hypothesis": "the system runs on three cor components that all have to stay in sink",
+ "errors": 2,
+ "words": 14,
+ "wer": 0.14285714285714285
+ },
+ "semantic": {
+ "normalized_reference": "the system runs on 3 core components that all have to stay in sync",
+ "normalized_hypothesis": "the system runs on 3 cor components that all have to stay in sink",
+ "errors": 2,
+ "words": 14,
+ "wer": 0.14285714285714285
+ }
+ },
+ {
+ "id": "release500_0015",
+ "category": "conversational",
+ "text": "We improvised the entire second half and somehow nobody noticed.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0015.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WE IMPROVISED THE ENTIRE SECOND HALF AND SOMEHOW NOBODY NOTICED",
+ "raw": {
+ "normalized_reference": "we improvised the entire second half and somehow nobody noticed.",
+ "normalized_hypothesis": "we improvised the entire second half and somehow nobody noticed",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "we improvised the entire 2nd half and somehow nobody noticed",
+ "normalized_hypothesis": "we improvised the entire 2nd half and somehow nobody noticed",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0016",
+ "category": "descriptive",
+ "text": "There's a smell in old buildings that's impossible to describe but immediately familiar.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0016.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THERE'S A SMELL IN OLD BUILDINGS THAT'S IMPOSSIBLE TO DESCRIBE BUT IMMEDIATELY FAMILIAR",
+ "raw": {
+ "normalized_reference": "there's a smell in old buildings that's impossible to describe but immediately familiar.",
+ "normalized_hypothesis": "there's a smell in old buildings that's impossible to describe but immediately familiar",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "there is a smell in old buildings that is impossible to describe but immediately familiar",
+ "normalized_hypothesis": "there is a smell in old buildings that is impossible to describe but immediately familiar",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0017",
+ "category": "emotional",
+ "text": "I specifically asked for one thing. One thing. And it still didn't happen.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0017.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I SPECIFICALLY ASKED FOR ONE THING ONE THING AND IT STILL DIDN'T HAPPEN",
+ "raw": {
+ "normalized_reference": "i specifically asked for one thing. one thing. and it still didn't happen.",
+ "normalized_hypothesis": "i specifically asked for one thing one thing and it still didn't happen",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i specifically asked for one thing one thing and it still did not happen",
+ "normalized_hypothesis": "i specifically asked for one thing one thing and it still did not happen",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0018",
+ "category": "homographs",
+ "text": "They refuse to place the refuse beside the produce aisle.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0018.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THEY REFUSED TO PLACE THE REFUSE BESIDE THE PRODUCE ISLE",
+ "raw": {
+ "normalized_reference": "they refuse to place the refuse beside the produce aisle.",
+ "normalized_hypothesis": "they refused to place the refuse beside the produce isle",
+ "errors": 2,
+ "words": 10,
+ "wer": 0.2
+ },
+ "semantic": {
+ "normalized_reference": "they refuse to place the refuse beside the produce aisle",
+ "normalized_hypothesis": "they refused to place the refuse beside the produce isle",
+ "errors": 2,
+ "words": 10,
+ "wer": 0.2
+ }
+ },
+ {
+ "id": "release500_0019",
+ "category": "long",
+ "text": "There were days last year when I genuinely didn't know how I was going to get through it, and somehow I did, and I still don't fully understand how.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0019.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THERE WERE DAYS LAST YEAR WHEN I GENUINELY DIDN'T KNOW HOW I WAS GOING TO GET THROUGH IT AND SOMEHOW I DID AND I STILL DON'T FULLY UNDERSTAND HOW",
+ "raw": {
+ "normalized_reference": "there were days last year when i genuinely didn't know how i was going to get through it, and somehow i did, and i still don't fully understand how.",
+ "normalized_hypothesis": "there were days last year when i genuinely didn't know how i was going to get through it and somehow i did and i still don't fully understand how",
+ "errors": 0,
+ "words": 29,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "there were days last year when i genuinely did not know how i was going to get through it and somehow i did and i still do not fully understand how",
+ "normalized_hypothesis": "there were days last year when i genuinely did not know how i was going to get through it and somehow i did and i still do not fully understand how",
+ "errors": 0,
+ "words": 31,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0020",
+ "category": "long_range",
+ "text": "Before announcing the winner, the judge thanked the volunteers who had worked through the storm to repair the damaged stage.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0020.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "BEFORE ANNOUNCING THE WINNER THE JUDGE THANKED THE VOLUNTEERS WHO HAD WORKED THROUGH THE STORM TO REPAIR THE DAMAGED STAGE",
+ "raw": {
+ "normalized_reference": "before announcing the winner, the judge thanked the volunteers who had worked through the storm to repair the damaged stage.",
+ "normalized_hypothesis": "before announcing the winner the judge thanked the volunteers who had worked through the storm to repair the damaged stage",
+ "errors": 0,
+ "words": 20,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "before announcing the winner the judge thanked the volunteers who had worked through the storm to repair the damaged stage",
+ "normalized_hypothesis": "before announcing the winner the judge thanked the volunteers who had worked through the storm to repair the damaged stage",
+ "errors": 0,
+ "words": 20,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0021",
+ "category": "names_places",
+ "text": "Professor Nguyen mailed the fluorescent poster to Reykjavik on Thursday.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0021.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "PROFESSOR NEWYAN MAILED THE FORESCENT POSTER TO REKJHEVIK ON THURSDAY",
+ "raw": {
+ "normalized_reference": "professor nguyen mailed the fluorescent poster to reykjavik on thursday.",
+ "normalized_hypothesis": "professor newyan mailed the forescent poster to rekjhevik on thursday",
+ "errors": 3,
+ "words": 10,
+ "wer": 0.3
+ },
+ "semantic": {
+ "normalized_reference": "professor nguyen mailed the fluorescent poster to reykjavik on thursday",
+ "normalized_hypothesis": "professor newyan mailed the forescent poster to rekjhevik on thursday",
+ "errors": 3,
+ "words": 10,
+ "wer": 0.3
+ }
+ },
+ {
+ "id": "release500_0022",
+ "category": "narrative",
+ "text": "The plan was simple. The execution was anything but.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0022.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE PLAN WAS SIMPLE THE EXECUTION WAS ANYTHING BUT",
+ "raw": {
+ "normalized_reference": "the plan was simple. the execution was anything but.",
+ "normalized_hypothesis": "the plan was simple the execution was anything but",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the plan was simple the execution was anything but",
+ "normalized_hypothesis": "the plan was simple the execution was anything but",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0023",
+ "category": "numbers",
+ "text": "Her score improved from eighty-two percent to ninety-seven percent in two weeks.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0023.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HER SCORE OL PROVE FROM EIGHTY TWO PER CENT TO NINETY SEVEN PER CENT IN TWO WEEKS",
+ "raw": {
+ "normalized_reference": "her score improved from eighty-two percent to ninety-seven percent in two weeks.",
+ "normalized_hypothesis": "her score ol prove from eighty two per cent to ninety seven per cent in two weeks",
+ "errors": 6,
+ "words": 14,
+ "wer": 0.42857142857142855
+ },
+ "semantic": {
+ "normalized_reference": "her score improved from 82% to 97% in 2 weeks",
+ "normalized_hypothesis": "her score ol prove from 82% to 97% in 2 weeks",
+ "errors": 2,
+ "words": 10,
+ "wer": 0.2
+ }
+ },
+ {
+ "id": "release500_0024",
+ "category": "punctuation",
+ "text": "Why would anyone pack spoons, batteries, and a violin for one short trip?",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0024.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WHY WOULD ANY ONE PACK SPOONS BATTERIES AND A VIOLIN FOR ONE SHORT TRIP",
+ "raw": {
+ "normalized_reference": "why would anyone pack spoons, batteries, and a violin for one short trip?",
+ "normalized_hypothesis": "why would any one pack spoons batteries and a violin for one short trip",
+ "errors": 2,
+ "words": 13,
+ "wer": 0.15384615384615385
+ },
+ "semantic": {
+ "normalized_reference": "why would anyone pack spoons batteries and a violin for one short trip",
+ "normalized_hypothesis": "why would any one pack spoons batteries and a violin for one short trip",
+ "errors": 2,
+ "words": 13,
+ "wer": 0.15384615384615385
+ }
+ },
+ {
+ "id": "release500_0025",
+ "category": "question",
+ "text": "How long have you been sitting on this?",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0025.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HOW LONG HAVE YOU BEEN SITTING ON THIS",
+ "raw": {
+ "normalized_reference": "how long have you been sitting on this?",
+ "normalized_hypothesis": "how long have you been sitting on this",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "how long have you been sitting on this",
+ "normalized_hypothesis": "how long have you been sitting on this",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0026",
+ "category": "rare_words",
+ "text": "A mischievous chameleon rested beside the chrysanthemum and the bougainvillea.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0026.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "A MISCHIEVOUS CAMELIAN RESTED BESIDE THE CHRYSANTHEMUM AND THE BARGAN VILLIA",
+ "raw": {
+ "normalized_reference": "a mischievous chameleon rested beside the chrysanthemum and the bougainvillea.",
+ "normalized_hypothesis": "a mischievous camelian rested beside the chrysanthemum and the bargan villia",
+ "errors": 3,
+ "words": 10,
+ "wer": 0.3
+ },
+ "semantic": {
+ "normalized_reference": "a mischievous chameleon rested beside the chrysanthemum and the bougainvillea",
+ "normalized_hypothesis": "a mischievous camelian rested beside the chrysanthemum and the bargan villia",
+ "errors": 3,
+ "words": 10,
+ "wer": 0.3
+ }
+ },
+ {
+ "id": "release500_0027",
+ "category": "short",
+ "text": "We don't talk about that.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0027.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WE DON'T TALK ABOUT THAT",
+ "raw": {
+ "normalized_reference": "we don't talk about that.",
+ "normalized_hypothesis": "we don't talk about that",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "we do not talk about that",
+ "normalized_hypothesis": "we do not talk about that",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0028",
+ "category": "technical",
+ "text": "Hash the UTF eight payload with S H A two fifty-six before opening the checkpoint.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0028.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HASH THE U T F A PEYLOAD WITH S H A TWO FIFTY SIX BEFORE OPENING THE CHECK BAIT",
+ "raw": {
+ "normalized_reference": "hash the utf eight payload with s h a two fifty-six before opening the checkpoint.",
+ "normalized_hypothesis": "hash the u t f a peyload with s h a two fifty six before opening the check bait",
+ "errors": 7,
+ "words": 16,
+ "wer": 0.4375
+ },
+ "semantic": {
+ "normalized_reference": "hash the utf 8 payload with s h a 256 before opening the checkpoint",
+ "normalized_hypothesis": "hash the u t f a peyload with s h a 256 before opening the check bait",
+ "errors": 7,
+ "words": 14,
+ "wer": 0.5
+ }
+ },
+ {
+ "id": "release500_0029",
+ "category": "conversational",
+ "text": "I thought I had it all figured out, and then this happened.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0029.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I THOUGHT I HAD IT ALL FIGURED OUT AND THEN THIS HAPPENED",
+ "raw": {
+ "normalized_reference": "i thought i had it all figured out, and then this happened.",
+ "normalized_hypothesis": "i thought i had it all figured out and then this happened",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i thought i had it all figured out and then this happened",
+ "normalized_hypothesis": "i thought i had it all figured out and then this happened",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0030",
+ "category": "descriptive",
+ "text": "The kind of quiet you only get at three in the morning in a city.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0030.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE KIND OF QUIET YOU ONLY GET AT THREE IN THE MORNING IN A CITY",
+ "raw": {
+ "normalized_reference": "the kind of quiet you only get at three in the morning in a city.",
+ "normalized_hypothesis": "the kind of quiet you only get at three in the morning in a city",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the kind of quiet you only get at 3 in the morning in a city",
+ "normalized_hypothesis": "the kind of quiet you only get at 3 in the morning in a city",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0031",
+ "category": "emotional",
+ "text": "The joy of it was almost overwhelming, like too much good all at once.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0031.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE JOY OF IT WAS ALMOST OVERWHELMING LIKE TOO MUCH COULD ALL AT ONCE",
+ "raw": {
+ "normalized_reference": "the joy of it was almost overwhelming, like too much good all at once.",
+ "normalized_hypothesis": "the joy of it was almost overwhelming like too much could all at once",
+ "errors": 1,
+ "words": 14,
+ "wer": 0.07142857142857142
+ },
+ "semantic": {
+ "normalized_reference": "the joy of it was almost overwhelming like too much good all at once",
+ "normalized_hypothesis": "the joy of it was almost overwhelming like too much could all at once",
+ "errors": 1,
+ "words": 14,
+ "wer": 0.07142857142857142
+ }
+ },
+ {
+ "id": "release500_0032",
+ "category": "homographs",
+ "text": "Please close the door while I stand close to the microphone.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0032.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "PLEASE CLOSE THE DOOR WHILE I STAND CLOSE TO THE MICROPHONE",
+ "raw": {
+ "normalized_reference": "please close the door while i stand close to the microphone.",
+ "normalized_hypothesis": "please close the door while i stand close to the microphone",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "please close the door while i stand close to the microphone",
+ "normalized_hypothesis": "please close the door while i stand close to the microphone",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0033",
+ "category": "long",
+ "text": "I think the most underrated skill in the world is knowing when to stop explaining yourself and just let people draw their own conclusions.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0033.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I THINK THE MOST UNDERRATED SKILL IN THE WORLD IS KNOWING WHEN TO STOP EXPLAINING YOURSELF AND JUST LET PEOPLE DRAW THEIR OWN CONCLUSIONS",
+ "raw": {
+ "normalized_reference": "i think the most underrated skill in the world is knowing when to stop explaining yourself and just let people draw their own conclusions.",
+ "normalized_hypothesis": "i think the most underrated skill in the world is knowing when to stop explaining yourself and just let people draw their own conclusions",
+ "errors": 0,
+ "words": 24,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i think the most underrated skill in the world is knowing when to stop explaining yourself and just let people draw their own conclusions",
+ "normalized_hypothesis": "i think the most underrated skill in the world is knowing when to stop explaining yourself and just let people draw their own conclusions",
+ "errors": 0,
+ "words": 24,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0034",
+ "category": "long_range",
+ "text": "When the final bus disappeared around the corner, Nadia remembered that her umbrella, notebook, and lunch were still inside the quiet office.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0034.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WHEN THE FINAL BUST DISAPPEARED AROUND THE CORNER NARTYA REMEMBERED THET HER UMBRELLA NOTE BOOK AND LUNCH WERE STILL INSIDE THE QUIET OFFICE",
+ "raw": {
+ "normalized_reference": "when the final bus disappeared around the corner, nadia remembered that her umbrella, notebook, and lunch were still inside the quiet office.",
+ "normalized_hypothesis": "when the final bust disappeared around the corner nartya remembered thet her umbrella note book and lunch were still inside the quiet office",
+ "errors": 5,
+ "words": 22,
+ "wer": 0.22727272727272727
+ },
+ "semantic": {
+ "normalized_reference": "when the final bus disappeared around the corner nadia remembered that her umbrella notebook and lunch were still inside the quiet office",
+ "normalized_hypothesis": "when the final bust disappeared around the corner nartya remembered thet her umbrella note book and lunch were still inside the quiet office",
+ "errors": 5,
+ "words": 22,
+ "wer": 0.22727272727272727
+ }
+ },
+ {
+ "id": "release500_0035",
+ "category": "names_places",
+ "text": "Aurelia and Joaquin crossed the Schuylkill River before noon.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0035.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "ORALYA AND CHOQUING CROSSED THE SCHIOKEL RIVER BEFORE NOON",
+ "raw": {
+ "normalized_reference": "aurelia and joaquin crossed the schuylkill river before noon.",
+ "normalized_hypothesis": "oralya and choquing crossed the schiokel river before noon",
+ "errors": 3,
+ "words": 9,
+ "wer": 0.3333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "aurelia and joaquin crossed the schuylkill river before noon",
+ "normalized_hypothesis": "oralya and choquing crossed the schiokel river before noon",
+ "errors": 3,
+ "words": 9,
+ "wer": 0.3333333333333333
+ }
+ },
+ {
+ "id": "release500_0036",
+ "category": "narrative",
+ "text": "The village had a name, but nobody used it. Everyone just called it home.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0036.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE VILLAGE HAD A NAME BUT NOBODY USED IT EVERY ONE JUST CALLED IT HOME",
+ "raw": {
+ "normalized_reference": "the village had a name, but nobody used it. everyone just called it home.",
+ "normalized_hypothesis": "the village had a name but nobody used it every one just called it home",
+ "errors": 2,
+ "words": 14,
+ "wer": 0.14285714285714285
+ },
+ "semantic": {
+ "normalized_reference": "the village had a name but nobody used it everyone just called it home",
+ "normalized_hypothesis": "the village had a name but nobody used it every one just called it home",
+ "errors": 2,
+ "words": 14,
+ "wer": 0.14285714285714285
+ }
+ },
+ {
+ "id": "release500_0037",
+ "category": "numbers",
+ "text": "Increase the gain by three decibels, wait five seconds, then reduce it by one.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0037.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "INCREASE THE GAME BY THREE DESIBLES WAIT FIVE SECONDS THEN REDUCE IT BY ONE",
+ "raw": {
+ "normalized_reference": "increase the gain by three decibels, wait five seconds, then reduce it by one.",
+ "normalized_hypothesis": "increase the game by three desibles wait five seconds then reduce it by one",
+ "errors": 2,
+ "words": 14,
+ "wer": 0.14285714285714285
+ },
+ "semantic": {
+ "normalized_reference": "increase the gain by 3 decibels wait 5 seconds then reduce it by one",
+ "normalized_hypothesis": "increase the game by 3 desibles wait 5 seconds then reduce it by one",
+ "errors": 2,
+ "words": 14,
+ "wer": 0.14285714285714285
+ }
+ },
+ {
+ "id": "release500_0038",
+ "category": "punctuation",
+ "text": "The result was unexpected: clear, stable, and surprisingly natural.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0038.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE RESULT WAS UNEXPECTED CLEAR STABLE AND SURPRISINGLY NATURAL",
+ "raw": {
+ "normalized_reference": "the result was unexpected: clear, stable, and surprisingly natural.",
+ "normalized_hypothesis": "the result was unexpected clear stable and surprisingly natural",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the result was unexpected clear stable and surprisingly natural",
+ "normalized_hypothesis": "the result was unexpected clear stable and surprisingly natural",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0039",
+ "category": "question",
+ "text": "Who told you that was the right way to do it?",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0039.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WHO TOLD YOU THAT WAS THE RIGHT WAY TO DO IT",
+ "raw": {
+ "normalized_reference": "who told you that was the right way to do it?",
+ "normalized_hypothesis": "who told you that was the right way to do it",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "who told you that was the right way to do it",
+ "normalized_hypothesis": "who told you that was the right way to do it",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0040",
+ "category": "rare_words",
+ "text": "The meticulous cartographer examined an iridescent archipelago through binoculars.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0040.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE MARTICULOUS CARTOGRAPHER EXAMINED AN IRUDESCENT ARCHIPELAGO THROUGH BENOCULARS",
+ "raw": {
+ "normalized_reference": "the meticulous cartographer examined an iridescent archipelago through binoculars.",
+ "normalized_hypothesis": "the marticulous cartographer examined an irudescent archipelago through benoculars",
+ "errors": 3,
+ "words": 9,
+ "wer": 0.3333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "the meticulous cartographer examined an iridescent archipelago through binoculars",
+ "normalized_hypothesis": "the marticulous cartographer examined an irudescent archipelago through benoculars",
+ "errors": 3,
+ "words": 9,
+ "wer": 0.3333333333333333
+ }
+ },
+ {
+ "id": "release500_0041",
+ "category": "short",
+ "text": "I mean. Yeah. Kind of.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0041.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I MEAN NE KIND OV",
+ "raw": {
+ "normalized_reference": "i mean. yeah. kind of.",
+ "normalized_hypothesis": "i mean ne kind ov",
+ "errors": 2,
+ "words": 5,
+ "wer": 0.4
+ },
+ "semantic": {
+ "normalized_reference": "i mean yeah kind of",
+ "normalized_hypothesis": "i mean ne kind ov",
+ "errors": 2,
+ "words": 5,
+ "wer": 0.4
+ }
+ },
+ {
+ "id": "release500_0042",
+ "category": "technical",
+ "text": "The logs are your best friend when something behaves unexpectedly.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0042.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE LOGS ARE YOUR BEST FRIEND WHEN SOMETHING BEHAVES UNEXPECTEDLY",
+ "raw": {
+ "normalized_reference": "the logs are your best friend when something behaves unexpectedly.",
+ "normalized_hypothesis": "the logs are your best friend when something behaves unexpectedly",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the logs are your best friend when something behaves unexpectedly",
+ "normalized_hypothesis": "the logs are your best friend when something behaves unexpectedly",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0043",
+ "category": "conversational",
+ "text": "He kept talking and I kept nodding and neither of us was listening.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0043.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE KEPT TALKING AND I KEPT NODDING AND NEITHER OF US WAS LISTENING",
+ "raw": {
+ "normalized_reference": "he kept talking and i kept nodding and neither of us was listening.",
+ "normalized_hypothesis": "he kept talking and i kept nodding and neither of us was listening",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he kept talking and i kept nodding and neither of us was listening",
+ "normalized_hypothesis": "he kept talking and i kept nodding and neither of us was listening",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0044",
+ "category": "descriptive",
+ "text": "It's one of those places where everyone seems to know each other already.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0044.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT'S ONE OF THOSE PLACES WHERE EVERY ONE SEEMS TO KNOW EACH OTHER ALREADY",
+ "raw": {
+ "normalized_reference": "it's one of those places where everyone seems to know each other already.",
+ "normalized_hypothesis": "it's one of those places where every one seems to know each other already",
+ "errors": 2,
+ "words": 13,
+ "wer": 0.15384615384615385
+ },
+ "semantic": {
+ "normalized_reference": "it is one of those places where everyone seems to know each other already",
+ "normalized_hypothesis": "it is one of those places where every one seems to know each other already",
+ "errors": 2,
+ "words": 14,
+ "wer": 0.14285714285714285
+ }
+ },
+ {
+ "id": "release500_0045",
+ "category": "emotional",
+ "text": "He said he was proud of me and nobody had ever said that to me before.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0045.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE SAID HE WAS PROUD OF ME AND NOBODY HAD EVER SAID THAT TO ME BEFORE",
+ "raw": {
+ "normalized_reference": "he said he was proud of me and nobody had ever said that to me before.",
+ "normalized_hypothesis": "he said he was proud of me and nobody had ever said that to me before",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he said he was proud of me and nobody had ever said that to me before",
+ "normalized_hypothesis": "he said he was proud of me and nobody had ever said that to me before",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0046",
+ "category": "homographs",
+ "text": "The minute detail took a minute to explain clearly.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0046.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE MINUTE DETAIL TOOK A MINUTE TO EXPLAIN CLEARLY",
+ "raw": {
+ "normalized_reference": "the minute detail took a minute to explain clearly.",
+ "normalized_hypothesis": "the minute detail took a minute to explain clearly",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the minute detail took a minute to explain clearly",
+ "normalized_hypothesis": "the minute detail took a minute to explain clearly",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0047",
+ "category": "long",
+ "text": "The strange thing about really wanting something is that getting it never quite feels the way you imagined it would, and not getting it never quite feels as bad.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0047.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE STRANGE THING ABOUT REALLY WANTING SOMETHING IS THAT GETTING IT NEVER QUITE FEELS THE WAY YOU IMAGINED IT WOULD AND NOT GETTING IT NEVER QUITE FEELS AS BAD",
+ "raw": {
+ "normalized_reference": "the strange thing about really wanting something is that getting it never quite feels the way you imagined it would, and not getting it never quite feels as bad.",
+ "normalized_hypothesis": "the strange thing about really wanting something is that getting it never quite feels the way you imagined it would and not getting it never quite feels as bad",
+ "errors": 0,
+ "words": 29,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the strange thing about really wanting something is that getting it never quite feels the way you imagined it would and not getting it never quite feels as bad",
+ "normalized_hypothesis": "the strange thing about really wanting something is that getting it never quite feels the way you imagined it would and not getting it never quite feels as bad",
+ "errors": 0,
+ "words": 29,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0048",
+ "category": "long_range",
+ "text": "Even though the recipe looked simple at first, preparing the sauce required patience, careful timing, and a much larger pan.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0048.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "EVEN THOUGH THE RECIPE LOOKED SIMPLE AT FIRST PREPARING THE SAUCE REQUIRED PATIENCE CAREFUL TIMING AND A MUCH LARGER PAN",
+ "raw": {
+ "normalized_reference": "even though the recipe looked simple at first, preparing the sauce required patience, careful timing, and a much larger pan.",
+ "normalized_hypothesis": "even though the recipe looked simple at first preparing the sauce required patience careful timing and a much larger pan",
+ "errors": 0,
+ "words": 20,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "even though the recipe looked simple at 1st preparing the sauce required patience careful timing and a much larger pan",
+ "normalized_hypothesis": "even though the recipe looked simple at 1st preparing the sauce required patience careful timing and a much larger pan",
+ "errors": 0,
+ "words": 20,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0049",
+ "category": "names_places",
+ "text": "Xiomara met Siobhan near the old library in Worcestershire.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0049.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SIAMARA MET CHAVONN NEAR THE OLD LIBRARY IN WORESTERSHIRE",
+ "raw": {
+ "normalized_reference": "xiomara met siobhan near the old library in worcestershire.",
+ "normalized_hypothesis": "siamara met chavonn near the old library in worestershire",
+ "errors": 3,
+ "words": 9,
+ "wer": 0.3333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "xiomara met siobhan near the old library in worcestershire",
+ "normalized_hypothesis": "siamara met chavonn near the old library in worestershire",
+ "errors": 3,
+ "words": 9,
+ "wer": 0.3333333333333333
+ }
+ },
+ {
+ "id": "release500_0050",
+ "category": "narrative",
+ "text": "She walked into the room and everyone went quiet, not because she was loud, but because she wasn't.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0050.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SHE WALKED INTO THE ROOM AND EVERYONE WENT QUIET NOT BECAUSE SHE WAS LOUD BUT BECAUSE SHE WASN'T",
+ "raw": {
+ "normalized_reference": "she walked into the room and everyone went quiet, not because she was loud, but because she wasn't.",
+ "normalized_hypothesis": "she walked into the room and everyone went quiet not because she was loud but because she wasn't",
+ "errors": 0,
+ "words": 18,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "she walked into the room and everyone went quiet not because she was loud but because she was not",
+ "normalized_hypothesis": "she walked into the room and everyone went quiet not because she was loud but because she was not",
+ "errors": 0,
+ "words": 19,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0051",
+ "category": "numbers",
+ "text": "At exactly seven forty-five, Morgan placed eighteen dollars and fifty cents beside the notebook.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0051.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "AT EXACTLY SEVEN FORTY FIVE MORGAN PLACED EIGHTEEN DOLLARS AND FIFTY CENTS BESIDE THE NOTEBOOK",
+ "raw": {
+ "normalized_reference": "at exactly seven forty-five, morgan placed eighteen dollars and fifty cents beside the notebook.",
+ "normalized_hypothesis": "at exactly seven forty five morgan placed eighteen dollars and fifty cents beside the notebook",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "at exactly 745 morgan placed $18.50 beside the notebook",
+ "normalized_hypothesis": "at exactly 745 morgan placed $18.50 beside the notebook",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0052",
+ "category": "punctuation",
+ "text": "Does punctuation work? Yes: commas, semicolons, and periods should create sensible boundaries.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0052.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "DOES PUNCTUATION WORK YES COMMERCE SEMICOLANS AND PERIODS SHOULD CREATE SENSIBLE BOUNDARIES",
+ "raw": {
+ "normalized_reference": "does punctuation work? yes: commas, semicolons, and periods should create sensible boundaries.",
+ "normalized_hypothesis": "does punctuation work yes commerce semicolans and periods should create sensible boundaries",
+ "errors": 2,
+ "words": 12,
+ "wer": 0.16666666666666666
+ },
+ "semantic": {
+ "normalized_reference": "does punctuation work yes commas semicolons and periods should create sensible boundaries",
+ "normalized_hypothesis": "does punctuation work yes commerce semicolans and periods should create sensible boundaries",
+ "errors": 2,
+ "words": 12,
+ "wer": 0.16666666666666666
+ }
+ },
+ {
+ "id": "release500_0053",
+ "category": "question",
+ "text": "Have you ever just stopped and thought about how strange all of this is?",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0053.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HAVE YOU EVER JUST STOPPED AND THOUGHT ABOUT HOW STRANGE ALL OF THIS IS",
+ "raw": {
+ "normalized_reference": "have you ever just stopped and thought about how strange all of this is?",
+ "normalized_hypothesis": "have you ever just stopped and thought about how strange all of this is",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "have you ever just stopped and thought about how strange all of this is",
+ "normalized_hypothesis": "have you ever just stopped and thought about how strange all of this is",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0054",
+ "category": "rare_words",
+ "text": "Her idiosyncratic explanation sounded plausible but remained fundamentally ambiguous.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0054.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HER IDIOSYNCRATIC EXPLANATIONS SOUNDED PLAUSIBLE BUT REMAINED FUNDAMENTALLY AMBIGUOUS",
+ "raw": {
+ "normalized_reference": "her idiosyncratic explanation sounded plausible but remained fundamentally ambiguous.",
+ "normalized_hypothesis": "her idiosyncratic explanations sounded plausible but remained fundamentally ambiguous",
+ "errors": 1,
+ "words": 9,
+ "wer": 0.1111111111111111
+ },
+ "semantic": {
+ "normalized_reference": "her idiosyncratic explanation sounded plausible but remained fundamentally ambiguous",
+ "normalized_hypothesis": "her idiosyncratic explanations sounded plausible but remained fundamentally ambiguous",
+ "errors": 1,
+ "words": 9,
+ "wer": 0.1111111111111111
+ }
+ },
+ {
+ "id": "release500_0055",
+ "category": "short",
+ "text": "That tracks.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0055.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "BAD TRACKS",
+ "raw": {
+ "normalized_reference": "that tracks.",
+ "normalized_hypothesis": "bad tracks",
+ "errors": 1,
+ "words": 2,
+ "wer": 0.5
+ },
+ "semantic": {
+ "normalized_reference": "that tracks",
+ "normalized_hypothesis": "bad tracks",
+ "errors": 1,
+ "words": 2,
+ "wer": 0.5
+ }
+ },
+ {
+ "id": "release500_0056",
+ "category": "technical",
+ "text": "The key thing to remember is that order matters here more than you'd think.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0056.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE KEY THING TO REMEMBER IS THAT ORDE MATTERS HEAR MORE THAN YOU'D THINK",
+ "raw": {
+ "normalized_reference": "the key thing to remember is that order matters here more than you'd think.",
+ "normalized_hypothesis": "the key thing to remember is that orde matters hear more than you'd think",
+ "errors": 2,
+ "words": 14,
+ "wer": 0.14285714285714285
+ },
+ "semantic": {
+ "normalized_reference": "the key thing to remember is that order matters here more than you would think",
+ "normalized_hypothesis": "the key thing to remember is that orde matters hear more than you would think",
+ "errors": 2,
+ "words": 15,
+ "wer": 0.13333333333333333
+ }
+ },
+ {
+ "id": "release500_0057",
+ "category": "conversational",
+ "text": "I would not say that again in this room if I were you.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0057.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I WOULD NOT SAY THAT AGAIN IN THIS ROOM IF I WERE YOU",
+ "raw": {
+ "normalized_reference": "i would not say that again in this room if i were you.",
+ "normalized_hypothesis": "i would not say that again in this room if i were you",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i would not say that again in this room if i were you",
+ "normalized_hypothesis": "i would not say that again in this room if i were you",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0058",
+ "category": "descriptive",
+ "text": "The room was filled with the quiet that means something is about to change.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0058.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE ROOM WAS FILLED WITH THE QUIET THAT MEANS SOMETHING IS ABOUT TO CHANGE",
+ "raw": {
+ "normalized_reference": "the room was filled with the quiet that means something is about to change.",
+ "normalized_hypothesis": "the room was filled with the quiet that means something is about to change",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the room was filled with the quiet that means something is about to change",
+ "normalized_hypothesis": "the room was filled with the quiet that means something is about to change",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0059",
+ "category": "emotional",
+ "text": "I didn't expect to feel that way about it, but I did.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0059.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I DIDN'T EXPECT TO FEEL THAT WAY ABOUT IT BUT I DID",
+ "raw": {
+ "normalized_reference": "i didn't expect to feel that way about it, but i did.",
+ "normalized_hypothesis": "i didn't expect to feel that way about it but i did",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i did not expect to feel that way about it but i did",
+ "normalized_hypothesis": "i did not expect to feel that way about it but i did",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0060",
+ "category": "homographs",
+ "text": "I object to the strange object that was left near the entrance.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0060.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I OBJECT TO THE STRANGE OBJECT THAT WAS LEFT NEAR THE ENTRANCE",
+ "raw": {
+ "normalized_reference": "i object to the strange object that was left near the entrance.",
+ "normalized_hypothesis": "i object to the strange object that was left near the entrance",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i object to the strange object that was left near the entrance",
+ "normalized_hypothesis": "i object to the strange object that was left near the entrance",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0061",
+ "category": "long",
+ "text": "I've learned more from the things that went wrong than from the things that worked, which is a frustrating way to have to learn things but apparently the only way that stuck.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0061.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I'VE LEARNED MORE FROM THE THINGS THAT WENT WRONG THAN FROM THE THINGS THAT WORKED WHICH IS A FRUSTRATING WAY TO HAVE TO LEARN THINGS BUT APPARENTLY THE ONLY WAY THAT STUCK",
+ "raw": {
+ "normalized_reference": "i've learned more from the things that went wrong than from the things that worked, which is a frustrating way to have to learn things but apparently the only way that stuck.",
+ "normalized_hypothesis": "i've learned more from the things that went wrong than from the things that worked which is a frustrating way to have to learn things but apparently the only way that stuck",
+ "errors": 0,
+ "words": 32,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i have learned more from the things that went wrong than from the things that worked which is a frustrating way to have to learn things but apparently the only way that stuck",
+ "normalized_hypothesis": "i have learned more from the things that went wrong than from the things that worked which is a frustrating way to have to learn things but apparently the only way that stuck",
+ "errors": 0,
+ "words": 33,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0062",
+ "category": "long_range",
+ "text": "Because every nearby hotel was full, the exhausted hikers followed the narrow road until they found a farmhouse with one empty room.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0062.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "BECAUSE EVERY NEARBY HOTEL WAS FULL THE EXHAUSTED HYKERS FOLLOWED THE NARROW ROAD UNTIL THEY FOUND A FARMHOUSE WITH ONE EMPTY ROOM",
+ "raw": {
+ "normalized_reference": "because every nearby hotel was full, the exhausted hikers followed the narrow road until they found a farmhouse with one empty room.",
+ "normalized_hypothesis": "because every nearby hotel was full the exhausted hykers followed the narrow road until they found a farmhouse with one empty room",
+ "errors": 1,
+ "words": 22,
+ "wer": 0.045454545454545456
+ },
+ "semantic": {
+ "normalized_reference": "because every nearby hotel was full the exhausted hikers followed the narrow road until they found a farmhouse with one empty room",
+ "normalized_hypothesis": "because every nearby hotel was full the exhausted hykers followed the narrow road until they found a farmhouse with one empty room",
+ "errors": 1,
+ "words": 22,
+ "wer": 0.045454545454545456
+ }
+ },
+ {
+ "id": "release500_0063",
+ "category": "names_places",
+ "text": "The itinerary sends Beatrice from Albuquerque to Des Moines and then to Boise.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0063.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE ITINERARY SENDS BEATRICE FROM ALBERKIRKY TO DE MOINE AND THEN TO BOISY",
+ "raw": {
+ "normalized_reference": "the itinerary sends beatrice from albuquerque to des moines and then to boise.",
+ "normalized_hypothesis": "the itinerary sends beatrice from alberkirky to de moine and then to boisy",
+ "errors": 4,
+ "words": 13,
+ "wer": 0.3076923076923077
+ },
+ "semantic": {
+ "normalized_reference": "the itinerary sends beatrice from albuquerque to des moines and then to boise",
+ "normalized_hypothesis": "the itinerary sends beatrice from alberkirky to de moine and then to boisy",
+ "errors": 4,
+ "words": 13,
+ "wer": 0.3076923076923077
+ }
+ },
+ {
+ "id": "release500_0064",
+ "category": "narrative",
+ "text": "We had been driving for six hours when we finally saw the sign for the town.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0064.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WE HAD BEEN DRIVING FOR SIX HOURS WHEN WE FINALLY SAW THE SIGN FOR THE TOWN",
+ "raw": {
+ "normalized_reference": "we had been driving for six hours when we finally saw the sign for the town.",
+ "normalized_hypothesis": "we had been driving for six hours when we finally saw the sign for the town",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "we had been driving for 6 hours when we finally saw the sign for the town",
+ "normalized_hypothesis": "we had been driving for 6 hours when we finally saw the sign for the town",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0065",
+ "category": "numbers",
+ "text": "Call extension zero four nine after eight thirty in the morning.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0065.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HALL EXTENSION ZERO FOUR NINE AFTER EIGHT THIRTY IN THE MORNING",
+ "raw": {
+ "normalized_reference": "call extension zero four nine after eight thirty in the morning.",
+ "normalized_hypothesis": "hall extension zero four nine after eight thirty in the morning",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ },
+ "semantic": {
+ "normalized_reference": "call extension 049 after 830 in the morning",
+ "normalized_hypothesis": "hall extension 049 after 830 in the morning",
+ "errors": 1,
+ "words": 8,
+ "wer": 0.125
+ }
+ },
+ {
+ "id": "release500_0066",
+ "category": "punctuation",
+ "text": "Stop - listen carefully - then answer the question without guessing.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0066.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "STOP LISTEN CAREFULLY THEN ANSWER THE QUESTION WITHOUT GUESSING",
+ "raw": {
+ "normalized_reference": "stop - listen carefully - then answer the question without guessing.",
+ "normalized_hypothesis": "stop listen carefully then answer the question without guessing",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "stop listen carefully then answer the question without guessing",
+ "normalized_hypothesis": "stop listen carefully then answer the question without guessing",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0067",
+ "category": "question",
+ "text": "Does it matter at this point, honestly?",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0067.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "DOES IT MATTER AT THIS POINT HONESTLY",
+ "raw": {
+ "normalized_reference": "does it matter at this point, honestly?",
+ "normalized_hypothesis": "does it matter at this point honestly",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "does it matter at this point honestly",
+ "normalized_hypothesis": "does it matter at this point honestly",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0068",
+ "category": "rare_words",
+ "text": "An entrepreneurial meteorologist described the unprecedented atmospheric phenomenon.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0068.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "AN ONTREPRINERIAL METEORALOGIST DESCRIBED THE UNPRECEDENTED ATMOSPHERIC PHENOMENON",
+ "raw": {
+ "normalized_reference": "an entrepreneurial meteorologist described the unprecedented atmospheric phenomenon.",
+ "normalized_hypothesis": "an ontreprinerial meteoralogist described the unprecedented atmospheric phenomenon",
+ "errors": 2,
+ "words": 8,
+ "wer": 0.25
+ },
+ "semantic": {
+ "normalized_reference": "an entrepreneurial meteorologist described the unprecedented atmospheric phenomenon",
+ "normalized_hypothesis": "an ontreprinerial meteoralogist described the unprecedented atmospheric phenomenon",
+ "errors": 2,
+ "words": 8,
+ "wer": 0.25
+ }
+ },
+ {
+ "id": "release500_0069",
+ "category": "short",
+ "text": "That's not what I expected.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0069.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THAT'S NOT WHAT I EXPECTED",
+ "raw": {
+ "normalized_reference": "that's not what i expected.",
+ "normalized_hypothesis": "that's not what i expected",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "that is not what i expected",
+ "normalized_hypothesis": "that is not what i expected",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0070",
+ "category": "technical",
+ "text": "There's a known issue with that version that causes failures under load.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0070.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THERE'S A KNOWN ISSUE WITH THAT VERSION THAT CAUSES FAILURES UNDER LOAD",
+ "raw": {
+ "normalized_reference": "there's a known issue with that version that causes failures under load.",
+ "normalized_hypothesis": "there's a known issue with that version that causes failures under load",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "there is a known issue with that version that causes failures under load",
+ "normalized_hypothesis": "there is a known issue with that version that causes failures under load",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0071",
+ "category": "conversational",
+ "text": "Don't look now but I think that's him over there.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0071.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "DON'T LOOK NOW BUT I THINK THAT'S HIM OVER THERE",
+ "raw": {
+ "normalized_reference": "don't look now but i think that's him over there.",
+ "normalized_hypothesis": "don't look now but i think that's him over there",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "do not look now but i think that is him over there",
+ "normalized_hypothesis": "do not look now but i think that is him over there",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0072",
+ "category": "descriptive",
+ "text": "The city at night looks completely different from how it feels during the day.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0072.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE CITY AT NIGHT LOOKS COMPLETELY DIFFERENT FROM HOW IT FEELS DURING THE DAY",
+ "raw": {
+ "normalized_reference": "the city at night looks completely different from how it feels during the day.",
+ "normalized_hypothesis": "the city at night looks completely different from how it feels during the day",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the city at night looks completely different from how it feels during the day",
+ "normalized_hypothesis": "the city at night looks completely different from how it feels during the day",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0073",
+ "category": "emotional",
+ "text": "It's the kind of happy that makes you nervous because you know it can't last.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0073.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT'S THE KIND OF HAPPY THAT MAKES YOU NERVOUS BECAUSE YOU KNOW IT CAN'T LAST",
+ "raw": {
+ "normalized_reference": "it's the kind of happy that makes you nervous because you know it can't last.",
+ "normalized_hypothesis": "it's the kind of happy that makes you nervous because you know it can't last",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it is the kind of happy that makes you nervous because you know it can not last",
+ "normalized_hypothesis": "it is the kind of happy that makes you nervous because you know it can not last",
+ "errors": 0,
+ "words": 17,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0074",
+ "category": "homographs",
+ "text": "The dove dove beneath the branch while the wind began to wind the loose cable.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0074.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE DOVE DOVE BENEATH THE BRANCH WHILE THE WIND BEGAN TO WIND THE LOOSE CABLE",
+ "raw": {
+ "normalized_reference": "the dove dove beneath the branch while the wind began to wind the loose cable.",
+ "normalized_hypothesis": "the dove dove beneath the branch while the wind began to wind the loose cable",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the dove dove beneath the branch while the wind began to wind the loose cable",
+ "normalized_hypothesis": "the dove dove beneath the branch while the wind began to wind the loose cable",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0075",
+ "category": "long",
+ "text": "I've thought a lot about what I would do differently if I could go back, and the honest answer is probably not as much as I'd like to think.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0075.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I'VE THOUGHT A LOT ABOUT WHAT I WOULD DO DIFFERENTLY IF I COULD GO BACK AND THE HONEST ANSWER IS PROBABLY NOT AS MUCH AS I'D LIKE TO THINK",
+ "raw": {
+ "normalized_reference": "i've thought a lot about what i would do differently if i could go back, and the honest answer is probably not as much as i'd like to think.",
+ "normalized_hypothesis": "i've thought a lot about what i would do differently if i could go back and the honest answer is probably not as much as i'd like to think",
+ "errors": 0,
+ "words": 29,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i have thought a lot about what i would do differently if i could go back and the honest answer is probably not as much as i would like to think",
+ "normalized_hypothesis": "i have thought a lot about what i would do differently if i could go back and the honest answer is probably not as much as i would like to think",
+ "errors": 0,
+ "words": 31,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0076",
+ "category": "long_range",
+ "text": "Although the weather changed without warning, the musicians finished packing before anyone opened the enormous glass door.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0076.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "ALTHOUGH THE WEATHER CHANGED WITHOUT WARNING THE MUSICIANS FINISHED PACKING BEFORE ANY ONE OPENED THE ENORMOUS GLASS DOOR",
+ "raw": {
+ "normalized_reference": "although the weather changed without warning, the musicians finished packing before anyone opened the enormous glass door.",
+ "normalized_hypothesis": "although the weather changed without warning the musicians finished packing before any one opened the enormous glass door",
+ "errors": 2,
+ "words": 17,
+ "wer": 0.11764705882352941
+ },
+ "semantic": {
+ "normalized_reference": "although the weather changed without warning the musicians finished packing before anyone opened the enormous glass door",
+ "normalized_hypothesis": "although the weather changed without warning the musicians finished packing before any one opened the enormous glass door",
+ "errors": 2,
+ "words": 17,
+ "wer": 0.11764705882352941
+ }
+ },
+ {
+ "id": "release500_0077",
+ "category": "names_places",
+ "text": "Saskatchewan looked unusually bright beneath the pale February sunrise.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0077.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SAS FICHUIN LOOKED UNUSUALLY BRIGHT BENEATH THE PALE FEBRUARY SUNRISE",
+ "raw": {
+ "normalized_reference": "saskatchewan looked unusually bright beneath the pale february sunrise.",
+ "normalized_hypothesis": "sas fichuin looked unusually bright beneath the pale february sunrise",
+ "errors": 2,
+ "words": 9,
+ "wer": 0.2222222222222222
+ },
+ "semantic": {
+ "normalized_reference": "saskatchewan looked unusually bright beneath the pale february sunrise",
+ "normalized_hypothesis": "sas fichuin looked unusually bright beneath the pale february sunrise",
+ "errors": 2,
+ "words": 9,
+ "wer": 0.2222222222222222
+ }
+ },
+ {
+ "id": "release500_0078",
+ "category": "narrative",
+ "text": "She handled it with a grace I genuinely did not expect.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0078.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SHE HANDLED IT WITH A GRACE I GENUINELY DID NOT EXPECT",
+ "raw": {
+ "normalized_reference": "she handled it with a grace i genuinely did not expect.",
+ "normalized_hypothesis": "she handled it with a grace i genuinely did not expect",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "she handled it with a grace i genuinely did not expect",
+ "normalized_hypothesis": "she handled it with a grace i genuinely did not expect",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0079",
+ "category": "numbers",
+ "text": "Apartment four hundred seven is across from room twenty-nine on the third floor.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0079.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "APARTMENT FOUR HUNDRED SEVEN IS ACROSS FROM ROOM TWENTY NINE ON THE THIRD FLOOR",
+ "raw": {
+ "normalized_reference": "apartment four hundred seven is across from room twenty-nine on the third floor.",
+ "normalized_hypothesis": "apartment four hundred seven is across from room twenty nine on the third floor",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "apartment 407 is across from room 29 on the 3rd floor",
+ "normalized_hypothesis": "apartment 407 is across from room 29 on the 3rd floor",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0080",
+ "category": "punctuation",
+ "text": "\"Not yet,\" Rowan whispered. \"Give me another minute... please.\"",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0080.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "NOT YET ROWEN WHISPERED GIVE ME ANOTHER MINUTE PLEASE",
+ "raw": {
+ "normalized_reference": "\"not yet,\" rowan whispered. \"give me another minute... please.\"",
+ "normalized_hypothesis": "not yet rowen whispered give me another minute please",
+ "errors": 1,
+ "words": 9,
+ "wer": 0.1111111111111111
+ },
+ "semantic": {
+ "normalized_reference": "not yet rowan whispered give me another minute please",
+ "normalized_hypothesis": "not yet rowen whispered give me another minute please",
+ "errors": 1,
+ "words": 9,
+ "wer": 0.1111111111111111
+ }
+ },
+ {
+ "id": "release500_0081",
+ "category": "question",
+ "text": "What are we actually trying to accomplish here?",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0081.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WHAT ARE WE ACTUALLY TRYING TO ACCOMPLISH HERE",
+ "raw": {
+ "normalized_reference": "what are we actually trying to accomplish here?",
+ "normalized_hypothesis": "what are we actually trying to accomplish here",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "what are we actually trying to accomplish here",
+ "normalized_hypothesis": "what are we actually trying to accomplish here",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0082",
+ "category": "rare_words",
+ "text": "The physician recommended a noninvasive procedure for the recurring neurological symptoms.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0082.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE PHYSICIAN RECOMMENDED A NON EVASIVE PROCEDURE FOR THE RECURRING NEUROLOGICAL SYMPTOMS",
+ "raw": {
+ "normalized_reference": "the physician recommended a noninvasive procedure for the recurring neurological symptoms.",
+ "normalized_hypothesis": "the physician recommended a non evasive procedure for the recurring neurological symptoms",
+ "errors": 2,
+ "words": 11,
+ "wer": 0.18181818181818182
+ },
+ "semantic": {
+ "normalized_reference": "the physician recommended a noninvasive procedure for the recurring neurological symptoms",
+ "normalized_hypothesis": "the physician recommended a non evasive procedure for the recurring neurological symptoms",
+ "errors": 2,
+ "words": 11,
+ "wer": 0.18181818181818182
+ }
+ },
+ {
+ "id": "release500_0083",
+ "category": "short",
+ "text": "Are you serious right now?",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0083.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "ARE YOU SERIOUS RIGHT NOW",
+ "raw": {
+ "normalized_reference": "are you serious right now?",
+ "normalized_hypothesis": "are you serious right now",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "are you serious right now",
+ "normalized_hypothesis": "are you serious right now",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0084",
+ "category": "technical",
+ "text": "The performance improvement is real, but it comes with tradeoffs.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0084.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE PERFORMANCE IMPROVEMENT IS REAL BUT IT COMES WITH TRADOS",
+ "raw": {
+ "normalized_reference": "the performance improvement is real, but it comes with tradeoffs.",
+ "normalized_hypothesis": "the performance improvement is real but it comes with trados",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ },
+ "semantic": {
+ "normalized_reference": "the performance improvement is real but it comes with tradeoffs",
+ "normalized_hypothesis": "the performance improvement is real but it comes with trados",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ }
+ },
+ {
+ "id": "release500_0085",
+ "category": "conversational",
+ "text": "I don't know. It's just been one of those weeks, you know?",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0085.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I DON'T KNOW IT'S JUST BEEN ONE OF THOSE WEEKS YOU KNOW",
+ "raw": {
+ "normalized_reference": "i don't know. it's just been one of those weeks, you know?",
+ "normalized_hypothesis": "i don't know it's just been one of those weeks you know",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i do not know it is just been one of those weeks you know",
+ "normalized_hypothesis": "i do not know it is just been one of those weeks you know",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0086",
+ "category": "descriptive",
+ "text": "The moment you stop expecting perfection, things get a lot more interesting.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0086.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE MOMENT YOU STOP EXPECTING PERFECTION THINGS GET A LOT MORE INTERESTING",
+ "raw": {
+ "normalized_reference": "the moment you stop expecting perfection, things get a lot more interesting.",
+ "normalized_hypothesis": "the moment you stop expecting perfection things get a lot more interesting",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the moment you stop expecting perfection things get a lot more interesting",
+ "normalized_hypothesis": "the moment you stop expecting perfection things get a lot more interesting",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0087",
+ "category": "emotional",
+ "text": "It's the kind of thing that stays with you even years later.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0087.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT'S THE KIND OF THING THAT STAYS WITH YOU EVEN YEARS LATER",
+ "raw": {
+ "normalized_reference": "it's the kind of thing that stays with you even years later.",
+ "normalized_hypothesis": "it's the kind of thing that stays with you even years later",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it is the kind of thing that stays with you even years later",
+ "normalized_hypothesis": "it is the kind of thing that stays with you even years later",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0088",
+ "category": "long",
+ "text": "She told me once that the most important skill she'd developed was learning to disagree with someone without making them feel they were wrong.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0088.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SHE TOLD ME ONCE THAT THE MOST IMPORTANT SKILL SHE DEVELOPED WAS LEARNING TO DISAGREE WITH SOMEONE WITHOUT MAKING THEM FEEL THEY WERE WRONG",
+ "raw": {
+ "normalized_reference": "she told me once that the most important skill she'd developed was learning to disagree with someone without making them feel they were wrong.",
+ "normalized_hypothesis": "she told me once that the most important skill she developed was learning to disagree with someone without making them feel they were wrong",
+ "errors": 1,
+ "words": 24,
+ "wer": 0.041666666666666664
+ },
+ "semantic": {
+ "normalized_reference": "she told me once that the most important skill she would developed was learning to disagree with someone without making them feel they were wrong",
+ "normalized_hypothesis": "she told me once that the most important skill she developed was learning to disagree with someone without making them feel they were wrong",
+ "errors": 1,
+ "words": 25,
+ "wer": 0.04
+ }
+ },
+ {
+ "id": "release500_0089",
+ "category": "narrative",
+ "text": "He stood at the edge of the cliff and wondered, not for the first time, how he'd gotten here.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0089.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE STOOD AT THE EDGE OF THE CLIFF AND WONDERED NOT FOR THE FIRST TIME HOW HE'D GOTTEN HERE",
+ "raw": {
+ "normalized_reference": "he stood at the edge of the cliff and wondered, not for the first time, how he'd gotten here.",
+ "normalized_hypothesis": "he stood at the edge of the cliff and wondered not for the first time how he'd gotten here",
+ "errors": 0,
+ "words": 19,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he stood at the edge of the cliff and wondered not for the 1st time how he would gotten here",
+ "normalized_hypothesis": "he stood at the edge of the cliff and wondered not for the 1st time how he would gotten here",
+ "errors": 0,
+ "words": 20,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0090",
+ "category": "question",
+ "text": "Why does asking for help feel so hard when you actually need it?",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0090.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WHY DOES ASKING FOR HELP FEEL SO HARD WHEN YOU ACTUALLY NEED IT",
+ "raw": {
+ "normalized_reference": "why does asking for help feel so hard when you actually need it?",
+ "normalized_hypothesis": "why does asking for help feel so hard when you actually need it",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "why does asking for help feel so hard when you actually need it",
+ "normalized_hypothesis": "why does asking for help feel so hard when you actually need it",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0091",
+ "category": "short",
+ "text": "That's fair.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0091.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THAT'S FAIR",
+ "raw": {
+ "normalized_reference": "that's fair.",
+ "normalized_hypothesis": "that's fair",
+ "errors": 0,
+ "words": 2,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "that is fair",
+ "normalized_hypothesis": "that is fair",
+ "errors": 0,
+ "words": 3,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0092",
+ "category": "technical",
+ "text": "If you run into that error, the fix is usually just restarting the service.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0092.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IF YOU RUN INTO THAT ERROR THE FIX IS USUALLY JUST RESTARTING THE SERVICE",
+ "raw": {
+ "normalized_reference": "if you run into that error, the fix is usually just restarting the service.",
+ "normalized_hypothesis": "if you run into that error the fix is usually just restarting the service",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "if you run into that error the fix is usually just restarting the service",
+ "normalized_hypothesis": "if you run into that error the fix is usually just restarting the service",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0093",
+ "category": "conversational",
+ "text": "It was a disaster in the best possible way.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0093.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT WAS A DISASTER IN THE BEST POSSIBLE WAY",
+ "raw": {
+ "normalized_reference": "it was a disaster in the best possible way.",
+ "normalized_hypothesis": "it was a disaster in the best possible way",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it was a disaster in the best possible way",
+ "normalized_hypothesis": "it was a disaster in the best possible way",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0094",
+ "category": "descriptive",
+ "text": "It had the look of a place that used to mean something to someone.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0094.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT HAD THE LOOK OF A PLACE THAT USED TO MEAN SOMETHING TO SOME ONE",
+ "raw": {
+ "normalized_reference": "it had the look of a place that used to mean something to someone.",
+ "normalized_hypothesis": "it had the look of a place that used to mean something to some one",
+ "errors": 2,
+ "words": 14,
+ "wer": 0.14285714285714285
+ },
+ "semantic": {
+ "normalized_reference": "it had the look of a place that used to mean something to someone",
+ "normalized_hypothesis": "it had the look of a place that used to mean something to some one",
+ "errors": 2,
+ "words": 14,
+ "wer": 0.14285714285714285
+ }
+ },
+ {
+ "id": "release500_0095",
+ "category": "emotional",
+ "text": "There's this feeling you get when something finally works after weeks of failure.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0095.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THERE'S THIS FEELING YOU GET WHEN SOMETHING FINALLY WORKS AFTER WEEKS OF FAILURE",
+ "raw": {
+ "normalized_reference": "there's this feeling you get when something finally works after weeks of failure.",
+ "normalized_hypothesis": "there's this feeling you get when something finally works after weeks of failure",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "there is this feeling you get when something finally works after weeks of failure",
+ "normalized_hypothesis": "there is this feeling you get when something finally works after weeks of failure",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0096",
+ "category": "long",
+ "text": "He had a theory that most arguments aren't actually about what they seem to be about, and the longer I live the more I think he was onto something.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0096.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE HAD A THEORY THAT MOST ARGUMENTS AREN'T ACTUALLY ABOUT WHAT THEY SEEM TO BE ABOUT AND THE LONGER I LIVE THE MORE I THINK HE WAS ON TO SOMETHING",
+ "raw": {
+ "normalized_reference": "he had a theory that most arguments aren't actually about what they seem to be about, and the longer i live the more i think he was onto something.",
+ "normalized_hypothesis": "he had a theory that most arguments aren't actually about what they seem to be about and the longer i live the more i think he was on to something",
+ "errors": 2,
+ "words": 29,
+ "wer": 0.06896551724137931
+ },
+ "semantic": {
+ "normalized_reference": "he had a theory that most arguments are not actually about what they seem to be about and the longer i live the more i think he was onto something",
+ "normalized_hypothesis": "he had a theory that most arguments are not actually about what they seem to be about and the longer i live the more i think he was on to something",
+ "errors": 2,
+ "words": 30,
+ "wer": 0.06666666666666667
+ }
+ },
+ {
+ "id": "release500_0097",
+ "category": "narrative",
+ "text": "The thing nobody tells you about starting over is how ordinary most of it feels.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0097.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE THING NOBODY TELLS YOU ABOUT STARTING OVER IS HOW ORDINARY MOST OF IT FEELS",
+ "raw": {
+ "normalized_reference": "the thing nobody tells you about starting over is how ordinary most of it feels.",
+ "normalized_hypothesis": "the thing nobody tells you about starting over is how ordinary most of it feels",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the thing nobody tells you about starting over is how ordinary most of it feels",
+ "normalized_hypothesis": "the thing nobody tells you about starting over is how ordinary most of it feels",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0098",
+ "category": "question",
+ "text": "What exactly are we waiting for at this point?",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0098.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WHAT EXACTLY I'LL BE WAITING FOR AT THIS POINT",
+ "raw": {
+ "normalized_reference": "what exactly are we waiting for at this point?",
+ "normalized_hypothesis": "what exactly i'll be waiting for at this point",
+ "errors": 2,
+ "words": 9,
+ "wer": 0.2222222222222222
+ },
+ "semantic": {
+ "normalized_reference": "what exactly are we waiting for at this point",
+ "normalized_hypothesis": "what exactly i will be waiting for at this point",
+ "errors": 3,
+ "words": 9,
+ "wer": 0.3333333333333333
+ }
+ },
+ {
+ "id": "release500_0099",
+ "category": "short",
+ "text": "I stand by what I said.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0099.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I STAND BY WHAT I SAID",
+ "raw": {
+ "normalized_reference": "i stand by what i said.",
+ "normalized_hypothesis": "i stand by what i said",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i stand by what i said",
+ "normalized_hypothesis": "i stand by what i said",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0100",
+ "category": "technical",
+ "text": "Double-check that value before you push anything to production.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0100.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "DOUBLE CHECK THAT VALUE BEFORE YOU PUSH ANYTHING TO PRODUCTION",
+ "raw": {
+ "normalized_reference": "double-check that value before you push anything to production.",
+ "normalized_hypothesis": "double check that value before you push anything to production",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "double check that value before you push anything to production",
+ "normalized_hypothesis": "double check that value before you push anything to production",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0101",
+ "category": "conversational",
+ "text": "I understand your point; I just do not agree with the conclusion.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0101.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I UNDERSTAND YOUR POINT I JUST DO NOT AGREE WITH THE CONCLUSION",
+ "raw": {
+ "normalized_reference": "i understand your point; i just do not agree with the conclusion.",
+ "normalized_hypothesis": "i understand your point i just do not agree with the conclusion",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i understand your point i just do not agree with the conclusion",
+ "normalized_hypothesis": "i understand your point i just do not agree with the conclusion",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0102",
+ "category": "descriptive",
+ "text": "The sky that day was a strange shade of orange I'd never seen before.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0102.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SKY THAT DAY WAS A STRANGE SHADE OF ORANGE I'D NEVER SEEN BEFORE",
+ "raw": {
+ "normalized_reference": "the sky that day was a strange shade of orange i'd never seen before.",
+ "normalized_hypothesis": "sky that day was a strange shade of orange i'd never seen before",
+ "errors": 1,
+ "words": 14,
+ "wer": 0.07142857142857142
+ },
+ "semantic": {
+ "normalized_reference": "the sky that day was a strange shade of orange i would never seen before",
+ "normalized_hypothesis": "sky that day was a strange shade of orange i would never seen before",
+ "errors": 1,
+ "words": 15,
+ "wer": 0.06666666666666667
+ }
+ },
+ {
+ "id": "release500_0103",
+ "category": "emotional",
+ "text": "It hurt more than I expected, and I thought I was prepared.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0103.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT HURT MORE THAN I EXPECTED AND I THOUGHT I WAS PREPARED",
+ "raw": {
+ "normalized_reference": "it hurt more than i expected, and i thought i was prepared.",
+ "normalized_hypothesis": "it hurt more than i expected and i thought i was prepared",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it hurt more than i expected and i thought i was prepared",
+ "normalized_hypothesis": "it hurt more than i expected and i thought i was prepared",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0104",
+ "category": "long",
+ "text": "The hardest part wasn't making the decision, it was sitting with the uncertainty of not knowing whether it was the right one for months afterward.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0104.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE HARDEST PART WAS MAKING THE DECISION IT WAS SITTING WITH THE UNCERTAINTY OF NOT KNOWING WHETHER IT WAS THE RIGHT ONE FOR MONTHS AFTERWARD",
+ "raw": {
+ "normalized_reference": "the hardest part wasn't making the decision, it was sitting with the uncertainty of not knowing whether it was the right one for months afterward.",
+ "normalized_hypothesis": "the hardest part was making the decision it was sitting with the uncertainty of not knowing whether it was the right one for months afterward",
+ "errors": 1,
+ "words": 25,
+ "wer": 0.04
+ },
+ "semantic": {
+ "normalized_reference": "the hardest part was not making the decision it was sitting with the uncertainty of not knowing whether it was the right one for months afterward",
+ "normalized_hypothesis": "the hardest part was making the decision it was sitting with the uncertainty of not knowing whether it was the right one for months afterward",
+ "errors": 1,
+ "words": 26,
+ "wer": 0.038461538461538464
+ }
+ },
+ {
+ "id": "release500_0105",
+ "category": "narrative",
+ "text": "There was this one day last summer when everything just clicked into place.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0105.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THERE WAS THIS ONE DAY LAST SUMMER WHEN EVERYTHING JUST CLICKED INTO PLACE",
+ "raw": {
+ "normalized_reference": "there was this one day last summer when everything just clicked into place.",
+ "normalized_hypothesis": "there was this one day last summer when everything just clicked into place",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "there was this one day last summer when everything just clicked into place",
+ "normalized_hypothesis": "there was this one day last summer when everything just clicked into place",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0106",
+ "category": "question",
+ "text": "Have you ever thought about why we keep doing things the same way when they stop working?",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0106.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HAVE YOU EVER THOUGHT ABOUT WHY WE KEEP DOING THINGS THE SAME WAY WHEN THEY STOP WORKING",
+ "raw": {
+ "normalized_reference": "have you ever thought about why we keep doing things the same way when they stop working?",
+ "normalized_hypothesis": "have you ever thought about why we keep doing things the same way when they stop working",
+ "errors": 0,
+ "words": 17,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "have you ever thought about why we keep doing things the same way when they stop working",
+ "normalized_hypothesis": "have you ever thought about why we keep doing things the same way when they stop working",
+ "errors": 0,
+ "words": 17,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0107",
+ "category": "short",
+ "text": "I can't believe you just said that.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0107.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I CAN'T BELIEVE YOU JUST SAID THAT",
+ "raw": {
+ "normalized_reference": "i can't believe you just said that.",
+ "normalized_hypothesis": "i can't believe you just said that",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i can not believe you just said that",
+ "normalized_hypothesis": "i can not believe you just said that",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0108",
+ "category": "technical",
+ "text": "You'll want to isolate that variable before drawing any conclusions.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0108.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "YOU WILL WANT TO ISOLATE THAT VARIABLE BEFORE DRAWING ANY CONCLUSIONS",
+ "raw": {
+ "normalized_reference": "you'll want to isolate that variable before drawing any conclusions.",
+ "normalized_hypothesis": "you will want to isolate that variable before drawing any conclusions",
+ "errors": 2,
+ "words": 10,
+ "wer": 0.2
+ },
+ "semantic": {
+ "normalized_reference": "you will want to isolate that variable before drawing any conclusions",
+ "normalized_hypothesis": "you will want to isolate that variable before drawing any conclusions",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0109",
+ "category": "conversational",
+ "text": "The meeting that could have been an email lasted two hours.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0109.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE MEETING THAT COULD HAVE BEEN IN EMALE LASTED TWO HOURS",
+ "raw": {
+ "normalized_reference": "the meeting that could have been an email lasted two hours.",
+ "normalized_hypothesis": "the meeting that could have been in emale lasted two hours",
+ "errors": 2,
+ "words": 11,
+ "wer": 0.18181818181818182
+ },
+ "semantic": {
+ "normalized_reference": "the meeting that could have been an email lasted 2 hours",
+ "normalized_hypothesis": "the meeting that could have been in emale lasted 2 hours",
+ "errors": 2,
+ "words": 11,
+ "wer": 0.18181818181818182
+ }
+ },
+ {
+ "id": "release500_0110",
+ "category": "descriptive",
+ "text": "There's a specific kind of tired that sleep doesn't fix.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0110.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THERE'S A SPECIFIC KIND OF TYRE THAT SLEEP DOESN'T FIX",
+ "raw": {
+ "normalized_reference": "there's a specific kind of tired that sleep doesn't fix.",
+ "normalized_hypothesis": "there's a specific kind of tyre that sleep doesn't fix",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ },
+ "semantic": {
+ "normalized_reference": "there is a specific kind of tired that sleep does not fix",
+ "normalized_hypothesis": "there is a specific kind of tyre that sleep does not fix",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ }
+ },
+ {
+ "id": "release500_0111",
+ "category": "emotional",
+ "text": "Something about that moment broke something open in me that I hadn't realized was closed.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0111.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SOMETHING ABOUT THAT MOMENT BROKE SOMETHING OPEN IN ME THAT I HADN'T REALIZED WAS CLOSED",
+ "raw": {
+ "normalized_reference": "something about that moment broke something open in me that i hadn't realized was closed.",
+ "normalized_hypothesis": "something about that moment broke something open in me that i hadn't realized was closed",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "something about that moment broke something open in me that i had not realized was closed",
+ "normalized_hypothesis": "something about that moment broke something open in me that i had not realized was closed",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0112",
+ "category": "long",
+ "text": "The thing about making something good is that you don't usually know it's good while you're making it. You just know you're not done yet.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0112.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE THING ABOUT MAKING SOMETHING GOOD IS THAT YOU DON'T USUALLY KNOW IT'S GOOD WHILE YOU'RE MAKING IT YOU JUST KNOW YOU'RE NOT DONE YET",
+ "raw": {
+ "normalized_reference": "the thing about making something good is that you don't usually know it's good while you're making it. you just know you're not done yet.",
+ "normalized_hypothesis": "the thing about making something good is that you don't usually know it's good while you're making it you just know you're not done yet",
+ "errors": 0,
+ "words": 25,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the thing about making something good is that you do not usually know it is good while you are making it you just know you are not done yet",
+ "normalized_hypothesis": "the thing about making something good is that you do not usually know it is good while you are making it you just know you are not done yet",
+ "errors": 0,
+ "words": 29,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0113",
+ "category": "narrative",
+ "text": "She opened the door slowly, as if she expected to find something she wasn't ready for.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0113.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SHE OPENED THE DOOR SLOWLY AS IF SHE EXPECTED TO FIND SOMETHING SHE WASN'T READY FOR",
+ "raw": {
+ "normalized_reference": "she opened the door slowly, as if she expected to find something she wasn't ready for.",
+ "normalized_hypothesis": "she opened the door slowly as if she expected to find something she wasn't ready for",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "she opened the door slowly as if she expected to find something she was not ready for",
+ "normalized_hypothesis": "she opened the door slowly as if she expected to find something she was not ready for",
+ "errors": 0,
+ "words": 17,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0114",
+ "category": "question",
+ "text": "What would you do differently if you could go back?",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0114.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WHAT WOULD YOU DO DIFFERENTLY IF YOU COULD GO BACK",
+ "raw": {
+ "normalized_reference": "what would you do differently if you could go back?",
+ "normalized_hypothesis": "what would you do differently if you could go back",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "what would you do differently if you could go back",
+ "normalized_hypothesis": "what would you do differently if you could go back",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0115",
+ "category": "short",
+ "text": "Okay. Okay okay okay.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0115.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "O K O K O K O K",
+ "raw": {
+ "normalized_reference": "okay. okay okay okay.",
+ "normalized_hypothesis": "o k o k o k o k",
+ "errors": 8,
+ "words": 4,
+ "wer": 2.0
+ },
+ "semantic": {
+ "normalized_reference": "okay okay okay okay",
+ "normalized_hypothesis": "0 k 0 k 0 k 0 k",
+ "errors": 8,
+ "words": 4,
+ "wer": 2.0
+ }
+ },
+ {
+ "id": "release500_0116",
+ "category": "technical",
+ "text": "The latency goes up significantly at scale, which is expected.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0116.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE LATENCY GOES UP SIGNIFICANTLY AT SCALE WHICH IS EXPECTED",
+ "raw": {
+ "normalized_reference": "the latency goes up significantly at scale, which is expected.",
+ "normalized_hypothesis": "the latency goes up significantly at scale which is expected",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the latency goes up significantly at scale which is expected",
+ "normalized_hypothesis": "the latency goes up significantly at scale which is expected",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0117",
+ "category": "conversational",
+ "text": "We ended up staying until two in the morning talking about nothing.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0117.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WE ENDED UP STAYING UNTIL TWO IN THE MORNING TALKING ABOUT NOTHING",
+ "raw": {
+ "normalized_reference": "we ended up staying until two in the morning talking about nothing.",
+ "normalized_hypothesis": "we ended up staying until two in the morning talking about nothing",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "we ended up staying until 2 in the morning talking about nothing",
+ "normalized_hypothesis": "we ended up staying until 2 in the morning talking about nothing",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0118",
+ "category": "descriptive",
+ "text": "The rain started slowly, and then all at once, the way it usually does.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0118.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE RAIN STARTED SLOWLY AND THEN ALL AT ONCE THE WAY IT USUALLY DOES",
+ "raw": {
+ "normalized_reference": "the rain started slowly, and then all at once, the way it usually does.",
+ "normalized_hypothesis": "the rain started slowly and then all at once the way it usually does",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the rain started slowly and then all at once the way it usually does",
+ "normalized_hypothesis": "the rain started slowly and then all at once the way it usually does",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0119",
+ "category": "emotional",
+ "text": "I honestly did not think I was going to make it through that year.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0119.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I HONESTLY DID NOT THINK I WAS GOING TO MAKE IT THROUGH THAT YEAR",
+ "raw": {
+ "normalized_reference": "i honestly did not think i was going to make it through that year.",
+ "normalized_hypothesis": "i honestly did not think i was going to make it through that year",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i honestly did not think i was going to make it through that year",
+ "normalized_hypothesis": "i honestly did not think i was going to make it through that year",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0120",
+ "category": "long",
+ "text": "There's an assumption that people who seem confident all the time actually feel confident, and in my experience that's almost never true.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0120.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THERE'S AN ASSUMPTION THAT PEOPLE WHO SEEM CONFIDENT ALL THE TIME ACTUALLY FEEL CONFIDENT AND IN MY EXPERIENCE THAT'S ALMOST NEVER TRUE",
+ "raw": {
+ "normalized_reference": "there's an assumption that people who seem confident all the time actually feel confident, and in my experience that's almost never true.",
+ "normalized_hypothesis": "there's an assumption that people who seem confident all the time actually feel confident and in my experience that's almost never true",
+ "errors": 0,
+ "words": 22,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "there is an assumption that people who seem confident all the time actually feel confident and in my experience that is almost never true",
+ "normalized_hypothesis": "there is an assumption that people who seem confident all the time actually feel confident and in my experience that is almost never true",
+ "errors": 0,
+ "words": 24,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0121",
+ "category": "narrative",
+ "text": "The moment I opened the door I knew something was different, even if I couldn't say what.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0121.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE MOMENT I OPENED THE DOOR I KNEW SOMETHING WAS DIFFERENT EVEN IF I COULDN'T SAY WHAT",
+ "raw": {
+ "normalized_reference": "the moment i opened the door i knew something was different, even if i couldn't say what.",
+ "normalized_hypothesis": "the moment i opened the door i knew something was different even if i couldn't say what",
+ "errors": 0,
+ "words": 17,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the moment i opened the door i knew something was different even if i could not say what",
+ "normalized_hypothesis": "the moment i opened the door i knew something was different even if i could not say what",
+ "errors": 0,
+ "words": 18,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0122",
+ "category": "question",
+ "text": "How do you explain something that doesn't make sense even to you?",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0122.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "DO YOU EXPLAIN SOMETHING THAT DOESN'T MAKE SENSE EVEN TO YOU",
+ "raw": {
+ "normalized_reference": "how do you explain something that doesn't make sense even to you?",
+ "normalized_hypothesis": "do you explain something that doesn't make sense even to you",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "how do you explain something that does not make sense even to you",
+ "normalized_hypothesis": "do you explain something that does not make sense even to you",
+ "errors": 1,
+ "words": 13,
+ "wer": 0.07692307692307693
+ }
+ },
+ {
+ "id": "release500_0123",
+ "category": "short",
+ "text": "Hold on a second.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0123.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HOLD ON A SECOND",
+ "raw": {
+ "normalized_reference": "hold on a second.",
+ "normalized_hypothesis": "hold on a second",
+ "errors": 0,
+ "words": 4,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "hold on a 2nd",
+ "normalized_hypothesis": "hold on a 2nd",
+ "errors": 0,
+ "words": 4,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0124",
+ "category": "technical",
+ "text": "A deterministic split prevents duplicate normalized text from leaking into evaluation.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0124.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "A DETERMINISTIC SPLIT PREVENTS DUPLICATE NORMALIZED TEXT FROM LEAKING INTO EVALUATION",
+ "raw": {
+ "normalized_reference": "a deterministic split prevents duplicate normalized text from leaking into evaluation.",
+ "normalized_hypothesis": "a deterministic split prevents duplicate normalized text from leaking into evaluation",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "a deterministic split prevents duplicate normalized text from leaking into evaluation",
+ "normalized_hypothesis": "a deterministic split prevents duplicate normalized text from leaking into evaluation",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0125",
+ "category": "conversational",
+ "text": "Honestly, that was much better than I expected, but the ending still felt rushed.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0125.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HONESTLY THAT WAS MUCH BETTER THAN I EXPECTED BUT THE ENDING STILL FELT RUSHED",
+ "raw": {
+ "normalized_reference": "honestly, that was much better than i expected, but the ending still felt rushed.",
+ "normalized_hypothesis": "honestly that was much better than i expected but the ending still felt rushed",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "honestly that was much better than i expected but the ending still felt rushed",
+ "normalized_hypothesis": "honestly that was much better than i expected but the ending still felt rushed",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0126",
+ "category": "descriptive",
+ "text": "She had a habit of tilting her head slightly when thinking hard about something.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0126.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SHE HAD A HABIT OF TILTING HER HEAD SLIGHTLY WHEN THINKING HARD ABOUT SOMETHING",
+ "raw": {
+ "normalized_reference": "she had a habit of tilting her head slightly when thinking hard about something.",
+ "normalized_hypothesis": "she had a habit of tilting her head slightly when thinking hard about something",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "she had a habit of tilting her head slightly when thinking hard about something",
+ "normalized_hypothesis": "she had a habit of tilting her head slightly when thinking hard about something",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0127",
+ "category": "emotional",
+ "text": "I cannot stress enough how much I did not want to be right about this.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0127.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I CANNOT STRESS ENOUGH HOW MUCH I DID NOT WANT TO BE RIGHT ABOUT THIS",
+ "raw": {
+ "normalized_reference": "i cannot stress enough how much i did not want to be right about this.",
+ "normalized_hypothesis": "i cannot stress enough how much i did not want to be right about this",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i cannot stress enough how much i did not want to be right about this",
+ "normalized_hypothesis": "i cannot stress enough how much i did not want to be right about this",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0128",
+ "category": "long",
+ "text": "I used to think the goal was to get to a point where nothing was hard anymore, and now I think that point doesn't exist and I'm not sure I'd want it to.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0128.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I USED TO THINK THE GOAL WAS TO GET TO A POINT WHERE NOTHING WAS HARD ANY MORE AND NOW I THINK THAT POINT DOESN'T EXIST AND I'M NOT SURE I WANTED TO",
+ "raw": {
+ "normalized_reference": "i used to think the goal was to get to a point where nothing was hard anymore, and now i think that point doesn't exist and i'm not sure i'd want it to.",
+ "normalized_hypothesis": "i used to think the goal was to get to a point where nothing was hard any more and now i think that point doesn't exist and i'm not sure i wanted to",
+ "errors": 5,
+ "words": 33,
+ "wer": 0.15151515151515152
+ },
+ "semantic": {
+ "normalized_reference": "i used to think the goal was to get to a point where nothing was hard anymore and now i think that point does not exist and i am not sure i would want it to",
+ "normalized_hypothesis": "i used to think the goal was to get to a point where nothing was hard any more and now i think that point does not exist and i am not sure i wanted to",
+ "errors": 5,
+ "words": 36,
+ "wer": 0.1388888888888889
+ }
+ },
+ {
+ "id": "release500_0129",
+ "category": "narrative",
+ "text": "So this is the part of the story where everything goes completely sideways.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0129.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SO THIS IS THE PART OF THE STORY WHERE EVERYTHING GOES COMPLETELY SIDEWAYS",
+ "raw": {
+ "normalized_reference": "so this is the part of the story where everything goes completely sideways.",
+ "normalized_hypothesis": "so this is the part of the story where everything goes completely sideways",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "so this is the part of the story where everything goes completely sideways",
+ "normalized_hypothesis": "so this is the part of the story where everything goes completely sideways",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0130",
+ "category": "question",
+ "text": "Is there a version of this where it actually works out okay?",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0130.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IS THERE A VERSION OF THIS WHERE IT ACTUALLY WORKS OUT OQ",
+ "raw": {
+ "normalized_reference": "is there a version of this where it actually works out okay?",
+ "normalized_hypothesis": "is there a version of this where it actually works out oq",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "is there a version of this where it actually works out okay",
+ "normalized_hypothesis": "is there a version of this where it actually works out oq",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ }
+ },
+ {
+ "id": "release500_0131",
+ "category": "short",
+ "text": "Oh no. Oh no no no.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0131.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "OH NO OH NO NO NO",
+ "raw": {
+ "normalized_reference": "oh no. oh no no no.",
+ "normalized_hypothesis": "oh no oh no no no",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "0 no 0 no no no",
+ "normalized_hypothesis": "0 no 0 no no no",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0132",
+ "category": "technical",
+ "text": "Always test on a copy before modifying the original.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0132.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "ALWAYS TEST ON A COPY BEFORE MODIFYING THE ORIGINAL",
+ "raw": {
+ "normalized_reference": "always test on a copy before modifying the original.",
+ "normalized_hypothesis": "always test on a copy before modifying the original",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "always test on a copy before modifying the original",
+ "normalized_hypothesis": "always test on a copy before modifying the original",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0133",
+ "category": "conversational",
+ "text": "I wasn't even supposed to be there. That's the wild part.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0133.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I WASN'T EVEN SUPPOSED TO BE THERE THAT'S THE WILD PART",
+ "raw": {
+ "normalized_reference": "i wasn't even supposed to be there. that's the wild part.",
+ "normalized_hypothesis": "i wasn't even supposed to be there that's the wild part",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i was not even supposed to be there that is the wild part",
+ "normalized_hypothesis": "i was not even supposed to be there that is the wild part",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0134",
+ "category": "descriptive",
+ "text": "It's a small town. Not the kind you leave. The kind you leave and come back to.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0134.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT'S A SMALL TOWN NOT THE KIND YOU LEAVE THE KIND YOU LEAVE AND COME BACK TO",
+ "raw": {
+ "normalized_reference": "it's a small town. not the kind you leave. the kind you leave and come back to.",
+ "normalized_hypothesis": "it's a small town not the kind you leave the kind you leave and come back to",
+ "errors": 0,
+ "words": 17,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it is a small town not the kind you leave the kind you leave and come back to",
+ "normalized_hypothesis": "it is a small town not the kind you leave the kind you leave and come back to",
+ "errors": 0,
+ "words": 18,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0135",
+ "category": "emotional",
+ "text": "She cried. I cried. We all just sort of stood there crying.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0135.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SHE CRIED I CRIED WE ALL JUST SORT OF STOOD THE CRYING",
+ "raw": {
+ "normalized_reference": "she cried. i cried. we all just sort of stood there crying.",
+ "normalized_hypothesis": "she cried i cried we all just sort of stood the crying",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "she cried i cried we all just sort of stood there crying",
+ "normalized_hypothesis": "she cried i cried we all just sort of stood the crying",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ }
+ },
+ {
+ "id": "release500_0136",
+ "category": "long",
+ "text": "There's a version of being busy that's actually just a very elaborate way of avoiding the one thing you know you should be doing.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0136.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THERE'S A VERSION OF BEING BUSY THAT'S ACTUALLY JUST A VERY ELABORATE WAY OF AVOIDING THE ONE THING YOU KNOW YOU SHOULD BE DOING",
+ "raw": {
+ "normalized_reference": "there's a version of being busy that's actually just a very elaborate way of avoiding the one thing you know you should be doing.",
+ "normalized_hypothesis": "there's a version of being busy that's actually just a very elaborate way of avoiding the one thing you know you should be doing",
+ "errors": 0,
+ "words": 24,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "there is a version of being busy that is actually just a very elaborate way of avoiding the one thing you know you should be doing",
+ "normalized_hypothesis": "there is a version of being busy that is actually just a very elaborate way of avoiding the one thing you know you should be doing",
+ "errors": 0,
+ "words": 26,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0137",
+ "category": "narrative",
+ "text": "The letter arrived on a Tuesday, which felt appropriate somehow.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0137.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE LETTER ARRIVED ON A TUESDAY WHICH FELT APPROPRIATE SOMEHOW",
+ "raw": {
+ "normalized_reference": "the letter arrived on a tuesday, which felt appropriate somehow.",
+ "normalized_hypothesis": "the letter arrived on a tuesday which felt appropriate somehow",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the letter arrived on a tuesday which felt appropriate somehow",
+ "normalized_hypothesis": "the letter arrived on a tuesday which felt appropriate somehow",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0138",
+ "category": "question",
+ "text": "What exactly did she say when you told her?",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0138.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WHAT EXACTLY DID SHE SAY WHEN YOU TOLD HER",
+ "raw": {
+ "normalized_reference": "what exactly did she say when you told her?",
+ "normalized_hypothesis": "what exactly did she say when you told her",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "what exactly did she say when you told her",
+ "normalized_hypothesis": "what exactly did she say when you told her",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0139",
+ "category": "short",
+ "text": "No way. Absolutely not.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0139.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "NO WAY ABSOLUTELY NOT",
+ "raw": {
+ "normalized_reference": "no way. absolutely not.",
+ "normalized_hypothesis": "no way absolutely not",
+ "errors": 0,
+ "words": 4,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "no way absolutely not",
+ "normalized_hypothesis": "no way absolutely not",
+ "errors": 0,
+ "words": 4,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0140",
+ "category": "technical",
+ "text": "The first thing you want to do is make sure everything is backed up.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0140.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE FIRST THING YOU WANT TO DO IS MAKE SURE EVERYTHING IS BACKED UP",
+ "raw": {
+ "normalized_reference": "the first thing you want to do is make sure everything is backed up.",
+ "normalized_hypothesis": "the first thing you want to do is make sure everything is backed up",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the 1st thing you want to do is make sure everything is backed up",
+ "normalized_hypothesis": "the 1st thing you want to do is make sure everything is backed up",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0141",
+ "category": "conversational",
+ "text": "I have the most insane idea and you need to just hear me out.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0141.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I HAVE THE MOST INSANE IDEA AND YOU NEED T JUST HEAR ME OUT",
+ "raw": {
+ "normalized_reference": "i have the most insane idea and you need to just hear me out.",
+ "normalized_hypothesis": "i have the most insane idea and you need t just hear me out",
+ "errors": 1,
+ "words": 14,
+ "wer": 0.07142857142857142
+ },
+ "semantic": {
+ "normalized_reference": "i have the most insane idea and you need to just hear me out",
+ "normalized_hypothesis": "i have the most insane idea and you need t just hear me out",
+ "errors": 1,
+ "words": 14,
+ "wer": 0.07142857142857142
+ }
+ },
+ {
+ "id": "release500_0142",
+ "category": "descriptive",
+ "text": "There's a coffee shop on the corner that's been there longer than anyone can remember.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0142.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THERE'S A COFFEE SHOP ON THE CORNER OF TIT'S BEEN THERE LONGER THAN ANY ONE CAN REMEMBER",
+ "raw": {
+ "normalized_reference": "there's a coffee shop on the corner that's been there longer than anyone can remember.",
+ "normalized_hypothesis": "there's a coffee shop on the corner of tit's been there longer than any one can remember",
+ "errors": 4,
+ "words": 15,
+ "wer": 0.26666666666666666
+ },
+ "semantic": {
+ "normalized_reference": "there is a coffee shop on the corner that has been there longer than anyone can remember",
+ "normalized_hypothesis": "there is a coffee shop on the corner of tit has been there longer than any one can remember",
+ "errors": 4,
+ "words": 17,
+ "wer": 0.23529411764705882
+ }
+ },
+ {
+ "id": "release500_0143",
+ "category": "emotional",
+ "text": "I kept it together until I got to the car, and then I completely fell apart.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0143.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I KEPT IT TOGETHER UNTIL I GOT TO THE CAR AND THEN I COMPLETELY FELL APART",
+ "raw": {
+ "normalized_reference": "i kept it together until i got to the car, and then i completely fell apart.",
+ "normalized_hypothesis": "i kept it together until i got to the car and then i completely fell apart",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i kept it together until i got to the car and then i completely fell apart",
+ "normalized_hypothesis": "i kept it together until i got to the car and then i completely fell apart",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0144",
+ "category": "long",
+ "text": "The older I get, the more I think that most of what we call wisdom is just pattern recognition built from having been wrong enough times to start noticing.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0144.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE OLDER I GET THE MORE I THINK THAT MOST OF WHAT WE CALL WISDOM IS JUST PATTERN RECOGNITION BUILT FROM HAVING BEEN WRONG ENOUGH TIMES TO START NOTICING",
+ "raw": {
+ "normalized_reference": "the older i get, the more i think that most of what we call wisdom is just pattern recognition built from having been wrong enough times to start noticing.",
+ "normalized_hypothesis": "the older i get the more i think that most of what we call wisdom is just pattern recognition built from having been wrong enough times to start noticing",
+ "errors": 0,
+ "words": 29,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the older i get the more i think that most of what we call wisdom is just pattern recognition built from having been wrong enough times to start noticing",
+ "normalized_hypothesis": "the older i get the more i think that most of what we call wisdom is just pattern recognition built from having been wrong enough times to start noticing",
+ "errors": 0,
+ "words": 29,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0145",
+ "category": "narrative",
+ "text": "I remember the exact moment I realized I had made a huge mistake.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0145.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I REMEMBER THE EXACT MOMENT I REALIZED I HAD MADE A HUGE MISTAKE",
+ "raw": {
+ "normalized_reference": "i remember the exact moment i realized i had made a huge mistake.",
+ "normalized_hypothesis": "i remember the exact moment i realized i had made a huge mistake",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i remember the exact moment i realized i had made a huge mistake",
+ "normalized_hypothesis": "i remember the exact moment i realized i had made a huge mistake",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0146",
+ "category": "question",
+ "text": "Do you think it would have gone differently if we'd left earlier?",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0146.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "DO YOU THINK IT WOULD HAVE GONE DIFFERENTLY IF WE'D LEFT EARLIER",
+ "raw": {
+ "normalized_reference": "do you think it would have gone differently if we'd left earlier?",
+ "normalized_hypothesis": "do you think it would have gone differently if we'd left earlier",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "do you think it would have gone differently if we would left earlier",
+ "normalized_hypothesis": "do you think it would have gone differently if we would left earlier",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0147",
+ "category": "short",
+ "text": "You have to be kidding me.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0147.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "YOU HAVE TO BE KIDDING ME",
+ "raw": {
+ "normalized_reference": "you have to be kidding me.",
+ "normalized_hypothesis": "you have to be kidding me",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "you have to be kidding me",
+ "normalized_hypothesis": "you have to be kidding me",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0148",
+ "category": "technical",
+ "text": "If the checksum fails, restart the worker process without deleting the cached tensors.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0148.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IF THE CHEXAM FAILS RESTART THE WORKER PROCESS WITHOUT DELEATING THE CASH TENSERS",
+ "raw": {
+ "normalized_reference": "if the checksum fails, restart the worker process without deleting the cached tensors.",
+ "normalized_hypothesis": "if the chexam fails restart the worker process without deleating the cash tensers",
+ "errors": 4,
+ "words": 13,
+ "wer": 0.3076923076923077
+ },
+ "semantic": {
+ "normalized_reference": "if the checksum fails restart the worker process without deleting the cached tensors",
+ "normalized_hypothesis": "if the chexam fails restart the worker process without deleating the cash tensers",
+ "errors": 4,
+ "words": 13,
+ "wer": 0.3076923076923077
+ }
+ },
+ {
+ "id": "release500_0149",
+ "category": "conversational",
+ "text": "I need you to understand how close that was to not working.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0149.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I NEED YOU TO UNDERSTAND HOW CLOSE THAT WAS TO NOT WORKING",
+ "raw": {
+ "normalized_reference": "i need you to understand how close that was to not working.",
+ "normalized_hypothesis": "i need you to understand how close that was to not working",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i need you to understand how close that was to not working",
+ "normalized_hypothesis": "i need you to understand how close that was to not working",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0150",
+ "category": "descriptive",
+ "text": "It was the sort of morning where the whole world looks rinsed clean.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0150.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT WAS THE SORT OF MORNING WHERE THE WHOLE WORLD LOOKS RINCED CLEAN",
+ "raw": {
+ "normalized_reference": "it was the sort of morning where the whole world looks rinsed clean.",
+ "normalized_hypothesis": "it was the sort of morning where the whole world looks rinced clean",
+ "errors": 1,
+ "words": 13,
+ "wer": 0.07692307692307693
+ },
+ "semantic": {
+ "normalized_reference": "it was the sort of morning where the whole world looks rinsed clean",
+ "normalized_hypothesis": "it was the sort of morning where the whole world looks rinced clean",
+ "errors": 1,
+ "words": 13,
+ "wer": 0.07692307692307693
+ }
+ },
+ {
+ "id": "release500_0151",
+ "category": "emotional",
+ "text": "She started laughing and I started laughing and we couldn't stop for a full minute.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0151.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SHE STARTED LAUGHING AND I STARTED LAUGHING AND WE COULDN'T STOP FOR A FULL MINUTE",
+ "raw": {
+ "normalized_reference": "she started laughing and i started laughing and we couldn't stop for a full minute.",
+ "normalized_hypothesis": "she started laughing and i started laughing and we couldn't stop for a full minute",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "she started laughing and i started laughing and we could not stop for a full minute",
+ "normalized_hypothesis": "she started laughing and i started laughing and we could not stop for a full minute",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0152",
+ "category": "long",
+ "text": "We spent three years building something we were genuinely proud of, and then the market shifted and none of it mattered the way we thought it would.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0152.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WE SPENT THREE YEARS BUILDING SOMETHING WE WERE GENUINELY PROUD OF AND THEN THE MARKET SHIFTED AND NONE OF IT MATTERED THE WAY WE THOUGHT IT WOULD",
+ "raw": {
+ "normalized_reference": "we spent three years building something we were genuinely proud of, and then the market shifted and none of it mattered the way we thought it would.",
+ "normalized_hypothesis": "we spent three years building something we were genuinely proud of and then the market shifted and none of it mattered the way we thought it would",
+ "errors": 0,
+ "words": 27,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "we spent 3 years building something we were genuinely proud of and then the market shifted and none of it mattered the way we thought it would",
+ "normalized_hypothesis": "we spent 3 years building something we were genuinely proud of and then the market shifted and none of it mattered the way we thought it would",
+ "errors": 0,
+ "words": 27,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0153",
+ "category": "narrative",
+ "text": "And somehow, despite all of that, it actually turned out okay.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0153.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "AND SOMEHOW DESPITE ALL OF THAT IT ACTUALLY TURNED OUT O K",
+ "raw": {
+ "normalized_reference": "and somehow, despite all of that, it actually turned out okay.",
+ "normalized_hypothesis": "and somehow despite all of that it actually turned out o k",
+ "errors": 2,
+ "words": 11,
+ "wer": 0.18181818181818182
+ },
+ "semantic": {
+ "normalized_reference": "and somehow despite all of that it actually turned out okay",
+ "normalized_hypothesis": "and somehow despite all of that it actually turned out 0 k",
+ "errors": 2,
+ "words": 11,
+ "wer": 0.18181818181818182
+ }
+ },
+ {
+ "id": "release500_0154",
+ "category": "question",
+ "text": "Is that something you actually want, or something you think you should want?",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0154.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IS THAT SOMETHING YOU ACTUALLY WANT OR SOMETHING YOU THINK YOU SHOULD WANT",
+ "raw": {
+ "normalized_reference": "is that something you actually want, or something you think you should want?",
+ "normalized_hypothesis": "is that something you actually want or something you think you should want",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "is that something you actually want or something you think you should want",
+ "normalized_hypothesis": "is that something you actually want or something you think you should want",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0155",
+ "category": "short",
+ "text": "This is fine.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0155.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THIS IS FINE",
+ "raw": {
+ "normalized_reference": "this is fine.",
+ "normalized_hypothesis": "this is fine",
+ "errors": 0,
+ "words": 3,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "this is fine",
+ "normalized_hypothesis": "this is fine",
+ "errors": 0,
+ "words": 3,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0156",
+ "category": "technical",
+ "text": "The tricky part isn't the setup, it's making sure all dependencies line up.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0156.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE TRICKY PART ISN'T THE SET UP IT'S MAKING SURE ALL DEPENDENCIES LINE UP",
+ "raw": {
+ "normalized_reference": "the tricky part isn't the setup, it's making sure all dependencies line up.",
+ "normalized_hypothesis": "the tricky part isn't the set up it's making sure all dependencies line up",
+ "errors": 2,
+ "words": 13,
+ "wer": 0.15384615384615385
+ },
+ "semantic": {
+ "normalized_reference": "the tricky part is not the setup it is making sure all dependencies line up",
+ "normalized_hypothesis": "the tricky part is not the set up it is making sure all dependencies line up",
+ "errors": 2,
+ "words": 15,
+ "wer": 0.13333333333333333
+ }
+ },
+ {
+ "id": "release500_0157",
+ "category": "conversational",
+ "text": "He called it a minor setback. I called it a disaster.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0157.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE CALLED IT A MINOR SETBACK I CALLED IT A DISASTER",
+ "raw": {
+ "normalized_reference": "he called it a minor setback. i called it a disaster.",
+ "normalized_hypothesis": "he called it a minor setback i called it a disaster",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he called it a minor setback i called it a disaster",
+ "normalized_hypothesis": "he called it a minor setback i called it a disaster",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0158",
+ "category": "descriptive",
+ "text": "The whole place smelled like pine and old wood and cold air.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0158.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE WHOLE PLACE SMELLED LIKE PINE AND OLD WOOD AND COLD AIR",
+ "raw": {
+ "normalized_reference": "the whole place smelled like pine and old wood and cold air.",
+ "normalized_hypothesis": "the whole place smelled like pine and old wood and cold air",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the whole place smelled like pine and old wood and cold air",
+ "normalized_hypothesis": "the whole place smelled like pine and old wood and cold air",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0159",
+ "category": "emotional",
+ "text": "I've never been so relieved in my entire life.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0159.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I'VE NEVER BEEN SO RELIEVED IN MY ENTIRE LIFE",
+ "raw": {
+ "normalized_reference": "i've never been so relieved in my entire life.",
+ "normalized_hypothesis": "i've never been so relieved in my entire life",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i have never been so relieved in my entire life",
+ "normalized_hypothesis": "i have never been so relieved in my entire life",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0160",
+ "category": "narrative",
+ "text": "I wrote it all down that night so I wouldn't forget, and I still haven't looked at it.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0160.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I WROTE IT ALL DOWN THAT NIGHT SO I WOULDN'T FORGET AND I STILL HAVEN'T LOOKED AT IT",
+ "raw": {
+ "normalized_reference": "i wrote it all down that night so i wouldn't forget, and i still haven't looked at it.",
+ "normalized_hypothesis": "i wrote it all down that night so i wouldn't forget and i still haven't looked at it",
+ "errors": 0,
+ "words": 18,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i wrote it all down that night so i would not forget and i still have not looked at it",
+ "normalized_hypothesis": "i wrote it all down that night so i would not forget and i still have not looked at it",
+ "errors": 0,
+ "words": 20,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0161",
+ "category": "short",
+ "text": "That's kind of wild actually.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0161.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THAT'S KIND OF WILD ACTUALLY",
+ "raw": {
+ "normalized_reference": "that's kind of wild actually.",
+ "normalized_hypothesis": "that's kind of wild actually",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "that is kind of wild actually",
+ "normalized_hypothesis": "that is kind of wild actually",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0162",
+ "category": "technical",
+ "text": "Version pinning will save you a lot of pain later down the line.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0162.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "VIRSION PINNING WILL SAVE YOU A LOT OF PAIN LATER DOWN THE LINE",
+ "raw": {
+ "normalized_reference": "version pinning will save you a lot of pain later down the line.",
+ "normalized_hypothesis": "virsion pinning will save you a lot of pain later down the line",
+ "errors": 1,
+ "words": 13,
+ "wer": 0.07692307692307693
+ },
+ "semantic": {
+ "normalized_reference": "version pinning will save you a lot of pain later down the line",
+ "normalized_hypothesis": "virsion pinning will save you a lot of pain later down the line",
+ "errors": 1,
+ "words": 13,
+ "wer": 0.07692307692307693
+ }
+ },
+ {
+ "id": "release500_0163",
+ "category": "conversational",
+ "text": "At some point you just have to commit and see what happens.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0163.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "AT SOME POINT YOU JUST HAVE TO COMMIT AND SEE WHAT HAPPENS",
+ "raw": {
+ "normalized_reference": "at some point you just have to commit and see what happens.",
+ "normalized_hypothesis": "at some point you just have to commit and see what happens",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "at some point you just have to commit and see what happens",
+ "normalized_hypothesis": "at some point you just have to commit and see what happens",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0164",
+ "category": "emotional",
+ "text": "Nobody told me it was going to be this hard, and honestly, I'm grateful.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0164.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "NOBODY TOLD ME IT WAS GOING TO BE AS HARD AND HONESTLY I'M GRATEFUL",
+ "raw": {
+ "normalized_reference": "nobody told me it was going to be this hard, and honestly, i'm grateful.",
+ "normalized_hypothesis": "nobody told me it was going to be as hard and honestly i'm grateful",
+ "errors": 1,
+ "words": 14,
+ "wer": 0.07142857142857142
+ },
+ "semantic": {
+ "normalized_reference": "nobody told me it was going to be this hard and honestly i am grateful",
+ "normalized_hypothesis": "nobody told me it was going to be as hard and honestly i am grateful",
+ "errors": 1,
+ "words": 15,
+ "wer": 0.06666666666666667
+ }
+ },
+ {
+ "id": "release500_0165",
+ "category": "narrative",
+ "text": "Chapter seven. In which everything that could go wrong did, and a few things that couldn't.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0165.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "CHAPTER SEVEN IN WHICH EVERYTHING THAT COULD GO WRONG DID AND A FEW THINGS THAT COULDN'T",
+ "raw": {
+ "normalized_reference": "chapter seven. in which everything that could go wrong did, and a few things that couldn't.",
+ "normalized_hypothesis": "chapter seven in which everything that could go wrong did and a few things that couldn't",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "chapter 7 in which everything that could go wrong did and a few things that could not",
+ "normalized_hypothesis": "chapter 7 in which everything that could go wrong did and a few things that could not",
+ "errors": 0,
+ "words": 17,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0166",
+ "category": "short",
+ "text": "Wait, what?",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0166.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WAIT WHAT",
+ "raw": {
+ "normalized_reference": "wait, what?",
+ "normalized_hypothesis": "wait what",
+ "errors": 0,
+ "words": 2,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "wait what",
+ "normalized_hypothesis": "wait what",
+ "errors": 0,
+ "words": 2,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0167",
+ "category": "technical",
+ "text": "The graphics processor remained idle because the data loader exhausted every central processing core.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0167.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE GRAPHIC'S PROCESSOR REMAINED IDLE BECAUSE THE DATA LODER EXHAUSTED EVERY CENTRAL PROCESSING CORE",
+ "raw": {
+ "normalized_reference": "the graphics processor remained idle because the data loader exhausted every central processing core.",
+ "normalized_hypothesis": "the graphic's processor remained idle because the data loder exhausted every central processing core",
+ "errors": 2,
+ "words": 14,
+ "wer": 0.14285714285714285
+ },
+ "semantic": {
+ "normalized_reference": "the graphics processor remained idle because the data loader exhausted every central processing core",
+ "normalized_hypothesis": "the graphic is processor remained idle because the data loder exhausted every central processing core",
+ "errors": 3,
+ "words": 14,
+ "wer": 0.21428571428571427
+ }
+ },
+ {
+ "id": "release500_0168",
+ "category": "conversational",
+ "text": "Could you speak a little more slowly? I am trying to write every detail down.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0168.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "COULD YOU SPEAK A LITTLE MORE SLOWLY I AM TRYING TO WRITE EVERY DETAIL DOWN",
+ "raw": {
+ "normalized_reference": "could you speak a little more slowly? i am trying to write every detail down.",
+ "normalized_hypothesis": "could you speak a little more slowly i am trying to write every detail down",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "could you speak a little more slowly i am trying to write every detail down",
+ "normalized_hypothesis": "could you speak a little more slowly i am trying to write every detail down",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0169",
+ "category": "emotional",
+ "text": "I was so excited I could barely form a complete sentence.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0169.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I WAS SO EXCITED I COULD BARELY FORM A COMPLETE SENTENCE",
+ "raw": {
+ "normalized_reference": "i was so excited i could barely form a complete sentence.",
+ "normalized_hypothesis": "i was so excited i could barely form a complete sentence",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i was so excited i could barely form a complete sentence",
+ "normalized_hypothesis": "i was so excited i could barely form a complete sentence",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0170",
+ "category": "narrative",
+ "text": "He looked at me like I had said something in a language he didn't speak.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0170.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE LOOKED AT ME LIKE I HAD SAID SOMETHING IN A LANGUAGE HE DIDN'T SPEAK",
+ "raw": {
+ "normalized_reference": "he looked at me like i had said something in a language he didn't speak.",
+ "normalized_hypothesis": "he looked at me like i had said something in a language he didn't speak",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he looked at me like i had said something in a language he did not speak",
+ "normalized_hypothesis": "he looked at me like i had said something in a language he did not speak",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0171",
+ "category": "short",
+ "text": "That's actually hilarious.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0171.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THAT'S ACTUALLY HILARIOUS",
+ "raw": {
+ "normalized_reference": "that's actually hilarious.",
+ "normalized_hypothesis": "that's actually hilarious",
+ "errors": 0,
+ "words": 3,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "that is actually hilarious",
+ "normalized_hypothesis": "that is actually hilarious",
+ "errors": 0,
+ "words": 4,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0172",
+ "category": "technical",
+ "text": "The encoder maps phonemes, stress, and phrase boundaries into a shared hidden representation.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0172.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE INCODO MAPS PHONIMS STRESS AND PHRASE BOUNDARIES INTO A SHARED HIDDEN REPRESENTATION",
+ "raw": {
+ "normalized_reference": "the encoder maps phonemes, stress, and phrase boundaries into a shared hidden representation.",
+ "normalized_hypothesis": "the incodo maps phonims stress and phrase boundaries into a shared hidden representation",
+ "errors": 2,
+ "words": 13,
+ "wer": 0.15384615384615385
+ },
+ "semantic": {
+ "normalized_reference": "the encoder maps phonemes stress and phrase boundaries into a shared hidden representation",
+ "normalized_hypothesis": "the incodo maps phonims stress and phrase boundaries into a shared hidden representation",
+ "errors": 2,
+ "words": 13,
+ "wer": 0.15384615384615385
+ }
+ },
+ {
+ "id": "release500_0173",
+ "category": "conversational",
+ "text": "She showed up three hours late and acted like nothing had happened.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0173.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SHE SHOWED UP THREE HOURS LATE AND ACTED LIKE NOTHING HAD HAPPENED",
+ "raw": {
+ "normalized_reference": "she showed up three hours late and acted like nothing had happened.",
+ "normalized_hypothesis": "she showed up three hours late and acted like nothing had happened",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "she showed up 3 hours late and acted like nothing had happened",
+ "normalized_hypothesis": "she showed up 3 hours late and acted like nothing had happened",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0174",
+ "category": "emotional",
+ "text": "Oh my god. This is the best thing I've ever tasted.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0174.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "O MY GOD THIS IS THE BEST THING I'VE EVER TASTED",
+ "raw": {
+ "normalized_reference": "oh my god. this is the best thing i've ever tasted.",
+ "normalized_hypothesis": "o my god this is the best thing i've ever tasted",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ },
+ "semantic": {
+ "normalized_reference": "0 my god this is the best thing i have ever tasted",
+ "normalized_hypothesis": "0 my god this is the best thing i have ever tasted",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0175",
+ "category": "narrative",
+ "text": "The note was short, which was somehow more frightening than a long one would have been.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0175.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE NOTE WAS SHORT WHICH WAS SOMEHOW MORE FRIGHTENING THAN A LONG ONE WOULD HAVE BEEN",
+ "raw": {
+ "normalized_reference": "the note was short, which was somehow more frightening than a long one would have been.",
+ "normalized_hypothesis": "the note was short which was somehow more frightening than a long one would have been",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the note was short which was somehow more frightening than a long one would have been",
+ "normalized_hypothesis": "the note was short which was somehow more frightening than a long one would have been",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0176",
+ "category": "short",
+ "text": "Noted.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0176.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "NOTED",
+ "raw": {
+ "normalized_reference": "noted.",
+ "normalized_hypothesis": "noted",
+ "errors": 0,
+ "words": 1,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "noted",
+ "normalized_hypothesis": "noted",
+ "errors": 0,
+ "words": 1,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0177",
+ "category": "technical",
+ "text": "It's a simple process once you understand the underlying structure.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0177.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT'S A SIMPLE PROCESS ONCE YOU UNDERSTAND THE UNDERLYING STRUCTURE",
+ "raw": {
+ "normalized_reference": "it's a simple process once you understand the underlying structure.",
+ "normalized_hypothesis": "it's a simple process once you understand the underlying structure",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it is a simple process once you understand the underlying structure",
+ "normalized_hypothesis": "it is a simple process once you understand the underlying structure",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0178",
+ "category": "conversational",
+ "text": "You have to hear this story because it's almost too good to be true.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0178.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "YOU HAVE TO HEAR THIS STORY BECAUSE IT'S ALMOST TOO GOOD TO BE TRUE",
+ "raw": {
+ "normalized_reference": "you have to hear this story because it's almost too good to be true.",
+ "normalized_hypothesis": "you have to hear this story because it's almost too good to be true",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "you have to hear this story because it is almost too good to be true",
+ "normalized_hypothesis": "you have to hear this story because it is almost too good to be true",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0179",
+ "category": "emotional",
+ "text": "I have never felt more seen in my entire life than in that moment.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0179.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I HAVE NEVER FELT MORE SEEN IN MY ENTIRE LIFE THAN IN THAT MOMENT",
+ "raw": {
+ "normalized_reference": "i have never felt more seen in my entire life than in that moment.",
+ "normalized_hypothesis": "i have never felt more seen in my entire life than in that moment",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i have never felt more seen in my entire life than in that moment",
+ "normalized_hypothesis": "i have never felt more seen in my entire life than in that moment",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0180",
+ "category": "narrative",
+ "text": "She knew. She had always known. She just waited for the right moment.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0180.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SHE KNEW SHE HAD ALWAYS KNOWN SHE JUST WAITED FOR THE RIGHT MOMENT",
+ "raw": {
+ "normalized_reference": "she knew. she had always known. she just waited for the right moment.",
+ "normalized_hypothesis": "she knew she had always known she just waited for the right moment",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "she knew she had always known she just waited for the right moment",
+ "normalized_hypothesis": "she knew she had always known she just waited for the right moment",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0181",
+ "category": "short",
+ "text": "Yeah, no. Hard pass.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0181.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "YEA NO HARD BAS",
+ "raw": {
+ "normalized_reference": "yeah, no. hard pass.",
+ "normalized_hypothesis": "yea no hard bas",
+ "errors": 2,
+ "words": 4,
+ "wer": 0.5
+ },
+ "semantic": {
+ "normalized_reference": "yeah no hard pass",
+ "normalized_hypothesis": "yea no hard bas",
+ "errors": 2,
+ "words": 4,
+ "wer": 0.5
+ }
+ },
+ {
+ "id": "release500_0182",
+ "category": "technical",
+ "text": "Please verify the sample rate, hop length, and mel filter configuration before inference.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0182.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "PLEASE VERIFY THE SAMPLE RATE HOP LENGTH AND MELL FILTER CONFIGURATION BEFORE INFERENCE",
+ "raw": {
+ "normalized_reference": "please verify the sample rate, hop length, and mel filter configuration before inference.",
+ "normalized_hypothesis": "please verify the sample rate hop length and mell filter configuration before inference",
+ "errors": 1,
+ "words": 13,
+ "wer": 0.07692307692307693
+ },
+ "semantic": {
+ "normalized_reference": "please verify the sample rate hop length and mel filter configuration before inference",
+ "normalized_hypothesis": "please verify the sample rate hop length and mell filter configuration before inference",
+ "errors": 1,
+ "words": 13,
+ "wer": 0.07692307692307693
+ }
+ },
+ {
+ "id": "release500_0183",
+ "category": "conversational",
+ "text": "That's the worst idea I've ever heard and I'm completely in.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0183.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THAT'S THE WORST IDEA I'VE EVER HEARD AND I'M COMPLETELY IN",
+ "raw": {
+ "normalized_reference": "that's the worst idea i've ever heard and i'm completely in.",
+ "normalized_hypothesis": "that's the worst idea i've ever heard and i'm completely in",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "that is the worst idea i have ever heard and i am completely in",
+ "normalized_hypothesis": "that is the worst idea i have ever heard and i am completely in",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0184",
+ "category": "emotional",
+ "text": "I remember thinking: this is it. This is the thing I've been waiting for.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0184.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I REMEMBER THINKING THIS IS IT THIS IS THE THING I'VE BEEN WAITING FOR",
+ "raw": {
+ "normalized_reference": "i remember thinking: this is it. this is the thing i've been waiting for.",
+ "normalized_hypothesis": "i remember thinking this is it this is the thing i've been waiting for",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i remember thinking this is it this is the thing i have been waiting for",
+ "normalized_hypothesis": "i remember thinking this is it this is the thing i have been waiting for",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0185",
+ "category": "narrative",
+ "text": "We never talked about it again, which I think was the right call.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0185.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WE NEVER TALKED ABOUT IT AGAIN WHICH I THINK WAS THE RIGHT CAR",
+ "raw": {
+ "normalized_reference": "we never talked about it again, which i think was the right call.",
+ "normalized_hypothesis": "we never talked about it again which i think was the right car",
+ "errors": 1,
+ "words": 13,
+ "wer": 0.07692307692307693
+ },
+ "semantic": {
+ "normalized_reference": "we never talked about it again which i think was the right call",
+ "normalized_hypothesis": "we never talked about it again which i think was the right car",
+ "errors": 1,
+ "words": 13,
+ "wer": 0.07692307692307693
+ }
+ },
+ {
+ "id": "release500_0186",
+ "category": "short",
+ "text": "In hindsight, obvious.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0186.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IN HINDSIGHT OBVIOUS",
+ "raw": {
+ "normalized_reference": "in hindsight, obvious.",
+ "normalized_hypothesis": "in hindsight obvious",
+ "errors": 0,
+ "words": 3,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "in hindsight obvious",
+ "normalized_hypothesis": "in hindsight obvious",
+ "errors": 0,
+ "words": 3,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0187",
+ "category": "technical",
+ "text": "Once it's configured correctly, it basically runs itself.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0187.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "ONCE ITS CONFIGURED CORRECTLY IT BASICALLY RUNS ITSELF",
+ "raw": {
+ "normalized_reference": "once it's configured correctly, it basically runs itself.",
+ "normalized_hypothesis": "once its configured correctly it basically runs itself",
+ "errors": 1,
+ "words": 8,
+ "wer": 0.125
+ },
+ "semantic": {
+ "normalized_reference": "once it is configured correctly it basically runs itself",
+ "normalized_hypothesis": "once its configured correctly it basically runs itself",
+ "errors": 2,
+ "words": 9,
+ "wer": 0.2222222222222222
+ }
+ },
+ {
+ "id": "release500_0188",
+ "category": "conversational",
+ "text": "You got engaged? When? How? Tell me everything right now.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0188.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "YOU GOT ENGAGED WHENG O TELL ME EVERYTHING RIGHT NOW",
+ "raw": {
+ "normalized_reference": "you got engaged? when? how? tell me everything right now.",
+ "normalized_hypothesis": "you got engaged wheng o tell me everything right now",
+ "errors": 2,
+ "words": 10,
+ "wer": 0.2
+ },
+ "semantic": {
+ "normalized_reference": "you got engaged when how tell me everything right now",
+ "normalized_hypothesis": "you got engaged wheng 0 tell me everything right now",
+ "errors": 2,
+ "words": 10,
+ "wer": 0.2
+ }
+ },
+ {
+ "id": "release500_0189",
+ "category": "emotional",
+ "text": "It's hard to explain how much that moment meant to me.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0189.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT'S HARD TO EXPLAIN HOW MUCH THAT MOMENT MEANT TO ME",
+ "raw": {
+ "normalized_reference": "it's hard to explain how much that moment meant to me.",
+ "normalized_hypothesis": "it's hard to explain how much that moment meant to me",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it is hard to explain how much that moment meant to me",
+ "normalized_hypothesis": "it is hard to explain how much that moment meant to me",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0190",
+ "category": "narrative",
+ "text": "He told the story three different times that night and somehow it got better each time.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0190.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE TOLD THE STORY THREE DIFFERENT TIMES THAT NIGHT AND SOMEHOW IT GOT BETTER EACH TIME",
+ "raw": {
+ "normalized_reference": "he told the story three different times that night and somehow it got better each time.",
+ "normalized_hypothesis": "he told the story three different times that night and somehow it got better each time",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he told the story 3 different times that night and somehow it got better each time",
+ "normalized_hypothesis": "he told the story 3 different times that night and somehow it got better each time",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0191",
+ "category": "short",
+ "text": "Right. Moving on.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0191.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "BIGHT MOVING ON",
+ "raw": {
+ "normalized_reference": "right. moving on.",
+ "normalized_hypothesis": "bight moving on",
+ "errors": 1,
+ "words": 3,
+ "wer": 0.3333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "right moving on",
+ "normalized_hypothesis": "bight moving on",
+ "errors": 1,
+ "words": 3,
+ "wer": 0.3333333333333333
+ }
+ },
+ {
+ "id": "release500_0192",
+ "category": "conversational",
+ "text": "No, I did not move your keys; check beside the small ceramic bowl.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0192.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "NO I DID NOT MOVE YOUR KEYS CHECK BESIDE THE SMALL SERAMIC BOWL",
+ "raw": {
+ "normalized_reference": "no, i did not move your keys; check beside the small ceramic bowl.",
+ "normalized_hypothesis": "no i did not move your keys check beside the small seramic bowl",
+ "errors": 1,
+ "words": 13,
+ "wer": 0.07692307692307693
+ },
+ "semantic": {
+ "normalized_reference": "no i did not move your keys check beside the small ceramic bowl",
+ "normalized_hypothesis": "no i did not move your keys check beside the small seramic bowl",
+ "errors": 1,
+ "words": 13,
+ "wer": 0.07692307692307693
+ }
+ },
+ {
+ "id": "release500_0193",
+ "category": "emotional",
+ "text": "I was devastated. I'm not going to pretend otherwise.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0193.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I WAS DEVASTATED I'M NOT GOING TO PRETEND OTHERWISE",
+ "raw": {
+ "normalized_reference": "i was devastated. i'm not going to pretend otherwise.",
+ "normalized_hypothesis": "i was devastated i'm not going to pretend otherwise",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i was devastated i am not going to pretend otherwise",
+ "normalized_hypothesis": "i was devastated i am not going to pretend otherwise",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0194",
+ "category": "narrative",
+ "text": "That was the last summer before everything changed, though we didn't know it yet.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0194.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THAT WAS LAST SUMMER BEFORE EVERYTHING CHANGED THOUGH WE DIDN'T KNOW IT YET",
+ "raw": {
+ "normalized_reference": "that was the last summer before everything changed, though we didn't know it yet.",
+ "normalized_hypothesis": "that was last summer before everything changed though we didn't know it yet",
+ "errors": 1,
+ "words": 14,
+ "wer": 0.07142857142857142
+ },
+ "semantic": {
+ "normalized_reference": "that was the last summer before everything changed though we did not know it yet",
+ "normalized_hypothesis": "that was last summer before everything changed though we did not know it yet",
+ "errors": 1,
+ "words": 15,
+ "wer": 0.06666666666666667
+ }
+ },
+ {
+ "id": "release500_0195",
+ "category": "short",
+ "text": "Bold of you to assume.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0195.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "BOLD OF YOU TO ASSUME",
+ "raw": {
+ "normalized_reference": "bold of you to assume.",
+ "normalized_hypothesis": "bold of you to assume",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "bold of you to assume",
+ "normalized_hypothesis": "bold of you to assume",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0196",
+ "category": "conversational",
+ "text": "It's one of those decisions that makes total sense in the moment.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0196.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT'S ONE OF THOSE DECISIONS THAT MAKES TOTAL SENSE IN THE MOMENT",
+ "raw": {
+ "normalized_reference": "it's one of those decisions that makes total sense in the moment.",
+ "normalized_hypothesis": "it's one of those decisions that makes total sense in the moment",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it is one of those decisions that makes total sense in the moment",
+ "normalized_hypothesis": "it is one of those decisions that makes total sense in the moment",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0197",
+ "category": "emotional",
+ "text": "I was furious. Not the loud kind. The quiet kind that lasts for days.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0197.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I WAS FURIOUS NOT THE LOUD KIND THE QUIET KIND THAT LASTS FOR DAYS",
+ "raw": {
+ "normalized_reference": "i was furious. not the loud kind. the quiet kind that lasts for days.",
+ "normalized_hypothesis": "i was furious not the loud kind the quiet kind that lasts for days",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i was furious not the loud kind the quiet kind that lasts for days",
+ "normalized_hypothesis": "i was furious not the loud kind the quiet kind that lasts for days",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0198",
+ "category": "narrative",
+ "text": "That's the part of the story I usually leave out.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0198.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THAT'S THE PART OF THE STORY I USUALLY LEAVE OUT",
+ "raw": {
+ "normalized_reference": "that's the part of the story i usually leave out.",
+ "normalized_hypothesis": "that's the part of the story i usually leave out",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "that is the part of the story i usually leave out",
+ "normalized_hypothesis": "that is the part of the story i usually leave out",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0199",
+ "category": "short",
+ "text": "I've said too much.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0199.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I'VE SAID TOO MUCH",
+ "raw": {
+ "normalized_reference": "i've said too much.",
+ "normalized_hypothesis": "i've said too much",
+ "errors": 0,
+ "words": 4,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i have said too much",
+ "normalized_hypothesis": "i have said too much",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0200",
+ "category": "conversational",
+ "text": "I've been faking that I understood this the entire time.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0200.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I'VE BEEN THAKING THAT I UNDERSTOOD THIS THE ENTIRE TIME",
+ "raw": {
+ "normalized_reference": "i've been faking that i understood this the entire time.",
+ "normalized_hypothesis": "i've been thaking that i understood this the entire time",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ },
+ "semantic": {
+ "normalized_reference": "i have been faking that i understood this the entire time",
+ "normalized_hypothesis": "i have been thaking that i understood this the entire time",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ }
+ },
+ {
+ "id": "release500_0201",
+ "category": "phonetic_balance",
+ "text": "Peace be unto you and grace before the Lord.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0201.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "PEACE BE UNTO YOU AND GRACE BEFORE THE LORD",
+ "raw": {
+ "normalized_reference": "peace be unto you and grace before the lord.",
+ "normalized_hypothesis": "peace be unto you and grace before the lord",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "peace be unto you and grace before the lord",
+ "normalized_hypothesis": "peace be unto you and grace before the lord",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0202",
+ "category": "phonetic_balance",
+ "text": "From now on we're pals.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0202.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "FROM NOW ON WERE PADDLS",
+ "raw": {
+ "normalized_reference": "from now on we're pals.",
+ "normalized_hypothesis": "from now on were paddls",
+ "errors": 2,
+ "words": 5,
+ "wer": 0.4
+ },
+ "semantic": {
+ "normalized_reference": "from now on we are pals",
+ "normalized_hypothesis": "from now on were paddls",
+ "errors": 3,
+ "words": 6,
+ "wer": 0.5
+ }
+ },
+ {
+ "id": "release500_0203",
+ "category": "phonetic_balance",
+ "text": "Here we allow our solicitors to look after our legal work.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0203.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HERE WE ALLOW OUR SOLICITORS TO LOOK AFTER OUR LEGAL WORK",
+ "raw": {
+ "normalized_reference": "here we allow our solicitors to look after our legal work.",
+ "normalized_hypothesis": "here we allow our solicitors to look after our legal work",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "here we allow our solicitors to look after our legal work",
+ "normalized_hypothesis": "here we allow our solicitors to look after our legal work",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0204",
+ "category": "phonetic_balance",
+ "text": "Her words sent a strange chill through Philip.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0204.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HER WORDS SENT A STRANGE CHILL THROUGH PHILIP",
+ "raw": {
+ "normalized_reference": "her words sent a strange chill through philip.",
+ "normalized_hypothesis": "her words sent a strange chill through philip",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "her words sent a strange chill through philip",
+ "normalized_hypothesis": "her words sent a strange chill through philip",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0205",
+ "category": "phonetic_balance",
+ "text": "It was introduced by Representative Dick of Ohio.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0205.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT WAS INTRODUCED BY REPRESENTATIVE DICK OF OHIO",
+ "raw": {
+ "normalized_reference": "it was introduced by representative dick of ohio.",
+ "normalized_hypothesis": "it was introduced by representative dick of ohio",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it was introduced by representative dick of ohio",
+ "normalized_hypothesis": "it was introduced by representative dick of ohio",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0206",
+ "category": "phonetic_balance",
+ "text": "It's the aurora borealis.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0206.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT'S THE ARA BORIALIS",
+ "raw": {
+ "normalized_reference": "it's the aurora borealis.",
+ "normalized_hypothesis": "it's the ara borialis",
+ "errors": 2,
+ "words": 4,
+ "wer": 0.5
+ },
+ "semantic": {
+ "normalized_reference": "it is the aurora borealis",
+ "normalized_hypothesis": "it is the ara borialis",
+ "errors": 2,
+ "words": 5,
+ "wer": 0.4
+ }
+ },
+ {
+ "id": "release500_0207",
+ "category": "phonetic_balance",
+ "text": "The boy grew and prospered.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0207.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE BOY GREW AND PROSPERED",
+ "raw": {
+ "normalized_reference": "the boy grew and prospered.",
+ "normalized_hypothesis": "the boy grew and prospered",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the boy grew and prospered",
+ "normalized_hypothesis": "the boy grew and prospered",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0208",
+ "category": "phonetic_balance",
+ "text": "His voice was passionately rebellious.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0208.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HIS VOICE WAS PASSIONATELY REBELLIOUS",
+ "raw": {
+ "normalized_reference": "his voice was passionately rebellious.",
+ "normalized_hypothesis": "his voice was passionately rebellious",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "his voice was passionately rebellious",
+ "normalized_hypothesis": "his voice was passionately rebellious",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0209",
+ "category": "phonetic_balance",
+ "text": "That's where they cut off the Scottish Chiefs and killed all hands.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0209.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THAT'S WHERE THEY CUT OFF THE SCOTTISH CHIEFS AND KILLED ALL HANDS",
+ "raw": {
+ "normalized_reference": "that's where they cut off the scottish chiefs and killed all hands.",
+ "normalized_hypothesis": "that's where they cut off the scottish chiefs and killed all hands",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "that is where they cut off the scottish chiefs and killed all hands",
+ "normalized_hypothesis": "that is where they cut off the scottish chiefs and killed all hands",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0210",
+ "category": "phonetic_balance",
+ "text": "And Tom King patiently endured.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0210.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "AND TOM KING PATIENTLY ENDURED",
+ "raw": {
+ "normalized_reference": "and tom king patiently endured.",
+ "normalized_hypothesis": "and tom king patiently endured",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "and tom king patiently endured",
+ "normalized_hypothesis": "and tom king patiently endured",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0211",
+ "category": "phonetic_balance",
+ "text": "From that moment his friendship for Belize turns to hatred and jealousy.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0211.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "FROM THAT MOMENT HIS FRIENDSHIP FOR BELIS TURNS TO HATRED AND JEALOUSY",
+ "raw": {
+ "normalized_reference": "from that moment his friendship for belize turns to hatred and jealousy.",
+ "normalized_hypothesis": "from that moment his friendship for belis turns to hatred and jealousy",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "from that moment his friendship for belize turns to hatred and jealousy",
+ "normalized_hypothesis": "from that moment his friendship for belis turns to hatred and jealousy",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ }
+ },
+ {
+ "id": "release500_0212",
+ "category": "phonetic_balance",
+ "text": "Then you don't believe in altruism.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0212.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THEN YOU DON'T BELIEVE IN ALTRUISM",
+ "raw": {
+ "normalized_reference": "then you don't believe in altruism.",
+ "normalized_hypothesis": "then you don't believe in altruism",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "then you do not believe in altruism",
+ "normalized_hypothesis": "then you do not believe in altruism",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0213",
+ "category": "phonetic_balance",
+ "text": "What Pascal glimpsed with the vision of a seer, I have lived.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0213.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WHAT PASSBLE GLIMPSED WITH THE VISION OF A SIER I HAVE LIVED",
+ "raw": {
+ "normalized_reference": "what pascal glimpsed with the vision of a seer, i have lived.",
+ "normalized_hypothesis": "what passble glimpsed with the vision of a sier i have lived",
+ "errors": 2,
+ "words": 12,
+ "wer": 0.16666666666666666
+ },
+ "semantic": {
+ "normalized_reference": "what pascal glimpsed with the vision of a seer i have lived",
+ "normalized_hypothesis": "what passble glimpsed with the vision of a sier i have lived",
+ "errors": 2,
+ "words": 12,
+ "wer": 0.16666666666666666
+ }
+ },
+ {
+ "id": "release500_0214",
+ "category": "phonetic_balance",
+ "text": "You were making them talk shop, Ruth charged him.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0214.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "YOU WERE MAKING THEM TALK SHARP RUF CHARGED HIM",
+ "raw": {
+ "normalized_reference": "you were making them talk shop, ruth charged him.",
+ "normalized_hypothesis": "you were making them talk sharp ruf charged him",
+ "errors": 2,
+ "words": 9,
+ "wer": 0.2222222222222222
+ },
+ "semantic": {
+ "normalized_reference": "you were making them talk shop ruth charged him",
+ "normalized_hypothesis": "you were making them talk sharp ruf charged him",
+ "errors": 2,
+ "words": 9,
+ "wer": 0.2222222222222222
+ }
+ },
+ {
+ "id": "release500_0215",
+ "category": "phonetic_balance",
+ "text": "Philip snatched at the letter which Gregson held out to him.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0215.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "PHILIP SNATCHED AT THE LETTER WHICH GREXEN HELD OUT TO HIM",
+ "raw": {
+ "normalized_reference": "philip snatched at the letter which gregson held out to him.",
+ "normalized_hypothesis": "philip snatched at the letter which grexen held out to him",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ },
+ "semantic": {
+ "normalized_reference": "philip snatched at the letter which gregson held out to him",
+ "normalized_hypothesis": "philip snatched at the letter which grexen held out to him",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ }
+ },
+ {
+ "id": "release500_0216",
+ "category": "phonetic_balance",
+ "text": "For a time the exciting thrill of his adventure was gone.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0216.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "FOR A TIME THE EXCITING THRILL OF HIS ADVENTURE WAS GONE",
+ "raw": {
+ "normalized_reference": "for a time the exciting thrill of his adventure was gone.",
+ "normalized_hypothesis": "for a time the exciting thrill of his adventure was gone",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "for a time the exciting thrill of his adventure was gone",
+ "normalized_hypothesis": "for a time the exciting thrill of his adventure was gone",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0217",
+ "category": "phonetic_balance",
+ "text": "Obviously, it was a disease that could be contracted by contact.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0217.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "OBVIOUSLY IT WAS A DISEASE THAT COULD BE CONTRACTED BY CONTACT",
+ "raw": {
+ "normalized_reference": "obviously, it was a disease that could be contracted by contact.",
+ "normalized_hypothesis": "obviously it was a disease that could be contracted by contact",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "obviously it was a disease that could be contracted by contact",
+ "normalized_hypothesis": "obviously it was a disease that could be contracted by contact",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0218",
+ "category": "phonetic_balance",
+ "text": "Blind with rage, he darted in.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0218.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "BLIND WITH RAGE HE DARTED IN",
+ "raw": {
+ "normalized_reference": "blind with rage, he darted in.",
+ "normalized_hypothesis": "blind with rage he darted in",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "blind with rage he darted in",
+ "normalized_hypothesis": "blind with rage he darted in",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0219",
+ "category": "phonetic_balance",
+ "text": "You live on an income which your father earned.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0219.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "YOU LIVE ON AN INCOME WHICH YOUR FATHER EARNED",
+ "raw": {
+ "normalized_reference": "you live on an income which your father earned.",
+ "normalized_hypothesis": "you live on an income which your father earned",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "you live on an income which your father earned",
+ "normalized_hypothesis": "you live on an income which your father earned",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0220",
+ "category": "phonetic_balance",
+ "text": "Something about it seemed to fascinate him, to challenge his presence.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0220.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SOMETHING ABOUT IT SEEMED TO FASCINATE HIM TO CHALLENGE HIS PRESENCE",
+ "raw": {
+ "normalized_reference": "something about it seemed to fascinate him, to challenge his presence.",
+ "normalized_hypothesis": "something about it seemed to fascinate him to challenge his presence",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "something about it seemed to fascinate him to challenge his presence",
+ "normalized_hypothesis": "something about it seemed to fascinate him to challenge his presence",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0221",
+ "category": "phonetic_balance",
+ "text": "Her beautiful hair was done up in shining coils.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0221.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HER BEAUTIFUL HAIR WAS DONE UP IN SHINING COILS",
+ "raw": {
+ "normalized_reference": "her beautiful hair was done up in shining coils.",
+ "normalized_hypothesis": "her beautiful hair was done up in shining coils",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "her beautiful hair was done up in shining coils",
+ "normalized_hypothesis": "her beautiful hair was done up in shining coils",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0222",
+ "category": "phonetic_balance",
+ "text": "It does, was her audacious answer.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0222.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT DOES WAS HER AUDACIOUS ANSWER",
+ "raw": {
+ "normalized_reference": "it does, was her audacious answer.",
+ "normalized_hypothesis": "it does was her audacious answer",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it does was her audacious answer",
+ "normalized_hypothesis": "it does was her audacious answer",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0223",
+ "category": "phonetic_balance",
+ "text": "He caught himself with a jerk.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0223.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE CAUGHT HIMSELF WITH A JERK",
+ "raw": {
+ "normalized_reference": "he caught himself with a jerk.",
+ "normalized_hypothesis": "he caught himself with a jerk",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he caught himself with a jerk",
+ "normalized_hypothesis": "he caught himself with a jerk",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0224",
+ "category": "phonetic_balance",
+ "text": "He did not believe in the burning of daylight for such a luxury.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0224.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE DID NOT BELIEVE IN THE BURNING OF DAYLIGHT FOR SUCH A LUXURY",
+ "raw": {
+ "normalized_reference": "he did not believe in the burning of daylight for such a luxury.",
+ "normalized_hypothesis": "he did not believe in the burning of daylight for such a luxury",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he did not believe in the burning of daylight for such a luxury",
+ "normalized_hypothesis": "he did not believe in the burning of daylight for such a luxury",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0225",
+ "category": "phonetic_balance",
+ "text": "He had comparatively no advantages at first.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0225.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE HAD COMPARATIVELY NO ADVANTAGES AT FIRST",
+ "raw": {
+ "normalized_reference": "he had comparatively no advantages at first.",
+ "normalized_hypothesis": "he had comparatively no advantages at first",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he had comparatively no advantages at 1st",
+ "normalized_hypothesis": "he had comparatively no advantages at 1st",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0226",
+ "category": "phonetic_balance",
+ "text": "Such men believe, when they come together.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0226.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SUCH MEN BELIEVE WHEN THEY COME TOGETHER",
+ "raw": {
+ "normalized_reference": "such men believe, when they come together.",
+ "normalized_hypothesis": "such men believe when they come together",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "such men believe when they come together",
+ "normalized_hypothesis": "such men believe when they come together",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0227",
+ "category": "phonetic_balance",
+ "text": "The ship should be in within a week or ten days.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0227.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE SHIP SHOULD BE IN WITHIN A WEEK OR TEN DAYS",
+ "raw": {
+ "normalized_reference": "the ship should be in within a week or ten days.",
+ "normalized_hypothesis": "the ship should be in within a week or ten days",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the ship should be in within a week or 10 days",
+ "normalized_hypothesis": "the ship should be in within a week or 10 days",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0228",
+ "category": "phonetic_balance",
+ "text": "Such a dog the wise driver kills, or turns loose.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0228.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SUCH A DOG THE WISE DRIVER KILLS OR TURNS LOOSE",
+ "raw": {
+ "normalized_reference": "such a dog the wise driver kills, or turns loose.",
+ "normalized_hypothesis": "such a dog the wise driver kills or turns loose",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "such a dog the wise driver kills or turns loose",
+ "normalized_hypothesis": "such a dog the wise driver kills or turns loose",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0229",
+ "category": "phonetic_balance",
+ "text": "You have heard always how he was the lover of the Princess Naomi.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0229.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "YOU HAVE HEARD ALWAYS HOW HE WAS THE LOVER OF THE PRINCESS NAOMI",
+ "raw": {
+ "normalized_reference": "you have heard always how he was the lover of the princess naomi.",
+ "normalized_hypothesis": "you have heard always how he was the lover of the princess naomi",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "you have heard always how he was the lover of the princess naomi",
+ "normalized_hypothesis": "you have heard always how he was the lover of the princess naomi",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0230",
+ "category": "phonetic_balance",
+ "text": "I want to know how all this is possible.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0230.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I WANT TO KNOW HOW ALL THIS IS POSSIBLE",
+ "raw": {
+ "normalized_reference": "i want to know how all this is possible.",
+ "normalized_hypothesis": "i want to know how all this is possible",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i want to know how all this is possible",
+ "normalized_hypothesis": "i want to know how all this is possible",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0231",
+ "category": "phonetic_balance",
+ "text": "It was my idea to a tee.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0231.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT WAS MY IDEA TO ATEA",
+ "raw": {
+ "normalized_reference": "it was my idea to a tee.",
+ "normalized_hypothesis": "it was my idea to atea",
+ "errors": 2,
+ "words": 7,
+ "wer": 0.2857142857142857
+ },
+ "semantic": {
+ "normalized_reference": "it was my idea to a tee",
+ "normalized_hypothesis": "it was my idea to atea",
+ "errors": 2,
+ "words": 7,
+ "wer": 0.2857142857142857
+ }
+ },
+ {
+ "id": "release500_0232",
+ "category": "phonetic_balance",
+ "text": "They likewise are disinclined to being eaten.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0232.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THEY LIKEWISE ARE DISINCLINED TO BE IN EDEN",
+ "raw": {
+ "normalized_reference": "they likewise are disinclined to being eaten.",
+ "normalized_hypothesis": "they likewise are disinclined to be in eden",
+ "errors": 3,
+ "words": 7,
+ "wer": 0.42857142857142855
+ },
+ "semantic": {
+ "normalized_reference": "they likewise are disinclined to being eaten",
+ "normalized_hypothesis": "they likewise are disinclined to be in eden",
+ "errors": 3,
+ "words": 7,
+ "wer": 0.42857142857142855
+ }
+ },
+ {
+ "id": "release500_0233",
+ "category": "phonetic_balance",
+ "text": "This state of mind comes of an undue prominence of the ego.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0233.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THIS STATE OF MIND COMES OF AN UNDUE PROMINENCE OF THE EGO",
+ "raw": {
+ "normalized_reference": "this state of mind comes of an undue prominence of the ego.",
+ "normalized_hypothesis": "this state of mind comes of an undue prominence of the ego",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "this state of mind comes of an undue prominence of the ego",
+ "normalized_hypothesis": "this state of mind comes of an undue prominence of the ego",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0234",
+ "category": "phonetic_balance",
+ "text": "Then it is as I said, Womble announced with finality.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0234.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THEN IT IS AS I SAID WAMBELL ANNOUNCED WITH FINALITY",
+ "raw": {
+ "normalized_reference": "then it is as i said, womble announced with finality.",
+ "normalized_hypothesis": "then it is as i said wambell announced with finality",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ },
+ "semantic": {
+ "normalized_reference": "then it is as i said womble announced with finality",
+ "normalized_hypothesis": "then it is as i said wambell announced with finality",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ }
+ },
+ {
+ "id": "release500_0235",
+ "category": "phonetic_balance",
+ "text": "I can see that knife now.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0235.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I CAN SEE THAT KNIFE NOW",
+ "raw": {
+ "normalized_reference": "i can see that knife now.",
+ "normalized_hypothesis": "i can see that knife now",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i can see that knife now",
+ "normalized_hypothesis": "i can see that knife now",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0236",
+ "category": "phonetic_balance",
+ "text": "If I meet her again I shall apologize, said Eileen.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0236.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IF I MEET HER AGAIN I SHALL APOLOGIZE SAID ILEEN",
+ "raw": {
+ "normalized_reference": "if i meet her again i shall apologize, said eileen.",
+ "normalized_hypothesis": "if i meet her again i shall apologize said ileen",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ },
+ "semantic": {
+ "normalized_reference": "if i meet her again i shall apologize said eileen",
+ "normalized_hypothesis": "if i meet her again i shall apologize said ileen",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ }
+ },
+ {
+ "id": "release500_0237",
+ "category": "phonetic_balance",
+ "text": "A wildly exciting time was his during the week preceding Thursday the eighteenth.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0237.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "A WILDLY EXCITING TIME WAS HIS DURING THE WEEK PRECEDING THURSDAY THE EIGHTEENTH",
+ "raw": {
+ "normalized_reference": "a wildly exciting time was his during the week preceding thursday the eighteenth.",
+ "normalized_hypothesis": "a wildly exciting time was his during the week preceding thursday the eighteenth",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "a wildly exciting time was his during the week preceding thursday the 18th",
+ "normalized_hypothesis": "a wildly exciting time was his during the week preceding thursday the 18th",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0238",
+ "category": "phonetic_balance",
+ "text": "In what bucolic school of fence he had been taught was beyond imagining.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0238.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IN WHAT BUCOLIC SCHOOL O FENCEHE HAD BEEN TAUGHT WAS BEYOND IMAGINING",
+ "raw": {
+ "normalized_reference": "in what bucolic school of fence he had been taught was beyond imagining.",
+ "normalized_hypothesis": "in what bucolic school o fencehe had been taught was beyond imagining",
+ "errors": 3,
+ "words": 13,
+ "wer": 0.23076923076923078
+ },
+ "semantic": {
+ "normalized_reference": "in what bucolic school of fence he had been taught was beyond imagining",
+ "normalized_hypothesis": "in what bucolic school 0 fencehe had been taught was beyond imagining",
+ "errors": 3,
+ "words": 13,
+ "wer": 0.23076923076923078
+ }
+ },
+ {
+ "id": "release500_0239",
+ "category": "phonetic_balance",
+ "text": "So she said, the irate skipper dashed on.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0239.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SO SHE SAID THE IRATE SKIPPER DASHED OM",
+ "raw": {
+ "normalized_reference": "so she said, the irate skipper dashed on.",
+ "normalized_hypothesis": "so she said the irate skipper dashed om",
+ "errors": 1,
+ "words": 8,
+ "wer": 0.125
+ },
+ "semantic": {
+ "normalized_reference": "so she said the irate skipper dashed on",
+ "normalized_hypothesis": "so she said the irate skipper dashed om",
+ "errors": 1,
+ "words": 8,
+ "wer": 0.125
+ }
+ },
+ {
+ "id": "release500_0240",
+ "category": "phonetic_balance",
+ "text": "It is not an attempt to smash the market.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0240.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT IS NOT AN ATTEMPT TO SMASH THE MARKET",
+ "raw": {
+ "normalized_reference": "it is not an attempt to smash the market.",
+ "normalized_hypothesis": "it is not an attempt to smash the market",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it is not an attempt to smash the market",
+ "normalized_hypothesis": "it is not an attempt to smash the market",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0241",
+ "category": "phonetic_balance",
+ "text": "He destroyed everything that had belonged to the woman.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0241.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE DESTROYED EVERYTHING THAT HAD BELONGED TO THE WOMAN",
+ "raw": {
+ "normalized_reference": "he destroyed everything that had belonged to the woman.",
+ "normalized_hypothesis": "he destroyed everything that had belonged to the woman",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he destroyed everything that had belonged to the woman",
+ "normalized_hypothesis": "he destroyed everything that had belonged to the woman",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0242",
+ "category": "phonetic_balance",
+ "text": "Did Thorpe go to see any one in Churchill.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0242.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "DID THORPE GO TO SEE ANY ONE IN CHURCHILL",
+ "raw": {
+ "normalized_reference": "did thorpe go to see any one in churchill.",
+ "normalized_hypothesis": "did thorpe go to see any one in churchill",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "did thorpe go to see any one in churchill",
+ "normalized_hypothesis": "did thorpe go to see any one in churchill",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0243",
+ "category": "phonetic_balance",
+ "text": "Many other similar disconcerting omissions will be noticed in the Manuscript.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0243.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "MANY OTHER SIMILAR DISCONCERTING OMISSIONS WILL BE NOTICED IN THE MANUSCRIPT",
+ "raw": {
+ "normalized_reference": "many other similar disconcerting omissions will be noticed in the manuscript.",
+ "normalized_hypothesis": "many other similar disconcerting omissions will be noticed in the manuscript",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "many other similar disconcerting omissions will be noticed in the manuscript",
+ "normalized_hypothesis": "many other similar disconcerting omissions will be noticed in the manuscript",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0244",
+ "category": "phonetic_balance",
+ "text": "The ringing of the big bell aroused him.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0244.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE RINGING OF THE BIG BELL AROUSED HIM",
+ "raw": {
+ "normalized_reference": "the ringing of the big bell aroused him.",
+ "normalized_hypothesis": "the ringing of the big bell aroused him",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the ringing of the big bell aroused him",
+ "normalized_hypothesis": "the ringing of the big bell aroused him",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0245",
+ "category": "phonetic_balance",
+ "text": "A sinewy hand, dripping with water, was clutching the rail.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0245.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "A SINEWY HAND DRIPPING WITH WATER WAS CLUTCHING THE RAIL",
+ "raw": {
+ "normalized_reference": "a sinewy hand, dripping with water, was clutching the rail.",
+ "normalized_hypothesis": "a sinewy hand dripping with water was clutching the rail",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "a sinewy hand dripping with water was clutching the rail",
+ "normalized_hypothesis": "a sinewy hand dripping with water was clutching the rail",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0246",
+ "category": "phonetic_balance",
+ "text": "Enters now the psychology of the situation.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0246.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "ENTERS NOW THE PSYCHOLOGY OF THE SITUATION",
+ "raw": {
+ "normalized_reference": "enters now the psychology of the situation.",
+ "normalized_hypothesis": "enters now the psychology of the situation",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "enters now the psychology of the situation",
+ "normalized_hypothesis": "enters now the psychology of the situation",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0247",
+ "category": "phonetic_balance",
+ "text": "Wash your hands of me.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0247.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WASH YOUR HANDS OF ME",
+ "raw": {
+ "normalized_reference": "wash your hands of me.",
+ "normalized_hypothesis": "wash your hands of me",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "wash your hands of me",
+ "normalized_hypothesis": "wash your hands of me",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0248",
+ "category": "phonetic_balance",
+ "text": "He confessed that the sketch had startled him.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0248.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE CONFESSED THAT THE SKETCH HAD STARTLED HIM",
+ "raw": {
+ "normalized_reference": "he confessed that the sketch had startled him.",
+ "normalized_hypothesis": "he confessed that the sketch had startled him",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he confessed that the sketch had startled him",
+ "normalized_hypothesis": "he confessed that the sketch had startled him",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0249",
+ "category": "phonetic_balance",
+ "text": "Her face was against his breast.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0249.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HER FACE WAS AGAINST HIS BREAST",
+ "raw": {
+ "normalized_reference": "her face was against his breast.",
+ "normalized_hypothesis": "her face was against his breast",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "her face was against his breast",
+ "normalized_hypothesis": "her face was against his breast",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0250",
+ "category": "phonetic_balance",
+ "text": "He was a pariah; a wanderer without a friend or a home.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0250.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE WAS A PARIAH A WANDERER WITHOUT A FRIEND OR A HOME",
+ "raw": {
+ "normalized_reference": "he was a pariah; a wanderer without a friend or a home.",
+ "normalized_hypothesis": "he was a pariah a wanderer without a friend or a home",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he was a pariah a wanderer without a friend or a home",
+ "normalized_hypothesis": "he was a pariah a wanderer without a friend or a home",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0251",
+ "category": "phonetic_balance",
+ "text": "A cry of joy burst from Philip's lips.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0251.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "A CRY OF JOY BURST FROM PHILIP'S LIPS",
+ "raw": {
+ "normalized_reference": "a cry of joy burst from philip's lips.",
+ "normalized_hypothesis": "a cry of joy burst from philip's lips",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "a cry of joy burst from philip is lips",
+ "normalized_hypothesis": "a cry of joy burst from philip is lips",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0252",
+ "category": "phonetic_balance",
+ "text": "And the big Persian knew of his existence before he did of hers.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0252.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "AND THE BIG PERSIAN KNEW OF HIS EXISTENCE BEFORE HE DID OF HERS",
+ "raw": {
+ "normalized_reference": "and the big persian knew of his existence before he did of hers.",
+ "normalized_hypothesis": "and the big persian knew of his existence before he did of hers",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "and the big persian knew of his existence before he did of hers",
+ "normalized_hypothesis": "and the big persian knew of his existence before he did of hers",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0253",
+ "category": "phonetic_balance",
+ "text": "The men stared into each other's face.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0253.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE MEN STARED INTO EACH OTHER'S FACE",
+ "raw": {
+ "normalized_reference": "the men stared into each other's face.",
+ "normalized_hypothesis": "the men stared into each other's face",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the men stared into each other is face",
+ "normalized_hypothesis": "the men stared into each other is face",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0254",
+ "category": "phonetic_balance",
+ "text": "It is the fire, partly, she said.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0254.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT IS THE FIRE PARTLY SHE SAID",
+ "raw": {
+ "normalized_reference": "it is the fire, partly, she said.",
+ "normalized_hypothesis": "it is the fire partly she said",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it is the fire partly she said",
+ "normalized_hypothesis": "it is the fire partly she said",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0255",
+ "category": "phonetic_balance",
+ "text": "It seemed nearer to him since he had seen and talked with Gregson.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0255.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT SEEMED NEARER TO HIM SINCE HE HAD SEEN AND TALKED WITH GRIGSON",
+ "raw": {
+ "normalized_reference": "it seemed nearer to him since he had seen and talked with gregson.",
+ "normalized_hypothesis": "it seemed nearer to him since he had seen and talked with grigson",
+ "errors": 1,
+ "words": 13,
+ "wer": 0.07692307692307693
+ },
+ "semantic": {
+ "normalized_reference": "it seemed nearer to him since he had seen and talked with gregson",
+ "normalized_hypothesis": "it seemed nearer to him since he had seen and talked with grigson",
+ "errors": 1,
+ "words": 13,
+ "wer": 0.07692307692307693
+ }
+ },
+ {
+ "id": "release500_0256",
+ "category": "phonetic_balance",
+ "text": "Among my minor afflictions, I may mention a new and mysterious one.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0256.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "AMONG MY MINOR AFFLICTIONS I MAY MENTION A NEW AND MYSTERIOUS ONE",
+ "raw": {
+ "normalized_reference": "among my minor afflictions, i may mention a new and mysterious one.",
+ "normalized_hypothesis": "among my minor afflictions i may mention a new and mysterious one",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "among my minor afflictions i may mention a new and mysterious one",
+ "normalized_hypothesis": "among my minor afflictions i may mention a new and mysterious one",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0257",
+ "category": "phonetic_balance",
+ "text": "Between him and all domestic animals there must be no hostilities.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0257.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "BETWEEN HIM AND ALL DOMESTIC ANIMALS THERE MUST BE NO HOSTILITIES",
+ "raw": {
+ "normalized_reference": "between him and all domestic animals there must be no hostilities.",
+ "normalized_hypothesis": "between him and all domestic animals there must be no hostilities",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "between him and all domestic animals there must be no hostilities",
+ "normalized_hypothesis": "between him and all domestic animals there must be no hostilities",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0258",
+ "category": "phonetic_balance",
+ "text": "With these arguments he convinced himself that he should go on alone.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0258.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WITH THESE ARGUMENTS HE CONVINCED HIMSELF THAT HE SHOULD GO ON ALONE",
+ "raw": {
+ "normalized_reference": "with these arguments he convinced himself that he should go on alone.",
+ "normalized_hypothesis": "with these arguments he convinced himself that he should go on alone",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "with these arguments he convinced himself that he should go on alone",
+ "normalized_hypothesis": "with these arguments he convinced himself that he should go on alone",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0259",
+ "category": "phonetic_balance",
+ "text": "We are consumed in our own flesh-pots.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0259.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WE ARE CONSUMED IN OUR OWN FLESH PARTS",
+ "raw": {
+ "normalized_reference": "we are consumed in our own flesh-pots.",
+ "normalized_hypothesis": "we are consumed in our own flesh parts",
+ "errors": 1,
+ "words": 8,
+ "wer": 0.125
+ },
+ "semantic": {
+ "normalized_reference": "we are consumed in our own flesh pots",
+ "normalized_hypothesis": "we are consumed in our own flesh parts",
+ "errors": 1,
+ "words": 8,
+ "wer": 0.125
+ }
+ },
+ {
+ "id": "release500_0260",
+ "category": "phonetic_balance",
+ "text": "Mercedes screamed, cried, laughed, and manifested the chaotic abandonment of hysteria.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0260.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "MERCEDES SCREAMED CRIED LAUGHED AND MANIFESTED THE CHAOTIC ABANDONMENT OF HYSTERIA",
+ "raw": {
+ "normalized_reference": "mercedes screamed, cried, laughed, and manifested the chaotic abandonment of hysteria.",
+ "normalized_hypothesis": "mercedes screamed cried laughed and manifested the chaotic abandonment of hysteria",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "mercedes screamed cried laughed and manifested the chaotic abandonment of hysteria",
+ "normalized_hypothesis": "mercedes screamed cried laughed and manifested the chaotic abandonment of hysteria",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0261",
+ "category": "phonetic_balance",
+ "text": "Philip made no effort to follow.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0261.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "PHILIP MADE NO EFFORT TO FOLLOW",
+ "raw": {
+ "normalized_reference": "philip made no effort to follow.",
+ "normalized_hypothesis": "philip made no effort to follow",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "philip made no effort to follow",
+ "normalized_hypothesis": "philip made no effort to follow",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0262",
+ "category": "phonetic_balance",
+ "text": "She said with chattering teeth.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0262.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SHE SAID WITH CHATTERING TEETH",
+ "raw": {
+ "normalized_reference": "she said with chattering teeth.",
+ "normalized_hypothesis": "she said with chattering teeth",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "she said with chattering teeth",
+ "normalized_hypothesis": "she said with chattering teeth",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0263",
+ "category": "phonetic_balance",
+ "text": "Halfway around the track one donkey got into an argument with its rider.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0263.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HALFWAY AROUND THE TRACK ONE DONKEY GOT INTO AN ARGUMENT WITH ITS RIDER",
+ "raw": {
+ "normalized_reference": "halfway around the track one donkey got into an argument with its rider.",
+ "normalized_hypothesis": "halfway around the track one donkey got into an argument with its rider",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "halfway around the track one donkey got into an argument with its rider",
+ "normalized_hypothesis": "halfway around the track one donkey got into an argument with its rider",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0264",
+ "category": "phonetic_balance",
+ "text": "Earth and gravel seemed to fill the pan.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0264.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "EARTH AND GRAVEL SEEMED TO FILL THE PAN",
+ "raw": {
+ "normalized_reference": "earth and gravel seemed to fill the pan.",
+ "normalized_hypothesis": "earth and gravel seemed to fill the pan",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "earth and gravel seemed to fill the pan",
+ "normalized_hypothesis": "earth and gravel seemed to fill the pan",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0265",
+ "category": "phonetic_balance",
+ "text": "Her achievements with cocoanuts were a revelation.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0265.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HER ACHIEVEMENTS WITH COCOANUTS WERE A REVELATION",
+ "raw": {
+ "normalized_reference": "her achievements with cocoanuts were a revelation.",
+ "normalized_hypothesis": "her achievements with cocoanuts were a revelation",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "her achievements with cocoanuts were a revelation",
+ "normalized_hypothesis": "her achievements with cocoanuts were a revelation",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0266",
+ "category": "phonetic_balance",
+ "text": "For a full minute he crouched and listened.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0266.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "FOR A FULL MINUTE HE CROUCHED AND LISTENED",
+ "raw": {
+ "normalized_reference": "for a full minute he crouched and listened.",
+ "normalized_hypothesis": "for a full minute he crouched and listened",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "for a full minute he crouched and listened",
+ "normalized_hypothesis": "for a full minute he crouched and listened",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0267",
+ "category": "phonetic_balance",
+ "text": "It was edged with ice.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0267.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT WAS EDGED WITH ICE",
+ "raw": {
+ "normalized_reference": "it was edged with ice.",
+ "normalized_hypothesis": "it was edged with ice",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it was edged with ice",
+ "normalized_hypothesis": "it was edged with ice",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0268",
+ "category": "phonetic_balance",
+ "text": "The Portuguese boy passed the Hawaiian.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0268.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE PORTUGUESE BOY PASSED THE HAWAIAN",
+ "raw": {
+ "normalized_reference": "the portuguese boy passed the hawaiian.",
+ "normalized_hypothesis": "the portuguese boy passed the hawaian",
+ "errors": 1,
+ "words": 6,
+ "wer": 0.16666666666666666
+ },
+ "semantic": {
+ "normalized_reference": "the portuguese boy passed the hawaiian",
+ "normalized_hypothesis": "the portuguese boy passed the hawaian",
+ "errors": 1,
+ "words": 6,
+ "wer": 0.16666666666666666
+ }
+ },
+ {
+ "id": "release500_0269",
+ "category": "phonetic_balance",
+ "text": "But he did not broach it, preferring to mature it carefully.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0269.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "BUT HE DID NOT BROOCH IT PREFERRING TO MATURE IT CAREFULLY",
+ "raw": {
+ "normalized_reference": "but he did not broach it, preferring to mature it carefully.",
+ "normalized_hypothesis": "but he did not brooch it preferring to mature it carefully",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ },
+ "semantic": {
+ "normalized_reference": "but he did not broach it preferring to mature it carefully",
+ "normalized_hypothesis": "but he did not brooch it preferring to mature it carefully",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ }
+ },
+ {
+ "id": "release500_0270",
+ "category": "phonetic_balance",
+ "text": "Come on, Del Mar challenged.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0270.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "CAMON DELMAR CHALLENGED",
+ "raw": {
+ "normalized_reference": "come on, del mar challenged.",
+ "normalized_hypothesis": "camon delmar challenged",
+ "errors": 4,
+ "words": 5,
+ "wer": 0.8
+ },
+ "semantic": {
+ "normalized_reference": "come on del mar challenged",
+ "normalized_hypothesis": "camon delmar challenged",
+ "errors": 4,
+ "words": 5,
+ "wer": 0.8
+ }
+ },
+ {
+ "id": "release500_0271",
+ "category": "phonetic_balance",
+ "text": "For a full half minute Jeanne looked at him without speaking.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0271.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "FOR A FULL HALF MINUTEJEAN LOOKED AT HIM WITHOUT SPEAKING",
+ "raw": {
+ "normalized_reference": "for a full half minute jeanne looked at him without speaking.",
+ "normalized_hypothesis": "for a full half minutejean looked at him without speaking",
+ "errors": 2,
+ "words": 11,
+ "wer": 0.18181818181818182
+ },
+ "semantic": {
+ "normalized_reference": "for a full half minute jeanne looked at him without speaking",
+ "normalized_hypothesis": "for a full half minutejean looked at him without speaking",
+ "errors": 2,
+ "words": 11,
+ "wer": 0.18181818181818182
+ }
+ },
+ {
+ "id": "release500_0272",
+ "category": "phonetic_balance",
+ "text": "He bore no grudges and had few enemies.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0272.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE BORE NO GRUDGES AND HAD FEW ENEMIES",
+ "raw": {
+ "normalized_reference": "he bore no grudges and had few enemies.",
+ "normalized_hypothesis": "he bore no grudges and had few enemies",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he bore no grudges and had few enemies",
+ "normalized_hypothesis": "he bore no grudges and had few enemies",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0273",
+ "category": "phonetic_balance",
+ "text": "I'll go over tomorrow afternoon.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0273.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I'LL GO OVER TO MORROW AFTERNOON",
+ "raw": {
+ "normalized_reference": "i'll go over tomorrow afternoon.",
+ "normalized_hypothesis": "i'll go over to morrow afternoon",
+ "errors": 2,
+ "words": 5,
+ "wer": 0.4
+ },
+ "semantic": {
+ "normalized_reference": "i will go over tomorrow afternoon",
+ "normalized_hypothesis": "i will go over to morrow afternoon",
+ "errors": 2,
+ "words": 6,
+ "wer": 0.3333333333333333
+ }
+ },
+ {
+ "id": "release500_0274",
+ "category": "phonetic_balance",
+ "text": "Well, there are better men in Hawaii, that's all.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0274.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WELL THERE ARE BETTER MEN IN HAWAI THAT'S ALL",
+ "raw": {
+ "normalized_reference": "well, there are better men in hawaii, that's all.",
+ "normalized_hypothesis": "well there are better men in hawai that's all",
+ "errors": 1,
+ "words": 9,
+ "wer": 0.1111111111111111
+ },
+ "semantic": {
+ "normalized_reference": "well there are better men in hawaii that is all",
+ "normalized_hypothesis": "well there are better men in hawai that is all",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ }
+ },
+ {
+ "id": "release500_0275",
+ "category": "phonetic_balance",
+ "text": "He was an athlete and a giant.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0275.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE WAS AN ATHLETE AND A GIANT",
+ "raw": {
+ "normalized_reference": "he was an athlete and a giant.",
+ "normalized_hypothesis": "he was an athlete and a giant",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he was an athlete and a giant",
+ "normalized_hypothesis": "he was an athlete and a giant",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0276",
+ "category": "phonetic_balance",
+ "text": "McVeigh when he returns from a trip to Honolulu.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0276.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "MC VAY WHEN HE RETURNS FROM A TRIP TO HONNOLOLOU",
+ "raw": {
+ "normalized_reference": "mcveigh when he returns from a trip to honolulu.",
+ "normalized_hypothesis": "mc vay when he returns from a trip to honnololou",
+ "errors": 3,
+ "words": 9,
+ "wer": 0.3333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "mcveigh when he returns from a trip to honolulu",
+ "normalized_hypothesis": "mc vay when he returns from a trip to honnololou",
+ "errors": 3,
+ "words": 9,
+ "wer": 0.3333333333333333
+ }
+ },
+ {
+ "id": "release500_0277",
+ "category": "phonetic_balance",
+ "text": "You're a devil for fighting, and will surely win.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0277.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "YOU'RE A DEVIL FOR FIGHTING AND WE'LL SURELY WIN",
+ "raw": {
+ "normalized_reference": "you're a devil for fighting, and will surely win.",
+ "normalized_hypothesis": "you're a devil for fighting and we'll surely win",
+ "errors": 1,
+ "words": 9,
+ "wer": 0.1111111111111111
+ },
+ "semantic": {
+ "normalized_reference": "you are a devil for fighting and will surely win",
+ "normalized_hypothesis": "you are a devil for fighting and we will surely win",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ }
+ },
+ {
+ "id": "release500_0278",
+ "category": "phonetic_balance",
+ "text": "All an appearance can know is mirage.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0278.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "ALL IN APPEARANCE CAN KNOW ISMAROCHE",
+ "raw": {
+ "normalized_reference": "all an appearance can know is mirage.",
+ "normalized_hypothesis": "all in appearance can know ismaroche",
+ "errors": 3,
+ "words": 7,
+ "wer": 0.42857142857142855
+ },
+ "semantic": {
+ "normalized_reference": "all an appearance can know is mirage",
+ "normalized_hypothesis": "all in appearance can know ismaroche",
+ "errors": 3,
+ "words": 7,
+ "wer": 0.42857142857142855
+ }
+ },
+ {
+ "id": "release500_0279",
+ "category": "phonetic_balance",
+ "text": "You fired me out of your house, in short.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0279.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "YOU FIRED ME OUT OF YOUR HOUSE IN SHORT",
+ "raw": {
+ "normalized_reference": "you fired me out of your house, in short.",
+ "normalized_hypothesis": "you fired me out of your house in short",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "you fired me out of your house in short",
+ "normalized_hypothesis": "you fired me out of your house in short",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0280",
+ "category": "phonetic_balance",
+ "text": "He made sure that the magazine was loaded, and resumed his paddling.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0280.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE MADE SURE THAT THE MAGAZINE WAS LOADED AND RESUMED HIS PADDLING",
+ "raw": {
+ "normalized_reference": "he made sure that the magazine was loaded, and resumed his paddling.",
+ "normalized_hypothesis": "he made sure that the magazine was loaded and resumed his paddling",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he made sure that the magazine was loaded and resumed his paddling",
+ "normalized_hypothesis": "he made sure that the magazine was loaded and resumed his paddling",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0281",
+ "category": "phonetic_balance",
+ "text": "Change chairs, Daylight commanded.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0281.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "CHANGED CHAIRS DAYLIGHT COMMANDED",
+ "raw": {
+ "normalized_reference": "change chairs, daylight commanded.",
+ "normalized_hypothesis": "changed chairs daylight commanded",
+ "errors": 1,
+ "words": 4,
+ "wer": 0.25
+ },
+ "semantic": {
+ "normalized_reference": "change chairs daylight commanded",
+ "normalized_hypothesis": "changed chairs daylight commanded",
+ "errors": 1,
+ "words": 4,
+ "wer": 0.25
+ }
+ },
+ {
+ "id": "release500_0282",
+ "category": "phonetic_balance",
+ "text": "By that answer my professional medical prestige stood or fell.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0282.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "BY THAT ANSWER MY PROFESSIONAL MEDICAL PRESTIGE STOOD OR FELL",
+ "raw": {
+ "normalized_reference": "by that answer my professional medical prestige stood or fell.",
+ "normalized_hypothesis": "by that answer my professional medical prestige stood or fell",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "by that answer my professional medical prestige stood or fell",
+ "normalized_hypothesis": "by that answer my professional medical prestige stood or fell",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0283",
+ "category": "phonetic_balance",
+ "text": "The moon had already begun its westward decline.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0283.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE MOON HAD ALREADY BEGUN ITS WESTWARD DECLINE",
+ "raw": {
+ "normalized_reference": "the moon had already begun its westward decline.",
+ "normalized_hypothesis": "the moon had already begun its westward decline",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the moon had already begun its westward decline",
+ "normalized_hypothesis": "the moon had already begun its westward decline",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0284",
+ "category": "phonetic_balance",
+ "text": "At sea, Monday, March 16, 1908.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0284.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "AT C MONDAY MARCH SIXTEENTH NINETEEN HUNDRED EIGHT",
+ "raw": {
+ "normalized_reference": "at sea, monday, march 16, 1908.",
+ "normalized_hypothesis": "at c monday march sixteenth nineteen hundred eight",
+ "errors": 5,
+ "words": 6,
+ "wer": 0.8333333333333334
+ },
+ "semantic": {
+ "normalized_reference": "at sea monday march 16 1908",
+ "normalized_hypothesis": "at c monday march 16th 1908",
+ "errors": 2,
+ "words": 6,
+ "wer": 0.3333333333333333
+ }
+ },
+ {
+ "id": "release500_0285",
+ "category": "phonetic_balance",
+ "text": "And wherever I ranged, the way lay along alcohol-drenched roads.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0285.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "AND WHEREVER I RANGED THE WAY LAY ALONG ALCOHOL DRENCHED ROADS",
+ "raw": {
+ "normalized_reference": "and wherever i ranged, the way lay along alcohol-drenched roads.",
+ "normalized_hypothesis": "and wherever i ranged the way lay along alcohol drenched roads",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "and wherever i ranged the way lay along alcohol drenched roads",
+ "normalized_hypothesis": "and wherever i ranged the way lay along alcohol drenched roads",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0286",
+ "category": "phonetic_balance",
+ "text": "The date was nearly eighteen years old.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0286.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE DATE WAS NEARLY EIGHTEEN YEARS OLD",
+ "raw": {
+ "normalized_reference": "the date was nearly eighteen years old.",
+ "normalized_hypothesis": "the date was nearly eighteen years old",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the date was nearly 18 years old",
+ "normalized_hypothesis": "the date was nearly 18 years old",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0287",
+ "category": "phonetic_balance",
+ "text": "Very early in my life, I separated from my mother.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0287.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "VERY EARLY IN MY LIFE I SEPARATED FROM MY MOTHER",
+ "raw": {
+ "normalized_reference": "very early in my life, i separated from my mother.",
+ "normalized_hypothesis": "very early in my life i separated from my mother",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "very early in my life i separated from my mother",
+ "normalized_hypothesis": "very early in my life i separated from my mother",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0288",
+ "category": "phonetic_balance",
+ "text": "There was pride and strength, the ring of triumph in his voice.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0288.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THERE WAS PRIDE AND STRENGTH THE RING OF TRIUMPH IN HIS VOICE",
+ "raw": {
+ "normalized_reference": "there was pride and strength, the ring of triumph in his voice.",
+ "normalized_hypothesis": "there was pride and strength the ring of triumph in his voice",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "there was pride and strength the ring of triumph in his voice",
+ "normalized_hypothesis": "there was pride and strength the ring of triumph in his voice",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0289",
+ "category": "phonetic_balance",
+ "text": "From my earliest recollection my sleep was a period of terror.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0289.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "FROM MY EARLIEST RECOLLECTION MY SLEEP WAS A PERIOD OF TERROR",
+ "raw": {
+ "normalized_reference": "from my earliest recollection my sleep was a period of terror.",
+ "normalized_hypothesis": "from my earliest recollection my sleep was a period of terror",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "from my earliest recollection my sleep was a period of terror",
+ "normalized_hypothesis": "from my earliest recollection my sleep was a period of terror",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0290",
+ "category": "phonetic_balance",
+ "text": "We'll have to watch our chances.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0290.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WE'LL HAVE TO WATCH OUR CHANCES",
+ "raw": {
+ "normalized_reference": "we'll have to watch our chances.",
+ "normalized_hypothesis": "we'll have to watch our chances",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "we will have to watch our chances",
+ "normalized_hypothesis": "we will have to watch our chances",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0291",
+ "category": "phonetic_balance",
+ "text": "Close beside him gleamed the white fangs of the wolf-dog.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0291.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "CLOSE BESIDE HIM GLEAMED THE WHITE FANGS OF THE WOLF DOG",
+ "raw": {
+ "normalized_reference": "close beside him gleamed the white fangs of the wolf-dog.",
+ "normalized_hypothesis": "close beside him gleamed the white fangs of the wolf dog",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "close beside him gleamed the white fangs of the wolf dog",
+ "normalized_hypothesis": "close beside him gleamed the white fangs of the wolf dog",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0292",
+ "category": "phonetic_balance",
+ "text": "Would you be satisfied with that one hundredth part of me.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0292.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WOULD YOU BE SATISFIED WITH THAT ONE HUNDREDTH PART OF ME",
+ "raw": {
+ "normalized_reference": "would you be satisfied with that one hundredth part of me.",
+ "normalized_hypothesis": "would you be satisfied with that one hundredth part of me",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "would you be satisfied with that 100th part of me",
+ "normalized_hypothesis": "would you be satisfied with that 100th part of me",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0293",
+ "category": "phonetic_balance",
+ "text": "Two gigantic owls were tearing at the carcass.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0293.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "TWO GIGANTIC OWLS WERE TEARING AT THE CARCASE",
+ "raw": {
+ "normalized_reference": "two gigantic owls were tearing at the carcass.",
+ "normalized_hypothesis": "two gigantic owls were tearing at the carcase",
+ "errors": 1,
+ "words": 8,
+ "wer": 0.125
+ },
+ "semantic": {
+ "normalized_reference": "2 gigantic owls were tearing at the carcass",
+ "normalized_hypothesis": "2 gigantic owls were tearing at the carcase",
+ "errors": 1,
+ "words": 8,
+ "wer": 0.125
+ }
+ },
+ {
+ "id": "release500_0294",
+ "category": "phonetic_balance",
+ "text": "The lush vegetation of that sheltered spot make a natural shield.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0294.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE LUSH VEGETATION OF THAT SHELTERED SPOT MAKE A NATURAL SHIELD",
+ "raw": {
+ "normalized_reference": "the lush vegetation of that sheltered spot make a natural shield.",
+ "normalized_hypothesis": "the lush vegetation of that sheltered spot make a natural shield",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the lush vegetation of that sheltered spot make a natural shield",
+ "normalized_hypothesis": "the lush vegetation of that sheltered spot make a natural shield",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0295",
+ "category": "phonetic_balance",
+ "text": "Below him the shadow was broken into a pool of rippling starlight.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0295.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "BELOW HIM THE SHADOW WAS BROKEN INTO A POOL OF RIPPLING STARLIGHT",
+ "raw": {
+ "normalized_reference": "below him the shadow was broken into a pool of rippling starlight.",
+ "normalized_hypothesis": "below him the shadow was broken into a pool of rippling starlight",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "below him the shadow was broken into a pool of rippling starlight",
+ "normalized_hypothesis": "below him the shadow was broken into a pool of rippling starlight",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0296",
+ "category": "phonetic_balance",
+ "text": "Lots of men take women buggy riding.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0296.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "LOTS OF MEN TAKE WOMEN BUGGY RIDING",
+ "raw": {
+ "normalized_reference": "lots of men take women buggy riding.",
+ "normalized_hypothesis": "lots of men take women buggy riding",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "lots of men take women buggy riding",
+ "normalized_hypothesis": "lots of men take women buggy riding",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0297",
+ "category": "phonetic_balance",
+ "text": "But it contributed to the smash.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0297.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "BUT IT CONTRIBUTED TO THE SMASH",
+ "raw": {
+ "normalized_reference": "but it contributed to the smash.",
+ "normalized_hypothesis": "but it contributed to the smash",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "but it contributed to the smash",
+ "normalized_hypothesis": "but it contributed to the smash",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0298",
+ "category": "phonetic_balance",
+ "text": "No man ate of the seal meat or the oil.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0298.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "NO MAN ATE OF THE SEAL MEAT OR THE OIL",
+ "raw": {
+ "normalized_reference": "no man ate of the seal meat or the oil.",
+ "normalized_hypothesis": "no man ate of the seal meat or the oil",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "no man ate of the seal meat or the oil",
+ "normalized_hypothesis": "no man ate of the seal meat or the oil",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0299",
+ "category": "phonetic_balance",
+ "text": "You're going in for grab sharing.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0299.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "YOU'RE GOING IN FOR GRABB SHARING",
+ "raw": {
+ "normalized_reference": "you're going in for grab sharing.",
+ "normalized_hypothesis": "you're going in for grabb sharing",
+ "errors": 1,
+ "words": 6,
+ "wer": 0.16666666666666666
+ },
+ "semantic": {
+ "normalized_reference": "you are going in for grab sharing",
+ "normalized_hypothesis": "you are going in for grabb sharing",
+ "errors": 1,
+ "words": 7,
+ "wer": 0.14285714285714285
+ }
+ },
+ {
+ "id": "release500_0300",
+ "category": "phonetic_balance",
+ "text": "He it was that lived to found the family of the Patino.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0300.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE IT WAS THAT LIVED TO FOUND THE FAMILY OF THE PATINO",
+ "raw": {
+ "normalized_reference": "he it was that lived to found the family of the patino.",
+ "normalized_hypothesis": "he it was that lived to found the family of the patino",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he it was that lived to found the family of the patino",
+ "normalized_hypothesis": "he it was that lived to found the family of the patino",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0301",
+ "category": "phonetic_balance",
+ "text": "For such countries nothing remained but reorganization.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0301.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "FOR SUCH COUNTRIES NOTHING REMAINED BUT REORGANIZATION",
+ "raw": {
+ "normalized_reference": "for such countries nothing remained but reorganization.",
+ "normalized_hypothesis": "for such countries nothing remained but reorganization",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "for such countries nothing remained but reorganization",
+ "normalized_hypothesis": "for such countries nothing remained but reorganization",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0302",
+ "category": "phonetic_balance",
+ "text": "Stand off butcher and baker and all the rest.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0302.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "STAND OFF BUTCHER AND BAKER AND ALL THE REST",
+ "raw": {
+ "normalized_reference": "stand off butcher and baker and all the rest.",
+ "normalized_hypothesis": "stand off butcher and baker and all the rest",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "stand off butcher and baker and all the rest",
+ "normalized_hypothesis": "stand off butcher and baker and all the rest",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0303",
+ "category": "phonetic_balance",
+ "text": "All eyes, however, were staring at him in certitude of expectancy.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0303.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "ALL EYES HOWEVER WERE STARING AT HIM IN CERTITUDE OF EXPECTANCY",
+ "raw": {
+ "normalized_reference": "all eyes, however, were staring at him in certitude of expectancy.",
+ "normalized_hypothesis": "all eyes however were staring at him in certitude of expectancy",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "all eyes however were staring at him in certitude of expectancy",
+ "normalized_hypothesis": "all eyes however were staring at him in certitude of expectancy",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0304",
+ "category": "phonetic_balance",
+ "text": "Two days ago Jeanne learned where her father's men were hiding.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0304.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "TWO DAYS AGO JEAN LEARNED WHERE HER FATHER'S MEN WERE HIDING",
+ "raw": {
+ "normalized_reference": "two days ago jeanne learned where her father's men were hiding.",
+ "normalized_hypothesis": "two days ago jean learned where her father's men were hiding",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ },
+ "semantic": {
+ "normalized_reference": "2 days ago jeanne learned where her father is men were hiding",
+ "normalized_hypothesis": "2 days ago jean learned where her father is men were hiding",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ }
+ },
+ {
+ "id": "release500_0305",
+ "category": "phonetic_balance",
+ "text": "His previous wives had never lived long enough to bear him children.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0305.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HIS PREVIOUS WIVES HAD NEVER LIVED LONG ENOUGH TO BEAR HIM CHILDREN",
+ "raw": {
+ "normalized_reference": "his previous wives had never lived long enough to bear him children.",
+ "normalized_hypothesis": "his previous wives had never lived long enough to bear him children",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "his previous wives had never lived long enough to bear him children",
+ "normalized_hypothesis": "his previous wives had never lived long enough to bear him children",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0306",
+ "category": "phonetic_balance",
+ "text": "But this little defect did not worry him.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0306.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "BUT THIS LITTLE DEFECT DID NOT WORRY HIM",
+ "raw": {
+ "normalized_reference": "but this little defect did not worry him.",
+ "normalized_hypothesis": "but this little defect did not worry him",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "but this little defect did not worry him",
+ "normalized_hypothesis": "but this little defect did not worry him",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0307",
+ "category": "phonetic_balance",
+ "text": "Eli Harding asked, as Shunk started to follow.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0307.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "BILLY HARDING ASKED AS SHUNK STARTED TO FOLLOW",
+ "raw": {
+ "normalized_reference": "eli harding asked, as shunk started to follow.",
+ "normalized_hypothesis": "billy harding asked as shunk started to follow",
+ "errors": 1,
+ "words": 8,
+ "wer": 0.125
+ },
+ "semantic": {
+ "normalized_reference": "eli harding asked as shunk started to follow",
+ "normalized_hypothesis": "billy harding asked as shunk started to follow",
+ "errors": 1,
+ "words": 8,
+ "wer": 0.125
+ }
+ },
+ {
+ "id": "release500_0308",
+ "category": "phonetic_balance",
+ "text": "And this was their sole conversation throughout the meal.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0308.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "AND THIS WAS THEIR SOLE CONVERSATION THROUGHOUT THE MEAL",
+ "raw": {
+ "normalized_reference": "and this was their sole conversation throughout the meal.",
+ "normalized_hypothesis": "and this was their sole conversation throughout the meal",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "and this was their sole conversation throughout the meal",
+ "normalized_hypothesis": "and this was their sole conversation throughout the meal",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0309",
+ "category": "phonetic_balance",
+ "text": "Then, again, Tudor had such an irritating way about him.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0309.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THEN AGAIN CUDER HAD SUCH AN IRRITATING WAY ABOUT HIM",
+ "raw": {
+ "normalized_reference": "then, again, tudor had such an irritating way about him.",
+ "normalized_hypothesis": "then again cuder had such an irritating way about him",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ },
+ "semantic": {
+ "normalized_reference": "then again tudor had such an irritating way about him",
+ "normalized_hypothesis": "then again cuder had such an irritating way about him",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ }
+ },
+ {
+ "id": "release500_0310",
+ "category": "phonetic_balance",
+ "text": "In it there was something that was almost tragedy.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0310.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IN IT THERE WAS SOMETHING THAT WAS ALMOST TRAGEDY",
+ "raw": {
+ "normalized_reference": "in it there was something that was almost tragedy.",
+ "normalized_hypothesis": "in it there was something that was almost tragedy",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "in it there was something that was almost tragedy",
+ "normalized_hypothesis": "in it there was something that was almost tragedy",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0311",
+ "category": "phonetic_balance",
+ "text": "Thirty pounds, said the captain with finality.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0311.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THIRTY POUNDS SAID THE CAPTAIN WITH FINALITY",
+ "raw": {
+ "normalized_reference": "thirty pounds, said the captain with finality.",
+ "normalized_hypothesis": "thirty pounds said the captain with finality",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "\u00a330 said the captain with finality",
+ "normalized_hypothesis": "\u00a330 said the captain with finality",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0312",
+ "category": "phonetic_balance",
+ "text": "It was over when he made his way through the ring of spectators.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0312.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT WAS OVER WHEN HE MADE HIS WAY THROUGH THE RING OF SPECTATORS",
+ "raw": {
+ "normalized_reference": "it was over when he made his way through the ring of spectators.",
+ "normalized_hypothesis": "it was over when he made his way through the ring of spectators",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it was over when he made his way through the ring of spectators",
+ "normalized_hypothesis": "it was over when he made his way through the ring of spectators",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0313",
+ "category": "phonetic_balance",
+ "text": "She had died from cold and starvation.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0313.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SHE HAD DIED FROM COLD AND STARVATION",
+ "raw": {
+ "normalized_reference": "she had died from cold and starvation.",
+ "normalized_hypothesis": "she had died from cold and starvation",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "she had died from cold and starvation",
+ "normalized_hypothesis": "she had died from cold and starvation",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0314",
+ "category": "phonetic_balance",
+ "text": "We fished sharks on Niihau together.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0314.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WE FISHED SHARKS ON NIHOW TOGETHER",
+ "raw": {
+ "normalized_reference": "we fished sharks on niihau together.",
+ "normalized_hypothesis": "we fished sharks on nihow together",
+ "errors": 1,
+ "words": 6,
+ "wer": 0.16666666666666666
+ },
+ "semantic": {
+ "normalized_reference": "we fished sharks on niihau together",
+ "normalized_hypothesis": "we fished sharks on nihow together",
+ "errors": 1,
+ "words": 6,
+ "wer": 0.16666666666666666
+ }
+ },
+ {
+ "id": "release500_0315",
+ "category": "phonetic_balance",
+ "text": "His slim hands gripped the edges of the table.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0315.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HIS SLIM HANDS GRIPPED THE EDGES OF THE TABLE",
+ "raw": {
+ "normalized_reference": "his slim hands gripped the edges of the table.",
+ "normalized_hypothesis": "his slim hands gripped the edges of the table",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "his slim hands gripped the edges of the table",
+ "normalized_hypothesis": "his slim hands gripped the edges of the table",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0316",
+ "category": "phonetic_balance",
+ "text": "We had been chased by them ourselves, more than once.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0316.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WE HAD BEEN CHASED BY THEM OURSELVES MORE THAN ONCE",
+ "raw": {
+ "normalized_reference": "we had been chased by them ourselves, more than once.",
+ "normalized_hypothesis": "we had been chased by them ourselves more than once",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "we had been chased by them ourselves more than once",
+ "normalized_hypothesis": "we had been chased by them ourselves more than once",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0317",
+ "category": "phonetic_balance",
+ "text": "Or have they already devised one.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0317.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "OR HAVE THEY ALREADY DEVISED ONE",
+ "raw": {
+ "normalized_reference": "or have they already devised one.",
+ "normalized_hypothesis": "or have they already devised one",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "or have they already devised one",
+ "normalized_hypothesis": "or have they already devised one",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0318",
+ "category": "phonetic_balance",
+ "text": "I cannot follow you, she said.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0318.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I CANNOT FOLLOW YOU SHE SAID",
+ "raw": {
+ "normalized_reference": "i cannot follow you, she said.",
+ "normalized_hypothesis": "i cannot follow you she said",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i cannot follow you she said",
+ "normalized_hypothesis": "i cannot follow you she said",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0319",
+ "category": "phonetic_balance",
+ "text": "He went down in midstream, searching the shadows of both shores.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0319.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE WENT DOWN IN MID STREAM SEARCHING THE SHADOWS OF BOTH SHORES",
+ "raw": {
+ "normalized_reference": "he went down in midstream, searching the shadows of both shores.",
+ "normalized_hypothesis": "he went down in mid stream searching the shadows of both shores",
+ "errors": 2,
+ "words": 11,
+ "wer": 0.18181818181818182
+ },
+ "semantic": {
+ "normalized_reference": "he went down in midstream searching the shadows of both shores",
+ "normalized_hypothesis": "he went down in mid stream searching the shadows of both shores",
+ "errors": 2,
+ "words": 11,
+ "wer": 0.18181818181818182
+ }
+ },
+ {
+ "id": "release500_0320",
+ "category": "phonetic_balance",
+ "text": "I was sick once -- typhoid.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0320.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I WAS SICK ONCE TYPHED",
+ "raw": {
+ "normalized_reference": "i was sick once -- typhoid.",
+ "normalized_hypothesis": "i was sick once typhed",
+ "errors": 1,
+ "words": 5,
+ "wer": 0.2
+ },
+ "semantic": {
+ "normalized_reference": "i was sick once typhoid",
+ "normalized_hypothesis": "i was sick once typhed",
+ "errors": 1,
+ "words": 5,
+ "wer": 0.2
+ }
+ },
+ {
+ "id": "release500_0321",
+ "category": "phonetic_balance",
+ "text": "Why not like any railroad station or ferry depot.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0321.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WHY NOT LIKE ANY RAILROAD STATION OR FERRY DEPOT",
+ "raw": {
+ "normalized_reference": "why not like any railroad station or ferry depot.",
+ "normalized_hypothesis": "why not like any railroad station or ferry depot",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "why not like any railroad station or ferry depot",
+ "normalized_hypothesis": "why not like any railroad station or ferry depot",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0322",
+ "category": "phonetic_balance",
+ "text": "Red-Eye swung back and forth on the branch farther down.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0322.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "RED EYE SWUNG BACK AND FORTH ON THE BRANCH FARTHER DOWN",
+ "raw": {
+ "normalized_reference": "red-eye swung back and forth on the branch farther down.",
+ "normalized_hypothesis": "red eye swung back and forth on the branch farther down",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "red eye swung back and forth on the branch farther down",
+ "normalized_hypothesis": "red eye swung back and forth on the branch farther down",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0323",
+ "category": "phonetic_balance",
+ "text": "Within himself he called it no longer his own.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0323.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WITHIN HIMSELF HE CALLED IT NO LONGER HIS OWN",
+ "raw": {
+ "normalized_reference": "within himself he called it no longer his own.",
+ "normalized_hypothesis": "within himself he called it no longer his own",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "within himself he called it no longer his own",
+ "normalized_hypothesis": "within himself he called it no longer his own",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0324",
+ "category": "phonetic_balance",
+ "text": "It lasted as a deterrent for two days.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0324.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT LASTED AS A DETERRANT FOR TWO DAYS",
+ "raw": {
+ "normalized_reference": "it lasted as a deterrent for two days.",
+ "normalized_hypothesis": "it lasted as a deterrant for two days",
+ "errors": 1,
+ "words": 8,
+ "wer": 0.125
+ },
+ "semantic": {
+ "normalized_reference": "it lasted as a deterrent for 2 days",
+ "normalized_hypothesis": "it lasted as a deterrant for 2 days",
+ "errors": 1,
+ "words": 8,
+ "wer": 0.125
+ }
+ },
+ {
+ "id": "release500_0325",
+ "category": "phonetic_balance",
+ "text": "He loved to play Chinese lottery.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0325.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE LOVED TO PLAY CHINESE LOTTERY",
+ "raw": {
+ "normalized_reference": "he loved to play chinese lottery.",
+ "normalized_hypothesis": "he loved to play chinese lottery",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he loved to play chinese lottery",
+ "normalized_hypothesis": "he loved to play chinese lottery",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0326",
+ "category": "phonetic_balance",
+ "text": "Lord, but I'm glad to see you again, Phil.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0326.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "LORD BUT I'M GLAD TO SEE YOU AGAIN PHIL",
+ "raw": {
+ "normalized_reference": "lord, but i'm glad to see you again, phil.",
+ "normalized_hypothesis": "lord but i'm glad to see you again phil",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "lord but i am glad to see you again phil",
+ "normalized_hypothesis": "lord but i am glad to see you again phil",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0327",
+ "category": "phonetic_balance",
+ "text": "There is not an iota of truth in it, certainly not.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0327.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THERE IS NOT AN IOTA OF TRUTH IN IT CERTAINLY NOT",
+ "raw": {
+ "normalized_reference": "there is not an iota of truth in it, certainly not.",
+ "normalized_hypothesis": "there is not an iota of truth in it certainly not",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "there is not an iota of truth in it certainly not",
+ "normalized_hypothesis": "there is not an iota of truth in it certainly not",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0328",
+ "category": "phonetic_balance",
+ "text": "She added, with genuine sympathy in her face and voice.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0328.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SHE ADDED WITH GENUINE SYMPATHY IN HER FACE AND VOICE",
+ "raw": {
+ "normalized_reference": "she added, with genuine sympathy in her face and voice.",
+ "normalized_hypothesis": "she added with genuine sympathy in her face and voice",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "she added with genuine sympathy in her face and voice",
+ "normalized_hypothesis": "she added with genuine sympathy in her face and voice",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0329",
+ "category": "phonetic_balance",
+ "text": "We have plenty of capital ourselves, and yet we want more.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0329.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WE HAVE PLENTY OF CAPITAL OURSELVES AND YET WE WANT MORE",
+ "raw": {
+ "normalized_reference": "we have plenty of capital ourselves, and yet we want more.",
+ "normalized_hypothesis": "we have plenty of capital ourselves and yet we want more",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "we have plenty of capital ourselves and yet we want more",
+ "normalized_hypothesis": "we have plenty of capital ourselves and yet we want more",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0330",
+ "category": "phonetic_balance",
+ "text": "And now put yourself in my place for a moment.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0330.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "AND NOW PUT YOURSELF IN MY PLACE FOR A MOMENT",
+ "raw": {
+ "normalized_reference": "and now put yourself in my place for a moment.",
+ "normalized_hypothesis": "and now put yourself in my place for a moment",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "and now put yourself in my place for a moment",
+ "normalized_hypothesis": "and now put yourself in my place for a moment",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0331",
+ "category": "phonetic_balance",
+ "text": "Fast, but endure.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0331.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "FAST BUT ENDURE",
+ "raw": {
+ "normalized_reference": "fast, but endure.",
+ "normalized_hypothesis": "fast but endure",
+ "errors": 0,
+ "words": 3,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "fast but endure",
+ "normalized_hypothesis": "fast but endure",
+ "errors": 0,
+ "words": 3,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0332",
+ "category": "phonetic_balance",
+ "text": "You were looking squeamish this afternoon, he began.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0332.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "YOU WERE LOOKING SQUEAMISH THIS AFTERNOON HE BEGAN",
+ "raw": {
+ "normalized_reference": "you were looking squeamish this afternoon, he began.",
+ "normalized_hypothesis": "you were looking squeamish this afternoon he began",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "you were looking squeamish this afternoon he began",
+ "normalized_hypothesis": "you were looking squeamish this afternoon he began",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0333",
+ "category": "phonetic_balance",
+ "text": "It is the nearest refuge.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0333.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT IS THE NEAREST REFUGE",
+ "raw": {
+ "normalized_reference": "it is the nearest refuge.",
+ "normalized_hypothesis": "it is the nearest refuge",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it is the nearest refuge",
+ "normalized_hypothesis": "it is the nearest refuge",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0334",
+ "category": "phonetic_balance",
+ "text": "I want to die in it.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0334.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I WANT TO DIE IN IT",
+ "raw": {
+ "normalized_reference": "i want to die in it.",
+ "normalized_hypothesis": "i want to die in it",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i want to die in it",
+ "normalized_hypothesis": "i want to die in it",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0335",
+ "category": "phonetic_balance",
+ "text": "He stopped, and Philip nodded at the horrified question in his eyes.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0335.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE STOPPED AND PHILIP NODDED AT THE HORRIFIED QUESTION IN HIS EYES",
+ "raw": {
+ "normalized_reference": "he stopped, and philip nodded at the horrified question in his eyes.",
+ "normalized_hypothesis": "he stopped and philip nodded at the horrified question in his eyes",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he stopped and philip nodded at the horrified question in his eyes",
+ "normalized_hypothesis": "he stopped and philip nodded at the horrified question in his eyes",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0336",
+ "category": "phonetic_balance",
+ "text": "She turned, fearing that Jacques might see what was in her face.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0336.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SHE TURNED FEARING THAT SHACK MIGHT SEE WHAT WAS IN HER FACE",
+ "raw": {
+ "normalized_reference": "she turned, fearing that jacques might see what was in her face.",
+ "normalized_hypothesis": "she turned fearing that shack might see what was in her face",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "she turned fearing that jacques might see what was in her face",
+ "normalized_hypothesis": "she turned fearing that shack might see what was in her face",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ }
+ },
+ {
+ "id": "release500_0337",
+ "category": "phonetic_balance",
+ "text": "It was a curious coincidence.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0337.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT WAS A CURIOUS COINCIDENCE",
+ "raw": {
+ "normalized_reference": "it was a curious coincidence.",
+ "normalized_hypothesis": "it was a curious coincidence",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it was a curious coincidence",
+ "normalized_hypothesis": "it was a curious coincidence",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0338",
+ "category": "phonetic_balance",
+ "text": "Daylight made no answer, and the door closed behind him.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0338.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "DAYLIGHT MADE NO ANSWER AND THE DOOR CLOSED BEHIND HIM",
+ "raw": {
+ "normalized_reference": "daylight made no answer, and the door closed behind him.",
+ "normalized_hypothesis": "daylight made no answer and the door closed behind him",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "daylight made no answer and the door closed behind him",
+ "normalized_hypothesis": "daylight made no answer and the door closed behind him",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0339",
+ "category": "phonetic_balance",
+ "text": "And he did hurt my arm.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0339.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "AND HE DID HURT MY ARM",
+ "raw": {
+ "normalized_reference": "and he did hurt my arm.",
+ "normalized_hypothesis": "and he did hurt my arm",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "and he did hurt my arm",
+ "normalized_hypothesis": "and he did hurt my arm",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0340",
+ "category": "phonetic_balance",
+ "text": "Dunham, can your boy go along with Jesse.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0340.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "DUNHAM CAN YOUR BOY GO ALONG WITH JESSE",
+ "raw": {
+ "normalized_reference": "dunham, can your boy go along with jesse.",
+ "normalized_hypothesis": "dunham can your boy go along with jesse",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "dunham can your boy go along with jesse",
+ "normalized_hypothesis": "dunham can your boy go along with jesse",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0341",
+ "category": "phonetic_balance",
+ "text": "The creative joy, I murmured.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0341.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "CREATIVE JOY I MURMURED",
+ "raw": {
+ "normalized_reference": "the creative joy, i murmured.",
+ "normalized_hypothesis": "creative joy i murmured",
+ "errors": 1,
+ "words": 5,
+ "wer": 0.2
+ },
+ "semantic": {
+ "normalized_reference": "the creative joy i murmured",
+ "normalized_hypothesis": "creative joy i murmured",
+ "errors": 1,
+ "words": 5,
+ "wer": 0.2
+ }
+ },
+ {
+ "id": "release500_0342",
+ "category": "phonetic_balance",
+ "text": "He pulled, and the log crashed down to break his back.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0342.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE PULLED AND THE LOG CRASHED DOWN TO BREAK HIS BACK",
+ "raw": {
+ "normalized_reference": "he pulled, and the log crashed down to break his back.",
+ "normalized_hypothesis": "he pulled and the log crashed down to break his back",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he pulled and the log crashed down to break his back",
+ "normalized_hypothesis": "he pulled and the log crashed down to break his back",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0343",
+ "category": "phonetic_balance",
+ "text": "By golly, the boy wins.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0343.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "BY GOLLY THE BOY WINS",
+ "raw": {
+ "normalized_reference": "by golly, the boy wins.",
+ "normalized_hypothesis": "by golly the boy wins",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "by golly the boy wins",
+ "normalized_hypothesis": "by golly the boy wins",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0344",
+ "category": "phonetic_balance",
+ "text": "The boy threw back his head with pride.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0344.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE BOY THREW BACK HIS HEAD WITH PRIDE",
+ "raw": {
+ "normalized_reference": "the boy threw back his head with pride.",
+ "normalized_hypothesis": "the boy threw back his head with pride",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the boy threw back his head with pride",
+ "normalized_hypothesis": "the boy threw back his head with pride",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0345",
+ "category": "phonetic_balance",
+ "text": "The last one I knew was an overseer.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0345.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE LAST ONE I KNEW WAS AN OVERSEER",
+ "raw": {
+ "normalized_reference": "the last one i knew was an overseer.",
+ "normalized_hypothesis": "the last one i knew was an overseer",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the last one i knew was an overseer",
+ "normalized_hypothesis": "the last one i knew was an overseer",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0346",
+ "category": "phonetic_balance",
+ "text": "Two years ago I gave up civilization for this.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0346.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "TWO YEARS AGO I GAVE UP CIVILIZATION FOR THIS",
+ "raw": {
+ "normalized_reference": "two years ago i gave up civilization for this.",
+ "normalized_hypothesis": "two years ago i gave up civilization for this",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "2 years ago i gave up civilization for this",
+ "normalized_hypothesis": "2 years ago i gave up civilization for this",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0347",
+ "category": "phonetic_balance",
+ "text": "The river bared its bosom, and snorting steamboats challenged the wilderness.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0347.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE RIVER BARED ITS BOSOM AND SNORTING STEAMBOATS CHALLENGED THE WILDERNESS",
+ "raw": {
+ "normalized_reference": "the river bared its bosom, and snorting steamboats challenged the wilderness.",
+ "normalized_hypothesis": "the river bared its bosom and snorting steamboats challenged the wilderness",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the river bared its bosom and snorting steamboats challenged the wilderness",
+ "normalized_hypothesis": "the river bared its bosom and snorting steamboats challenged the wilderness",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0348",
+ "category": "phonetic_balance",
+ "text": "Mab, she said.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0348.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "BAB SHE SAID",
+ "raw": {
+ "normalized_reference": "mab, she said.",
+ "normalized_hypothesis": "bab she said",
+ "errors": 1,
+ "words": 3,
+ "wer": 0.3333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "mab she said",
+ "normalized_hypothesis": "bab she said",
+ "errors": 1,
+ "words": 3,
+ "wer": 0.3333333333333333
+ }
+ },
+ {
+ "id": "release500_0349",
+ "category": "phonetic_balance",
+ "text": "The regret in his voice was provocative of a second burst of laughter.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0349.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE REGRET IN HIS VOICE WAS PROVOCATIVE OF A SECOND BURST OF LAUGHTER",
+ "raw": {
+ "normalized_reference": "the regret in his voice was provocative of a second burst of laughter.",
+ "normalized_hypothesis": "the regret in his voice was provocative of a second burst of laughter",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the regret in his voice was provocative of a 2nd burst of laughter",
+ "normalized_hypothesis": "the regret in his voice was provocative of a 2nd burst of laughter",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0350",
+ "category": "phonetic_balance",
+ "text": "And each year something happened, and I did not go.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0350.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "AND EACH YEAR SOMETHING HAPPENED AND I DID NOT GO",
+ "raw": {
+ "normalized_reference": "and each year something happened, and i did not go.",
+ "normalized_hypothesis": "and each year something happened and i did not go",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "and each year something happened and i did not go",
+ "normalized_hypothesis": "and each year something happened and i did not go",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0351",
+ "category": "phonetic_balance",
+ "text": "Gad, your letter came just in time.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0351.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "GAD YOUR LETTER CAME JUST IN TIME",
+ "raw": {
+ "normalized_reference": "gad, your letter came just in time.",
+ "normalized_hypothesis": "gad your letter came just in time",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "gad your letter came just in time",
+ "normalized_hypothesis": "gad your letter came just in time",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0352",
+ "category": "phonetic_balance",
+ "text": "But if Pierre did not return, until tomorrow.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0352.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "BUT IF PIERRE DID NOT RETURN UNTIL TO MORROW",
+ "raw": {
+ "normalized_reference": "but if pierre did not return, until tomorrow.",
+ "normalized_hypothesis": "but if pierre did not return until to morrow",
+ "errors": 2,
+ "words": 8,
+ "wer": 0.25
+ },
+ "semantic": {
+ "normalized_reference": "but if pierre did not return until tomorrow",
+ "normalized_hypothesis": "but if pierre did not return until to morrow",
+ "errors": 2,
+ "words": 8,
+ "wer": 0.25
+ }
+ },
+ {
+ "id": "release500_0353",
+ "category": "phonetic_balance",
+ "text": "Billinger may arrive in time.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0353.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "BILINGE MAY ARRIVE IN TIME",
+ "raw": {
+ "normalized_reference": "billinger may arrive in time.",
+ "normalized_hypothesis": "bilinge may arrive in time",
+ "errors": 1,
+ "words": 5,
+ "wer": 0.2
+ },
+ "semantic": {
+ "normalized_reference": "billinger may arrive in time",
+ "normalized_hypothesis": "bilinge may arrive in time",
+ "errors": 1,
+ "words": 5,
+ "wer": 0.2
+ }
+ },
+ {
+ "id": "release500_0354",
+ "category": "phonetic_balance",
+ "text": "In such a tumbling of values was no time to sell.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0354.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IN SUCH A TUMBLING OF VALUES WAS NO TIME TO SELL",
+ "raw": {
+ "normalized_reference": "in such a tumbling of values was no time to sell.",
+ "normalized_hypothesis": "in such a tumbling of values was no time to sell",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "in such a tumbling of values was no time to sell",
+ "normalized_hypothesis": "in such a tumbling of values was no time to sell",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0355",
+ "category": "phonetic_balance",
+ "text": "There's not an iota of truth in it.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0355.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THERE'S NOT A NIOTA OF TRUTH IN IT",
+ "raw": {
+ "normalized_reference": "there's not an iota of truth in it.",
+ "normalized_hypothesis": "there's not a niota of truth in it",
+ "errors": 2,
+ "words": 8,
+ "wer": 0.25
+ },
+ "semantic": {
+ "normalized_reference": "there is not an iota of truth in it",
+ "normalized_hypothesis": "there is not a niota of truth in it",
+ "errors": 2,
+ "words": 9,
+ "wer": 0.2222222222222222
+ }
+ },
+ {
+ "id": "release500_0356",
+ "category": "phonetic_balance",
+ "text": "It is dog eat dog, and you ate them up.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0356.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT IS DOG EAT DOG AND YOU ATE THEM UP",
+ "raw": {
+ "normalized_reference": "it is dog eat dog, and you ate them up.",
+ "normalized_hypothesis": "it is dog eat dog and you ate them up",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it is dog eat dog and you ate them up",
+ "normalized_hypothesis": "it is dog eat dog and you ate them up",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0357",
+ "category": "phonetic_balance",
+ "text": "You can take a vacation on pay.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0357.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "YOU CAN TAKE A VACATION ON PAY",
+ "raw": {
+ "normalized_reference": "you can take a vacation on pay.",
+ "normalized_hypothesis": "you can take a vacation on pay",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "you can take a vacation on pay",
+ "normalized_hypothesis": "you can take a vacation on pay",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0358",
+ "category": "phonetic_balance",
+ "text": "I have seen myself that one man contemplated by Pascal's philosophic eye.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0358.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I HAVE SEEN MYSELF THAT ONE MAN CONTEMPLATED BY PASCOLL'S PHILOSOPHIC EYE",
+ "raw": {
+ "normalized_reference": "i have seen myself that one man contemplated by pascal's philosophic eye.",
+ "normalized_hypothesis": "i have seen myself that one man contemplated by pascoll's philosophic eye",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "i have seen myself that one man contemplated by pascal is philosophic eye",
+ "normalized_hypothesis": "i have seen myself that one man contemplated by pascoll is philosophic eye",
+ "errors": 1,
+ "words": 13,
+ "wer": 0.07692307692307693
+ }
+ },
+ {
+ "id": "release500_0359",
+ "category": "phonetic_balance",
+ "text": "Ah, I had forgotten, he exclaimed.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0359.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "AH I HAD FORGOTTEN HE EXCLAIMED",
+ "raw": {
+ "normalized_reference": "ah, i had forgotten, he exclaimed.",
+ "normalized_hypothesis": "ah i had forgotten he exclaimed",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "ah i had forgotten he exclaimed",
+ "normalized_hypothesis": "ah i had forgotten he exclaimed",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0360",
+ "category": "phonetic_balance",
+ "text": "Providence had delivered him through the maelstrom.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0360.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "PROVIDENCE HAD DELIVERED HIM THROUGH THE MAILSTROM",
+ "raw": {
+ "normalized_reference": "providence had delivered him through the maelstrom.",
+ "normalized_hypothesis": "providence had delivered him through the mailstrom",
+ "errors": 1,
+ "words": 7,
+ "wer": 0.14285714285714285
+ },
+ "semantic": {
+ "normalized_reference": "providence had delivered him through the maelstrom",
+ "normalized_hypothesis": "providence had delivered him through the mailstrom",
+ "errors": 1,
+ "words": 7,
+ "wer": 0.14285714285714285
+ }
+ },
+ {
+ "id": "release500_0361",
+ "category": "phonetic_balance",
+ "text": "She was his now, forever.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0361.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SHE WAS HIS NOW FOR EVER",
+ "raw": {
+ "normalized_reference": "she was his now, forever.",
+ "normalized_hypothesis": "she was his now for ever",
+ "errors": 2,
+ "words": 5,
+ "wer": 0.4
+ },
+ "semantic": {
+ "normalized_reference": "she was his now forever",
+ "normalized_hypothesis": "she was his now for ever",
+ "errors": 2,
+ "words": 5,
+ "wer": 0.4
+ }
+ },
+ {
+ "id": "release500_0362",
+ "category": "phonetic_balance",
+ "text": "It was unobtrusive, yet it was there.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0362.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT WAS UNOBTRUSIVE YET IT WAS THERE",
+ "raw": {
+ "normalized_reference": "it was unobtrusive, yet it was there.",
+ "normalized_hypothesis": "it was unobtrusive yet it was there",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it was unobtrusive yet it was there",
+ "normalized_hypothesis": "it was unobtrusive yet it was there",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0363",
+ "category": "phonetic_balance",
+ "text": "Men like Joe Goose dated existence from drunk to drunk.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0363.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "MEN LIKE JOE GOOSE DATED EXISTENCE FROM DRUNK TO DRUNK",
+ "raw": {
+ "normalized_reference": "men like joe goose dated existence from drunk to drunk.",
+ "normalized_hypothesis": "men like joe goose dated existence from drunk to drunk",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "men like joe goose dated existence from drunk to drunk",
+ "normalized_hypothesis": "men like joe goose dated existence from drunk to drunk",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0364",
+ "category": "phonetic_balance",
+ "text": "I was completely lost in my work.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0364.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I WAS COMPLETELY LOST IN MY WORK",
+ "raw": {
+ "normalized_reference": "i was completely lost in my work.",
+ "normalized_hypothesis": "i was completely lost in my work",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i was completely lost in my work",
+ "normalized_hypothesis": "i was completely lost in my work",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0365",
+ "category": "phonetic_balance",
+ "text": "The very thing, Ernest agreed.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0365.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "EVERY THING ERNEST AGREED",
+ "raw": {
+ "normalized_reference": "the very thing, ernest agreed.",
+ "normalized_hypothesis": "every thing ernest agreed",
+ "errors": 2,
+ "words": 5,
+ "wer": 0.4
+ },
+ "semantic": {
+ "normalized_reference": "the very thing ernest agreed",
+ "normalized_hypothesis": "every thing ernest agreed",
+ "errors": 2,
+ "words": 5,
+ "wer": 0.4
+ }
+ },
+ {
+ "id": "release500_0366",
+ "category": "phonetic_balance",
+ "text": "He was the leader, and Tudor was his lieutenant.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0366.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE WAS THE LEADER AND TUDOR WAS HIS LIEUTENANT",
+ "raw": {
+ "normalized_reference": "he was the leader, and tudor was his lieutenant.",
+ "normalized_hypothesis": "he was the leader and tudor was his lieutenant",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he was the leader and tudor was his lieutenant",
+ "normalized_hypothesis": "he was the leader and tudor was his lieutenant",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0367",
+ "category": "phonetic_balance",
+ "text": "The Fire-Men wore animal skins around their waists and across their shoulders.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0367.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE FIREMEN WORE ANIMAL SKINS AROUND THEIR WAISTS AND ACROSS THEIR SHOULDERS",
+ "raw": {
+ "normalized_reference": "the fire-men wore animal skins around their waists and across their shoulders.",
+ "normalized_hypothesis": "the firemen wore animal skins around their waists and across their shoulders",
+ "errors": 2,
+ "words": 13,
+ "wer": 0.15384615384615385
+ },
+ "semantic": {
+ "normalized_reference": "the fire men wore animal skins around their waists and across their shoulders",
+ "normalized_hypothesis": "the firemen wore animal skins around their waists and across their shoulders",
+ "errors": 2,
+ "words": 13,
+ "wer": 0.15384615384615385
+ }
+ },
+ {
+ "id": "release500_0368",
+ "category": "phonetic_balance",
+ "text": "We could throw stones with our feet.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0368.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WE COULD THROW STONES WITH OUR FEET",
+ "raw": {
+ "normalized_reference": "we could throw stones with our feet.",
+ "normalized_hypothesis": "we could throw stones with our feet",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "we could throw stones with our feet",
+ "normalized_hypothesis": "we could throw stones with our feet",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0369",
+ "category": "phonetic_balance",
+ "text": "He turned sharply, and faced Gregson across the table.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0369.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE TURNED SHARPLY AND FACED GREGSON ACROSS THE TABLE",
+ "raw": {
+ "normalized_reference": "he turned sharply, and faced gregson across the table.",
+ "normalized_hypothesis": "he turned sharply and faced gregson across the table",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he turned sharply and faced gregson across the table",
+ "normalized_hypothesis": "he turned sharply and faced gregson across the table",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0370",
+ "category": "phonetic_balance",
+ "text": "They are to attack your camp tomorrow night.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0370.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THEY ARE TO ATTACK YOUR CAMP TO MORROW NIGHT",
+ "raw": {
+ "normalized_reference": "they are to attack your camp tomorrow night.",
+ "normalized_hypothesis": "they are to attack your camp to morrow night",
+ "errors": 2,
+ "words": 8,
+ "wer": 0.25
+ },
+ "semantic": {
+ "normalized_reference": "they are to attack your camp tomorrow night",
+ "normalized_hypothesis": "they are to attack your camp to morrow night",
+ "errors": 2,
+ "words": 8,
+ "wer": 0.25
+ }
+ },
+ {
+ "id": "release500_0371",
+ "category": "phonetic_balance",
+ "text": "And watch out for wet feet, was his parting advice.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0371.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "AND WATCH OUT FOR WED FEET WAS HIS PARTING ADVICE",
+ "raw": {
+ "normalized_reference": "and watch out for wet feet, was his parting advice.",
+ "normalized_hypothesis": "and watch out for wed feet was his parting advice",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ },
+ "semantic": {
+ "normalized_reference": "and watch out for wet feet was his parting advice",
+ "normalized_hypothesis": "and watch out for wed feet was his parting advice",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ }
+ },
+ {
+ "id": "release500_0372",
+ "category": "phonetic_balance",
+ "text": "Two weeks passed, and in that time Thorpe left camp three times.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0372.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "TWO WEEKS PASSED AND IN THAT TIME FORKE LEFT CAMP THREE TIMES",
+ "raw": {
+ "normalized_reference": "two weeks passed, and in that time thorpe left camp three times.",
+ "normalized_hypothesis": "two weeks passed and in that time forke left camp three times",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "2 weeks passed and in that time thorpe left camp 3 times",
+ "normalized_hypothesis": "2 weeks passed and in that time forke left camp 3 times",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ }
+ },
+ {
+ "id": "release500_0373",
+ "category": "phonetic_balance",
+ "text": "But ever his gaze returned to that Crouched Venus on the piano.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0373.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "BUT EVER HIS GAZE RETURNED TO THEIR CROUCHED VENUS ON THE PIANO",
+ "raw": {
+ "normalized_reference": "but ever his gaze returned to that crouched venus on the piano.",
+ "normalized_hypothesis": "but ever his gaze returned to their crouched venus on the piano",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "but ever his gaze returned to that crouched venus on the piano",
+ "normalized_hypothesis": "but ever his gaze returned to their crouched venus on the piano",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ }
+ },
+ {
+ "id": "release500_0374",
+ "category": "phonetic_balance",
+ "text": "But she had become an automaton.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0374.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "BUT SHE HAD BECOME AN AUTOMATON",
+ "raw": {
+ "normalized_reference": "but she had become an automaton.",
+ "normalized_hypothesis": "but she had become an automaton",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "but she had become an automaton",
+ "normalized_hypothesis": "but she had become an automaton",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0375",
+ "category": "phonetic_balance",
+ "text": "I do not blame you for anything; remember that.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0375.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I DO NOT BLAME YOU FOR ANYTHING REMEMBER THAT",
+ "raw": {
+ "normalized_reference": "i do not blame you for anything; remember that.",
+ "normalized_hypothesis": "i do not blame you for anything remember that",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i do not blame you for anything remember that",
+ "normalized_hypothesis": "i do not blame you for anything remember that",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0376",
+ "category": "phonetic_balance",
+ "text": "The next thing to watch out for is bed sores.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0376.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE NEXT THING TO WATCH OUT FOR IS BED SORES",
+ "raw": {
+ "normalized_reference": "the next thing to watch out for is bed sores.",
+ "normalized_hypothesis": "the next thing to watch out for is bed sores",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the next thing to watch out for is bed sores",
+ "normalized_hypothesis": "the next thing to watch out for is bed sores",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0377",
+ "category": "phonetic_balance",
+ "text": "He will follow us soon.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0377.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE WILL FOLLOW US SOON",
+ "raw": {
+ "normalized_reference": "he will follow us soon.",
+ "normalized_hypothesis": "he will follow us soon",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he will follow us soon",
+ "normalized_hypothesis": "he will follow us soon",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0378",
+ "category": "phonetic_balance",
+ "text": "His beady black eyes saw bargains where other men saw bankruptcy.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0378.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HIS BEADY BLACK EYES SAW BARGAINS WHERE OTHER MEN SAW BANKRUPTCY",
+ "raw": {
+ "normalized_reference": "his beady black eyes saw bargains where other men saw bankruptcy.",
+ "normalized_hypothesis": "his beady black eyes saw bargains where other men saw bankruptcy",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "his beady black eyes saw bargains where other men saw bankruptcy",
+ "normalized_hypothesis": "his beady black eyes saw bargains where other men saw bankruptcy",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0379",
+ "category": "phonetic_balance",
+ "text": "They robbed me a few years later.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0379.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THEY ROBBED ME A FEW YEARS LATER",
+ "raw": {
+ "normalized_reference": "they robbed me a few years later.",
+ "normalized_hypothesis": "they robbed me a few years later",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "they robbed me a few years later",
+ "normalized_hypothesis": "they robbed me a few years later",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0380",
+ "category": "phonetic_balance",
+ "text": "Gregson held a lighted match until it burnt his fingertips.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0380.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "GREGSON HELD A LIGHTED MATCH UNTIL IT BURNT HIS FINGER TIPS",
+ "raw": {
+ "normalized_reference": "gregson held a lighted match until it burnt his fingertips.",
+ "normalized_hypothesis": "gregson held a lighted match until it burnt his finger tips",
+ "errors": 2,
+ "words": 10,
+ "wer": 0.2
+ },
+ "semantic": {
+ "normalized_reference": "gregson held a lighted match until it burnt his fingertips",
+ "normalized_hypothesis": "gregson held a lighted match until it burnt his finger tips",
+ "errors": 2,
+ "words": 10,
+ "wer": 0.2
+ }
+ },
+ {
+ "id": "release500_0381",
+ "category": "phonetic_balance",
+ "text": "A burst of laughter was his reward.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0381.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "A BURST OF LAUGHTER WAS HIS REWARD",
+ "raw": {
+ "normalized_reference": "a burst of laughter was his reward.",
+ "normalized_hypothesis": "a burst of laughter was his reward",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "a burst of laughter was his reward",
+ "normalized_hypothesis": "a burst of laughter was his reward",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0382",
+ "category": "phonetic_balance",
+ "text": "He cried, and swung the club wildly.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0382.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE CRIED AND SWUNG THE CLUB WILDLY",
+ "raw": {
+ "normalized_reference": "he cried, and swung the club wildly.",
+ "normalized_hypothesis": "he cried and swung the club wildly",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he cried and swung the club wildly",
+ "normalized_hypothesis": "he cried and swung the club wildly",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0383",
+ "category": "phonetic_balance",
+ "text": "Sandel would never become a world champion.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0383.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SANDALL WOULD NEVER BECOME A WORLD CHAMPION",
+ "raw": {
+ "normalized_reference": "sandel would never become a world champion.",
+ "normalized_hypothesis": "sandall would never become a world champion",
+ "errors": 1,
+ "words": 7,
+ "wer": 0.14285714285714285
+ },
+ "semantic": {
+ "normalized_reference": "sandel would never become a world champion",
+ "normalized_hypothesis": "sandall would never become a world champion",
+ "errors": 1,
+ "words": 7,
+ "wer": 0.14285714285714285
+ }
+ },
+ {
+ "id": "release500_0384",
+ "category": "phonetic_balance",
+ "text": "And after the bath a shave would not be bad.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0384.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "AND AFTER THE BATHASHEF WOULD NOT BE BAD",
+ "raw": {
+ "normalized_reference": "and after the bath a shave would not be bad.",
+ "normalized_hypothesis": "and after the bathashef would not be bad",
+ "errors": 3,
+ "words": 10,
+ "wer": 0.3
+ },
+ "semantic": {
+ "normalized_reference": "and after the bath a shave would not be bad",
+ "normalized_hypothesis": "and after the bathashef would not be bad",
+ "errors": 3,
+ "words": 10,
+ "wer": 0.3
+ }
+ },
+ {
+ "id": "release500_0385",
+ "category": "phonetic_balance",
+ "text": "They saw each other for the first time in Boston.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0385.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THEY SAW EACH OTHER FOR THE FIRST TIME IN BOSTON",
+ "raw": {
+ "normalized_reference": "they saw each other for the first time in boston.",
+ "normalized_hypothesis": "they saw each other for the first time in boston",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "they saw each other for the 1st time in boston",
+ "normalized_hypothesis": "they saw each other for the 1st time in boston",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0386",
+ "category": "phonetic_balance",
+ "text": "I had been born with no organic, chemical predisposition toward alcohol.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0386.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I HAD BEEN BORN WITH NO ORGANIC CHEMICAL PREDISPOSITION TOWARD ALCOHOL",
+ "raw": {
+ "normalized_reference": "i had been born with no organic, chemical predisposition toward alcohol.",
+ "normalized_hypothesis": "i had been born with no organic chemical predisposition toward alcohol",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i had been born with no organic chemical predisposition toward alcohol",
+ "normalized_hypothesis": "i had been born with no organic chemical predisposition toward alcohol",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0387",
+ "category": "phonetic_balance",
+ "text": "They had no fixed values, to be altered by adjectives and adverbs.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0387.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THEY HAD NO FIXED VALUES TO BE ALTERED BY ADJECTIVES AND ADVERBS",
+ "raw": {
+ "normalized_reference": "they had no fixed values, to be altered by adjectives and adverbs.",
+ "normalized_hypothesis": "they had no fixed values to be altered by adjectives and adverbs",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "they had no fixed values to be altered by adjectives and adverbs",
+ "normalized_hypothesis": "they had no fixed values to be altered by adjectives and adverbs",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0388",
+ "category": "phonetic_balance",
+ "text": "Nor was Elam Harnish an exception.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0388.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "NOR WAS THE LAMB HARNISH AN EXCEPTION",
+ "raw": {
+ "normalized_reference": "nor was elam harnish an exception.",
+ "normalized_hypothesis": "nor was the lamb harnish an exception",
+ "errors": 2,
+ "words": 6,
+ "wer": 0.3333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "nor was elam harnish an exception",
+ "normalized_hypothesis": "nor was the lamb harnish an exception",
+ "errors": 2,
+ "words": 6,
+ "wer": 0.3333333333333333
+ }
+ },
+ {
+ "id": "release500_0389",
+ "category": "phonetic_balance",
+ "text": "He was a wise hyena.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0389.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE WAS A WISE HYENA",
+ "raw": {
+ "normalized_reference": "he was a wise hyena.",
+ "normalized_hypothesis": "he was a wise hyena",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he was a wise hyena",
+ "normalized_hypothesis": "he was a wise hyena",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0390",
+ "category": "phonetic_balance",
+ "text": "And now behold the perversity of things.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0390.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "AND NOW BEHOLD THE PERVERSITY OF THINGS",
+ "raw": {
+ "normalized_reference": "and now behold the perversity of things.",
+ "normalized_hypothesis": "and now behold the perversity of things",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "and now behold the perversity of things",
+ "normalized_hypothesis": "and now behold the perversity of things",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0391",
+ "category": "phonetic_balance",
+ "text": "The Fire People, like ourselves, lived in caves.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0391.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE FIRE PEOPLE LIKE OURSELVES LIVED IN CAVES",
+ "raw": {
+ "normalized_reference": "the fire people, like ourselves, lived in caves.",
+ "normalized_hypothesis": "the fire people like ourselves lived in caves",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the fire people like ourselves lived in caves",
+ "normalized_hypothesis": "the fire people like ourselves lived in caves",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0392",
+ "category": "phonetic_balance",
+ "text": "To her the bridge was tambo, which is the native for taboo.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0392.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WHERE THE BRIDGE WAS TAMBEU WHICH IS THE NATIVE FOR TABOO",
+ "raw": {
+ "normalized_reference": "to her the bridge was tambo, which is the native for taboo.",
+ "normalized_hypothesis": "where the bridge was tambeu which is the native for taboo",
+ "errors": 3,
+ "words": 12,
+ "wer": 0.25
+ },
+ "semantic": {
+ "normalized_reference": "to her the bridge was tambo which is the native for taboo",
+ "normalized_hypothesis": "where the bridge was tambeu which is the native for taboo",
+ "errors": 3,
+ "words": 12,
+ "wer": 0.25
+ }
+ },
+ {
+ "id": "release500_0393",
+ "category": "phonetic_balance",
+ "text": "The voyage was our idea of a good time.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0393.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE VOYAGE WAS OUR IDEA OF A GOOD TIME",
+ "raw": {
+ "normalized_reference": "the voyage was our idea of a good time.",
+ "normalized_hypothesis": "the voyage was our idea of a good time",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the voyage was our idea of a good time",
+ "normalized_hypothesis": "the voyage was our idea of a good time",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0394",
+ "category": "phonetic_balance",
+ "text": "Of course much grumbling went on, and little outbursts were continually occurring.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0394.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "OF COURSE MUCH GRUMBLING WENT ON AND LITTLE OUTBURSTS WERE CONTINUALLY OCCURRING",
+ "raw": {
+ "normalized_reference": "of course much grumbling went on, and little outbursts were continually occurring.",
+ "normalized_hypothesis": "of course much grumbling went on and little outbursts were continually occurring",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "of course much grumbling went on and little outbursts were continually occurring",
+ "normalized_hypothesis": "of course much grumbling went on and little outbursts were continually occurring",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0395",
+ "category": "phonetic_balance",
+ "text": "Get down and dig in.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0395.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "GET DOWN AND DIG IN",
+ "raw": {
+ "normalized_reference": "get down and dig in.",
+ "normalized_hypothesis": "get down and dig in",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "get down and dig in",
+ "normalized_hypothesis": "get down and dig in",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0396",
+ "category": "phonetic_balance",
+ "text": "It was put together in a casual, helter-skelter sort of way.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0396.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT WAS PUT TOGETHER IN A CASUAL HELTESCELTER SORT OF WAY",
+ "raw": {
+ "normalized_reference": "it was put together in a casual, helter-skelter sort of way.",
+ "normalized_hypothesis": "it was put together in a casual heltescelter sort of way",
+ "errors": 2,
+ "words": 12,
+ "wer": 0.16666666666666666
+ },
+ "semantic": {
+ "normalized_reference": "it was put together in a casual helter skelter sort of way",
+ "normalized_hypothesis": "it was put together in a casual heltescelter sort of way",
+ "errors": 2,
+ "words": 12,
+ "wer": 0.16666666666666666
+ }
+ },
+ {
+ "id": "release500_0397",
+ "category": "phonetic_balance",
+ "text": "He deluged me, overwhelmed me with argument.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0397.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE DELUGED ME OVERWHELMED ME WITH ARGUMENT",
+ "raw": {
+ "normalized_reference": "he deluged me, overwhelmed me with argument.",
+ "normalized_hypothesis": "he deluged me overwhelmed me with argument",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he deluged me overwhelmed me with argument",
+ "normalized_hypothesis": "he deluged me overwhelmed me with argument",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0398",
+ "category": "phonetic_balance",
+ "text": "The girl faced him, her eyes shining with sudden fear.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0398.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE GIRL FACED HIM HER EYES SHINING WITH SUDDEN FEAR",
+ "raw": {
+ "normalized_reference": "the girl faced him, her eyes shining with sudden fear.",
+ "normalized_hypothesis": "the girl faced him her eyes shining with sudden fear",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the girl faced him her eyes shining with sudden fear",
+ "normalized_hypothesis": "the girl faced him her eyes shining with sudden fear",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0399",
+ "category": "phonetic_balance",
+ "text": "It will break our hearts and our backs to hoist anchor by hand.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0399.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT WILL BREAK OUR HEARTS AND OUR BACKS TO HOIST ANCHOR BY HAND",
+ "raw": {
+ "normalized_reference": "it will break our hearts and our backs to hoist anchor by hand.",
+ "normalized_hypothesis": "it will break our hearts and our backs to hoist anchor by hand",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it will break our hearts and our backs to hoist anchor by hand",
+ "normalized_hypothesis": "it will break our hearts and our backs to hoist anchor by hand",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0400",
+ "category": "phonetic_balance",
+ "text": "He made no reply as he waited for Whittemore to continue.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0400.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE MADE NO REPLY AS HE WAITED FOR WHITTERMORE TO CONTINUE",
+ "raw": {
+ "normalized_reference": "he made no reply as he waited for whittemore to continue.",
+ "normalized_hypothesis": "he made no reply as he waited for whittermore to continue",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ },
+ "semantic": {
+ "normalized_reference": "he made no reply as he waited for whittemore to continue",
+ "normalized_hypothesis": "he made no reply as he waited for whittermore to continue",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ }
+ },
+ {
+ "id": "release500_0401",
+ "category": "phonetic_balance",
+ "text": "Who the devil gave it to you to be judge and jury.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0401.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WHO THE DEVIL GAVE IT TO YOU TO BE JUDGE AND JURY",
+ "raw": {
+ "normalized_reference": "who the devil gave it to you to be judge and jury.",
+ "normalized_hypothesis": "who the devil gave it to you to be judge and jury",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "who the devil gave it to you to be judge and jury",
+ "normalized_hypothesis": "who the devil gave it to you to be judge and jury",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0402",
+ "category": "phonetic_balance",
+ "text": "Daylight found himself charmed and made curious by the little man.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0402.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "DAYLIGHT FOUND HIMSELF CHARMED AND MADE CURIOUS BY THE LITTLE MAN",
+ "raw": {
+ "normalized_reference": "daylight found himself charmed and made curious by the little man.",
+ "normalized_hypothesis": "daylight found himself charmed and made curious by the little man",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "daylight found himself charmed and made curious by the little man",
+ "normalized_hypothesis": "daylight found himself charmed and made curious by the little man",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0403",
+ "category": "phonetic_balance",
+ "text": "It's a Yankee, Joan cried.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0403.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT'S A YANKEE JOAN CRIED",
+ "raw": {
+ "normalized_reference": "it's a yankee, joan cried.",
+ "normalized_hypothesis": "it's a yankee joan cried",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it is a yankee joan cried",
+ "normalized_hypothesis": "it is a yankee joan cried",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0404",
+ "category": "phonetic_balance",
+ "text": "But what they want with your toothbrush is more than I can imagine.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0404.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "BUT WHAT THEY WANT WITH YOUR TOOTH BRUSH IS MORE THAN I CAN IMAGINE",
+ "raw": {
+ "normalized_reference": "but what they want with your toothbrush is more than i can imagine.",
+ "normalized_hypothesis": "but what they want with your tooth brush is more than i can imagine",
+ "errors": 2,
+ "words": 13,
+ "wer": 0.15384615384615385
+ },
+ "semantic": {
+ "normalized_reference": "but what they want with your toothbrush is more than i can imagine",
+ "normalized_hypothesis": "but what they want with your tooth brush is more than i can imagine",
+ "errors": 2,
+ "words": 13,
+ "wer": 0.15384615384615385
+ }
+ },
+ {
+ "id": "release500_0405",
+ "category": "phonetic_balance",
+ "text": "They were babbling and chattering all together.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0405.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THEY WERE BABBLING AND CHATTERING ALTOGETHER",
+ "raw": {
+ "normalized_reference": "they were babbling and chattering all together.",
+ "normalized_hypothesis": "they were babbling and chattering altogether",
+ "errors": 2,
+ "words": 7,
+ "wer": 0.2857142857142857
+ },
+ "semantic": {
+ "normalized_reference": "they were babbling and chattering all together",
+ "normalized_hypothesis": "they were babbling and chattering altogether",
+ "errors": 2,
+ "words": 7,
+ "wer": 0.2857142857142857
+ }
+ },
+ {
+ "id": "release500_0406",
+ "category": "phonetic_balance",
+ "text": "Men of Selden's stamp don't stop at women and children.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0406.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "MEN OF SELDON'S STAMP DON'T STOP AT WOMEN AND CHILDREN",
+ "raw": {
+ "normalized_reference": "men of selden's stamp don't stop at women and children.",
+ "normalized_hypothesis": "men of seldon's stamp don't stop at women and children",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ },
+ "semantic": {
+ "normalized_reference": "men of selden is stamp do not stop at women and children",
+ "normalized_hypothesis": "men of seldon is stamp do not stop at women and children",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ }
+ },
+ {
+ "id": "release500_0407",
+ "category": "phonetic_balance",
+ "text": "Cameron looked at his hands with their long, sinewy fingers.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0407.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "CAMERON LOOKED AT HIS HANDS WITH HER LONG SINEWY FINGERS",
+ "raw": {
+ "normalized_reference": "cameron looked at his hands with their long, sinewy fingers.",
+ "normalized_hypothesis": "cameron looked at his hands with her long sinewy fingers",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ },
+ "semantic": {
+ "normalized_reference": "cameron looked at his hands with their long sinewy fingers",
+ "normalized_hypothesis": "cameron looked at his hands with her long sinewy fingers",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ }
+ },
+ {
+ "id": "release500_0408",
+ "category": "phonetic_balance",
+ "text": "Saxon waited, for she knew a fresh idea had struck Billy.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0408.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SAXON WAITED FOR SHE KNEW A FRESH IDEA HAD STRUCK BILLY",
+ "raw": {
+ "normalized_reference": "saxon waited, for she knew a fresh idea had struck billy.",
+ "normalized_hypothesis": "saxon waited for she knew a fresh idea had struck billy",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "saxon waited for she knew a fresh idea had struck billy",
+ "normalized_hypothesis": "saxon waited for she knew a fresh idea had struck billy",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0409",
+ "category": "phonetic_balance",
+ "text": "He saw only the effect in a general, sketchy way.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0409.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE SAW ONLY THE EFFECT IN A GENERAL SKETCHY WAY",
+ "raw": {
+ "normalized_reference": "he saw only the effect in a general, sketchy way.",
+ "normalized_hypothesis": "he saw only the effect in a general sketchy way",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he saw only the effect in a general sketchy way",
+ "normalized_hypothesis": "he saw only the effect in a general sketchy way",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0410",
+ "category": "phonetic_balance",
+ "text": "He began to follow the footprints of the dog.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0410.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE BEGAN TO FOLLOW THE FOOTPRINTS OF THE DOG",
+ "raw": {
+ "normalized_reference": "he began to follow the footprints of the dog.",
+ "normalized_hypothesis": "he began to follow the footprints of the dog",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he began to follow the footprints of the dog",
+ "normalized_hypothesis": "he began to follow the footprints of the dog",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0411",
+ "category": "phonetic_balance",
+ "text": "Do you value your hide.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0411.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "DO YOU VALUE YOUR HIDE",
+ "raw": {
+ "normalized_reference": "do you value your hide.",
+ "normalized_hypothesis": "do you value your hide",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "do you value your hide",
+ "normalized_hypothesis": "do you value your hide",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0412",
+ "category": "phonetic_balance",
+ "text": "A rising tide of fat had submerged them.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0412.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "A RISING TIDE OF FAT HAD SUBMERGED THEM",
+ "raw": {
+ "normalized_reference": "a rising tide of fat had submerged them.",
+ "normalized_hypothesis": "a rising tide of fat had submerged them",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "a rising tide of fat had submerged them",
+ "normalized_hypothesis": "a rising tide of fat had submerged them",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0413",
+ "category": "phonetic_balance",
+ "text": "There was nothing on the rock.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0413.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THERE WAS NOTHING ON THE ROCK",
+ "raw": {
+ "normalized_reference": "there was nothing on the rock.",
+ "normalized_hypothesis": "there was nothing on the rock",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "there was nothing on the rock",
+ "normalized_hypothesis": "there was nothing on the rock",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0414",
+ "category": "phonetic_balance",
+ "text": "Whatever he guessed he locked away in the taboo room of Naomi.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0414.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WHATEVER HE GUESSED HE LOCKED AWAY IN THE TABOO ROOM OF NAOMI",
+ "raw": {
+ "normalized_reference": "whatever he guessed he locked away in the taboo room of naomi.",
+ "normalized_hypothesis": "whatever he guessed he locked away in the taboo room of naomi",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "whatever he guessed he locked away in the taboo room of naomi",
+ "normalized_hypothesis": "whatever he guessed he locked away in the taboo room of naomi",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0415",
+ "category": "phonetic_balance",
+ "text": "Will we ever forget it.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0415.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WILL WE EVER FORGET IT",
+ "raw": {
+ "normalized_reference": "will we ever forget it.",
+ "normalized_hypothesis": "will we ever forget it",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "will we ever forget it",
+ "normalized_hypothesis": "will we ever forget it",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0416",
+ "category": "phonetic_balance",
+ "text": "Jeanne and Pierre both gazed toward the great rock.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0416.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "JEAN AND PIERRE BOTH GAZED TOWARD THE GREAT ROCK",
+ "raw": {
+ "normalized_reference": "jeanne and pierre both gazed toward the great rock.",
+ "normalized_hypothesis": "jean and pierre both gazed toward the great rock",
+ "errors": 1,
+ "words": 9,
+ "wer": 0.1111111111111111
+ },
+ "semantic": {
+ "normalized_reference": "jeanne and pierre both gazed toward the great rock",
+ "normalized_hypothesis": "jean and pierre both gazed toward the great rock",
+ "errors": 1,
+ "words": 9,
+ "wer": 0.1111111111111111
+ }
+ },
+ {
+ "id": "release500_0417",
+ "category": "phonetic_balance",
+ "text": "Does the old boy often go off at half-cock that way.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0417.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "DOES THE OLD BOY OFTEN GO OFF AT HALF COCK THAT WAY",
+ "raw": {
+ "normalized_reference": "does the old boy often go off at half-cock that way.",
+ "normalized_hypothesis": "does the old boy often go off at half cock that way",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "does the old boy often go off at half cock that way",
+ "normalized_hypothesis": "does the old boy often go off at half cock that way",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0418",
+ "category": "phonetic_balance",
+ "text": "The temperature dropped to fifty below zero and remained there the whole trip.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0418.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE TEMPERATURE DROPPED TO FIFTY BELOW ZERO AND REMAINED THERE THE WHOLE TRIP",
+ "raw": {
+ "normalized_reference": "the temperature dropped to fifty below zero and remained there the whole trip.",
+ "normalized_hypothesis": "the temperature dropped to fifty below zero and remained there the whole trip",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the temperature dropped to 50 below 0 and remained there the whole trip",
+ "normalized_hypothesis": "the temperature dropped to 50 below 0 and remained there the whole trip",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0419",
+ "category": "phonetic_balance",
+ "text": "It's worth eight dollars.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0419.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT'S WORTH EIGHT DOLLARS",
+ "raw": {
+ "normalized_reference": "it's worth eight dollars.",
+ "normalized_hypothesis": "it's worth eight dollars",
+ "errors": 0,
+ "words": 4,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it is worth $8",
+ "normalized_hypothesis": "it is worth $8",
+ "errors": 0,
+ "words": 4,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0420",
+ "category": "phonetic_balance",
+ "text": "But I did not enjoy it long.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0420.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "BUT I DID NOT ENJOY IT LONG",
+ "raw": {
+ "normalized_reference": "but i did not enjoy it long.",
+ "normalized_hypothesis": "but i did not enjoy it long",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "but i did not enjoy it long",
+ "normalized_hypothesis": "but i did not enjoy it long",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0421",
+ "category": "phonetic_balance",
+ "text": "Shorty turned to their employers.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0421.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SHORTLY TURNED TO THEIR EMPLOYERS",
+ "raw": {
+ "normalized_reference": "shorty turned to their employers.",
+ "normalized_hypothesis": "shortly turned to their employers",
+ "errors": 1,
+ "words": 5,
+ "wer": 0.2
+ },
+ "semantic": {
+ "normalized_reference": "shorty turned to their employers",
+ "normalized_hypothesis": "shortly turned to their employers",
+ "errors": 1,
+ "words": 5,
+ "wer": 0.2
+ }
+ },
+ {
+ "id": "release500_0422",
+ "category": "phonetic_balance",
+ "text": "It was the third or fourth time that Philip had heard MacDougall swear.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0422.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT WAS THE THIRD OR FOURTH TIME THAT PHILIP HAD HEARD MACDUGALL SWEAR",
+ "raw": {
+ "normalized_reference": "it was the third or fourth time that philip had heard macdougall swear.",
+ "normalized_hypothesis": "it was the third or fourth time that philip had heard macdugall swear",
+ "errors": 1,
+ "words": 13,
+ "wer": 0.07692307692307693
+ },
+ "semantic": {
+ "normalized_reference": "it was the 3rd or 4th time that philip had heard macdougall swear",
+ "normalized_hypothesis": "it was the 3rd or 4th time that philip had heard macdugall swear",
+ "errors": 1,
+ "words": 13,
+ "wer": 0.07692307692307693
+ }
+ },
+ {
+ "id": "release500_0423",
+ "category": "phonetic_balance",
+ "text": "Ernest saw in the affair the most sinister import.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0423.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "ERNEST SAW IN THE AFFAIR THE MOST SINISTER IMPORT",
+ "raw": {
+ "normalized_reference": "ernest saw in the affair the most sinister import.",
+ "normalized_hypothesis": "ernest saw in the affair the most sinister import",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "ernest saw in the affair the most sinister import",
+ "normalized_hypothesis": "ernest saw in the affair the most sinister import",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0424",
+ "category": "phonetic_balance",
+ "text": "Also, at regular intervals, he would mutter.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0424.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "ALSO AT REGULAR INTERVALS HE WOULD MUTTER",
+ "raw": {
+ "normalized_reference": "also, at regular intervals, he would mutter.",
+ "normalized_hypothesis": "also at regular intervals he would mutter",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "also at regular intervals he would mutter",
+ "normalized_hypothesis": "also at regular intervals he would mutter",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0425",
+ "category": "phonetic_balance",
+ "text": "He may anticipate the day of his death.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0425.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE MAY ANTICIPATE THE DAY OF HIS DEATH",
+ "raw": {
+ "normalized_reference": "he may anticipate the day of his death.",
+ "normalized_hypothesis": "he may anticipate the day of his death",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he may anticipate the day of his death",
+ "normalized_hypothesis": "he may anticipate the day of his death",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0426",
+ "category": "phonetic_balance",
+ "text": "I defy any man to get a Solomon Island sore in California.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0426.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I DEFY ANY MAN TO GET A SOLAMMON ISLAND SORE IN CALIFORNIA",
+ "raw": {
+ "normalized_reference": "i defy any man to get a solomon island sore in california.",
+ "normalized_hypothesis": "i defy any man to get a solammon island sore in california",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "i defy any man to get a solomon island sore in california",
+ "normalized_hypothesis": "i defy any man to get a solammon island sore in california",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ }
+ },
+ {
+ "id": "release500_0427",
+ "category": "phonetic_balance",
+ "text": "Now our figuring was all right.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0427.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "NOW OUR FIGURING WAS ALL RIGHT",
+ "raw": {
+ "normalized_reference": "now our figuring was all right.",
+ "normalized_hypothesis": "now our figuring was all right",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "now our figuring was all right",
+ "normalized_hypothesis": "now our figuring was all right",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0428",
+ "category": "phonetic_balance",
+ "text": "MacDougall tapped his forehead suspiciously with a stubby forefinger.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0428.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "MAC DOUBLE TAPPED HIS FOREHEAD SUSPICIOUSLY WITH A STUBBY FOREFINGER",
+ "raw": {
+ "normalized_reference": "macdougall tapped his forehead suspiciously with a stubby forefinger.",
+ "normalized_hypothesis": "mac double tapped his forehead suspiciously with a stubby forefinger",
+ "errors": 2,
+ "words": 9,
+ "wer": 0.2222222222222222
+ },
+ "semantic": {
+ "normalized_reference": "macdougall tapped his forehead suspiciously with a stubby forefinger",
+ "normalized_hypothesis": "mac double tapped his forehead suspiciously with a stubby forefinger",
+ "errors": 2,
+ "words": 9,
+ "wer": 0.2222222222222222
+ }
+ },
+ {
+ "id": "release500_0429",
+ "category": "phonetic_balance",
+ "text": "O'Brien took off his coat and bared his right arm.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0429.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "OBRIAN TOOK OFF HIS COAT AND BARED HIS RIGHT ARM",
+ "raw": {
+ "normalized_reference": "o'brien took off his coat and bared his right arm.",
+ "normalized_hypothesis": "obrian took off his coat and bared his right arm",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ },
+ "semantic": {
+ "normalized_reference": "0 brien took off his coat and bared his right arm",
+ "normalized_hypothesis": "obrian took off his coat and bared his right arm",
+ "errors": 2,
+ "words": 11,
+ "wer": 0.18181818181818182
+ }
+ },
+ {
+ "id": "release500_0430",
+ "category": "phonetic_balance",
+ "text": "But it won't continue, she said with easy confidence.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0430.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "BUT IT WON'T CONTINUE SHE SAID WITH EASY CONFIDENCE",
+ "raw": {
+ "normalized_reference": "but it won't continue, she said with easy confidence.",
+ "normalized_hypothesis": "but it won't continue she said with easy confidence",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "but it will not continue she said with easy confidence",
+ "normalized_hypothesis": "but it will not continue she said with easy confidence",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0431",
+ "category": "phonetic_balance",
+ "text": "I was still weak from my prolonged immersion.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0431.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I WAS STILL WEAK FROM MY PROLONGED IMMERSION",
+ "raw": {
+ "normalized_reference": "i was still weak from my prolonged immersion.",
+ "normalized_hypothesis": "i was still weak from my prolonged immersion",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i was still weak from my prolonged immersion",
+ "normalized_hypothesis": "i was still weak from my prolonged immersion",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0432",
+ "category": "phonetic_balance",
+ "text": "Oolong was two hundred and fifty miles from the nearest land.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0432.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "OULON WAS TWO HUNDRED AND FIFTY MILES FROM THE NEAREST LAND",
+ "raw": {
+ "normalized_reference": "oolong was two hundred and fifty miles from the nearest land.",
+ "normalized_hypothesis": "oulon was two hundred and fifty miles from the nearest land",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ },
+ "semantic": {
+ "normalized_reference": "oolong was 250 miles from the nearest land",
+ "normalized_hypothesis": "oulon was 250 miles from the nearest land",
+ "errors": 1,
+ "words": 8,
+ "wer": 0.125
+ }
+ },
+ {
+ "id": "release500_0433",
+ "category": "phonetic_balance",
+ "text": "I just do appreciate it without being able to express my feelings.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0433.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I JUST DO APPRECIATE IT WITHOUT BEING ABLE TO EXPRESS MY FEELINGS",
+ "raw": {
+ "normalized_reference": "i just do appreciate it without being able to express my feelings.",
+ "normalized_hypothesis": "i just do appreciate it without being able to express my feelings",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i just do appreciate it without being able to express my feelings",
+ "normalized_hypothesis": "i just do appreciate it without being able to express my feelings",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0434",
+ "category": "phonetic_balance",
+ "text": "Now just what do you want to know.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0434.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "NOW JUST WHAT DO YOU WANT TO KNOW",
+ "raw": {
+ "normalized_reference": "now just what do you want to know.",
+ "normalized_hypothesis": "now just what do you want to know",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "now just what do you want to know",
+ "normalized_hypothesis": "now just what do you want to know",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0435",
+ "category": "phonetic_balance",
+ "text": "But she swung obediently on her heel into the wind.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0435.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "BUT SHE SWUNG OBEDIENTLY ON HER HEEL INTO THE WIND",
+ "raw": {
+ "normalized_reference": "but she swung obediently on her heel into the wind.",
+ "normalized_hypothesis": "but she swung obediently on her heel into the wind",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "but she swung obediently on her heel into the wind",
+ "normalized_hypothesis": "but she swung obediently on her heel into the wind",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0436",
+ "category": "phonetic_balance",
+ "text": "Ask him, I laughed, then turned to Pasquini.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0436.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "ASK HIM I LAUGHED THEN TURNED TO PESQUINI",
+ "raw": {
+ "normalized_reference": "ask him, i laughed, then turned to pasquini.",
+ "normalized_hypothesis": "ask him i laughed then turned to pesquini",
+ "errors": 1,
+ "words": 8,
+ "wer": 0.125
+ },
+ "semantic": {
+ "normalized_reference": "ask him i laughed then turned to pasquini",
+ "normalized_hypothesis": "ask him i laughed then turned to pesquini",
+ "errors": 1,
+ "words": 8,
+ "wer": 0.125
+ }
+ },
+ {
+ "id": "release500_0437",
+ "category": "phonetic_balance",
+ "text": "He did not rush in.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0437.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE DID NOT RUSH IN",
+ "raw": {
+ "normalized_reference": "he did not rush in.",
+ "normalized_hypothesis": "he did not rush in",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he did not rush in",
+ "normalized_hypothesis": "he did not rush in",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0438",
+ "category": "phonetic_balance",
+ "text": "They obeyed him, and went here and there at his commands.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0438.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THEY OBEYED HIM AND WENT HERE AND THERE AT HIS COMMANDS",
+ "raw": {
+ "normalized_reference": "they obeyed him, and went here and there at his commands.",
+ "normalized_hypothesis": "they obeyed him and went here and there at his commands",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "they obeyed him and went here and there at his commands",
+ "normalized_hypothesis": "they obeyed him and went here and there at his commands",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0439",
+ "category": "phonetic_balance",
+ "text": "Tom Spink has a harpoon.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0439.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "TOM SMINK HAS A HARPOON",
+ "raw": {
+ "normalized_reference": "tom spink has a harpoon.",
+ "normalized_hypothesis": "tom smink has a harpoon",
+ "errors": 1,
+ "words": 5,
+ "wer": 0.2
+ },
+ "semantic": {
+ "normalized_reference": "tom spink has a harpoon",
+ "normalized_hypothesis": "tom smink has a harpoon",
+ "errors": 1,
+ "words": 5,
+ "wer": 0.2
+ }
+ },
+ {
+ "id": "release500_0440",
+ "category": "phonetic_balance",
+ "text": "It was my reports from the north which chiefly induced people to buy.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0440.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT WAS MY REPORTS FROM THE NORTH WHICH CHIEFLY INDUCED PEOPLE TO BUY",
+ "raw": {
+ "normalized_reference": "it was my reports from the north which chiefly induced people to buy.",
+ "normalized_hypothesis": "it was my reports from the north which chiefly induced people to buy",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it was my reports from the north which chiefly induced people to buy",
+ "normalized_hypothesis": "it was my reports from the north which chiefly induced people to buy",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0441",
+ "category": "phonetic_balance",
+ "text": "Cherokee identified himself with his instinct.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0441.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SCHEROQY IDENTIFIED HIMSELF WITH HIS INSTINCT",
+ "raw": {
+ "normalized_reference": "cherokee identified himself with his instinct.",
+ "normalized_hypothesis": "scheroqy identified himself with his instinct",
+ "errors": 1,
+ "words": 6,
+ "wer": 0.16666666666666666
+ },
+ "semantic": {
+ "normalized_reference": "cherokee identified himself with his instinct",
+ "normalized_hypothesis": "scheroqy identified himself with his instinct",
+ "errors": 1,
+ "words": 6,
+ "wer": 0.16666666666666666
+ }
+ },
+ {
+ "id": "release500_0442",
+ "category": "phonetic_balance",
+ "text": "I graduated last of my class.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0442.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I GRADUATED LAST OF MY CLASS",
+ "raw": {
+ "normalized_reference": "i graduated last of my class.",
+ "normalized_hypothesis": "i graduated last of my class",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i graduated last of my class",
+ "normalized_hypothesis": "i graduated last of my class",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0443",
+ "category": "phonetic_balance",
+ "text": "It was a gigantic inadequacy.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0443.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT WAS A GIGANTIC INADEQUACY",
+ "raw": {
+ "normalized_reference": "it was a gigantic inadequacy.",
+ "normalized_hypothesis": "it was a gigantic inadequacy",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it was a gigantic inadequacy",
+ "normalized_hypothesis": "it was a gigantic inadequacy",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0444",
+ "category": "phonetic_balance",
+ "text": "Did I possess too much vitality.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0444.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "DID I POSSESS TOO MUCH VITALITY",
+ "raw": {
+ "normalized_reference": "did i possess too much vitality.",
+ "normalized_hypothesis": "did i possess too much vitality",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "did i possess too much vitality",
+ "normalized_hypothesis": "did i possess too much vitality",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0445",
+ "category": "phonetic_balance",
+ "text": "Therefore, hurrah for the game.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0445.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THEREFORE HURRAH FOR THE GAME",
+ "raw": {
+ "normalized_reference": "therefore, hurrah for the game.",
+ "normalized_hypothesis": "therefore hurrah for the game",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "therefore hurrah for the game",
+ "normalized_hypothesis": "therefore hurrah for the game",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0446",
+ "category": "phonetic_balance",
+ "text": "He had fulfilled his duty and paid properly.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0446.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE HAD FULFILLED HIS DUTY AND PAID PROPERLY",
+ "raw": {
+ "normalized_reference": "he had fulfilled his duty and paid properly.",
+ "normalized_hypothesis": "he had fulfilled his duty and paid properly",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he had fulfilled his duty and paid properly",
+ "normalized_hypothesis": "he had fulfilled his duty and paid properly",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0447",
+ "category": "phonetic_balance",
+ "text": "How can you manage all alone, Mr Young.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0447.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "OW CAN YOU MANAGE ALL ALONE MISTER YOUNG",
+ "raw": {
+ "normalized_reference": "how can you manage all alone, mr young.",
+ "normalized_hypothesis": "ow can you manage all alone mister young",
+ "errors": 2,
+ "words": 8,
+ "wer": 0.25
+ },
+ "semantic": {
+ "normalized_reference": "how can you manage all alone mister young",
+ "normalized_hypothesis": "ow can you manage all alone mister young",
+ "errors": 1,
+ "words": 8,
+ "wer": 0.125
+ }
+ },
+ {
+ "id": "release500_0448",
+ "category": "phonetic_balance",
+ "text": "Red-Eye never committed a more outrageous deed.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0448.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "RED EYE NEVER COMMITTED A MORE OUTRAGEOUS DEED",
+ "raw": {
+ "normalized_reference": "red-eye never committed a more outrageous deed.",
+ "normalized_hypothesis": "red eye never committed a more outrageous deed",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "red eye never committed a more outrageous deed",
+ "normalized_hypothesis": "red eye never committed a more outrageous deed",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0449",
+ "category": "phonetic_balance",
+ "text": "He would keep his faith with Gregson for the promised day or two.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0449.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE WOULD KEEP HIS FAITH WITH GRIGON FOR THE PROMISED DAY OR TWO",
+ "raw": {
+ "normalized_reference": "he would keep his faith with gregson for the promised day or two.",
+ "normalized_hypothesis": "he would keep his faith with grigon for the promised day or two",
+ "errors": 1,
+ "words": 13,
+ "wer": 0.07692307692307693
+ },
+ "semantic": {
+ "normalized_reference": "he would keep his faith with gregson for the promised day or 2",
+ "normalized_hypothesis": "he would keep his faith with grigon for the promised day or 2",
+ "errors": 1,
+ "words": 13,
+ "wer": 0.07692307692307693
+ }
+ },
+ {
+ "id": "release500_0450",
+ "category": "phonetic_balance",
+ "text": "There was one chance, and only one, of saving Jeanne.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0450.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THERE WAS ONE CHANCE AND ONLY ONE OF SAVING JEAN",
+ "raw": {
+ "normalized_reference": "there was one chance, and only one, of saving jeanne.",
+ "normalized_hypothesis": "there was one chance and only one of saving jean",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ },
+ "semantic": {
+ "normalized_reference": "there was one chance and only one of saving jeanne",
+ "normalized_hypothesis": "there was one chance and only one of saving jean",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ }
+ },
+ {
+ "id": "release500_0451",
+ "category": "phonetic_balance",
+ "text": "I was beneath the water, suffocating and drowning.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0451.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I WAS BENEATH THE WATER SUFFOCATING AND DROWNING",
+ "raw": {
+ "normalized_reference": "i was beneath the water, suffocating and drowning.",
+ "normalized_hypothesis": "i was beneath the water suffocating and drowning",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i was beneath the water suffocating and drowning",
+ "normalized_hypothesis": "i was beneath the water suffocating and drowning",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0452",
+ "category": "phonetic_balance",
+ "text": "Give them their choice between a fine or an official whipping.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0452.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "GIVE THEM THEIR CHOICE BETWEEN A FINE OR AN OFFICIAL WHIPPING",
+ "raw": {
+ "normalized_reference": "give them their choice between a fine or an official whipping.",
+ "normalized_hypothesis": "give them their choice between a fine or an official whipping",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "give them their choice between a fine or an official whipping",
+ "normalized_hypothesis": "give them their choice between a fine or an official whipping",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0453",
+ "category": "phonetic_balance",
+ "text": "They must have been swept away by the chaotic currents.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0453.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THEY MUST HAVE BEEN SWEPT AWAY BY THE CHAOTIC CURRENTS",
+ "raw": {
+ "normalized_reference": "they must have been swept away by the chaotic currents.",
+ "normalized_hypothesis": "they must have been swept away by the chaotic currents",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "they must have been swept away by the chaotic currents",
+ "normalized_hypothesis": "they must have been swept away by the chaotic currents",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0454",
+ "category": "phonetic_balance",
+ "text": "You yellow giant thing of the frost.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0454.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "YOU YELLOW GIANT THING OF THE FROST",
+ "raw": {
+ "normalized_reference": "you yellow giant thing of the frost.",
+ "normalized_hypothesis": "you yellow giant thing of the frost",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "you yellow giant thing of the frost",
+ "normalized_hypothesis": "you yellow giant thing of the frost",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0455",
+ "category": "phonetic_balance",
+ "text": "He had peeled off his shirt and was wildly waving it.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0455.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE HAD PEELED OFF HIS SHIRT AND WAS WILDLY WAVING IT",
+ "raw": {
+ "normalized_reference": "he had peeled off his shirt and was wildly waving it.",
+ "normalized_hypothesis": "he had peeled off his shirt and was wildly waving it",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he had peeled off his shirt and was wildly waving it",
+ "normalized_hypothesis": "he had peeled off his shirt and was wildly waving it",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0456",
+ "category": "phonetic_balance",
+ "text": "To these he gave castor oil.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0456.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "TOTHESE HE GAVE CASTOR OIL",
+ "raw": {
+ "normalized_reference": "to these he gave castor oil.",
+ "normalized_hypothesis": "tothese he gave castor oil",
+ "errors": 2,
+ "words": 6,
+ "wer": 0.3333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "to these he gave castor oil",
+ "normalized_hypothesis": "tothese he gave castor oil",
+ "errors": 2,
+ "words": 6,
+ "wer": 0.3333333333333333
+ }
+ },
+ {
+ "id": "release500_0457",
+ "category": "phonetic_balance",
+ "text": "Tudor surveyed him with withering disgust.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0457.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "JUDA SURVEYED HIM WITH WITHERING DISGUST",
+ "raw": {
+ "normalized_reference": "tudor surveyed him with withering disgust.",
+ "normalized_hypothesis": "juda surveyed him with withering disgust",
+ "errors": 1,
+ "words": 6,
+ "wer": 0.16666666666666666
+ },
+ "semantic": {
+ "normalized_reference": "tudor surveyed him with withering disgust",
+ "normalized_hypothesis": "juda surveyed him with withering disgust",
+ "errors": 1,
+ "words": 6,
+ "wer": 0.16666666666666666
+ }
+ },
+ {
+ "id": "release500_0458",
+ "category": "phonetic_balance",
+ "text": "After all, it was simply a mistake in judgment.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0458.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "AFTER ALL IT WAS SIMPLY A MISTAKEN JUDGMENT",
+ "raw": {
+ "normalized_reference": "after all, it was simply a mistake in judgment.",
+ "normalized_hypothesis": "after all it was simply a mistaken judgment",
+ "errors": 2,
+ "words": 9,
+ "wer": 0.2222222222222222
+ },
+ "semantic": {
+ "normalized_reference": "after all it was simply a mistake in judgment",
+ "normalized_hypothesis": "after all it was simply a mistaken judgment",
+ "errors": 2,
+ "words": 9,
+ "wer": 0.2222222222222222
+ }
+ },
+ {
+ "id": "release500_0459",
+ "category": "phonetic_balance",
+ "text": "That is the strange part of it.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0459.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THAT IS THE STRANGE PART OF IT",
+ "raw": {
+ "normalized_reference": "that is the strange part of it.",
+ "normalized_hypothesis": "that is the strange part of it",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "that is the strange part of it",
+ "normalized_hypothesis": "that is the strange part of it",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0460",
+ "category": "phonetic_balance",
+ "text": "Not a wheel moved in his empire.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0460.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "NOT A WEEL MOVE IN HIS EMPIRE",
+ "raw": {
+ "normalized_reference": "not a wheel moved in his empire.",
+ "normalized_hypothesis": "not a weel move in his empire",
+ "errors": 2,
+ "words": 7,
+ "wer": 0.2857142857142857
+ },
+ "semantic": {
+ "normalized_reference": "not a wheel moved in his empire",
+ "normalized_hypothesis": "not a weel move in his empire",
+ "errors": 2,
+ "words": 7,
+ "wer": 0.2857142857142857
+ }
+ },
+ {
+ "id": "release500_0461",
+ "category": "phonetic_balance",
+ "text": "Its diameter was not more than two hundred yards.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0461.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "ITS DIAMETER WAS NOT MORE THAN TWO HUNDRED YARDS",
+ "raw": {
+ "normalized_reference": "its diameter was not more than two hundred yards.",
+ "normalized_hypothesis": "its diameter was not more than two hundred yards",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "its diameter was not more than 200 yards",
+ "normalized_hypothesis": "its diameter was not more than 200 yards",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0462",
+ "category": "phonetic_balance",
+ "text": "Nimrod replied, with a slight manifestation of sensitiveness.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0462.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "NIAMERAB REPLIED WITH A SLIGHT MANIFESTATION OF SENSITIVENESS",
+ "raw": {
+ "normalized_reference": "nimrod replied, with a slight manifestation of sensitiveness.",
+ "normalized_hypothesis": "niamerab replied with a slight manifestation of sensitiveness",
+ "errors": 1,
+ "words": 8,
+ "wer": 0.125
+ },
+ "semantic": {
+ "normalized_reference": "nimrod replied with a slight manifestation of sensitiveness",
+ "normalized_hypothesis": "niamerab replied with a slight manifestation of sensitiveness",
+ "errors": 1,
+ "words": 8,
+ "wer": 0.125
+ }
+ },
+ {
+ "id": "release500_0463",
+ "category": "phonetic_balance",
+ "text": "I tried to read George Moore last night, and was dreadfully bored.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0463.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I TRIED TO READ GEORGE MOORE LAST NIGHT AND WAS DREADFULLY BORED",
+ "raw": {
+ "normalized_reference": "i tried to read george moore last night, and was dreadfully bored.",
+ "normalized_hypothesis": "i tried to read george moore last night and was dreadfully bored",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i tried to read george moore last night and was dreadfully bored",
+ "normalized_hypothesis": "i tried to read george moore last night and was dreadfully bored",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0464",
+ "category": "phonetic_balance",
+ "text": "Yes, it was a man who asked, a stranger.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0464.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "YES IT WAS A MAN WHO ASKED A STRANGER",
+ "raw": {
+ "normalized_reference": "yes, it was a man who asked, a stranger.",
+ "normalized_hypothesis": "yes it was a man who asked a stranger",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "yes it was a man who asked a stranger",
+ "normalized_hypothesis": "yes it was a man who asked a stranger",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0465",
+ "category": "phonetic_balance",
+ "text": "We were now good friends.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0465.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WE WERE NOW GOOD FRIENDS",
+ "raw": {
+ "normalized_reference": "we were now good friends.",
+ "normalized_hypothesis": "we were now good friends",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "we were now good friends",
+ "normalized_hypothesis": "we were now good friends",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0466",
+ "category": "phonetic_balance",
+ "text": "Yet, in accordance with Ernest's test of truth, it worked.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0466.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "YET IN ACCORDANCE WITH ERNEST'S TEST OF TRUTH IT WORKED",
+ "raw": {
+ "normalized_reference": "yet, in accordance with ernest's test of truth, it worked.",
+ "normalized_hypothesis": "yet in accordance with ernest's test of truth it worked",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "yet in accordance with ernest is test of truth it worked",
+ "normalized_hypothesis": "yet in accordance with ernest is test of truth it worked",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0467",
+ "category": "phonetic_balance",
+ "text": "It's the strap hangers that'll keep us from going under.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0467.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT'S THE STRAP HANNERS THAT'LL KEEP US FROM GOING UNDER",
+ "raw": {
+ "normalized_reference": "it's the strap hangers that'll keep us from going under.",
+ "normalized_hypothesis": "it's the strap hanners that'll keep us from going under",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ },
+ "semantic": {
+ "normalized_reference": "it is the strap hangers that will keep us from going under",
+ "normalized_hypothesis": "it is the strap hanners that will keep us from going under",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ }
+ },
+ {
+ "id": "release500_0468",
+ "category": "phonetic_balance",
+ "text": "I never allow what can't be changed to annoy me.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0468.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I NEVER ALLOW WHAT CAN'T BE CHANGED TO ANNOY ME",
+ "raw": {
+ "normalized_reference": "i never allow what can't be changed to annoy me.",
+ "normalized_hypothesis": "i never allow what can't be changed to annoy me",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i never allow what can not be changed to annoy me",
+ "normalized_hypothesis": "i never allow what can not be changed to annoy me",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0469",
+ "category": "phonetic_balance",
+ "text": "After all superfluous flesh is gone what is left is stringy and resistant.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0469.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "AFTER ALL SUPERFLUOUS FLESH IS GONE WHAT IS LEFT IS STRINGY AND RESISTANT",
+ "raw": {
+ "normalized_reference": "after all superfluous flesh is gone what is left is stringy and resistant.",
+ "normalized_hypothesis": "after all superfluous flesh is gone what is left is stringy and resistant",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "after all superfluous flesh is gone what is left is stringy and resistant",
+ "normalized_hypothesis": "after all superfluous flesh is gone what is left is stringy and resistant",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0470",
+ "category": "phonetic_balance",
+ "text": "Then he shouted, Shut up.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0470.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THEN HE SHOUTED SHUT UP",
+ "raw": {
+ "normalized_reference": "then he shouted, shut up.",
+ "normalized_hypothesis": "then he shouted shut up",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "then he shouted shut up",
+ "normalized_hypothesis": "then he shouted shut up",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0471",
+ "category": "phonetic_balance",
+ "text": "He had worshiped her, as Dante might have worshiped Beatrice.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0471.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE HAD WORSHIPPED HER AS DANTES MIGHT HAVE WORSHIPPED BEATRICE",
+ "raw": {
+ "normalized_reference": "he had worshiped her, as dante might have worshiped beatrice.",
+ "normalized_hypothesis": "he had worshipped her as dantes might have worshipped beatrice",
+ "errors": 3,
+ "words": 10,
+ "wer": 0.3
+ },
+ "semantic": {
+ "normalized_reference": "he had worshiped her as dante might have worshiped beatrice",
+ "normalized_hypothesis": "he had worshipped her as dantes might have worshipped beatrice",
+ "errors": 3,
+ "words": 10,
+ "wer": 0.3
+ }
+ },
+ {
+ "id": "release500_0472",
+ "category": "phonetic_balance",
+ "text": "He looked like one who had passed through an uncomfortable hour or two.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0472.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE LOOKED LIKE ONE WHO HAD PASSED THROUGH AN UNCOMFORTABLE HOUR OR TWO",
+ "raw": {
+ "normalized_reference": "he looked like one who had passed through an uncomfortable hour or two.",
+ "normalized_hypothesis": "he looked like one who had passed through an uncomfortable hour or two",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he looked like one who had passed through an uncomfortable hour or 2",
+ "normalized_hypothesis": "he looked like one who had passed through an uncomfortable hour or 2",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0473",
+ "category": "phonetic_balance",
+ "text": "The history of our westward-faring race is written in it.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0473.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE HISTORY OF OUR WESTWARD FARING RACE IS WRITTEN IN IT",
+ "raw": {
+ "normalized_reference": "the history of our westward-faring race is written in it.",
+ "normalized_hypothesis": "the history of our westward faring race is written in it",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the history of our westward faring race is written in it",
+ "normalized_hypothesis": "the history of our westward faring race is written in it",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0474",
+ "category": "phonetic_balance",
+ "text": "He may desire to escape pain, or to enjoy pleasure.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0474.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE MAY DESIRE TO ESCAPE PAIN OR TO ENJOY PLEASURE",
+ "raw": {
+ "normalized_reference": "he may desire to escape pain, or to enjoy pleasure.",
+ "normalized_hypothesis": "he may desire to escape pain or to enjoy pleasure",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he may desire to escape pain or to enjoy pleasure",
+ "normalized_hypothesis": "he may desire to escape pain or to enjoy pleasure",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0475",
+ "category": "phonetic_balance",
+ "text": "And the air was growing chilly.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0475.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "AND THE AIR WAS GROWING CHILLY",
+ "raw": {
+ "normalized_reference": "and the air was growing chilly.",
+ "normalized_hypothesis": "and the air was growing chilly",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "and the air was growing chilly",
+ "normalized_hypothesis": "and the air was growing chilly",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0476",
+ "category": "phonetic_balance",
+ "text": "Dig in; you're sure good, was Daylight's answer.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0476.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "DEGIN YOURE SURE GOOD WAS DEYLIGHT'S ANSWER",
+ "raw": {
+ "normalized_reference": "dig in; you're sure good, was daylight's answer.",
+ "normalized_hypothesis": "degin youre sure good was deylight's answer",
+ "errors": 4,
+ "words": 8,
+ "wer": 0.5
+ },
+ "semantic": {
+ "normalized_reference": "dig in you are sure good was daylight is answer",
+ "normalized_hypothesis": "degin youre sure good was deylight is answer",
+ "errors": 5,
+ "words": 10,
+ "wer": 0.5
+ }
+ },
+ {
+ "id": "release500_0477",
+ "category": "phonetic_balance",
+ "text": "My idea was that he would have more influence over the natives.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0477.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "MY IDEA WAS THAT HE WOULD HAVE MORE INFLUENCE OVER THE NATIVES",
+ "raw": {
+ "normalized_reference": "my idea was that he would have more influence over the natives.",
+ "normalized_hypothesis": "my idea was that he would have more influence over the natives",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "my idea was that he would have more influence over the natives",
+ "normalized_hypothesis": "my idea was that he would have more influence over the natives",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0478",
+ "category": "phonetic_balance",
+ "text": "Now run along, and tell them to hurry.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0478.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "NOW RUN ALONG AND TELL THEM TO HURRY",
+ "raw": {
+ "normalized_reference": "now run along, and tell them to hurry.",
+ "normalized_hypothesis": "now run along and tell them to hurry",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "now run along and tell them to hurry",
+ "normalized_hypothesis": "now run along and tell them to hurry",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0479",
+ "category": "phonetic_balance",
+ "text": "I have no idea, replied Philip.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0479.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I HAVE NO IDEA REPLIED PHILIP",
+ "raw": {
+ "normalized_reference": "i have no idea, replied philip.",
+ "normalized_hypothesis": "i have no idea replied philip",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i have no idea replied philip",
+ "normalized_hypothesis": "i have no idea replied philip",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0480",
+ "category": "phonetic_balance",
+ "text": "For the rest, he was a mere automaton.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0480.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "FOR THE REST HE WAS A MERE AUTOMATON",
+ "raw": {
+ "normalized_reference": "for the rest, he was a mere automaton.",
+ "normalized_hypothesis": "for the rest he was a mere automaton",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "for the rest he was a mere automaton",
+ "normalized_hypothesis": "for the rest he was a mere automaton",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0481",
+ "category": "phonetic_balance",
+ "text": "Why, the average review is more nauseating than cod liver oil.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0481.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WHY THE AVERAGE REVIEW IS MORE NAUSEATING THAN CARD LIVER OIL",
+ "raw": {
+ "normalized_reference": "why, the average review is more nauseating than cod liver oil.",
+ "normalized_hypothesis": "why the average review is more nauseating than card liver oil",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ },
+ "semantic": {
+ "normalized_reference": "why the average review is more nauseating than cod liver oil",
+ "normalized_hypothesis": "why the average review is more nauseating than card liver oil",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ }
+ },
+ {
+ "id": "release500_0482",
+ "category": "phonetic_balance",
+ "text": "I have been robbed, sir, I amended.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0482.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IHAVE BEEN ROBBED SIR I AMMENDED",
+ "raw": {
+ "normalized_reference": "i have been robbed, sir, i amended.",
+ "normalized_hypothesis": "ihave been robbed sir i ammended",
+ "errors": 3,
+ "words": 7,
+ "wer": 0.42857142857142855
+ },
+ "semantic": {
+ "normalized_reference": "i have been robbed sir i amended",
+ "normalized_hypothesis": "ihave been robbed sir i ammended",
+ "errors": 3,
+ "words": 7,
+ "wer": 0.42857142857142855
+ }
+ },
+ {
+ "id": "release500_0483",
+ "category": "phonetic_balance",
+ "text": "He moved away as quietly as he had come.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0483.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE MOVED AWAY AS QUIETLY AS HE HAD COME",
+ "raw": {
+ "normalized_reference": "he moved away as quietly as he had come.",
+ "normalized_hypothesis": "he moved away as quietly as he had come",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he moved away as quietly as he had come",
+ "normalized_hypothesis": "he moved away as quietly as he had come",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0484",
+ "category": "phonetic_balance",
+ "text": "A trickle of fresh blood ran over his face.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0484.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "A TRICKLE OF FRESH BLOOD RAN OVER HIS FACE",
+ "raw": {
+ "normalized_reference": "a trickle of fresh blood ran over his face.",
+ "normalized_hypothesis": "a trickle of fresh blood ran over his face",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "a trickle of fresh blood ran over his face",
+ "normalized_hypothesis": "a trickle of fresh blood ran over his face",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0485",
+ "category": "phonetic_balance",
+ "text": "Suppose you saw me at work through the window.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0485.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SUPPOSE YOU SAW ME AT WORK THROUGH THE WINDOW",
+ "raw": {
+ "normalized_reference": "suppose you saw me at work through the window.",
+ "normalized_hypothesis": "suppose you saw me at work through the window",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "suppose you saw me at work through the window",
+ "normalized_hypothesis": "suppose you saw me at work through the window",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0486",
+ "category": "phonetic_balance",
+ "text": "But who was Eileen's double.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0486.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "BUT WHO WAS ILEEN STUBBLE",
+ "raw": {
+ "normalized_reference": "but who was eileen's double.",
+ "normalized_hypothesis": "but who was ileen stubble",
+ "errors": 2,
+ "words": 5,
+ "wer": 0.4
+ },
+ "semantic": {
+ "normalized_reference": "but who was eileen is double",
+ "normalized_hypothesis": "but who was ileen stubble",
+ "errors": 3,
+ "words": 6,
+ "wer": 0.5
+ }
+ },
+ {
+ "id": "release500_0487",
+ "category": "phonetic_balance",
+ "text": "Darkness hid him from Jeanne.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0487.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "DARKNESS HID HIM FROM JEAN",
+ "raw": {
+ "normalized_reference": "darkness hid him from jeanne.",
+ "normalized_hypothesis": "darkness hid him from jean",
+ "errors": 1,
+ "words": 5,
+ "wer": 0.2
+ },
+ "semantic": {
+ "normalized_reference": "darkness hid him from jeanne",
+ "normalized_hypothesis": "darkness hid him from jean",
+ "errors": 1,
+ "words": 5,
+ "wer": 0.2
+ }
+ },
+ {
+ "id": "release500_0488",
+ "category": "phonetic_balance",
+ "text": "By this time Charley was as enraged as the Greek.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0488.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "BY THIS TIME CHARLEY WAS AS ENRAGED AS THE GREEK",
+ "raw": {
+ "normalized_reference": "by this time charley was as enraged as the greek.",
+ "normalized_hypothesis": "by this time charley was as enraged as the greek",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "by this time charley was as enraged as the greek",
+ "normalized_hypothesis": "by this time charley was as enraged as the greek",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0489",
+ "category": "phonetic_balance",
+ "text": "I am writing these lines in Honolulu, Hawaii.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0489.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I AM WRITING THESE LINES IN HANONLOLU HAWA'I",
+ "raw": {
+ "normalized_reference": "i am writing these lines in honolulu, hawaii.",
+ "normalized_hypothesis": "i am writing these lines in hanonlolu hawa'i",
+ "errors": 2,
+ "words": 8,
+ "wer": 0.25
+ },
+ "semantic": {
+ "normalized_reference": "i am writing these lines in honolulu hawaii",
+ "normalized_hypothesis": "i am writing these lines in hanonlolu hawa i",
+ "errors": 3,
+ "words": 8,
+ "wer": 0.375
+ }
+ },
+ {
+ "id": "release500_0490",
+ "category": "phonetic_balance",
+ "text": "Nobody knows how the natives got them.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0490.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "NOBODY KNOWS HOW THE NATIVES GOT THEM",
+ "raw": {
+ "normalized_reference": "nobody knows how the natives got them.",
+ "normalized_hypothesis": "nobody knows how the natives got them",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "nobody knows how the natives got them",
+ "normalized_hypothesis": "nobody knows how the natives got them",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0491",
+ "category": "phonetic_balance",
+ "text": "And how would we ever find ourselves.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0491.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "AND HOW WOULD WE EVER FIND OURSELVES",
+ "raw": {
+ "normalized_reference": "and how would we ever find ourselves.",
+ "normalized_hypothesis": "and how would we ever find ourselves",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "and how would we ever find ourselves",
+ "normalized_hypothesis": "and how would we ever find ourselves",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0492",
+ "category": "phonetic_balance",
+ "text": "They are coming ashore, whoever they are.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0492.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THEY ARE COMING ASHORE WHOEVER THEY ARE",
+ "raw": {
+ "normalized_reference": "they are coming ashore, whoever they are.",
+ "normalized_hypothesis": "they are coming ashore whoever they are",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "they are coming ashore whoever they are",
+ "normalized_hypothesis": "they are coming ashore whoever they are",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0493",
+ "category": "phonetic_balance",
+ "text": "This is my fifth voyage.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0493.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THIS IS MY FIFTH VOYAGE",
+ "raw": {
+ "normalized_reference": "this is my fifth voyage.",
+ "normalized_hypothesis": "this is my fifth voyage",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "this is my 5th voyage",
+ "normalized_hypothesis": "this is my 5th voyage",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0494",
+ "category": "phonetic_balance",
+ "text": "And yet if she came he had no words to say.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0494.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "AND YET IF SHE CAME HE HAD NO WORDS TO SAY",
+ "raw": {
+ "normalized_reference": "and yet if she came he had no words to say.",
+ "normalized_hypothesis": "and yet if she came he had no words to say",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "and yet if she came he had no words to say",
+ "normalized_hypothesis": "and yet if she came he had no words to say",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0495",
+ "category": "phonetic_balance",
+ "text": "You must sleep, he urged.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0495.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "YOU MUST SLEEP HE URGED",
+ "raw": {
+ "normalized_reference": "you must sleep, he urged.",
+ "normalized_hypothesis": "you must sleep he urged",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "you must sleep he urged",
+ "normalized_hypothesis": "you must sleep he urged",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0496",
+ "category": "phonetic_balance",
+ "text": "Could the incident have anything to do with Jeanne and Pierre.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0496.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "COULD THE INCIDENT HAVE ANYTHING TO DO WITH JEAN AND PIERRE",
+ "raw": {
+ "normalized_reference": "could the incident have anything to do with jeanne and pierre.",
+ "normalized_hypothesis": "could the incident have anything to do with jean and pierre",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ },
+ "semantic": {
+ "normalized_reference": "could the incident have anything to do with jeanne and pierre",
+ "normalized_hypothesis": "could the incident have anything to do with jean and pierre",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ }
+ },
+ {
+ "id": "release500_0497",
+ "category": "phonetic_balance",
+ "text": "Each day she became a more vital part of him.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0497.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "EACH DAY SHE BECAME A MORE VITAL PART OF HIM",
+ "raw": {
+ "normalized_reference": "each day she became a more vital part of him.",
+ "normalized_hypothesis": "each day she became a more vital part of him",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "each day she became a more vital part of him",
+ "normalized_hypothesis": "each day she became a more vital part of him",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0498",
+ "category": "phonetic_balance",
+ "text": "But why continue the tirade, for tirade it was.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0498.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "BUT WHY CONTINUE THE TARADE FOR TARADE IT WAS",
+ "raw": {
+ "normalized_reference": "but why continue the tirade, for tirade it was.",
+ "normalized_hypothesis": "but why continue the tarade for tarade it was",
+ "errors": 2,
+ "words": 9,
+ "wer": 0.2222222222222222
+ },
+ "semantic": {
+ "normalized_reference": "but why continue the tirade for tirade it was",
+ "normalized_hypothesis": "but why continue the tarade for tarade it was",
+ "errors": 2,
+ "words": 9,
+ "wer": 0.2222222222222222
+ }
+ },
+ {
+ "id": "release500_0499",
+ "category": "phonetic_balance",
+ "text": "That is what distinguishes all of us from the lower animals.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0499.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THAT IS WHAT DISTINGUISHES ALL OF US FROM THE LOWER ANIMALS",
+ "raw": {
+ "normalized_reference": "that is what distinguishes all of us from the lower animals.",
+ "normalized_hypothesis": "that is what distinguishes all of us from the lower animals",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "that is what distinguishes all of us from the lower animals",
+ "normalized_hypothesis": "that is what distinguishes all of us from the lower animals",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0500",
+ "category": "phonetic_balance",
+ "text": "There was something pathetic in the girl's attitude now.",
+ "model": "piper-danny-low",
+ "audio": "benchmark/audio/piper-danny-low/release500_0500.wav",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THERE WAS SOMETHING PATHETIC IN THE GIRL'S ATTITUDE NOW",
+ "raw": {
+ "normalized_reference": "there was something pathetic in the girl's attitude now.",
+ "normalized_hypothesis": "there was something pathetic in the girl's attitude now",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "there was something pathetic in the girl is attitude now",
+ "normalized_hypothesis": "there was something pathetic in the girl is attitude now",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0001",
+ "category": "conversational",
+ "text": "It wasn't until later that I realized what had actually happened.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0001.wav",
+ "render_seconds": 76.20106935000513,
+ "audio_seconds": 4.31891156462585,
+ "rtf": 17.64358177049325,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT WASN'T UNTIL LATER THAT I REALIZED WHAT HAD ACTUALLY HAPPENED",
+ "raw": {
+ "normalized_reference": "it wasn't until later that i realized what had actually happened.",
+ "normalized_hypothesis": "it wasn't until later that i realized what had actually happened",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it was not until later that i realized what had actually happened",
+ "normalized_hypothesis": "it was not until later that i realized what had actually happened",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0002",
+ "category": "descriptive",
+ "text": "The light at that time of afternoon is perfect for about twenty minutes.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0002.wav",
+ "render_seconds": 89.0924288610695,
+ "audio_seconds": 4.806530612244898,
+ "rtf": 18.535704034444656,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE LIGHT AT THAT TIME OF AFTERNOON IS PERFECT FOR ABOUT TWENTY MINUTES",
+ "raw": {
+ "normalized_reference": "the light at that time of afternoon is perfect for about twenty minutes.",
+ "normalized_hypothesis": "the light at that time of afternoon is perfect for about twenty minutes",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the light at that time of afternoon is perfect for about 20 minutes",
+ "normalized_hypothesis": "the light at that time of afternoon is perfect for about 20 minutes",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0003",
+ "category": "emotional",
+ "text": "That was the scariest thing I've ever done, and also maybe the best.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0003.wav",
+ "render_seconds": 76.90116830705665,
+ "audio_seconds": 4.527891156462585,
+ "rtf": 16.983881822622184,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THAT WAS THE SCARIEST THING I'VE EVER DONE AND ALSO MAYBE THE BEST",
+ "raw": {
+ "normalized_reference": "that was the scariest thing i've ever done, and also maybe the best.",
+ "normalized_hypothesis": "that was the scariest thing i've ever done and also maybe the best",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "that was the scariest thing i have ever done and also maybe the best",
+ "normalized_hypothesis": "that was the scariest thing i have ever done and also maybe the best",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0004",
+ "category": "homographs",
+ "text": "After I record the record, please present the present to the bass player.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0004.wav",
+ "render_seconds": 78.39508487901185,
+ "audio_seconds": 5.015510204081632,
+ "rtf": 15.63053043242017,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "AFTER I REQUIRED THE RECORD PLEASE PRESENT THE PRESENT TO THE BASE PLAYER",
+ "raw": {
+ "normalized_reference": "after i record the record, please present the present to the bass player.",
+ "normalized_hypothesis": "after i required the record please present the present to the base player",
+ "errors": 2,
+ "words": 13,
+ "wer": 0.15384615384615385
+ },
+ "semantic": {
+ "normalized_reference": "after i record the record please present the present to the bass player",
+ "normalized_hypothesis": "after i required the record please present the present to the base player",
+ "errors": 2,
+ "words": 13,
+ "wer": 0.15384615384615385
+ }
+ },
+ {
+ "id": "release500_0005",
+ "category": "long",
+ "text": "If I'm being completely honest, I had no idea what I was doing for most of it, and I think that might be the most normal thing about the whole experience.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0005.wav",
+ "render_seconds": 108.40687325003091,
+ "audio_seconds": 9.47374149659864,
+ "rtf": 11.442878538426688,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IF I 'M BEING COMPLETELY HONEST I HAD NO IDEA WHAT I WAS DOING FOR MOST OF IT AND I THINK THAT MIGHT BE THE MOST NORMAL THING ABOUT THE WHOLE EXPERIENCE",
+ "raw": {
+ "normalized_reference": "if i'm being completely honest, i had no idea what i was doing for most of it, and i think that might be the most normal thing about the whole experience.",
+ "normalized_hypothesis": "if i 'm being completely honest i had no idea what i was doing for most of it and i think that might be the most normal thing about the whole experience",
+ "errors": 2,
+ "words": 31,
+ "wer": 0.06451612903225806
+ },
+ "semantic": {
+ "normalized_reference": "if i am being completely honest i had no idea what i was doing for most of it and i think that might be the most normal thing about the whole experience",
+ "normalized_hypothesis": "if i am being completely honest i had no idea what i was doing for most of it and i think that might be the most normal thing about the whole experience",
+ "errors": 0,
+ "words": 32,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0006",
+ "category": "long_range",
+ "text": "The committee postponed its decision after discovering that three pages of the report had been printed in the wrong order.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0006.wav",
+ "render_seconds": 122.28961312398314,
+ "audio_seconds": 7.9412244897959186,
+ "rtf": 15.399339646060788,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE COMMITTEE POSTPONED ITS DECISION AFTER DISCOVERING THAT THREE PAGES OF THE REPORT HAD BEEN PRINTED IN THE WRONG ORDER",
+ "raw": {
+ "normalized_reference": "the committee postponed its decision after discovering that three pages of the report had been printed in the wrong order.",
+ "normalized_hypothesis": "the committee postponed its decision after discovering that three pages of the report had been printed in the wrong order",
+ "errors": 0,
+ "words": 20,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the committee postponed its decision after discovering that 3 pages of the report had been printed in the wrong order",
+ "normalized_hypothesis": "the committee postponed its decision after discovering that 3 pages of the report had been printed in the wrong order",
+ "errors": 0,
+ "words": 20,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0007",
+ "category": "names_places",
+ "text": "Gwendolyn photographed the eucalyptus trees outside Ljubljana.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0007.wav",
+ "render_seconds": 75.72085215803236,
+ "audio_seconds": 4.527891156462585,
+ "rtf": 16.723205028892362,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "GWENDOLEN PHOTOGRAPHED THE EUCALYPTUS TREES OUTSIDE ILBRAJANA",
+ "raw": {
+ "normalized_reference": "gwendolyn photographed the eucalyptus trees outside ljubljana.",
+ "normalized_hypothesis": "gwendolen photographed the eucalyptus trees outside ilbrajana",
+ "errors": 2,
+ "words": 7,
+ "wer": 0.2857142857142857
+ },
+ "semantic": {
+ "normalized_reference": "gwendolyn photographed the eucalyptus trees outside ljubljana",
+ "normalized_hypothesis": "gwendolen photographed the eucalyptus trees outside ilbrajana",
+ "errors": 2,
+ "words": 7,
+ "wer": 0.2857142857142857
+ }
+ },
+ {
+ "id": "release500_0008",
+ "category": "narrative",
+ "text": "He had this theory, and the more time passes, the more I think he was right.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0008.wav",
+ "render_seconds": 92.60325879510492,
+ "audio_seconds": 5.294149659863946,
+ "rtf": 17.491620891859096,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "YOU HAD THIS THEORY AND THE MORE TIME PASSES THE MORE I THINK HE WAS RIGHT",
+ "raw": {
+ "normalized_reference": "he had this theory, and the more time passes, the more i think he was right.",
+ "normalized_hypothesis": "you had this theory and the more time passes the more i think he was right",
+ "errors": 1,
+ "words": 16,
+ "wer": 0.0625
+ },
+ "semantic": {
+ "normalized_reference": "he had this theory and the more time passes the more i think he was right",
+ "normalized_hypothesis": "you had this theory and the more time passes the more i think he was right",
+ "errors": 1,
+ "words": 16,
+ "wer": 0.0625
+ }
+ },
+ {
+ "id": "release500_0009",
+ "category": "numbers",
+ "text": "The package weighs twelve point six kilograms and arrived on July twenty-first.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0009.wav",
+ "render_seconds": 94.99756314198021,
+ "audio_seconds": 5.7817687074829935,
+ "rtf": 16.430536735070465,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE PACKEG WEIGHS TWELVE POINT SIX KILOGRAMS AND ARRIVED ON JULY TWENTY FIRST",
+ "raw": {
+ "normalized_reference": "the package weighs twelve point six kilograms and arrived on july twenty-first.",
+ "normalized_hypothesis": "the packeg weighs twelve point six kilograms and arrived on july twenty first",
+ "errors": 1,
+ "words": 13,
+ "wer": 0.07692307692307693
+ },
+ "semantic": {
+ "normalized_reference": "the package weighs 12.6 kilograms and arrived on july 21st",
+ "normalized_hypothesis": "the packeg weighs 12.6 kilograms and arrived on july 21st",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ }
+ },
+ {
+ "id": "release500_0010",
+ "category": "punctuation",
+ "text": "First, close the window; second, turn off the lamp; finally, lock the door.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0010.wav",
+ "render_seconds": 90.09958808403462,
+ "audio_seconds": 5.224489795918367,
+ "rtf": 17.245624281709752,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "FIRST CLOSE THE WINDOW SECOND TURN OFF THE LAMP FINALLY LOCK THE DOOR",
+ "raw": {
+ "normalized_reference": "first, close the window; second, turn off the lamp; finally, lock the door.",
+ "normalized_hypothesis": "first close the window second turn off the lamp finally lock the door",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "1st close the window 2nd turn off the lamp finally lock the door",
+ "normalized_hypothesis": "1st close the window 2nd turn off the lamp finally lock the door",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0011",
+ "category": "question",
+ "text": "When did you first realize something was wrong?",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0011.wav",
+ "render_seconds": 31.208350338041782,
+ "audio_seconds": 3.1346938775510202,
+ "rtf": 9.955788844818017,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WHEN DID YOR FIRST REALIZE SOMETHING WAS WRONG",
+ "raw": {
+ "normalized_reference": "when did you first realize something was wrong?",
+ "normalized_hypothesis": "when did yor first realize something was wrong",
+ "errors": 1,
+ "words": 8,
+ "wer": 0.125
+ },
+ "semantic": {
+ "normalized_reference": "when did you 1st realize something was wrong",
+ "normalized_hypothesis": "when did yor 1st realize something was wrong",
+ "errors": 1,
+ "words": 8,
+ "wer": 0.125
+ }
+ },
+ {
+ "id": "release500_0012",
+ "category": "rare_words",
+ "text": "The archaeologist catalogued a Byzantine reliquary and several anachronistic fragments.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0012.wav",
+ "render_seconds": 103.6040112740593,
+ "audio_seconds": 6.478367346938776,
+ "rtf": 15.992302647520495,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE ARCHEOLOGIST CATALOGUED A BIZANTINE RILIQUARY AND SEVERAL ANACRINISTIC FRAGMENTS",
+ "raw": {
+ "normalized_reference": "the archaeologist catalogued a byzantine reliquary and several anachronistic fragments.",
+ "normalized_hypothesis": "the archeologist catalogued a bizantine riliquary and several anacrinistic fragments",
+ "errors": 4,
+ "words": 10,
+ "wer": 0.4
+ },
+ "semantic": {
+ "normalized_reference": "the archaeologist catalogued a byzantine reliquary and several anachronistic fragments",
+ "normalized_hypothesis": "the archeologist catalogued a bizantine riliquary and several anacrinistic fragments",
+ "errors": 4,
+ "words": 10,
+ "wer": 0.4
+ }
+ },
+ {
+ "id": "release500_0013",
+ "category": "short",
+ "text": "I have no idea.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0013.wav",
+ "render_seconds": 20.591476502013393,
+ "audio_seconds": 1.5325170068027212,
+ "rtf": 13.436377156409662,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "V NO IBOM",
+ "raw": {
+ "normalized_reference": "i have no idea.",
+ "normalized_hypothesis": "v no ibom",
+ "errors": 3,
+ "words": 4,
+ "wer": 0.75
+ },
+ "semantic": {
+ "normalized_reference": "i have no idea",
+ "normalized_hypothesis": "v no ibom",
+ "errors": 3,
+ "words": 4,
+ "wer": 0.75
+ }
+ },
+ {
+ "id": "release500_0014",
+ "category": "technical",
+ "text": "The system runs on three core components that all have to stay in sync.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0014.wav",
+ "render_seconds": 94.01606339204591,
+ "audio_seconds": 4.9458503401360545,
+ "rtf": 19.00907971862724,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE SYSTEM RUNS ON THREE CORPS COMPONENTS THAT ALL HAVE TO STAY AND SINK",
+ "raw": {
+ "normalized_reference": "the system runs on three core components that all have to stay in sync.",
+ "normalized_hypothesis": "the system runs on three corps components that all have to stay and sink",
+ "errors": 3,
+ "words": 14,
+ "wer": 0.21428571428571427
+ },
+ "semantic": {
+ "normalized_reference": "the system runs on 3 core components that all have to stay in sync",
+ "normalized_hypothesis": "the system runs on 3 corps components that all have to stay and sink",
+ "errors": 3,
+ "words": 14,
+ "wer": 0.21428571428571427
+ }
+ },
+ {
+ "id": "release500_0015",
+ "category": "conversational",
+ "text": "We improvised the entire second half and somehow nobody noticed.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0015.wav",
+ "render_seconds": 81.10544405097608,
+ "audio_seconds": 4.597551020408163,
+ "rtf": 17.641010114070614,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WE IMPROVISED THE ENTIRE SECOND HALF AND SOMEHOW NOBODY NOTICED",
+ "raw": {
+ "normalized_reference": "we improvised the entire second half and somehow nobody noticed.",
+ "normalized_hypothesis": "we improvised the entire second half and somehow nobody noticed",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "we improvised the entire 2nd half and somehow nobody noticed",
+ "normalized_hypothesis": "we improvised the entire 2nd half and somehow nobody noticed",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0016",
+ "category": "descriptive",
+ "text": "There's a smell in old buildings that's impossible to describe but immediately familiar.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0016.wav",
+ "render_seconds": 92.49169414199423,
+ "audio_seconds": 5.851428571428571,
+ "rtf": 15.806686010594717,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THERE'S A SMELL IN OLD BUILDINGS THAT'S IMPOSSIBLE TO DESCRIBE BUT IMMEDIATELY FAMILIAR",
+ "raw": {
+ "normalized_reference": "there's a smell in old buildings that's impossible to describe but immediately familiar.",
+ "normalized_hypothesis": "there's a smell in old buildings that's impossible to describe but immediately familiar",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "there is a smell in old buildings that is impossible to describe but immediately familiar",
+ "normalized_hypothesis": "there is a smell in old buildings that is impossible to describe but immediately familiar",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0017",
+ "category": "emotional",
+ "text": "I specifically asked for one thing. One thing. And it still didn't happen.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0017.wav",
+ "render_seconds": 99.40190737403464,
+ "audio_seconds": 5.712108843537415,
+ "rtf": 17.401963109735963,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I SPECIFICALLY ASKED FOR ONE THING ONE THING NIT'S DILLTEN HAPPEN",
+ "raw": {
+ "normalized_reference": "i specifically asked for one thing. one thing. and it still didn't happen.",
+ "normalized_hypothesis": "i specifically asked for one thing one thing nit's dillten happen",
+ "errors": 4,
+ "words": 13,
+ "wer": 0.3076923076923077
+ },
+ "semantic": {
+ "normalized_reference": "i specifically asked for one thing one thing and it still did not happen",
+ "normalized_hypothesis": "i specifically asked for one thing one thing nit is dillten happen",
+ "errors": 5,
+ "words": 14,
+ "wer": 0.35714285714285715
+ }
+ },
+ {
+ "id": "release500_0018",
+ "category": "homographs",
+ "text": "They refuse to place the refuse beside the produce aisle.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0018.wav",
+ "render_seconds": 74.38826269307174,
+ "audio_seconds": 3.9706122448979593,
+ "rtf": 18.734708429073372,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THEY REFUSED TO PLACE THE REFUSE BESIDE THE PRODUCE ISLE",
+ "raw": {
+ "normalized_reference": "they refuse to place the refuse beside the produce aisle.",
+ "normalized_hypothesis": "they refused to place the refuse beside the produce isle",
+ "errors": 2,
+ "words": 10,
+ "wer": 0.2
+ },
+ "semantic": {
+ "normalized_reference": "they refuse to place the refuse beside the produce aisle",
+ "normalized_hypothesis": "they refused to place the refuse beside the produce isle",
+ "errors": 2,
+ "words": 10,
+ "wer": 0.2
+ }
+ },
+ {
+ "id": "release500_0019",
+ "category": "long",
+ "text": "There were days last year when I genuinely didn't know how I was going to get through it, and somehow I did, and I still don't fully understand how.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0019.wav",
+ "render_seconds": 99.78391260211356,
+ "audio_seconds": 8.916462585034013,
+ "rtf": 11.19097530556541,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THERE WERE DAYS LAST YEAR WHEN I GENUALLY DIDN'T KNOW HOW I WAS GOING TO GET THROUGH IT AND SOMEHOW I DID AND I STILL DON'T FULLY UNDERSTAND HOW",
+ "raw": {
+ "normalized_reference": "there were days last year when i genuinely didn't know how i was going to get through it, and somehow i did, and i still don't fully understand how.",
+ "normalized_hypothesis": "there were days last year when i genually didn't know how i was going to get through it and somehow i did and i still don't fully understand how",
+ "errors": 1,
+ "words": 29,
+ "wer": 0.034482758620689655
+ },
+ "semantic": {
+ "normalized_reference": "there were days last year when i genuinely did not know how i was going to get through it and somehow i did and i still do not fully understand how",
+ "normalized_hypothesis": "there were days last year when i genually did not know how i was going to get through it and somehow i did and i still do not fully understand how",
+ "errors": 1,
+ "words": 31,
+ "wer": 0.03225806451612903
+ }
+ },
+ {
+ "id": "release500_0020",
+ "category": "long_range",
+ "text": "Before announcing the winner, the judge thanked the volunteers who had worked through the storm to repair the damaged stage.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0020.wav",
+ "render_seconds": 125.08898339990992,
+ "audio_seconds": 7.801904761904762,
+ "rtf": 16.033133858631032,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "BEFORE ANNOUNCING THE WINNER THE JUDGE THANKED THE VOLUNTEERS WHO HAD WORKED THROUGH THE STORM TO REPAIR THE DAMAGED STAGE",
+ "raw": {
+ "normalized_reference": "before announcing the winner, the judge thanked the volunteers who had worked through the storm to repair the damaged stage.",
+ "normalized_hypothesis": "before announcing the winner the judge thanked the volunteers who had worked through the storm to repair the damaged stage",
+ "errors": 0,
+ "words": 20,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "before announcing the winner the judge thanked the volunteers who had worked through the storm to repair the damaged stage",
+ "normalized_hypothesis": "before announcing the winner the judge thanked the volunteers who had worked through the storm to repair the damaged stage",
+ "errors": 0,
+ "words": 20,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0021",
+ "category": "names_places",
+ "text": "Professor Nguyen mailed the fluorescent poster to Reykjavik on Thursday.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0021.wav",
+ "render_seconds": 90.39663011499215,
+ "audio_seconds": 5.154829931972789,
+ "rtf": 17.536297280014576,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "PROFESSOR INGWEN MAILED THE FLORESCENT POSTER TO RAT JUVIK ON THURSDAY",
+ "raw": {
+ "normalized_reference": "professor nguyen mailed the fluorescent poster to reykjavik on thursday.",
+ "normalized_hypothesis": "professor ingwen mailed the florescent poster to rat juvik on thursday",
+ "errors": 4,
+ "words": 10,
+ "wer": 0.4
+ },
+ "semantic": {
+ "normalized_reference": "professor nguyen mailed the fluorescent poster to reykjavik on thursday",
+ "normalized_hypothesis": "professor ingwen mailed the florescent poster to rat juvik on thursday",
+ "errors": 4,
+ "words": 10,
+ "wer": 0.4
+ }
+ },
+ {
+ "id": "release500_0022",
+ "category": "narrative",
+ "text": "The plan was simple. The execution was anything but.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0022.wav",
+ "render_seconds": 79.6991829710314,
+ "audio_seconds": 3.9706122448979593,
+ "rtf": 20.072265448090764,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE PLAN WAS SIMPLE THE EXECUTION WAS ANYTHING BUT",
+ "raw": {
+ "normalized_reference": "the plan was simple. the execution was anything but.",
+ "normalized_hypothesis": "the plan was simple the execution was anything but",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the plan was simple the execution was anything but",
+ "normalized_hypothesis": "the plan was simple the execution was anything but",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0023",
+ "category": "numbers",
+ "text": "Her score improved from eighty-two percent to ninety-seven percent in two weeks.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0023.wav",
+ "render_seconds": 55.29268986696843,
+ "audio_seconds": 5.50312925170068,
+ "rtf": 10.047499765679834,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HER SCORE IMPROVED FROM EIGHTY TWO PER CENT TO NINETY SEVEN PER CENT IN TWO WEEKS",
+ "raw": {
+ "normalized_reference": "her score improved from eighty-two percent to ninety-seven percent in two weeks.",
+ "normalized_hypothesis": "her score improved from eighty two per cent to ninety seven per cent in two weeks",
+ "errors": 4,
+ "words": 14,
+ "wer": 0.2857142857142857
+ },
+ "semantic": {
+ "normalized_reference": "her score improved from 82% to 97% in 2 weeks",
+ "normalized_hypothesis": "her score improved from 82% to 97% in 2 weeks",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0024",
+ "category": "punctuation",
+ "text": "Why would anyone pack spoons, batteries, and a violin for one short trip?",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0024.wav",
+ "render_seconds": 88.19932402996346,
+ "audio_seconds": 5.294149659863946,
+ "rtf": 16.659771577411377,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WHY WOULD ANYONE PACK SPOONS BATTERIES AND A BIOLIN FOR ONE SHORT TRIP",
+ "raw": {
+ "normalized_reference": "why would anyone pack spoons, batteries, and a violin for one short trip?",
+ "normalized_hypothesis": "why would anyone pack spoons batteries and a biolin for one short trip",
+ "errors": 1,
+ "words": 13,
+ "wer": 0.07692307692307693
+ },
+ "semantic": {
+ "normalized_reference": "why would anyone pack spoons batteries and a violin for one short trip",
+ "normalized_hypothesis": "why would anyone pack spoons batteries and a biolin for one short trip",
+ "errors": 1,
+ "words": 13,
+ "wer": 0.07692307692307693
+ }
+ },
+ {
+ "id": "release500_0025",
+ "category": "question",
+ "text": "How long have you been sitting on this?",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0025.wav",
+ "render_seconds": 35.49992396996822,
+ "audio_seconds": 2.5774149659863945,
+ "rtf": 13.77346078860148,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HOW LONG HAVE YOU BEEN SHILLINNG",
+ "raw": {
+ "normalized_reference": "how long have you been sitting on this?",
+ "normalized_hypothesis": "how long have you been shillinng",
+ "errors": 3,
+ "words": 8,
+ "wer": 0.375
+ },
+ "semantic": {
+ "normalized_reference": "how long have you been sitting on this",
+ "normalized_hypothesis": "how long have you been shillinng",
+ "errors": 3,
+ "words": 8,
+ "wer": 0.375
+ }
+ },
+ {
+ "id": "release500_0026",
+ "category": "rare_words",
+ "text": "A mischievous chameleon rested beside the chrysanthemum and the bougainvillea.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0026.wav",
+ "render_seconds": 99.68826815101784,
+ "audio_seconds": 5.572789115646258,
+ "rtf": 17.88839772729446,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "A MISCHIEVOUS CHAMLEON RESTED BESIDE THE CHRYSANTHEMUM AND THE BAGANBILLIA",
+ "raw": {
+ "normalized_reference": "a mischievous chameleon rested beside the chrysanthemum and the bougainvillea.",
+ "normalized_hypothesis": "a mischievous chamleon rested beside the chrysanthemum and the baganbillia",
+ "errors": 2,
+ "words": 10,
+ "wer": 0.2
+ },
+ "semantic": {
+ "normalized_reference": "a mischievous chameleon rested beside the chrysanthemum and the bougainvillea",
+ "normalized_hypothesis": "a mischievous chamleon rested beside the chrysanthemum and the baganbillia",
+ "errors": 2,
+ "words": 10,
+ "wer": 0.2
+ }
+ },
+ {
+ "id": "release500_0027",
+ "category": "short",
+ "text": "We don't talk about that.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0027.wav",
+ "render_seconds": 50.900528709054925,
+ "audio_seconds": 1.9504761904761905,
+ "rtf": 26.096462472904136,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WRIHT ON TACK ABOUT THAT",
+ "raw": {
+ "normalized_reference": "we don't talk about that.",
+ "normalized_hypothesis": "wriht on tack about that",
+ "errors": 3,
+ "words": 5,
+ "wer": 0.6
+ },
+ "semantic": {
+ "normalized_reference": "we do not talk about that",
+ "normalized_hypothesis": "wriht on tack about that",
+ "errors": 4,
+ "words": 6,
+ "wer": 0.6666666666666666
+ }
+ },
+ {
+ "id": "release500_0028",
+ "category": "technical",
+ "text": "Hash the UTF eight payload with S H A two fifty-six before opening the checkpoint.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0028.wav",
+ "render_seconds": 98.01105747499969,
+ "audio_seconds": 5.7817687074829935,
+ "rtf": 16.95174304502183,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HASH THE U T F PAY LOAD WITH S A A FIFTY SIX BEFORE OPENING OF THE CHECK JOINT",
+ "raw": {
+ "normalized_reference": "hash the utf eight payload with s h a two fifty-six before opening the checkpoint.",
+ "normalized_hypothesis": "hash the u t f pay load with s a a fifty six before opening of the check joint",
+ "errors": 10,
+ "words": 16,
+ "wer": 0.625
+ },
+ "semantic": {
+ "normalized_reference": "hash the utf 8 payload with s h a 256 before opening the checkpoint",
+ "normalized_hypothesis": "hash the u t f pay load with s a a 56 before opening of the check joint",
+ "errors": 10,
+ "words": 14,
+ "wer": 0.7142857142857143
+ }
+ },
+ {
+ "id": "release500_0029",
+ "category": "conversational",
+ "text": "I thought I had it all figured out, and then this happened.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0029.wav",
+ "render_seconds": 71.49475576600526,
+ "audio_seconds": 4.179591836734694,
+ "rtf": 17.10567886979618,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I THOUGHT I HAD IT ALL FIGURED OUT AND THEN THIS HAPPENED",
+ "raw": {
+ "normalized_reference": "i thought i had it all figured out, and then this happened.",
+ "normalized_hypothesis": "i thought i had it all figured out and then this happened",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i thought i had it all figured out and then this happened",
+ "normalized_hypothesis": "i thought i had it all figured out and then this happened",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0030",
+ "category": "descriptive",
+ "text": "The kind of quiet you only get at three in the morning in a city.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0030.wav",
+ "render_seconds": 74.69945722492412,
+ "audio_seconds": 4.179591836734694,
+ "rtf": 17.872428730572665,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE KIND OF QUIET YOU ONLY GET AT THREE IN THE MORNING IN A CITY",
+ "raw": {
+ "normalized_reference": "the kind of quiet you only get at three in the morning in a city.",
+ "normalized_hypothesis": "the kind of quiet you only get at three in the morning in a city",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the kind of quiet you only get at 3 in the morning in a city",
+ "normalized_hypothesis": "the kind of quiet you only get at 3 in the morning in a city",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0031",
+ "category": "emotional",
+ "text": "The joy of it was almost overwhelming, like too much good all at once.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0031.wav",
+ "render_seconds": 74.61474312399514,
+ "audio_seconds": 4.876190476190477,
+ "rtf": 15.301851617225564,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE JOY OF IT WAS ALMOST OVERWHELMING LIKE TOO MUCH GOOD ALL AT ONCE",
+ "raw": {
+ "normalized_reference": "the joy of it was almost overwhelming, like too much good all at once.",
+ "normalized_hypothesis": "the joy of it was almost overwhelming like too much good all at once",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the joy of it was almost overwhelming like too much good all at once",
+ "normalized_hypothesis": "the joy of it was almost overwhelming like too much good all at once",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0032",
+ "category": "homographs",
+ "text": "Please close the door while I stand close to the microphone.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0032.wav",
+ "render_seconds": 82.40504369803239,
+ "audio_seconds": 4.1099319727891155,
+ "rtf": 20.050220841516754,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "PLEASE CLOSE THE DOOR WHILE I STAND CLOSE TO THE MICROPHONE",
+ "raw": {
+ "normalized_reference": "please close the door while i stand close to the microphone.",
+ "normalized_hypothesis": "please close the door while i stand close to the microphone",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "please close the door while i stand close to the microphone",
+ "normalized_hypothesis": "please close the door while i stand close to the microphone",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0033",
+ "category": "long",
+ "text": "I think the most underrated skill in the world is knowing when to stop explaining yourself and just let people draw their own conclusions.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0033.wav",
+ "render_seconds": 132.40289115905762,
+ "audio_seconds": 8.916462585034013,
+ "rtf": 14.849262237839866,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I THINK THE MOST UNDERRATED SKILL IN THE WORLD IS KNOWING WHEN TO STOP EXPLAINING YOURSELF AND JUST LET PEOPLE DRAW THEIR OWN CONCLUSIONS",
+ "raw": {
+ "normalized_reference": "i think the most underrated skill in the world is knowing when to stop explaining yourself and just let people draw their own conclusions.",
+ "normalized_hypothesis": "i think the most underrated skill in the world is knowing when to stop explaining yourself and just let people draw their own conclusions",
+ "errors": 0,
+ "words": 24,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i think the most underrated skill in the world is knowing when to stop explaining yourself and just let people draw their own conclusions",
+ "normalized_hypothesis": "i think the most underrated skill in the world is knowing when to stop explaining yourself and just let people draw their own conclusions",
+ "errors": 0,
+ "words": 24,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0034",
+ "category": "long_range",
+ "text": "When the final bus disappeared around the corner, Nadia remembered that her umbrella, notebook, and lunch were still inside the quiet office.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0034.wav",
+ "render_seconds": 125.5994538969826,
+ "audio_seconds": 9.404081632653062,
+ "rtf": 13.355844706927401,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WHEN THE FINAL BUST DISAPPEARED AROUND THE CORNER NATTIA REMEMBERED THAT HER UMBRELLA NOTEBOOK AND LUNCH WERE STILL INSIDE THE QUIET OFFICE",
+ "raw": {
+ "normalized_reference": "when the final bus disappeared around the corner, nadia remembered that her umbrella, notebook, and lunch were still inside the quiet office.",
+ "normalized_hypothesis": "when the final bust disappeared around the corner nattia remembered that her umbrella notebook and lunch were still inside the quiet office",
+ "errors": 2,
+ "words": 22,
+ "wer": 0.09090909090909091
+ },
+ "semantic": {
+ "normalized_reference": "when the final bus disappeared around the corner nadia remembered that her umbrella notebook and lunch were still inside the quiet office",
+ "normalized_hypothesis": "when the final bust disappeared around the corner nattia remembered that her umbrella notebook and lunch were still inside the quiet office",
+ "errors": 2,
+ "words": 22,
+ "wer": 0.09090909090909091
+ }
+ },
+ {
+ "id": "release500_0035",
+ "category": "names_places",
+ "text": "Aurelia and Joaquin crossed the Schuylkill River before noon.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0035.wav",
+ "render_seconds": 80.80838598497212,
+ "audio_seconds": 4.31891156462585,
+ "rtf": 18.710359028148474,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "ARELIA AND JOKAN CROSSED THE SHRILKELL RIVER BEFORE NOON",
+ "raw": {
+ "normalized_reference": "aurelia and joaquin crossed the schuylkill river before noon.",
+ "normalized_hypothesis": "arelia and jokan crossed the shrilkell river before noon",
+ "errors": 3,
+ "words": 9,
+ "wer": 0.3333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "aurelia and joaquin crossed the schuylkill river before noon",
+ "normalized_hypothesis": "arelia and jokan crossed the shrilkell river before noon",
+ "errors": 3,
+ "words": 9,
+ "wer": 0.3333333333333333
+ }
+ },
+ {
+ "id": "release500_0036",
+ "category": "narrative",
+ "text": "The village had a name, but nobody used it. Everyone just called it home.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0036.wav",
+ "render_seconds": 91.90063758695032,
+ "audio_seconds": 5.224489795918367,
+ "rtf": 17.59035641312721,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE VILLAGE HAD A NAME BUT NOBODY USED IT EVERYONE JUST CALLED IT HOME",
+ "raw": {
+ "normalized_reference": "the village had a name, but nobody used it. everyone just called it home.",
+ "normalized_hypothesis": "the village had a name but nobody used it everyone just called it home",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the village had a name but nobody used it everyone just called it home",
+ "normalized_hypothesis": "the village had a name but nobody used it everyone just called it home",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0037",
+ "category": "numbers",
+ "text": "Increase the gain by three decibels, wait five seconds, then reduce it by one.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0037.wav",
+ "render_seconds": 50.48591187398415,
+ "audio_seconds": 5.50312925170068,
+ "rtf": 9.174037091420676,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "INCREASE THE GAIN BY THREE DISSIBALS WAIT FIVE SECONDS THEN REDUCE IT BY ONE",
+ "raw": {
+ "normalized_reference": "increase the gain by three decibels, wait five seconds, then reduce it by one.",
+ "normalized_hypothesis": "increase the gain by three dissibals wait five seconds then reduce it by one",
+ "errors": 1,
+ "words": 14,
+ "wer": 0.07142857142857142
+ },
+ "semantic": {
+ "normalized_reference": "increase the gain by 3 decibels wait 5 seconds then reduce it by one",
+ "normalized_hypothesis": "increase the gain by 3 dissibals wait 5 seconds then reduce it by one",
+ "errors": 1,
+ "words": 14,
+ "wer": 0.07142857142857142
+ }
+ },
+ {
+ "id": "release500_0038",
+ "category": "punctuation",
+ "text": "The result was unexpected: clear, stable, and surprisingly natural.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0038.wav",
+ "render_seconds": 86.60410415800288,
+ "audio_seconds": 5.433469387755102,
+ "rtf": 15.939006549512248,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE RESULT WAS UNEXPECTED CLEAR STABLE AND SURPRISINGLY NATURAL",
+ "raw": {
+ "normalized_reference": "the result was unexpected: clear, stable, and surprisingly natural.",
+ "normalized_hypothesis": "the result was unexpected clear stable and surprisingly natural",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the result was unexpected clear stable and surprisingly natural",
+ "normalized_hypothesis": "the result was unexpected clear stable and surprisingly natural",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0039",
+ "category": "question",
+ "text": "Who told you that was the right way to do it?",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0039.wav",
+ "render_seconds": 63.40366674901452,
+ "audio_seconds": 2.9257142857142857,
+ "rtf": 21.671175158354572,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WHO TOLD YOU THAT WAS THE RIGHT WAY TO DO IT",
+ "raw": {
+ "normalized_reference": "who told you that was the right way to do it?",
+ "normalized_hypothesis": "who told you that was the right way to do it",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "who told you that was the right way to do it",
+ "normalized_hypothesis": "who told you that was the right way to do it",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0040",
+ "category": "rare_words",
+ "text": "The meticulous cartographer examined an iridescent archipelago through binoculars.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0040.wav",
+ "render_seconds": 89.78799257206265,
+ "audio_seconds": 5.851428571428571,
+ "rtf": 15.344627636827113,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE METICULOUS CARTOGRAPHER EXAMINED AN IRIDESCENT ARCHIPELAGO THROUGH BONOCULARS",
+ "raw": {
+ "normalized_reference": "the meticulous cartographer examined an iridescent archipelago through binoculars.",
+ "normalized_hypothesis": "the meticulous cartographer examined an iridescent archipelago through bonoculars",
+ "errors": 1,
+ "words": 9,
+ "wer": 0.1111111111111111
+ },
+ "semantic": {
+ "normalized_reference": "the meticulous cartographer examined an iridescent archipelago through binoculars",
+ "normalized_hypothesis": "the meticulous cartographer examined an iridescent archipelago through bonoculars",
+ "errors": 1,
+ "words": 9,
+ "wer": 0.1111111111111111
+ }
+ },
+ {
+ "id": "release500_0041",
+ "category": "short",
+ "text": "I mean. Yeah. Kind of.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0041.wav",
+ "render_seconds": 46.79784265195485,
+ "audio_seconds": 2.2987755102040817,
+ "rtf": 20.357726295683484,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IM AR KIND OF",
+ "raw": {
+ "normalized_reference": "i mean. yeah. kind of.",
+ "normalized_hypothesis": "im ar kind of",
+ "errors": 3,
+ "words": 5,
+ "wer": 0.6
+ },
+ "semantic": {
+ "normalized_reference": "i mean yeah kind of",
+ "normalized_hypothesis": "im ar kind of",
+ "errors": 3,
+ "words": 5,
+ "wer": 0.6
+ }
+ },
+ {
+ "id": "release500_0042",
+ "category": "technical",
+ "text": "The logs are your best friend when something behaves unexpectedly.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0042.wav",
+ "render_seconds": 87.90835851500742,
+ "audio_seconds": 4.527891156462585,
+ "rtf": 19.414856823476697,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE LOGS ARE YOUR BEST FRIEND WHEN SOMETHING BEHAVES UNEXPECTEDLY",
+ "raw": {
+ "normalized_reference": "the logs are your best friend when something behaves unexpectedly.",
+ "normalized_hypothesis": "the logs are your best friend when something behaves unexpectedly",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the logs are your best friend when something behaves unexpectedly",
+ "normalized_hypothesis": "the logs are your best friend when something behaves unexpectedly",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0043",
+ "category": "conversational",
+ "text": "He kept talking and I kept nodding and neither of us was listening.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0043.wav",
+ "render_seconds": 50.20885578193702,
+ "audio_seconds": 4.527891156462585,
+ "rtf": 11.088794771551594,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE KEPT TALKING AND I KEPT NODDING AND NEITHER OF US WAS LISTENING",
+ "raw": {
+ "normalized_reference": "he kept talking and i kept nodding and neither of us was listening.",
+ "normalized_hypothesis": "he kept talking and i kept nodding and neither of us was listening",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he kept talking and i kept nodding and neither of us was listening",
+ "normalized_hypothesis": "he kept talking and i kept nodding and neither of us was listening",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0044",
+ "category": "descriptive",
+ "text": "It's one of those places where everyone seems to know each other already.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0044.wav",
+ "render_seconds": 76.58786594006233,
+ "audio_seconds": 4.876190476190477,
+ "rtf": 15.706495944739343,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT'S ONE OF THOSE PLACES WHERE EVERYONE SEEMS TO KNOW EACH OTHER ALREADY",
+ "raw": {
+ "normalized_reference": "it's one of those places where everyone seems to know each other already.",
+ "normalized_hypothesis": "it's one of those places where everyone seems to know each other already",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it is one of those places where everyone seems to know each other already",
+ "normalized_hypothesis": "it is one of those places where everyone seems to know each other already",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0045",
+ "category": "emotional",
+ "text": "He said he was proud of me and nobody had ever said that to me before.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0045.wav",
+ "render_seconds": 90.09963908197824,
+ "audio_seconds": 4.876190476190477,
+ "rtf": 18.47746504610882,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE SAID HE WAS PROUD OF ME AND NOBODY HAD EVER SAID THAT TO ME BEFORE",
+ "raw": {
+ "normalized_reference": "he said he was proud of me and nobody had ever said that to me before.",
+ "normalized_hypothesis": "he said he was proud of me and nobody had ever said that to me before",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he said he was proud of me and nobody had ever said that to me before",
+ "normalized_hypothesis": "he said he was proud of me and nobody had ever said that to me before",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0046",
+ "category": "homographs",
+ "text": "The minute detail took a minute to explain clearly.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0046.wav",
+ "render_seconds": 64.10040538094472,
+ "audio_seconds": 3.622312925170068,
+ "rtf": 17.695987813624686,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "A MINUTE DETAIL TOOK A MINUTE TO EXPLAIN CLEARLY",
+ "raw": {
+ "normalized_reference": "the minute detail took a minute to explain clearly.",
+ "normalized_hypothesis": "a minute detail took a minute to explain clearly",
+ "errors": 1,
+ "words": 9,
+ "wer": 0.1111111111111111
+ },
+ "semantic": {
+ "normalized_reference": "the minute detail took a minute to explain clearly",
+ "normalized_hypothesis": "a minute detail took a minute to explain clearly",
+ "errors": 1,
+ "words": 9,
+ "wer": 0.1111111111111111
+ }
+ },
+ {
+ "id": "release500_0047",
+ "category": "long",
+ "text": "The strange thing about really wanting something is that getting it never quite feels the way you imagined it would, and not getting it never quite feels as bad.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0047.wav",
+ "render_seconds": 159.90445923490915,
+ "audio_seconds": 9.961360544217687,
+ "rtf": 16.052471800925787,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE STRANGE THING ABOUT REALLY WANTING SOMETHING IS THAT GETTING IT NEVER QUITE FEELS THE WAY YOU IMAGINED IT WUD AND NOT GETTING IT NEVER QUITE FEELS AS BAD",
+ "raw": {
+ "normalized_reference": "the strange thing about really wanting something is that getting it never quite feels the way you imagined it would, and not getting it never quite feels as bad.",
+ "normalized_hypothesis": "the strange thing about really wanting something is that getting it never quite feels the way you imagined it wud and not getting it never quite feels as bad",
+ "errors": 1,
+ "words": 29,
+ "wer": 0.034482758620689655
+ },
+ "semantic": {
+ "normalized_reference": "the strange thing about really wanting something is that getting it never quite feels the way you imagined it would and not getting it never quite feels as bad",
+ "normalized_hypothesis": "the strange thing about really wanting something is that getting it never quite feels the way you imagined it wud and not getting it never quite feels as bad",
+ "errors": 1,
+ "words": 29,
+ "wer": 0.034482758620689655
+ }
+ },
+ {
+ "id": "release500_0048",
+ "category": "long_range",
+ "text": "Even though the recipe looked simple at first, preparing the sauce required patience, careful timing, and a much larger pan.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0048.wav",
+ "render_seconds": 120.39180352794938,
+ "audio_seconds": 8.289523809523809,
+ "rtf": 14.523367842870732,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "EVEN THOUGH THE RECIPE LOOKED SIMPLE AT FIRST PREPARING THE SAUCE REQUIRED PATIENCE CAREFUL TIMING AND A MUCH LARGER PAN",
+ "raw": {
+ "normalized_reference": "even though the recipe looked simple at first, preparing the sauce required patience, careful timing, and a much larger pan.",
+ "normalized_hypothesis": "even though the recipe looked simple at first preparing the sauce required patience careful timing and a much larger pan",
+ "errors": 0,
+ "words": 20,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "even though the recipe looked simple at 1st preparing the sauce required patience careful timing and a much larger pan",
+ "normalized_hypothesis": "even though the recipe looked simple at 1st preparing the sauce required patience careful timing and a much larger pan",
+ "errors": 0,
+ "words": 20,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0049",
+ "category": "names_places",
+ "text": "Xiomara met Siobhan near the old library in Worcestershire.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0049.wav",
+ "render_seconds": 48.40293605194893,
+ "audio_seconds": 4.667210884353741,
+ "rtf": 10.370848297044795,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SAMARIM AT SEALPUN NEAR THE OLD LIBRARY IN WORCESTERSHIRE",
+ "raw": {
+ "normalized_reference": "xiomara met siobhan near the old library in worcestershire.",
+ "normalized_hypothesis": "samarim at sealpun near the old library in worcestershire",
+ "errors": 3,
+ "words": 9,
+ "wer": 0.3333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "xiomara met siobhan near the old library in worcestershire",
+ "normalized_hypothesis": "samarim at sealpun near the old library in worcestershire",
+ "errors": 3,
+ "words": 9,
+ "wer": 0.3333333333333333
+ }
+ },
+ {
+ "id": "release500_0050",
+ "category": "narrative",
+ "text": "She walked into the room and everyone went quiet, not because she was loud, but because she wasn't.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0050.wav",
+ "render_seconds": 103.90908648003824,
+ "audio_seconds": 6.408707482993197,
+ "rtf": 16.21373525875257,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SHE WALKED INTO THE ROOM AND EVERYONE WENT QUIET NOT BECAUSE SHE WAS LOUD BUT BECAUSE SHE WASN'T",
+ "raw": {
+ "normalized_reference": "she walked into the room and everyone went quiet, not because she was loud, but because she wasn't.",
+ "normalized_hypothesis": "she walked into the room and everyone went quiet not because she was loud but because she wasn't",
+ "errors": 0,
+ "words": 18,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "she walked into the room and everyone went quiet not because she was loud but because she was not",
+ "normalized_hypothesis": "she walked into the room and everyone went quiet not because she was loud but because she was not",
+ "errors": 0,
+ "words": 19,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0051",
+ "category": "numbers",
+ "text": "At exactly seven forty-five, Morgan placed eighteen dollars and fifty cents beside the notebook.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0051.wav",
+ "render_seconds": 110.0988628109917,
+ "audio_seconds": 6.826666666666667,
+ "rtf": 16.127763107078863,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "AT EXACTLY SEVEN FORTY FIVE MORGAN PLACED EIGHTEEN DOLLARS AND FIFTY CENTS BESIDE THE NOTEBOOK",
+ "raw": {
+ "normalized_reference": "at exactly seven forty-five, morgan placed eighteen dollars and fifty cents beside the notebook.",
+ "normalized_hypothesis": "at exactly seven forty five morgan placed eighteen dollars and fifty cents beside the notebook",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "at exactly 745 morgan placed $18.50 beside the notebook",
+ "normalized_hypothesis": "at exactly 745 morgan placed $18.50 beside the notebook",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0052",
+ "category": "punctuation",
+ "text": "Does punctuation work? Yes: commas, semicolons, and periods should create sensible boundaries.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0052.wav",
+ "render_seconds": 105.61094658903312,
+ "audio_seconds": 7.453605442176871,
+ "rtf": 14.169108816979291,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "TE PENSELLITION YERK YES COBAS SEMICALONS AND PERIODS SHOULD CREATE SENSIBLE BOUNDARIES",
+ "raw": {
+ "normalized_reference": "does punctuation work? yes: commas, semicolons, and periods should create sensible boundaries.",
+ "normalized_hypothesis": "te pensellition yerk yes cobas semicalons and periods should create sensible boundaries",
+ "errors": 5,
+ "words": 12,
+ "wer": 0.4166666666666667
+ },
+ "semantic": {
+ "normalized_reference": "does punctuation work yes commas semicolons and periods should create sensible boundaries",
+ "normalized_hypothesis": "te pensellition yerk yes cobas semicalons and periods should create sensible boundaries",
+ "errors": 5,
+ "words": 12,
+ "wer": 0.4166666666666667
+ }
+ },
+ {
+ "id": "release500_0053",
+ "category": "question",
+ "text": "Have you ever just stopped and thought about how strange all of this is?",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0053.wav",
+ "render_seconds": 90.99906517192721,
+ "audio_seconds": 4.876190476190477,
+ "rtf": 18.661917662211636,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HAVE YOU EVER JUST STOPPED AND THOUGHT ABOUT HOW STRANGE ALL OF THIS IS",
+ "raw": {
+ "normalized_reference": "have you ever just stopped and thought about how strange all of this is?",
+ "normalized_hypothesis": "have you ever just stopped and thought about how strange all of this is",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "have you ever just stopped and thought about how strange all of this is",
+ "normalized_hypothesis": "have you ever just stopped and thought about how strange all of this is",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0054",
+ "category": "rare_words",
+ "text": "Her idiosyncratic explanation sounded plausible but remained fundamentally ambiguous.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0054.wav",
+ "render_seconds": 114.90803999104537,
+ "audio_seconds": 6.339047619047619,
+ "rtf": 18.127019529837384,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HER IDIOSYNCRATIC EXPLANATIONS SOUNDED PLAUSIBLE BUT REMAINED FUNDAMENTALLY AMBIGUOUS",
+ "raw": {
+ "normalized_reference": "her idiosyncratic explanation sounded plausible but remained fundamentally ambiguous.",
+ "normalized_hypothesis": "her idiosyncratic explanations sounded plausible but remained fundamentally ambiguous",
+ "errors": 1,
+ "words": 9,
+ "wer": 0.1111111111111111
+ },
+ "semantic": {
+ "normalized_reference": "her idiosyncratic explanation sounded plausible but remained fundamentally ambiguous",
+ "normalized_hypothesis": "her idiosyncratic explanations sounded plausible but remained fundamentally ambiguous",
+ "errors": 1,
+ "words": 9,
+ "wer": 0.1111111111111111
+ }
+ },
+ {
+ "id": "release500_0055",
+ "category": "short",
+ "text": "That tracks.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0055.wav",
+ "render_seconds": 23.48697709396947,
+ "audio_seconds": 1.4628571428571429,
+ "rtf": 16.055550747830694,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THAT TRACK",
+ "raw": {
+ "normalized_reference": "that tracks.",
+ "normalized_hypothesis": "that track",
+ "errors": 1,
+ "words": 2,
+ "wer": 0.5
+ },
+ "semantic": {
+ "normalized_reference": "that tracks",
+ "normalized_hypothesis": "that track",
+ "errors": 1,
+ "words": 2,
+ "wer": 0.5
+ }
+ },
+ {
+ "id": "release500_0056",
+ "category": "technical",
+ "text": "The key thing to remember is that order matters here more than you'd think.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0056.wav",
+ "render_seconds": 93.39177715405822,
+ "audio_seconds": 4.597551020408163,
+ "rtf": 20.313374824879496,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE KEY THING TO REMEMBER IS THAT ORDER MATTERS HERE MORE THAN YOU'D THINK",
+ "raw": {
+ "normalized_reference": "the key thing to remember is that order matters here more than you'd think.",
+ "normalized_hypothesis": "the key thing to remember is that order matters here more than you'd think",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the key thing to remember is that order matters here more than you would think",
+ "normalized_hypothesis": "the key thing to remember is that order matters here more than you would think",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0057",
+ "category": "conversational",
+ "text": "I would not say that again in this room if I were you.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0057.wav",
+ "render_seconds": 77.50085372989997,
+ "audio_seconds": 3.7616326530612243,
+ "rtf": 20.602983033664817,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I WOULD NOT SAY THAT AGAIN IN THIS ROOM IF I WERE YOU",
+ "raw": {
+ "normalized_reference": "i would not say that again in this room if i were you.",
+ "normalized_hypothesis": "i would not say that again in this room if i were you",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i would not say that again in this room if i were you",
+ "normalized_hypothesis": "i would not say that again in this room if i were you",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0058",
+ "category": "descriptive",
+ "text": "The room was filled with the quiet that means something is about to change.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0058.wav",
+ "render_seconds": 88.89165852591395,
+ "audio_seconds": 4.73687074829932,
+ "rtf": 18.76590332506513,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE ROOM WAS FILLED WITH THE QUIET THAT MEANS SOMETHING IS ABOUT TO CHANGE",
+ "raw": {
+ "normalized_reference": "the room was filled with the quiet that means something is about to change.",
+ "normalized_hypothesis": "the room was filled with the quiet that means something is about to change",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the room was filled with the quiet that means something is about to change",
+ "normalized_hypothesis": "the room was filled with the quiet that means something is about to change",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0059",
+ "category": "emotional",
+ "text": "I didn't expect to feel that way about it, but I did.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0059.wav",
+ "render_seconds": 32.98780106299091,
+ "audio_seconds": 3.4829931972789114,
+ "rtf": 9.471106945819656,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I DIDN'T EXPECT TO FEEL THAT WAY ABOUT IT BUT I DID",
+ "raw": {
+ "normalized_reference": "i didn't expect to feel that way about it, but i did.",
+ "normalized_hypothesis": "i didn't expect to feel that way about it but i did",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i did not expect to feel that way about it but i did",
+ "normalized_hypothesis": "i did not expect to feel that way about it but i did",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0060",
+ "category": "homographs",
+ "text": "I object to the strange object that was left near the entrance.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0060.wav",
+ "render_seconds": 75.39093819109257,
+ "audio_seconds": 4.1099319727891155,
+ "rtf": 18.343597580261203,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I OBJECT TO THE STRANGE OBJECT THAT WAS LEFT NEAR THE ENTRANCE",
+ "raw": {
+ "normalized_reference": "i object to the strange object that was left near the entrance.",
+ "normalized_hypothesis": "i object to the strange object that was left near the entrance",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i object to the strange object that was left near the entrance",
+ "normalized_hypothesis": "i object to the strange object that was left near the entrance",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0061",
+ "category": "long",
+ "text": "I've learned more from the things that went wrong than from the things that worked, which is a frustrating way to have to learn things but apparently the only way that stuck.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0061.wav",
+ "render_seconds": 106.90791040903423,
+ "audio_seconds": 10.65795918367347,
+ "rtf": 10.030805013102553,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I'VE LEARNED MORE FROM THE THINGS THAT WENT WRONG THAN FROM THE THINGS THAT WORKED WHICH IS A FRUSTRATING WAY TO HAVE TO LEARN THINGS BUT APPARENTLY THE ONLY WAY THAT STUCK",
+ "raw": {
+ "normalized_reference": "i've learned more from the things that went wrong than from the things that worked, which is a frustrating way to have to learn things but apparently the only way that stuck.",
+ "normalized_hypothesis": "i've learned more from the things that went wrong than from the things that worked which is a frustrating way to have to learn things but apparently the only way that stuck",
+ "errors": 0,
+ "words": 32,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i have learned more from the things that went wrong than from the things that worked which is a frustrating way to have to learn things but apparently the only way that stuck",
+ "normalized_hypothesis": "i have learned more from the things that went wrong than from the things that worked which is a frustrating way to have to learn things but apparently the only way that stuck",
+ "errors": 0,
+ "words": 33,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0062",
+ "category": "long_range",
+ "text": "Because every nearby hotel was full, the exhausted hikers followed the narrow road until they found a farmhouse with one empty room.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0062.wav",
+ "render_seconds": 118.20163273497019,
+ "audio_seconds": 8.916462585034013,
+ "rtf": 13.256561288483137,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "BECAUSE EVERY NEARBY HOTEL WAS FULL THE EXHAUSTED HYKERS FOLLOWED THE NARROW ROAD UNTIL THEY FOUND A FARMHOUSE WITH ONE EMPTY ROOM",
+ "raw": {
+ "normalized_reference": "because every nearby hotel was full, the exhausted hikers followed the narrow road until they found a farmhouse with one empty room.",
+ "normalized_hypothesis": "because every nearby hotel was full the exhausted hykers followed the narrow road until they found a farmhouse with one empty room",
+ "errors": 1,
+ "words": 22,
+ "wer": 0.045454545454545456
+ },
+ "semantic": {
+ "normalized_reference": "because every nearby hotel was full the exhausted hikers followed the narrow road until they found a farmhouse with one empty room",
+ "normalized_hypothesis": "because every nearby hotel was full the exhausted hykers followed the narrow road until they found a farmhouse with one empty room",
+ "errors": 1,
+ "words": 22,
+ "wer": 0.045454545454545456
+ }
+ },
+ {
+ "id": "release500_0063",
+ "category": "names_places",
+ "text": "The itinerary sends Beatrice from Albuquerque to Des Moines and then to Boise.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0063.wav",
+ "render_seconds": 92.99552207894158,
+ "audio_seconds": 5.363809523809524,
+ "rtf": 17.337588455768582,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE ITINERARY SENDS BEATRICE FROM HALBECURKET TO DIZMOINS AND THN THEN TO BOYS",
+ "raw": {
+ "normalized_reference": "the itinerary sends beatrice from albuquerque to des moines and then to boise.",
+ "normalized_hypothesis": "the itinerary sends beatrice from halbecurket to dizmoins and thn then to boys",
+ "errors": 5,
+ "words": 13,
+ "wer": 0.38461538461538464
+ },
+ "semantic": {
+ "normalized_reference": "the itinerary sends beatrice from albuquerque to des moines and then to boise",
+ "normalized_hypothesis": "the itinerary sends beatrice from halbecurket to dizmoins and thn then to boys",
+ "errors": 5,
+ "words": 13,
+ "wer": 0.38461538461538464
+ }
+ },
+ {
+ "id": "release500_0064",
+ "category": "narrative",
+ "text": "We had been driving for six hours when we finally saw the sign for the town.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0064.wav",
+ "render_seconds": 83.19974137598183,
+ "audio_seconds": 5.085170068027211,
+ "rtf": 16.361250511383414,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WE HAD BEEN DRIVING FOR SIX HOURS WHEN WE FINALLY SAW THE SIGN FOR THE TOWN",
+ "raw": {
+ "normalized_reference": "we had been driving for six hours when we finally saw the sign for the town.",
+ "normalized_hypothesis": "we had been driving for six hours when we finally saw the sign for the town",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "we had been driving for 6 hours when we finally saw the sign for the town",
+ "normalized_hypothesis": "we had been driving for 6 hours when we finally saw the sign for the town",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0065",
+ "category": "numbers",
+ "text": "Call extension zero four nine after eight thirty in the morning.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0065.wav",
+ "render_seconds": 62.2007240439998,
+ "audio_seconds": 4.458231292517007,
+ "rtf": 13.951883597515824,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "CALL EXTENSION ZERO FOUR NINE AFTER EIGHT THIRTY IN THE MORNING",
+ "raw": {
+ "normalized_reference": "call extension zero four nine after eight thirty in the morning.",
+ "normalized_hypothesis": "call extension zero four nine after eight thirty in the morning",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "call extension 049 after 830 in the morning",
+ "normalized_hypothesis": "call extension 049 after 830 in the morning",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0066",
+ "category": "punctuation",
+ "text": "Stop - listen carefully - then answer the question without guessing.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0066.wav",
+ "render_seconds": 83.50207139598206,
+ "audio_seconds": 4.597551020408163,
+ "rtf": 18.162293583110444,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "STOPPED LISTEN CAREFULLY THEN ANSWER THE QUESTION WITHOUT GUESSING",
+ "raw": {
+ "normalized_reference": "stop - listen carefully - then answer the question without guessing.",
+ "normalized_hypothesis": "stopped listen carefully then answer the question without guessing",
+ "errors": 1,
+ "words": 9,
+ "wer": 0.1111111111111111
+ },
+ "semantic": {
+ "normalized_reference": "stop listen carefully then answer the question without guessing",
+ "normalized_hypothesis": "stopped listen carefully then answer the question without guessing",
+ "errors": 1,
+ "words": 9,
+ "wer": 0.1111111111111111
+ }
+ },
+ {
+ "id": "release500_0067",
+ "category": "question",
+ "text": "Does it matter at this point, honestly?",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0067.wav",
+ "render_seconds": 25.000298880971968,
+ "audio_seconds": 2.716734693877551,
+ "rtf": 9.202333572473155,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "DO TAT MATTER AT THIS POINT ARDEFLAP",
+ "raw": {
+ "normalized_reference": "does it matter at this point, honestly?",
+ "normalized_hypothesis": "do tat matter at this point ardeflap",
+ "errors": 3,
+ "words": 7,
+ "wer": 0.42857142857142855
+ },
+ "semantic": {
+ "normalized_reference": "does it matter at this point honestly",
+ "normalized_hypothesis": "do tat matter at this point ardeflap",
+ "errors": 3,
+ "words": 7,
+ "wer": 0.42857142857142855
+ }
+ },
+ {
+ "id": "release500_0068",
+ "category": "rare_words",
+ "text": "An entrepreneurial meteorologist described the unprecedented atmospheric phenomenon.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0068.wav",
+ "render_seconds": 86.80405418702867,
+ "audio_seconds": 5.851428571428571,
+ "rtf": 14.834677229228532,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "AN UNTERPENURIAL METEOROLOGIST DESCRIBED THE UNPRECEDENTED ATMOSPHERIC PHENOMENON",
+ "raw": {
+ "normalized_reference": "an entrepreneurial meteorologist described the unprecedented atmospheric phenomenon.",
+ "normalized_hypothesis": "an unterpenurial meteorologist described the unprecedented atmospheric phenomenon",
+ "errors": 1,
+ "words": 8,
+ "wer": 0.125
+ },
+ "semantic": {
+ "normalized_reference": "an entrepreneurial meteorologist described the unprecedented atmospheric phenomenon",
+ "normalized_hypothesis": "an unterpenurial meteorologist described the unprecedented atmospheric phenomenon",
+ "errors": 1,
+ "words": 8,
+ "wer": 0.125
+ }
+ },
+ {
+ "id": "release500_0069",
+ "category": "short",
+ "text": "That's not what I expected.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0069.wav",
+ "render_seconds": 42.80378484702669,
+ "audio_seconds": 2.089795918367347,
+ "rtf": 20.482279858440503,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THAT'S NOT WHAT I EXPECTED",
+ "raw": {
+ "normalized_reference": "that's not what i expected.",
+ "normalized_hypothesis": "that's not what i expected",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "that is not what i expected",
+ "normalized_hypothesis": "that is not what i expected",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0070",
+ "category": "technical",
+ "text": "There's a known issue with that version that causes failures under load.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0070.wav",
+ "render_seconds": 83.70731660805177,
+ "audio_seconds": 4.806530612244898,
+ "rtf": 17.415329966858597,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THERE'S A KNOWN ISSUE WITH THAT VERSION THAT CAUSES FERRIERS UNDER LOAD",
+ "raw": {
+ "normalized_reference": "there's a known issue with that version that causes failures under load.",
+ "normalized_hypothesis": "there's a known issue with that version that causes ferriers under load",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "there is a known issue with that version that causes failures under load",
+ "normalized_hypothesis": "there is a known issue with that version that causes ferriers under load",
+ "errors": 1,
+ "words": 13,
+ "wer": 0.07692307692307693
+ }
+ },
+ {
+ "id": "release500_0071",
+ "category": "conversational",
+ "text": "Don't look now but I think that's him over there.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0071.wav",
+ "render_seconds": 71.40638228901662,
+ "audio_seconds": 2.8560544217687074,
+ "rtf": 25.00175828050077,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "DON'T LOOK NOW BUT I HILL GOT SOME OVER THERE",
+ "raw": {
+ "normalized_reference": "don't look now but i think that's him over there.",
+ "normalized_hypothesis": "don't look now but i hill got some over there",
+ "errors": 3,
+ "words": 10,
+ "wer": 0.3
+ },
+ "semantic": {
+ "normalized_reference": "do not look now but i think that is him over there",
+ "normalized_hypothesis": "do not look now but i hill got some over there",
+ "errors": 4,
+ "words": 12,
+ "wer": 0.3333333333333333
+ }
+ },
+ {
+ "id": "release500_0072",
+ "category": "descriptive",
+ "text": "The city at night looks completely different from how it feels during the day.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0072.wav",
+ "render_seconds": 76.19589207996614,
+ "audio_seconds": 5.154829931972789,
+ "rtf": 14.781456049085493,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE CITY AT NIGHT LOOKS CONCRETELY DIFFERENT FROM HOW IT FEELS DURING THE DAY",
+ "raw": {
+ "normalized_reference": "the city at night looks completely different from how it feels during the day.",
+ "normalized_hypothesis": "the city at night looks concretely different from how it feels during the day",
+ "errors": 1,
+ "words": 14,
+ "wer": 0.07142857142857142
+ },
+ "semantic": {
+ "normalized_reference": "the city at night looks completely different from how it feels during the day",
+ "normalized_hypothesis": "the city at night looks concretely different from how it feels during the day",
+ "errors": 1,
+ "words": 14,
+ "wer": 0.07142857142857142
+ }
+ },
+ {
+ "id": "release500_0073",
+ "category": "emotional",
+ "text": "It's the kind of happy that makes you nervous because you know it can't last.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0073.wav",
+ "render_seconds": 43.68924387602601,
+ "audio_seconds": 4.876190476190477,
+ "rtf": 8.959708216763147,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT'S THE KIND OF HAPPY THAT MAKES YOU NERVOUS BECAUSE YOU KNOW IT CAN'T LAST",
+ "raw": {
+ "normalized_reference": "it's the kind of happy that makes you nervous because you know it can't last.",
+ "normalized_hypothesis": "it's the kind of happy that makes you nervous because you know it can't last",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it is the kind of happy that makes you nervous because you know it can not last",
+ "normalized_hypothesis": "it is the kind of happy that makes you nervous because you know it can not last",
+ "errors": 0,
+ "words": 17,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0074",
+ "category": "homographs",
+ "text": "The dove dove beneath the branch while the wind began to wind the loose cable.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0074.wav",
+ "render_seconds": 98.30428628996015,
+ "audio_seconds": 5.085170068027211,
+ "rtf": 19.331563148309264,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE DOVE DOVED BENEATH THE BRANCH WHILE THE WIND BEGAN TO WING THE LOOSE CABLE",
+ "raw": {
+ "normalized_reference": "the dove dove beneath the branch while the wind began to wind the loose cable.",
+ "normalized_hypothesis": "the dove doved beneath the branch while the wind began to wing the loose cable",
+ "errors": 2,
+ "words": 15,
+ "wer": 0.13333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "the dove dove beneath the branch while the wind began to wind the loose cable",
+ "normalized_hypothesis": "the dove doved beneath the branch while the wind began to wing the loose cable",
+ "errors": 2,
+ "words": 15,
+ "wer": 0.13333333333333333
+ }
+ },
+ {
+ "id": "release500_0075",
+ "category": "long",
+ "text": "I've thought a lot about what I would do differently if I could go back, and the honest answer is probably not as much as I'd like to think.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0075.wav",
+ "render_seconds": 117.79970723902807,
+ "audio_seconds": 8.70748299319728,
+ "rtf": 13.528560128232128,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I'VE THOUGHT A LOT ABOUT WHAT I WOULD DO DIFFERENTLY IF I COULD GO BACK AND THE HONDEST ANSWER PROBABLY NOT AS MUCH AS I'D LIKE TO THINK",
+ "raw": {
+ "normalized_reference": "i've thought a lot about what i would do differently if i could go back, and the honest answer is probably not as much as i'd like to think.",
+ "normalized_hypothesis": "i've thought a lot about what i would do differently if i could go back and the hondest answer probably not as much as i'd like to think",
+ "errors": 2,
+ "words": 29,
+ "wer": 0.06896551724137931
+ },
+ "semantic": {
+ "normalized_reference": "i have thought a lot about what i would do differently if i could go back and the honest answer is probably not as much as i would like to think",
+ "normalized_hypothesis": "i have thought a lot about what i would do differently if i could go back and the hondest answer probably not as much as i would like to think",
+ "errors": 2,
+ "words": 31,
+ "wer": 0.06451612903225806
+ }
+ },
+ {
+ "id": "release500_0076",
+ "category": "long_range",
+ "text": "Although the weather changed without warning, the musicians finished packing before anyone opened the enormous glass door.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0076.wav",
+ "render_seconds": 114.79581492196303,
+ "audio_seconds": 7.801904761904762,
+ "rtf": 14.713819051277,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "ALTHOUGH THE WEATHER CHANGED WITHOUT WARNING THE MUSICIANS FINISHED PACKING BEFORE ANY ONE OPENED THE ENORMOUS GLASS DOOR",
+ "raw": {
+ "normalized_reference": "although the weather changed without warning, the musicians finished packing before anyone opened the enormous glass door.",
+ "normalized_hypothesis": "although the weather changed without warning the musicians finished packing before any one opened the enormous glass door",
+ "errors": 2,
+ "words": 17,
+ "wer": 0.11764705882352941
+ },
+ "semantic": {
+ "normalized_reference": "although the weather changed without warning the musicians finished packing before anyone opened the enormous glass door",
+ "normalized_hypothesis": "although the weather changed without warning the musicians finished packing before any one opened the enormous glass door",
+ "errors": 2,
+ "words": 17,
+ "wer": 0.11764705882352941
+ }
+ },
+ {
+ "id": "release500_0077",
+ "category": "names_places",
+ "text": "Saskatchewan looked unusually bright beneath the pale February sunrise.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0077.wav",
+ "render_seconds": 88.89546964899637,
+ "audio_seconds": 5.154829931972789,
+ "rtf": 17.245082926523526,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SUSCACHUON LOOKED UNUSUALLY BRIGHT BENEATH THE PALE FEBRUARY SUNRISE",
+ "raw": {
+ "normalized_reference": "saskatchewan looked unusually bright beneath the pale february sunrise.",
+ "normalized_hypothesis": "suscachuon looked unusually bright beneath the pale february sunrise",
+ "errors": 1,
+ "words": 9,
+ "wer": 0.1111111111111111
+ },
+ "semantic": {
+ "normalized_reference": "saskatchewan looked unusually bright beneath the pale february sunrise",
+ "normalized_hypothesis": "suscachuon looked unusually bright beneath the pale february sunrise",
+ "errors": 1,
+ "words": 9,
+ "wer": 0.1111111111111111
+ }
+ },
+ {
+ "id": "release500_0078",
+ "category": "narrative",
+ "text": "She handled it with a grace I genuinely did not expect.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0078.wav",
+ "render_seconds": 65.6038538919529,
+ "audio_seconds": 3.900952380952381,
+ "rtf": 16.817394186169565,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SHE HANDLED IT WITH A GRACE I GENUINELY DID NOT EXPECT",
+ "raw": {
+ "normalized_reference": "she handled it with a grace i genuinely did not expect.",
+ "normalized_hypothesis": "she handled it with a grace i genuinely did not expect",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "she handled it with a grace i genuinely did not expect",
+ "normalized_hypothesis": "she handled it with a grace i genuinely did not expect",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0079",
+ "category": "numbers",
+ "text": "Apartment four hundred seven is across from room twenty-nine on the third floor.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0079.wav",
+ "render_seconds": 74.11673066602089,
+ "audio_seconds": 5.642448979591837,
+ "rtf": 13.135560628743574,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "APARTMENT FOUR HUNDRED SEVEN IS ACROSS FROM ROOM TWENTY NINE ON THE THIRD FLOOR",
+ "raw": {
+ "normalized_reference": "apartment four hundred seven is across from room twenty-nine on the third floor.",
+ "normalized_hypothesis": "apartment four hundred seven is across from room twenty nine on the third floor",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "apartment 407 is across from room 29 on the 3rd floor",
+ "normalized_hypothesis": "apartment 407 is across from room 29 on the 3rd floor",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0080",
+ "category": "punctuation",
+ "text": "\"Not yet,\" Rowan whispered. \"Give me another minute... please.\"",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0080.wav",
+ "render_seconds": 90.39526544196997,
+ "audio_seconds": 5.50312925170068,
+ "rtf": 16.426157065824743,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "NOT MUT RAYHAM WHISPERED GIVE ME ANOTHER MIPPET PLEASE",
+ "raw": {
+ "normalized_reference": "\"not yet,\" rowan whispered. \"give me another minute... please.\"",
+ "normalized_hypothesis": "not mut rayham whispered give me another mippet please",
+ "errors": 3,
+ "words": 9,
+ "wer": 0.3333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "not yet rowan whispered give me another minute please",
+ "normalized_hypothesis": "not mut rayham whispered give me another mippet please",
+ "errors": 3,
+ "words": 9,
+ "wer": 0.3333333333333333
+ }
+ },
+ {
+ "id": "release500_0081",
+ "category": "question",
+ "text": "What are we actually trying to accomplish here?",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0081.wav",
+ "render_seconds": 52.39817513502203,
+ "audio_seconds": 3.274013605442177,
+ "rtf": 16.004263100166717,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WHAT ARE WE ACTUALLY TRYING TO ACCOMPLISH HERE",
+ "raw": {
+ "normalized_reference": "what are we actually trying to accomplish here?",
+ "normalized_hypothesis": "what are we actually trying to accomplish here",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "what are we actually trying to accomplish here",
+ "normalized_hypothesis": "what are we actually trying to accomplish here",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0082",
+ "category": "rare_words",
+ "text": "The physician recommended a noninvasive procedure for the recurring neurological symptoms.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0082.wav",
+ "render_seconds": 98.7946310529951,
+ "audio_seconds": 6.408707482993197,
+ "rtf": 15.415687377707075,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE PHYSICIAN RECOMMENDED A NON ENVASIVE PROCEDURE FOR THE RECURRING NEROLOGICAL SYMPTOMS",
+ "raw": {
+ "normalized_reference": "the physician recommended a noninvasive procedure for the recurring neurological symptoms.",
+ "normalized_hypothesis": "the physician recommended a non envasive procedure for the recurring nerological symptoms",
+ "errors": 3,
+ "words": 11,
+ "wer": 0.2727272727272727
+ },
+ "semantic": {
+ "normalized_reference": "the physician recommended a noninvasive procedure for the recurring neurological symptoms",
+ "normalized_hypothesis": "the physician recommended a non envasive procedure for the recurring nerological symptoms",
+ "errors": 3,
+ "words": 11,
+ "wer": 0.2727272727272727
+ }
+ },
+ {
+ "id": "release500_0083",
+ "category": "short",
+ "text": "Are you serious right now?",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0083.wav",
+ "render_seconds": 44.70363724103663,
+ "audio_seconds": 2.020136054421769,
+ "rtf": 22.129023014656468,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "ARE THE SERAMT RIGHT NOW",
+ "raw": {
+ "normalized_reference": "are you serious right now?",
+ "normalized_hypothesis": "are the seramt right now",
+ "errors": 2,
+ "words": 5,
+ "wer": 0.4
+ },
+ "semantic": {
+ "normalized_reference": "are you serious right now",
+ "normalized_hypothesis": "are the seramt right now",
+ "errors": 2,
+ "words": 5,
+ "wer": 0.4
+ }
+ },
+ {
+ "id": "release500_0084",
+ "category": "technical",
+ "text": "The performance improvement is real, but it comes with tradeoffs.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0084.wav",
+ "render_seconds": 93.99984763294924,
+ "audio_seconds": 4.597551020408163,
+ "rtf": 20.44563447272067,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE PERFORMANCE IMPROVEMENT IS REAL BUT IT COMES WITH TRADE OFFS",
+ "raw": {
+ "normalized_reference": "the performance improvement is real, but it comes with tradeoffs.",
+ "normalized_hypothesis": "the performance improvement is real but it comes with trade offs",
+ "errors": 2,
+ "words": 10,
+ "wer": 0.2
+ },
+ "semantic": {
+ "normalized_reference": "the performance improvement is real but it comes with tradeoffs",
+ "normalized_hypothesis": "the performance improvement is real but it comes with trade offs",
+ "errors": 2,
+ "words": 10,
+ "wer": 0.2
+ }
+ },
+ {
+ "id": "release500_0085",
+ "category": "conversational",
+ "text": "I don't know. It's just been one of those weeks, you know?",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0085.wav",
+ "render_seconds": 34.8873905390501,
+ "audio_seconds": 3.900952380952381,
+ "rtf": 8.943300797363918,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I DON'T KNOW IT'S JUST BEEN ONE OF THOSE WEEKS YOU KNOW",
+ "raw": {
+ "normalized_reference": "i don't know. it's just been one of those weeks, you know?",
+ "normalized_hypothesis": "i don't know it's just been one of those weeks you know",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i do not know it is just been one of those weeks you know",
+ "normalized_hypothesis": "i do not know it is just been one of those weeks you know",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0086",
+ "category": "descriptive",
+ "text": "The moment you stop expecting perfection, things get a lot more interesting.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0086.wav",
+ "render_seconds": 87.6976898870198,
+ "audio_seconds": 5.154829931972789,
+ "rtf": 17.01272225162573,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE MOMENT YOU STOP EXPECTING PERFECTION THINGS GET A LOT MORE INTERESTING",
+ "raw": {
+ "normalized_reference": "the moment you stop expecting perfection, things get a lot more interesting.",
+ "normalized_hypothesis": "the moment you stop expecting perfection things get a lot more interesting",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the moment you stop expecting perfection things get a lot more interesting",
+ "normalized_hypothesis": "the moment you stop expecting perfection things get a lot more interesting",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0087",
+ "category": "emotional",
+ "text": "It's the kind of thing that stays with you even years later.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0087.wav",
+ "render_seconds": 75.40016950899735,
+ "audio_seconds": 3.8312925170068026,
+ "rtf": 19.680086856929353,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT'S THE KIND OF THING THAT STAYS WITH YOU EVEN YEARS LATER",
+ "raw": {
+ "normalized_reference": "it's the kind of thing that stays with you even years later.",
+ "normalized_hypothesis": "it's the kind of thing that stays with you even years later",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it is the kind of thing that stays with you even years later",
+ "normalized_hypothesis": "it is the kind of thing that stays with you even years later",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0088",
+ "category": "long",
+ "text": "She told me once that the most important skill she'd developed was learning to disagree with someone without making them feel they were wrong.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0088.wav",
+ "render_seconds": 117.30900525301695,
+ "audio_seconds": 8.70748299319728,
+ "rtf": 13.472206072026164,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SHE TOLD ME ONCE THAT THE MOST IMPORTANT SKILL SHE DEVELOPED WAS LEARNING TO DISAGREE WITH SOMEONE BYT MAKING THEM FEEL THEY WERE RONG",
+ "raw": {
+ "normalized_reference": "she told me once that the most important skill she'd developed was learning to disagree with someone without making them feel they were wrong.",
+ "normalized_hypothesis": "she told me once that the most important skill she developed was learning to disagree with someone byt making them feel they were rong",
+ "errors": 3,
+ "words": 24,
+ "wer": 0.125
+ },
+ "semantic": {
+ "normalized_reference": "she told me once that the most important skill she would developed was learning to disagree with someone without making them feel they were wrong",
+ "normalized_hypothesis": "she told me once that the most important skill she developed was learning to disagree with someone byt making them feel they were rong",
+ "errors": 3,
+ "words": 25,
+ "wer": 0.12
+ }
+ },
+ {
+ "id": "release500_0089",
+ "category": "narrative",
+ "text": "He stood at the edge of the cliff and wondered, not for the first time, how he'd gotten here.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0089.wav",
+ "render_seconds": 77.19590870197862,
+ "audio_seconds": 5.7817687074829935,
+ "rtf": 13.351607891555508,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE STOOD AT THE EDGE OF THE CLIFF AND WONDERED NOT FOR THE FIRST TIME HOW HE'D GOTTN HERE",
+ "raw": {
+ "normalized_reference": "he stood at the edge of the cliff and wondered, not for the first time, how he'd gotten here.",
+ "normalized_hypothesis": "he stood at the edge of the cliff and wondered not for the first time how he'd gottn here",
+ "errors": 1,
+ "words": 19,
+ "wer": 0.05263157894736842
+ },
+ "semantic": {
+ "normalized_reference": "he stood at the edge of the cliff and wondered not for the 1st time how he would gotten here",
+ "normalized_hypothesis": "he stood at the edge of the cliff and wondered not for the 1st time how he would gottn here",
+ "errors": 1,
+ "words": 20,
+ "wer": 0.05
+ }
+ },
+ {
+ "id": "release500_0090",
+ "category": "question",
+ "text": "Why does asking for help feel so hard when you actually need it?",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0090.wav",
+ "render_seconds": 79.99640146596357,
+ "audio_seconds": 4.249251700680272,
+ "rtf": 18.825997399296625,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WHY DOES ASKING FOR HELP FEEL SO HARD WHEN YOU ACTUALLY NEED IT",
+ "raw": {
+ "normalized_reference": "why does asking for help feel so hard when you actually need it?",
+ "normalized_hypothesis": "why does asking for help feel so hard when you actually need it",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "why does asking for help feel so hard when you actually need it",
+ "normalized_hypothesis": "why does asking for help feel so hard when you actually need it",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0091",
+ "category": "short",
+ "text": "That's fair.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0091.wav",
+ "render_seconds": 18.09566029999405,
+ "audio_seconds": 1.4628571428571429,
+ "rtf": 12.370080283199059,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THAT'S FAIL",
+ "raw": {
+ "normalized_reference": "that's fair.",
+ "normalized_hypothesis": "that's fail",
+ "errors": 1,
+ "words": 2,
+ "wer": 0.5
+ },
+ "semantic": {
+ "normalized_reference": "that is fair",
+ "normalized_hypothesis": "that is fail",
+ "errors": 1,
+ "words": 3,
+ "wer": 0.3333333333333333
+ }
+ },
+ {
+ "id": "release500_0092",
+ "category": "technical",
+ "text": "If you run into that error, the fix is usually just restarting the service.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0092.wav",
+ "render_seconds": 90.703427462955,
+ "audio_seconds": 5.363809523809524,
+ "rtf": 16.91026257743302,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IF YOU RUN INTO THAT ERROR THE FIX IS USUALLY JUST RESTARTING THE SERVICE",
+ "raw": {
+ "normalized_reference": "if you run into that error, the fix is usually just restarting the service.",
+ "normalized_hypothesis": "if you run into that error the fix is usually just restarting the service",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "if you run into that error the fix is usually just restarting the service",
+ "normalized_hypothesis": "if you run into that error the fix is usually just restarting the service",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0093",
+ "category": "conversational",
+ "text": "It was a disaster in the best possible way.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0093.wav",
+ "render_seconds": 65.69613418600056,
+ "audio_seconds": 3.2043537414965986,
+ "rtf": 20.502147854411692,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT WAS A DISASTER IN THE BEST POSSIBLE WAY",
+ "raw": {
+ "normalized_reference": "it was a disaster in the best possible way.",
+ "normalized_hypothesis": "it was a disaster in the best possible way",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it was a disaster in the best possible way",
+ "normalized_hypothesis": "it was a disaster in the best possible way",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0094",
+ "category": "descriptive",
+ "text": "It had the look of a place that used to mean something to someone.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0094.wav",
+ "render_seconds": 75.99187482800335,
+ "audio_seconds": 4.458231292517007,
+ "rtf": 17.045296630426776,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT HAD THE LOOK OF A PLACE THAT USED TO MEAN SOMETHING TO SOME ONE",
+ "raw": {
+ "normalized_reference": "it had the look of a place that used to mean something to someone.",
+ "normalized_hypothesis": "it had the look of a place that used to mean something to some one",
+ "errors": 2,
+ "words": 14,
+ "wer": 0.14285714285714285
+ },
+ "semantic": {
+ "normalized_reference": "it had the look of a place that used to mean something to someone",
+ "normalized_hypothesis": "it had the look of a place that used to mean something to some one",
+ "errors": 2,
+ "words": 14,
+ "wer": 0.14285714285714285
+ }
+ },
+ {
+ "id": "release500_0095",
+ "category": "emotional",
+ "text": "There's this feeling you get when something finally works after weeks of failure.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0095.wav",
+ "render_seconds": 84.90776875498705,
+ "audio_seconds": 5.015510204081632,
+ "rtf": 16.929039180478377,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THERE 'S THIS FEELING YOU GET WHEN SOMETHING FINALLY WORKS AFTER WEEKS OF FAILURE",
+ "raw": {
+ "normalized_reference": "there's this feeling you get when something finally works after weeks of failure.",
+ "normalized_hypothesis": "there 's this feeling you get when something finally works after weeks of failure",
+ "errors": 2,
+ "words": 13,
+ "wer": 0.15384615384615385
+ },
+ "semantic": {
+ "normalized_reference": "there is this feeling you get when something finally works after weeks of failure",
+ "normalized_hypothesis": "there is this feeling you get when something finally works after weeks of failure",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0096",
+ "category": "long",
+ "text": "He had a theory that most arguments aren't actually about what they seem to be about, and the longer I live the more I think he was onto something.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0096.wav",
+ "render_seconds": 129.39934604801238,
+ "audio_seconds": 9.125442176870749,
+ "rtf": 14.180063118035706,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE HAD A THEORY THAT MOST ARGUMENTS AREN'T ACTUALLY ABOUT WHAT THEY SEEM TO BE ABOUT AND THE LONGER I LIVE THE MORE I THINK HE WAS ON TO SOMETHING",
+ "raw": {
+ "normalized_reference": "he had a theory that most arguments aren't actually about what they seem to be about, and the longer i live the more i think he was onto something.",
+ "normalized_hypothesis": "he had a theory that most arguments aren't actually about what they seem to be about and the longer i live the more i think he was on to something",
+ "errors": 2,
+ "words": 29,
+ "wer": 0.06896551724137931
+ },
+ "semantic": {
+ "normalized_reference": "he had a theory that most arguments are not actually about what they seem to be about and the longer i live the more i think he was onto something",
+ "normalized_hypothesis": "he had a theory that most arguments are not actually about what they seem to be about and the longer i live the more i think he was on to something",
+ "errors": 2,
+ "words": 30,
+ "wer": 0.06666666666666667
+ }
+ },
+ {
+ "id": "release500_0097",
+ "category": "narrative",
+ "text": "The thing nobody tells you about starting over is how ordinary most of it feels.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0097.wav",
+ "render_seconds": 57.30049965600483,
+ "audio_seconds": 5.433469387755102,
+ "rtf": 10.545840156040551,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE THING NOBODY TELLS YOU ABOUT STARTING OVER IS HOW ORDINARY MOST OF IT FEELS",
+ "raw": {
+ "normalized_reference": "the thing nobody tells you about starting over is how ordinary most of it feels.",
+ "normalized_hypothesis": "the thing nobody tells you about starting over is how ordinary most of it feels",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the thing nobody tells you about starting over is how ordinary most of it feels",
+ "normalized_hypothesis": "the thing nobody tells you about starting over is how ordinary most of it feels",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0098",
+ "category": "question",
+ "text": "What exactly are we waiting for at this point?",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0098.wav",
+ "render_seconds": 66.59413642797153,
+ "audio_seconds": 3.2043537414965986,
+ "rtf": 20.782392270108303,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WHAT EXACTLY ARE WE WAIT OFF HERE AT THIS PINT",
+ "raw": {
+ "normalized_reference": "what exactly are we waiting for at this point?",
+ "normalized_hypothesis": "what exactly are we wait off here at this pint",
+ "errors": 4,
+ "words": 9,
+ "wer": 0.4444444444444444
+ },
+ "semantic": {
+ "normalized_reference": "what exactly are we waiting for at this point",
+ "normalized_hypothesis": "what exactly are we wait off here at this pint",
+ "errors": 4,
+ "words": 9,
+ "wer": 0.4444444444444444
+ }
+ },
+ {
+ "id": "release500_0099",
+ "category": "short",
+ "text": "I stand by what I said.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0099.wav",
+ "render_seconds": 58.70346640900243,
+ "audio_seconds": 2.089795918367347,
+ "rtf": 28.0905259183703,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I STAND BY WHAT I SAID",
+ "raw": {
+ "normalized_reference": "i stand by what i said.",
+ "normalized_hypothesis": "i stand by what i said",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i stand by what i said",
+ "normalized_hypothesis": "i stand by what i said",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0100",
+ "category": "technical",
+ "text": "Double-check that value before you push anything to production.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0100.wav",
+ "render_seconds": 70.89639390504453,
+ "audio_seconds": 4.31891156462585,
+ "rtf": 16.41533818050899,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "DOUBLE CHECK THAT VALUE BEFORE YOU PUSH ANYTHING TO PRODUCTION",
+ "raw": {
+ "normalized_reference": "double-check that value before you push anything to production.",
+ "normalized_hypothesis": "double check that value before you push anything to production",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "double check that value before you push anything to production",
+ "normalized_hypothesis": "double check that value before you push anything to production",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0101",
+ "category": "conversational",
+ "text": "I understand your point; I just do not agree with the conclusion.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0101.wav",
+ "render_seconds": 81.69200813397765,
+ "audio_seconds": 4.458231292517007,
+ "rtf": 18.32386046706347,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I UNDERSTAND YOUR POINT I JUST DO NOT AGREE WITH THE CONCLUSION",
+ "raw": {
+ "normalized_reference": "i understand your point; i just do not agree with the conclusion.",
+ "normalized_hypothesis": "i understand your point i just do not agree with the conclusion",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i understand your point i just do not agree with the conclusion",
+ "normalized_hypothesis": "i understand your point i just do not agree with the conclusion",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0102",
+ "category": "descriptive",
+ "text": "The sky that day was a strange shade of orange I'd never seen before.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0102.wav",
+ "render_seconds": 88.40734693396371,
+ "audio_seconds": 4.597551020408163,
+ "rtf": 19.229225851226126,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE SKY THAT DAY WAS A STRANGE SHADE OF ORANGE IHAD NEVER SEEN BEFORE",
+ "raw": {
+ "normalized_reference": "the sky that day was a strange shade of orange i'd never seen before.",
+ "normalized_hypothesis": "the sky that day was a strange shade of orange ihad never seen before",
+ "errors": 1,
+ "words": 14,
+ "wer": 0.07142857142857142
+ },
+ "semantic": {
+ "normalized_reference": "the sky that day was a strange shade of orange i would never seen before",
+ "normalized_hypothesis": "the sky that day was a strange shade of orange ihad never seen before",
+ "errors": 2,
+ "words": 15,
+ "wer": 0.13333333333333333
+ }
+ },
+ {
+ "id": "release500_0103",
+ "category": "emotional",
+ "text": "It hurt more than I expected, and I thought I was prepared.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0103.wav",
+ "render_seconds": 58.3985950849019,
+ "audio_seconds": 4.388571428571429,
+ "rtf": 13.306971536273219,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT WAR'T MORE THAN I EXPECTED AND I THOUGHT I WAS PREPARED",
+ "raw": {
+ "normalized_reference": "it hurt more than i expected, and i thought i was prepared.",
+ "normalized_hypothesis": "it war't more than i expected and i thought i was prepared",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "it hurt more than i expected and i thought i was prepared",
+ "normalized_hypothesis": "it war not more than i expected and i thought i was prepared",
+ "errors": 2,
+ "words": 12,
+ "wer": 0.16666666666666666
+ }
+ },
+ {
+ "id": "release500_0104",
+ "category": "long",
+ "text": "The hardest part wasn't making the decision, it was sitting with the uncertainty of not knowing whether it was the right one for months afterward.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0104.wav",
+ "render_seconds": 139.29743686399888,
+ "audio_seconds": 8.986122448979591,
+ "rtf": 15.50139536322662,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE HARDEST PART WAS IN MAKING THE DECISION IT WAS SITTING WITH THE UNCERTAINTY OF NOT KNOWING WHETHER IT WAS THE RIGHT ONE FOR MONTHS AFTERWARD",
+ "raw": {
+ "normalized_reference": "the hardest part wasn't making the decision, it was sitting with the uncertainty of not knowing whether it was the right one for months afterward.",
+ "normalized_hypothesis": "the hardest part was in making the decision it was sitting with the uncertainty of not knowing whether it was the right one for months afterward",
+ "errors": 2,
+ "words": 25,
+ "wer": 0.08
+ },
+ "semantic": {
+ "normalized_reference": "the hardest part was not making the decision it was sitting with the uncertainty of not knowing whether it was the right one for months afterward",
+ "normalized_hypothesis": "the hardest part was in making the decision it was sitting with the uncertainty of not knowing whether it was the right one for months afterward",
+ "errors": 1,
+ "words": 26,
+ "wer": 0.038461538461538464
+ }
+ },
+ {
+ "id": "release500_0105",
+ "category": "narrative",
+ "text": "There was this one day last summer when everything just clicked into place.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0105.wav",
+ "render_seconds": 79.4943403560901,
+ "audio_seconds": 5.154829931972789,
+ "rtf": 15.421331334915074,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THERE WAS THIS ONE DAY LAST SUMMER WHEN EVERYTHING JUST CLICKED INTO PLACE",
+ "raw": {
+ "normalized_reference": "there was this one day last summer when everything just clicked into place.",
+ "normalized_hypothesis": "there was this one day last summer when everything just clicked into place",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "there was this one day last summer when everything just clicked into place",
+ "normalized_hypothesis": "there was this one day last summer when everything just clicked into place",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0106",
+ "category": "question",
+ "text": "Have you ever thought about why we keep doing things the same way when they stop working?",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0106.wav",
+ "render_seconds": 89.80292898905464,
+ "audio_seconds": 5.433469387755102,
+ "rtf": 16.52773257385696,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HAVE YOU EVER THOUGHT ABOUT WHY WE KEEP DOING THINGS THE SAME WAY WHEN THEY STOP WORKING",
+ "raw": {
+ "normalized_reference": "have you ever thought about why we keep doing things the same way when they stop working?",
+ "normalized_hypothesis": "have you ever thought about why we keep doing things the same way when they stop working",
+ "errors": 0,
+ "words": 17,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "have you ever thought about why we keep doing things the same way when they stop working",
+ "normalized_hypothesis": "have you ever thought about why we keep doing things the same way when they stop working",
+ "errors": 0,
+ "words": 17,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0107",
+ "category": "short",
+ "text": "I can't believe you just said that.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0107.wav",
+ "render_seconds": 62.10698268702254,
+ "audio_seconds": 2.4380952380952383,
+ "rtf": 25.473567117724084,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I CAN'T BLAR HOUND YOU' SAID THAT",
+ "raw": {
+ "normalized_reference": "i can't believe you just said that.",
+ "normalized_hypothesis": "i can't blar hound you' said that",
+ "errors": 3,
+ "words": 7,
+ "wer": 0.42857142857142855
+ },
+ "semantic": {
+ "normalized_reference": "i can not believe you just said that",
+ "normalized_hypothesis": "i can not blar hound you said that",
+ "errors": 3,
+ "words": 8,
+ "wer": 0.375
+ }
+ },
+ {
+ "id": "release500_0108",
+ "category": "technical",
+ "text": "You'll want to isolate that variable before drawing any conclusions.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0108.wav",
+ "render_seconds": 91.70395563798957,
+ "audio_seconds": 4.597551020408163,
+ "rtf": 19.94626165776584,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "YOU'LL WANT TO ISOLATE THAT VARIABLE BEFORE DRAWING ANY CONCLUSIONS",
+ "raw": {
+ "normalized_reference": "you'll want to isolate that variable before drawing any conclusions.",
+ "normalized_hypothesis": "you'll want to isolate that variable before drawing any conclusions",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "you will want to isolate that variable before drawing any conclusions",
+ "normalized_hypothesis": "you will want to isolate that variable before drawing any conclusions",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0109",
+ "category": "conversational",
+ "text": "The meeting that could have been an email lasted two hours.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0109.wav",
+ "render_seconds": 80.20279090700205,
+ "audio_seconds": 3.9706122448979593,
+ "rtf": 20.199099272425475,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE MEETING THAT COULD HAVE BEEN AN EMALE LASTED TWO HOURS",
+ "raw": {
+ "normalized_reference": "the meeting that could have been an email lasted two hours.",
+ "normalized_hypothesis": "the meeting that could have been an emale lasted two hours",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ },
+ "semantic": {
+ "normalized_reference": "the meeting that could have been an email lasted 2 hours",
+ "normalized_hypothesis": "the meeting that could have been an emale lasted 2 hours",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ }
+ },
+ {
+ "id": "release500_0110",
+ "category": "descriptive",
+ "text": "There's a specific kind of tired that sleep doesn't fix.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0110.wav",
+ "render_seconds": 68.00529682496563,
+ "audio_seconds": 3.622312925170068,
+ "rtf": 18.773998334716698,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THERE'S A SPECIFIC KIND OF TIRE THAT SLEEP DOESN'T FIX",
+ "raw": {
+ "normalized_reference": "there's a specific kind of tired that sleep doesn't fix.",
+ "normalized_hypothesis": "there's a specific kind of tire that sleep doesn't fix",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ },
+ "semantic": {
+ "normalized_reference": "there is a specific kind of tired that sleep does not fix",
+ "normalized_hypothesis": "there is a specific kind of tire that sleep does not fix",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ }
+ },
+ {
+ "id": "release500_0111",
+ "category": "emotional",
+ "text": "Something about that moment broke something open in me that I hadn't realized was closed.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0111.wav",
+ "render_seconds": 96.19787661207374,
+ "audio_seconds": 5.642448979591837,
+ "rtf": 17.048958166925683,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SOMETHING ABOUT THAT MOMENT BROKE SOMETHING OPEN IN ME THAT I HADN'T REALIZED WAS CLOSED",
+ "raw": {
+ "normalized_reference": "something about that moment broke something open in me that i hadn't realized was closed.",
+ "normalized_hypothesis": "something about that moment broke something open in me that i hadn't realized was closed",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "something about that moment broke something open in me that i had not realized was closed",
+ "normalized_hypothesis": "something about that moment broke something open in me that i had not realized was closed",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0112",
+ "category": "long",
+ "text": "The thing about making something good is that you don't usually know it's good while you're making it. You just know you're not done yet.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0112.wav",
+ "render_seconds": 124.59800055297092,
+ "audio_seconds": 7.662585034013605,
+ "rtf": 16.260570029551424,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE THING ABOUT MAKING SOMETHING GOOD IS THAT YOU DON'T USUALLY KNOW IT'S GOOD WHILE YOU'RE MAKING IT YOU JUST KNOW YOU'RE NOT DONE YET",
+ "raw": {
+ "normalized_reference": "the thing about making something good is that you don't usually know it's good while you're making it. you just know you're not done yet.",
+ "normalized_hypothesis": "the thing about making something good is that you don't usually know it's good while you're making it you just know you're not done yet",
+ "errors": 0,
+ "words": 25,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the thing about making something good is that you do not usually know it is good while you are making it you just know you are not done yet",
+ "normalized_hypothesis": "the thing about making something good is that you do not usually know it is good while you are making it you just know you are not done yet",
+ "errors": 0,
+ "words": 29,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0113",
+ "category": "narrative",
+ "text": "She opened the door slowly, as if she expected to find something she wasn't ready for.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0113.wav",
+ "render_seconds": 88.18888535699807,
+ "audio_seconds": 5.921088435374149,
+ "rtf": 14.894032798114335,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SHE OPENED THE DOOR SLOWLY AS IF SHE EXPECTED TO FIND SOMETHING SHE WASN'T READY FOR",
+ "raw": {
+ "normalized_reference": "she opened the door slowly, as if she expected to find something she wasn't ready for.",
+ "normalized_hypothesis": "she opened the door slowly as if she expected to find something she wasn't ready for",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "she opened the door slowly as if she expected to find something she was not ready for",
+ "normalized_hypothesis": "she opened the door slowly as if she expected to find something she was not ready for",
+ "errors": 0,
+ "words": 17,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0114",
+ "category": "question",
+ "text": "What would you do differently if you could go back?",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0114.wav",
+ "render_seconds": 66.09143147303257,
+ "audio_seconds": 3.3436734693877552,
+ "rtf": 19.766114149039282,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WHAT WOULD YOU DO DIFFERENTLY IF YOU COULD GE BACK",
+ "raw": {
+ "normalized_reference": "what would you do differently if you could go back?",
+ "normalized_hypothesis": "what would you do differently if you could ge back",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ },
+ "semantic": {
+ "normalized_reference": "what would you do differently if you could go back",
+ "normalized_hypothesis": "what would you do differently if you could ge back",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ }
+ },
+ {
+ "id": "release500_0115",
+ "category": "short",
+ "text": "Okay. Okay okay okay.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0115.wav",
+ "render_seconds": 56.49792234203778,
+ "audio_seconds": 2.4380952380952383,
+ "rtf": 23.17297596060143,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "A TURN OTAO CAIRO PARLE",
+ "raw": {
+ "normalized_reference": "okay. okay okay okay.",
+ "normalized_hypothesis": "a turn otao cairo parle",
+ "errors": 5,
+ "words": 4,
+ "wer": 1.25
+ },
+ "semantic": {
+ "normalized_reference": "okay okay okay okay",
+ "normalized_hypothesis": "a turn otao cairo parle",
+ "errors": 5,
+ "words": 4,
+ "wer": 1.25
+ }
+ },
+ {
+ "id": "release500_0116",
+ "category": "technical",
+ "text": "The latency goes up significantly at scale, which is expected.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0116.wav",
+ "render_seconds": 85.29182029305957,
+ "audio_seconds": 4.597551020408163,
+ "rtf": 18.551576679509584,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE LATENCY GOES UP SIGNIFICANTLY AT SCALE WHICH IS EXPECTED",
+ "raw": {
+ "normalized_reference": "the latency goes up significantly at scale, which is expected.",
+ "normalized_hypothesis": "the latency goes up significantly at scale which is expected",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the latency goes up significantly at scale which is expected",
+ "normalized_hypothesis": "the latency goes up significantly at scale which is expected",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0117",
+ "category": "conversational",
+ "text": "We ended up staying until two in the morning talking about nothing.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0117.wav",
+ "render_seconds": 80.60033227689564,
+ "audio_seconds": 4.458231292517007,
+ "rtf": 18.078992988134246,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WE ENDED UP STAYING UNTIL TWO IN THE MORNING TALKING ABOUT NOTHING",
+ "raw": {
+ "normalized_reference": "we ended up staying until two in the morning talking about nothing.",
+ "normalized_hypothesis": "we ended up staying until two in the morning talking about nothing",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "we ended up staying until 2 in the morning talking about nothing",
+ "normalized_hypothesis": "we ended up staying until 2 in the morning talking about nothing",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0118",
+ "category": "descriptive",
+ "text": "The rain started slowly, and then all at once, the way it usually does.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0118.wav",
+ "render_seconds": 83.00951866898686,
+ "audio_seconds": 5.154829931972789,
+ "rtf": 16.103250691961925,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE RAIN STARTED SLOWLY AND THEN ALL AT ONCE THE WAY IT USUALLY DOES",
+ "raw": {
+ "normalized_reference": "the rain started slowly, and then all at once, the way it usually does.",
+ "normalized_hypothesis": "the rain started slowly and then all at once the way it usually does",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the rain started slowly and then all at once the way it usually does",
+ "normalized_hypothesis": "the rain started slowly and then all at once the way it usually does",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0119",
+ "category": "emotional",
+ "text": "I honestly did not think I was going to make it through that year.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0119.wav",
+ "render_seconds": 75.00318047497422,
+ "audio_seconds": 4.31891156462585,
+ "rtf": 17.36622279772746,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I HONESTLY DID NOT THINK I WAS GOING TO MAKE IT THROUGH THAT YEAR",
+ "raw": {
+ "normalized_reference": "i honestly did not think i was going to make it through that year.",
+ "normalized_hypothesis": "i honestly did not think i was going to make it through that year",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i honestly did not think i was going to make it through that year",
+ "normalized_hypothesis": "i honestly did not think i was going to make it through that year",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0120",
+ "category": "long",
+ "text": "There's an assumption that people who seem confident all the time actually feel confident, and in my experience that's almost never true.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0120.wav",
+ "render_seconds": 135.2956613280112,
+ "audio_seconds": 8.986122448979591,
+ "rtf": 15.05606696282828,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THERE'S AN ASSUMPTION THAT PEOPLE WHO SEEM CONFIDENT ALL THE TIME ACTUALLY FEEL CONFIDENT AND IN MY EXPERIENCE THAT'S ALMOST NEVER TRUE",
+ "raw": {
+ "normalized_reference": "there's an assumption that people who seem confident all the time actually feel confident, and in my experience that's almost never true.",
+ "normalized_hypothesis": "there's an assumption that people who seem confident all the time actually feel confident and in my experience that's almost never true",
+ "errors": 0,
+ "words": 22,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "there is an assumption that people who seem confident all the time actually feel confident and in my experience that is almost never true",
+ "normalized_hypothesis": "there is an assumption that people who seem confident all the time actually feel confident and in my experience that is almost never true",
+ "errors": 0,
+ "words": 24,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0121",
+ "category": "narrative",
+ "text": "The moment I opened the door I knew something was different, even if I couldn't say what.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0121.wav",
+ "render_seconds": 61.21020917000715,
+ "audio_seconds": 5.572789115646258,
+ "rtf": 10.983765561512513,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE MOMENT I OPENED THE DOOR I KNEW SOMETHING WAS DIFFERENT EVEN IF I COULDN'T SAY WHAT",
+ "raw": {
+ "normalized_reference": "the moment i opened the door i knew something was different, even if i couldn't say what.",
+ "normalized_hypothesis": "the moment i opened the door i knew something was different even if i couldn't say what",
+ "errors": 0,
+ "words": 17,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the moment i opened the door i knew something was different even if i could not say what",
+ "normalized_hypothesis": "the moment i opened the door i knew something was different even if i could not say what",
+ "errors": 0,
+ "words": 18,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0122",
+ "category": "question",
+ "text": "How do you explain something that doesn't make sense even to you?",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0122.wav",
+ "render_seconds": 76.20779749599751,
+ "audio_seconds": 4.040272108843538,
+ "rtf": 18.862045783795182,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HOW DO YOU EXPLAIN SOMETHING THAT DOESN'T MAKE SENSE EVEN TO YOU",
+ "raw": {
+ "normalized_reference": "how do you explain something that doesn't make sense even to you?",
+ "normalized_hypothesis": "how do you explain something that doesn't make sense even to you",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "how do you explain something that does not make sense even to you",
+ "normalized_hypothesis": "how do you explain something that does not make sense even to you",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0123",
+ "category": "short",
+ "text": "Hold on a second.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0123.wav",
+ "render_seconds": 54.59877931093797,
+ "audio_seconds": 1.6718367346938776,
+ "rtf": 32.65796125776319,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HOSANDERCRATON",
+ "raw": {
+ "normalized_reference": "hold on a second.",
+ "normalized_hypothesis": "hosandercraton",
+ "errors": 4,
+ "words": 4,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "hold on a 2nd",
+ "normalized_hypothesis": "hosandercraton",
+ "errors": 4,
+ "words": 4,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0124",
+ "category": "technical",
+ "text": "A deterministic split prevents duplicate normalized text from leaking into evaluation.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0124.wav",
+ "render_seconds": 96.39785736100748,
+ "audio_seconds": 6.339047619047619,
+ "rtf": 15.206993724317586,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "A DETERMINISTIC SPLIT PREVENTS DUPLICATE NORMALIZED TEXT FROM REEKING INTO EVALUATION",
+ "raw": {
+ "normalized_reference": "a deterministic split prevents duplicate normalized text from leaking into evaluation.",
+ "normalized_hypothesis": "a deterministic split prevents duplicate normalized text from reeking into evaluation",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ },
+ "semantic": {
+ "normalized_reference": "a deterministic split prevents duplicate normalized text from leaking into evaluation",
+ "normalized_hypothesis": "a deterministic split prevents duplicate normalized text from reeking into evaluation",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ }
+ },
+ {
+ "id": "release500_0125",
+ "category": "conversational",
+ "text": "Honestly, that was much better than I expected, but the ending still felt rushed.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0125.wav",
+ "render_seconds": 103.99647059303243,
+ "audio_seconds": 5.642448979591837,
+ "rtf": 18.431087453192234,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HONESTLY THAT WAS MUCH BETTER THAN I EXPECTED BUT THE ENDING STILL FELT RUSHED",
+ "raw": {
+ "normalized_reference": "honestly, that was much better than i expected, but the ending still felt rushed.",
+ "normalized_hypothesis": "honestly that was much better than i expected but the ending still felt rushed",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "honestly that was much better than i expected but the ending still felt rushed",
+ "normalized_hypothesis": "honestly that was much better than i expected but the ending still felt rushed",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0126",
+ "category": "descriptive",
+ "text": "She had a habit of tilting her head slightly when thinking hard about something.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0126.wav",
+ "render_seconds": 85.90281561494339,
+ "audio_seconds": 5.224489795918367,
+ "rtf": 16.44233580129776,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SHE HAD A HABIT OF TILTING HER HEAD SLIGHTLY WHEN THINKING HARD ABOUT SOMETHING",
+ "raw": {
+ "normalized_reference": "she had a habit of tilting her head slightly when thinking hard about something.",
+ "normalized_hypothesis": "she had a habit of tilting her head slightly when thinking hard about something",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "she had a habit of tilting her head slightly when thinking hard about something",
+ "normalized_hypothesis": "she had a habit of tilting her head slightly when thinking hard about something",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0127",
+ "category": "emotional",
+ "text": "I cannot stress enough how much I did not want to be right about this.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0127.wav",
+ "render_seconds": 42.48695929395035,
+ "audio_seconds": 4.667210884353741,
+ "rtf": 9.103286812340691,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I CANNOT STRESS ENOUGH HOW MUCH I DID NOT WANT TO BE RIGHT ABOUT THIS",
+ "raw": {
+ "normalized_reference": "i cannot stress enough how much i did not want to be right about this.",
+ "normalized_hypothesis": "i cannot stress enough how much i did not want to be right about this",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i cannot stress enough how much i did not want to be right about this",
+ "normalized_hypothesis": "i cannot stress enough how much i did not want to be right about this",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0128",
+ "category": "long",
+ "text": "I used to think the goal was to get to a point where nothing was hard anymore, and now I think that point doesn't exist and I'm not sure I'd want it to.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0128.wav",
+ "render_seconds": 130.00142709596548,
+ "audio_seconds": 9.264761904761905,
+ "rtf": 14.031815219034103,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I USED TO THINK THE GOAL WAS TIO TO A POINT WHERE NOTHING WAS HARD ANY MORE AND NOW I THINK THAT POINT DOESN'T EXIST AND I'M NOT SURE I'D WANT IT TO",
+ "raw": {
+ "normalized_reference": "i used to think the goal was to get to a point where nothing was hard anymore, and now i think that point doesn't exist and i'm not sure i'd want it to.",
+ "normalized_hypothesis": "i used to think the goal was tio to a point where nothing was hard any more and now i think that point doesn't exist and i'm not sure i'd want it to",
+ "errors": 4,
+ "words": 33,
+ "wer": 0.12121212121212122
+ },
+ "semantic": {
+ "normalized_reference": "i used to think the goal was to get to a point where nothing was hard anymore and now i think that point does not exist and i am not sure i would want it to",
+ "normalized_hypothesis": "i used to think the goal was tio to a point where nothing was hard any more and now i think that point does not exist and i am not sure i would want it to",
+ "errors": 4,
+ "words": 36,
+ "wer": 0.1111111111111111
+ }
+ },
+ {
+ "id": "release500_0129",
+ "category": "narrative",
+ "text": "So this is the part of the story where everything goes completely sideways.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0129.wav",
+ "render_seconds": 89.60730500996578,
+ "audio_seconds": 5.224489795918367,
+ "rtf": 17.151398224563764,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SO THI'S THE PART OF THE STORY WHERE EVERYTHING GOES COMPLETELY SIDEWAYS",
+ "raw": {
+ "normalized_reference": "so this is the part of the story where everything goes completely sideways.",
+ "normalized_hypothesis": "so thi's the part of the story where everything goes completely sideways",
+ "errors": 2,
+ "words": 13,
+ "wer": 0.15384615384615385
+ },
+ "semantic": {
+ "normalized_reference": "so this is the part of the story where everything goes completely sideways",
+ "normalized_hypothesis": "so thi is the part of the story where everything goes completely sideways",
+ "errors": 1,
+ "words": 13,
+ "wer": 0.07692307692307693
+ }
+ },
+ {
+ "id": "release500_0130",
+ "category": "question",
+ "text": "Is there a version of this where it actually works out okay?",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0130.wav",
+ "render_seconds": 74.49139826593455,
+ "audio_seconds": 4.179591836734694,
+ "rtf": 17.822648999173794,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IS THERE A VERSION OF THIS WHERE IT ACTUALLY WORKS OUT O QUT",
+ "raw": {
+ "normalized_reference": "is there a version of this where it actually works out okay?",
+ "normalized_hypothesis": "is there a version of this where it actually works out o qut",
+ "errors": 2,
+ "words": 12,
+ "wer": 0.16666666666666666
+ },
+ "semantic": {
+ "normalized_reference": "is there a version of this where it actually works out okay",
+ "normalized_hypothesis": "is there a version of this where it actually works out 0 qut",
+ "errors": 2,
+ "words": 12,
+ "wer": 0.16666666666666666
+ }
+ },
+ {
+ "id": "release500_0131",
+ "category": "short",
+ "text": "Oh no. Oh no no no.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0131.wav",
+ "render_seconds": 46.206697111018,
+ "audio_seconds": 2.089795918367347,
+ "rtf": 22.11062654726447,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "OH NO O NO NO KNO",
+ "raw": {
+ "normalized_reference": "oh no. oh no no no.",
+ "normalized_hypothesis": "oh no o no no kno",
+ "errors": 2,
+ "words": 6,
+ "wer": 0.3333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "0 no 0 no no no",
+ "normalized_hypothesis": "0 no 0 no no kno",
+ "errors": 1,
+ "words": 6,
+ "wer": 0.16666666666666666
+ }
+ },
+ {
+ "id": "release500_0132",
+ "category": "technical",
+ "text": "Always test on a copy before modifying the original.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0132.wav",
+ "render_seconds": 73.90473467798438,
+ "audio_seconds": 3.7616326530612243,
+ "rtf": 19.646983502719372,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "ALWAYS TEST ON A COPY BEFORE MODIFYING THE ORIGINAL",
+ "raw": {
+ "normalized_reference": "always test on a copy before modifying the original.",
+ "normalized_hypothesis": "always test on a copy before modifying the original",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "always test on a copy before modifying the original",
+ "normalized_hypothesis": "always test on a copy before modifying the original",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0133",
+ "category": "conversational",
+ "text": "I wasn't even supposed to be there. That's the wild part.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0133.wav",
+ "render_seconds": 43.90362838306464,
+ "audio_seconds": 3.9706122448979593,
+ "rtf": 11.05714325025785,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I WASN'T EVEN SUPPOSED TO BE THERE THAT'S THE WILD PART",
+ "raw": {
+ "normalized_reference": "i wasn't even supposed to be there. that's the wild part.",
+ "normalized_hypothesis": "i wasn't even supposed to be there that's the wild part",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i was not even supposed to be there that is the wild part",
+ "normalized_hypothesis": "i was not even supposed to be there that is the wild part",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0134",
+ "category": "descriptive",
+ "text": "It's a small town. Not the kind you leave. The kind you leave and come back to.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0134.wav",
+ "render_seconds": 97.39793498499785,
+ "audio_seconds": 5.7817687074829935,
+ "rtf": 16.845698939658654,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT'S A SMALL TOWN NOT THE KIND YOU LEAVE THA KIND YOU LEAVE AND COME BACK THROUGH",
+ "raw": {
+ "normalized_reference": "it's a small town. not the kind you leave. the kind you leave and come back to.",
+ "normalized_hypothesis": "it's a small town not the kind you leave tha kind you leave and come back through",
+ "errors": 2,
+ "words": 17,
+ "wer": 0.11764705882352941
+ },
+ "semantic": {
+ "normalized_reference": "it is a small town not the kind you leave the kind you leave and come back to",
+ "normalized_hypothesis": "it is a small town not the kind you leave tha kind you leave and come back through",
+ "errors": 2,
+ "words": 18,
+ "wer": 0.1111111111111111
+ }
+ },
+ {
+ "id": "release500_0135",
+ "category": "emotional",
+ "text": "She cried. I cried. We all just sort of stood there crying.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0135.wav",
+ "render_seconds": 79.69221897004172,
+ "audio_seconds": 4.806530612244898,
+ "rtf": 16.579987812211467,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SISID I CRIED WE ALL JUST SORT OF STOOD THERE CRYING",
+ "raw": {
+ "normalized_reference": "she cried. i cried. we all just sort of stood there crying.",
+ "normalized_hypothesis": "sisid i cried we all just sort of stood there crying",
+ "errors": 2,
+ "words": 12,
+ "wer": 0.16666666666666666
+ },
+ "semantic": {
+ "normalized_reference": "she cried i cried we all just sort of stood there crying",
+ "normalized_hypothesis": "sisid i cried we all just sort of stood there crying",
+ "errors": 2,
+ "words": 12,
+ "wer": 0.16666666666666666
+ }
+ },
+ {
+ "id": "release500_0136",
+ "category": "long",
+ "text": "There's a version of being busy that's actually just a very elaborate way of avoiding the one thing you know you should be doing.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0136.wav",
+ "render_seconds": 101.89901606296189,
+ "audio_seconds": 8.150204081632653,
+ "rtf": 12.502633681603397,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THERE'S A VERSION OF BEING BUSY THAT'S ACTUALLY JUST A VERY ELABORATE WAY OF AVOIDING THE ONE THING YOU KNOW YOU SHOULD BE DOING",
+ "raw": {
+ "normalized_reference": "there's a version of being busy that's actually just a very elaborate way of avoiding the one thing you know you should be doing.",
+ "normalized_hypothesis": "there's a version of being busy that's actually just a very elaborate way of avoiding the one thing you know you should be doing",
+ "errors": 0,
+ "words": 24,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "there is a version of being busy that is actually just a very elaborate way of avoiding the one thing you know you should be doing",
+ "normalized_hypothesis": "there is a version of being busy that is actually just a very elaborate way of avoiding the one thing you know you should be doing",
+ "errors": 0,
+ "words": 26,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0137",
+ "category": "narrative",
+ "text": "The letter arrived on a Tuesday, which felt appropriate somehow.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0137.wav",
+ "render_seconds": 100.79218495893292,
+ "audio_seconds": 4.667210884353741,
+ "rtf": 21.595806886898234,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE LETTER ARRIVED ON A TUESDAY WHICH FELT APPROPRIATE SOMEHOW",
+ "raw": {
+ "normalized_reference": "the letter arrived on a tuesday, which felt appropriate somehow.",
+ "normalized_hypothesis": "the letter arrived on a tuesday which felt appropriate somehow",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the letter arrived on a tuesday which felt appropriate somehow",
+ "normalized_hypothesis": "the letter arrived on a tuesday which felt appropriate somehow",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0138",
+ "category": "question",
+ "text": "What exactly did she say when you told her?",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0138.wav",
+ "render_seconds": 64.09573250508402,
+ "audio_seconds": 2.9257142857142857,
+ "rtf": 21.907721071073638,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WHAT EXACTLY DID SHE SAY WHEN YOU TOLD HER",
+ "raw": {
+ "normalized_reference": "what exactly did she say when you told her?",
+ "normalized_hypothesis": "what exactly did she say when you told her",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "what exactly did she say when you told her",
+ "normalized_hypothesis": "what exactly did she say when you told her",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0139",
+ "category": "short",
+ "text": "No way. Absolutely not.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0139.wav",
+ "render_seconds": 23.296498080948368,
+ "audio_seconds": 2.5774149659863945,
+ "rtf": 9.03870676176998,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "OH WELL I'LL TURN IT PON THEM OP",
+ "raw": {
+ "normalized_reference": "no way. absolutely not.",
+ "normalized_hypothesis": "oh well i'll turn it pon them op",
+ "errors": 8,
+ "words": 4,
+ "wer": 2.0
+ },
+ "semantic": {
+ "normalized_reference": "no way absolutely not",
+ "normalized_hypothesis": "0 well i will turn it pon them op",
+ "errors": 9,
+ "words": 4,
+ "wer": 2.25
+ }
+ },
+ {
+ "id": "release500_0140",
+ "category": "technical",
+ "text": "The first thing you want to do is make sure everything is backed up.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0140.wav",
+ "render_seconds": 76.69429815898184,
+ "audio_seconds": 4.597551020408163,
+ "rtf": 16.6815545534007,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE FIRST THING YOU WANT TO DO IS MAKE SURE EVERYTHING IS BACKED UF",
+ "raw": {
+ "normalized_reference": "the first thing you want to do is make sure everything is backed up.",
+ "normalized_hypothesis": "the first thing you want to do is make sure everything is backed uf",
+ "errors": 1,
+ "words": 14,
+ "wer": 0.07142857142857142
+ },
+ "semantic": {
+ "normalized_reference": "the 1st thing you want to do is make sure everything is backed up",
+ "normalized_hypothesis": "the 1st thing you want to do is make sure everything is backed uf",
+ "errors": 1,
+ "words": 14,
+ "wer": 0.07142857142857142
+ }
+ },
+ {
+ "id": "release500_0141",
+ "category": "conversational",
+ "text": "I have the most insane idea and you need to just hear me out.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0141.wav",
+ "render_seconds": 74.30298421508633,
+ "audio_seconds": 4.179591836734694,
+ "rtf": 17.777569465523584,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I HAVE THE MOST INSANE IDEA AND YOU NEED TO JUST HEAR ME OUT",
+ "raw": {
+ "normalized_reference": "i have the most insane idea and you need to just hear me out.",
+ "normalized_hypothesis": "i have the most insane idea and you need to just hear me out",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i have the most insane idea and you need to just hear me out",
+ "normalized_hypothesis": "i have the most insane idea and you need to just hear me out",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0142",
+ "category": "descriptive",
+ "text": "There's a coffee shop on the corner that's been there longer than anyone can remember.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0142.wav",
+ "render_seconds": 83.41060081892647,
+ "audio_seconds": 5.085170068027211,
+ "rtf": 16.402716074997578,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THERE'S A COFFEE SHOP ON THE CORNER THAT'S BEEN THERE LONGER THAN ANY ONE CAN REMEMBER",
+ "raw": {
+ "normalized_reference": "there's a coffee shop on the corner that's been there longer than anyone can remember.",
+ "normalized_hypothesis": "there's a coffee shop on the corner that's been there longer than any one can remember",
+ "errors": 2,
+ "words": 15,
+ "wer": 0.13333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "there is a coffee shop on the corner that has been there longer than anyone can remember",
+ "normalized_hypothesis": "there is a coffee shop on the corner that has been there longer than any one can remember",
+ "errors": 2,
+ "words": 17,
+ "wer": 0.11764705882352941
+ }
+ },
+ {
+ "id": "release500_0143",
+ "category": "emotional",
+ "text": "I kept it together until I got to the car, and then I completely fell apart.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0143.wav",
+ "render_seconds": 96.30345823802054,
+ "audio_seconds": 5.294149659863946,
+ "rtf": 18.190543227010973,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I KEPT IT TOGETHER UNTIL I GOT TO THE CAR AND THEN I COMPLETELY FELL APART",
+ "raw": {
+ "normalized_reference": "i kept it together until i got to the car, and then i completely fell apart.",
+ "normalized_hypothesis": "i kept it together until i got to the car and then i completely fell apart",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i kept it together until i got to the car and then i completely fell apart",
+ "normalized_hypothesis": "i kept it together until i got to the car and then i completely fell apart",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0144",
+ "category": "long",
+ "text": "The older I get, the more I think that most of what we call wisdom is just pattern recognition built from having been wrong enough times to start noticing.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0144.wav",
+ "render_seconds": 148.19997806602623,
+ "audio_seconds": 9.682721088435374,
+ "rtf": 15.305612617823922,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE OLDER I GET THE MORE I THINK THAT MOST OF WHAT WE CALL WISDOM IS JUST PATTERN RECOGNITION BUILT FROM HAVING BEEN WRONG ENOUGH TIMES TO START NOTICING",
+ "raw": {
+ "normalized_reference": "the older i get, the more i think that most of what we call wisdom is just pattern recognition built from having been wrong enough times to start noticing.",
+ "normalized_hypothesis": "the older i get the more i think that most of what we call wisdom is just pattern recognition built from having been wrong enough times to start noticing",
+ "errors": 0,
+ "words": 29,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the older i get the more i think that most of what we call wisdom is just pattern recognition built from having been wrong enough times to start noticing",
+ "normalized_hypothesis": "the older i get the more i think that most of what we call wisdom is just pattern recognition built from having been wrong enough times to start noticing",
+ "errors": 0,
+ "words": 29,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0145",
+ "category": "narrative",
+ "text": "I remember the exact moment I realized I had made a huge mistake.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0145.wav",
+ "render_seconds": 42.1013664790662,
+ "audio_seconds": 4.527891156462585,
+ "rtf": 9.298228474192804,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I REMEMBER THE EXACT MOMENT I REALIZED I HAD MADE A HUGE MISTAKE",
+ "raw": {
+ "normalized_reference": "i remember the exact moment i realized i had made a huge mistake.",
+ "normalized_hypothesis": "i remember the exact moment i realized i had made a huge mistake",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i remember the exact moment i realized i had made a huge mistake",
+ "normalized_hypothesis": "i remember the exact moment i realized i had made a huge mistake",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0146",
+ "category": "question",
+ "text": "Do you think it would have gone differently if we'd left earlier?",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0146.wav",
+ "render_seconds": 79.8013374650618,
+ "audio_seconds": 3.9706122448979593,
+ "rtf": 20.09799309101577,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "DO YOU THINK IT WOULD HAVE GONE DIFFERENTLY IF WE'D LEFT EARLIER",
+ "raw": {
+ "normalized_reference": "do you think it would have gone differently if we'd left earlier?",
+ "normalized_hypothesis": "do you think it would have gone differently if we'd left earlier",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "do you think it would have gone differently if we would left earlier",
+ "normalized_hypothesis": "do you think it would have gone differently if we would left earlier",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0147",
+ "category": "short",
+ "text": "You have to be kidding me.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0147.wav",
+ "render_seconds": 50.69574046693742,
+ "audio_seconds": 1.9504761904761905,
+ "rtf": 25.991468501115378,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "YOU'VE ASKED ABOUT HRITHENEMI",
+ "raw": {
+ "normalized_reference": "you have to be kidding me.",
+ "normalized_hypothesis": "you've asked about hrithenemi",
+ "errors": 6,
+ "words": 6,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "you have to be kidding me",
+ "normalized_hypothesis": "you have asked about hrithenemi",
+ "errors": 4,
+ "words": 6,
+ "wer": 0.6666666666666666
+ }
+ },
+ {
+ "id": "release500_0148",
+ "category": "technical",
+ "text": "If the checksum fails, restart the worker process without deleting the cached tensors.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0148.wav",
+ "render_seconds": 106.50020606000908,
+ "audio_seconds": 5.851428571428571,
+ "rtf": 18.200718809083583,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IF THE CHEKSUMB FAILS RESTART THE WORKER PROCESS WITHOUT DELEAVING THE CATCH TENSERS",
+ "raw": {
+ "normalized_reference": "if the checksum fails, restart the worker process without deleting the cached tensors.",
+ "normalized_hypothesis": "if the cheksumb fails restart the worker process without deleaving the catch tensers",
+ "errors": 4,
+ "words": 13,
+ "wer": 0.3076923076923077
+ },
+ "semantic": {
+ "normalized_reference": "if the checksum fails restart the worker process without deleting the cached tensors",
+ "normalized_hypothesis": "if the cheksumb fails restart the worker process without deleaving the catch tensers",
+ "errors": 4,
+ "words": 13,
+ "wer": 0.3076923076923077
+ }
+ },
+ {
+ "id": "release500_0149",
+ "category": "conversational",
+ "text": "I need you to understand how close that was to not working.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0149.wav",
+ "render_seconds": 68.99980113690253,
+ "audio_seconds": 3.900952380952381,
+ "rtf": 17.68793730316105,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I NEED YOU TO UNDERSTAND HOW CLOSE THAT WAS TO NOT WORKING",
+ "raw": {
+ "normalized_reference": "i need you to understand how close that was to not working.",
+ "normalized_hypothesis": "i need you to understand how close that was to not working",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i need you to understand how close that was to not working",
+ "normalized_hypothesis": "i need you to understand how close that was to not working",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0150",
+ "category": "descriptive",
+ "text": "It was the sort of morning where the whole world looks rinsed clean.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0150.wav",
+ "render_seconds": 87.79981729900464,
+ "audio_seconds": 4.527891156462585,
+ "rtf": 19.390885130639546,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT WAS THE SORT OF MORNING WHERE THE WHOLE WORLD LOOKS RENCED CLEAN",
+ "raw": {
+ "normalized_reference": "it was the sort of morning where the whole world looks rinsed clean.",
+ "normalized_hypothesis": "it was the sort of morning where the whole world looks renced clean",
+ "errors": 1,
+ "words": 13,
+ "wer": 0.07692307692307693
+ },
+ "semantic": {
+ "normalized_reference": "it was the sort of morning where the whole world looks rinsed clean",
+ "normalized_hypothesis": "it was the sort of morning where the whole world looks renced clean",
+ "errors": 1,
+ "words": 13,
+ "wer": 0.07692307692307693
+ }
+ },
+ {
+ "id": "release500_0151",
+ "category": "emotional",
+ "text": "She started laughing and I started laughing and we couldn't stop for a full minute.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0151.wav",
+ "render_seconds": 49.10540115297772,
+ "audio_seconds": 5.433469387755102,
+ "rtf": 9.037577585997251,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SHE STARTED LAUGHING AND I STARTED LAUGHING AND WE COULDN'T STOP FOR A FULL MINUTE",
+ "raw": {
+ "normalized_reference": "she started laughing and i started laughing and we couldn't stop for a full minute.",
+ "normalized_hypothesis": "she started laughing and i started laughing and we couldn't stop for a full minute",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "she started laughing and i started laughing and we could not stop for a full minute",
+ "normalized_hypothesis": "she started laughing and i started laughing and we could not stop for a full minute",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0152",
+ "category": "long",
+ "text": "We spent three years building something we were genuinely proud of, and then the market shifted and none of it mattered the way we thought it would.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0152.wav",
+ "render_seconds": 136.8979239440523,
+ "audio_seconds": 9.543401360544218,
+ "rtf": 14.344772767289925,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WE SPENT THREE YEARS BUILDING SOMETHING WE WERE GENUINELY PROUD OF AND THEN THE MARKET SHIFTED AND NONE OF IT MATTERED THE WAY WE THOUGHT IT WOULD",
+ "raw": {
+ "normalized_reference": "we spent three years building something we were genuinely proud of, and then the market shifted and none of it mattered the way we thought it would.",
+ "normalized_hypothesis": "we spent three years building something we were genuinely proud of and then the market shifted and none of it mattered the way we thought it would",
+ "errors": 0,
+ "words": 27,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "we spent 3 years building something we were genuinely proud of and then the market shifted and none of it mattered the way we thought it would",
+ "normalized_hypothesis": "we spent 3 years building something we were genuinely proud of and then the market shifted and none of it mattered the way we thought it would",
+ "errors": 0,
+ "words": 27,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0153",
+ "category": "narrative",
+ "text": "And somehow, despite all of that, it actually turned out okay.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0153.wav",
+ "render_seconds": 75.70023390301503,
+ "audio_seconds": 4.527891156462585,
+ "rtf": 16.718651417883425,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "AND SHANHA DESPITE ALL OF THAT IT ACTUALLY TURNED OUT OPERN",
+ "raw": {
+ "normalized_reference": "and somehow, despite all of that, it actually turned out okay.",
+ "normalized_hypothesis": "and shanha despite all of that it actually turned out opern",
+ "errors": 2,
+ "words": 11,
+ "wer": 0.18181818181818182
+ },
+ "semantic": {
+ "normalized_reference": "and somehow despite all of that it actually turned out okay",
+ "normalized_hypothesis": "and shanha despite all of that it actually turned out opern",
+ "errors": 2,
+ "words": 11,
+ "wer": 0.18181818181818182
+ }
+ },
+ {
+ "id": "release500_0154",
+ "category": "question",
+ "text": "Is that something you actually want, or something you think you should want?",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0154.wav",
+ "render_seconds": 92.69187824998517,
+ "audio_seconds": 4.876190476190477,
+ "rtf": 19.00907659423524,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IS THAT SOMETHING YOU ACTUALLY WANT OR SOMETHING YOU THINK YOU SHOULD WANT",
+ "raw": {
+ "normalized_reference": "is that something you actually want, or something you think you should want?",
+ "normalized_hypothesis": "is that something you actually want or something you think you should want",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "is that something you actually want or something you think you should want",
+ "normalized_hypothesis": "is that something you actually want or something you think you should want",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0155",
+ "category": "short",
+ "text": "This is fine.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0155.wav",
+ "render_seconds": 28.795869291061535,
+ "audio_seconds": 1.5325170068027212,
+ "rtf": 18.789918260769024,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "NRRVAN",
+ "raw": {
+ "normalized_reference": "this is fine.",
+ "normalized_hypothesis": "nrrvan",
+ "errors": 3,
+ "words": 3,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "this is fine",
+ "normalized_hypothesis": "nrrvan",
+ "errors": 3,
+ "words": 3,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0156",
+ "category": "technical",
+ "text": "The tricky part isn't the setup, it's making sure all dependencies line up.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0156.wav",
+ "render_seconds": 89.29555551894009,
+ "audio_seconds": 4.9458503401360545,
+ "rtf": 18.054641644592035,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE TRICKY PART ISN'T THE SET UP IT'S MAKING SURE ALL DEPENDENCIES LINE UP",
+ "raw": {
+ "normalized_reference": "the tricky part isn't the setup, it's making sure all dependencies line up.",
+ "normalized_hypothesis": "the tricky part isn't the set up it's making sure all dependencies line up",
+ "errors": 2,
+ "words": 13,
+ "wer": 0.15384615384615385
+ },
+ "semantic": {
+ "normalized_reference": "the tricky part is not the setup it is making sure all dependencies line up",
+ "normalized_hypothesis": "the tricky part is not the set up it is making sure all dependencies line up",
+ "errors": 2,
+ "words": 15,
+ "wer": 0.13333333333333333
+ }
+ },
+ {
+ "id": "release500_0157",
+ "category": "conversational",
+ "text": "He called it a minor setback. I called it a disaster.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0157.wav",
+ "render_seconds": 61.599523954093456,
+ "audio_seconds": 4.1099319727891155,
+ "rtf": 14.987966798946866,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE CALLED IT A MINOR SET BACK I CALLED IT A DISASTER",
+ "raw": {
+ "normalized_reference": "he called it a minor setback. i called it a disaster.",
+ "normalized_hypothesis": "he called it a minor set back i called it a disaster",
+ "errors": 2,
+ "words": 11,
+ "wer": 0.18181818181818182
+ },
+ "semantic": {
+ "normalized_reference": "he called it a minor setback i called it a disaster",
+ "normalized_hypothesis": "he called it a minor set back i called it a disaster",
+ "errors": 2,
+ "words": 11,
+ "wer": 0.18181818181818182
+ }
+ },
+ {
+ "id": "release500_0158",
+ "category": "descriptive",
+ "text": "The whole place smelled like pine and old wood and cold air.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0158.wav",
+ "render_seconds": 81.99821299104951,
+ "audio_seconds": 4.31891156462585,
+ "rtf": 18.985851357239603,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE WHOLE PLACE SMELLED LIKE PINE AND OLD WOOD AND COLD AIR",
+ "raw": {
+ "normalized_reference": "the whole place smelled like pine and old wood and cold air.",
+ "normalized_hypothesis": "the whole place smelled like pine and old wood and cold air",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the whole place smelled like pine and old wood and cold air",
+ "normalized_hypothesis": "the whole place smelled like pine and old wood and cold air",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0159",
+ "category": "emotional",
+ "text": "I've never been so relieved in my entire life.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0159.wav",
+ "render_seconds": 58.11096977791749,
+ "audio_seconds": 3.2043537414965986,
+ "rtf": 18.135004579980194,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I'VE NEVER BEEN SO RELIEVED IN MY ENTIRE LIFE",
+ "raw": {
+ "normalized_reference": "i've never been so relieved in my entire life.",
+ "normalized_hypothesis": "i've never been so relieved in my entire life",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i have never been so relieved in my entire life",
+ "normalized_hypothesis": "i have never been so relieved in my entire life",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0160",
+ "category": "narrative",
+ "text": "I wrote it all down that night so I wouldn't forget, and I still haven't looked at it.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0160.wav",
+ "render_seconds": 99.69554700597655,
+ "audio_seconds": 5.363809523809524,
+ "rtf": 18.58670531894094,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I RAINE IT ALL DOWN THAT NIGHT SO I WOULDN'T FORGET AND I STILL HAVEN'T LOOKED AT IT",
+ "raw": {
+ "normalized_reference": "i wrote it all down that night so i wouldn't forget, and i still haven't looked at it.",
+ "normalized_hypothesis": "i raine it all down that night so i wouldn't forget and i still haven't looked at it",
+ "errors": 1,
+ "words": 18,
+ "wer": 0.05555555555555555
+ },
+ "semantic": {
+ "normalized_reference": "i wrote it all down that night so i would not forget and i still have not looked at it",
+ "normalized_hypothesis": "i raine it all down that night so i would not forget and i still have not looked at it",
+ "errors": 1,
+ "words": 20,
+ "wer": 0.05
+ }
+ },
+ {
+ "id": "release500_0161",
+ "category": "short",
+ "text": "That's kind of wild actually.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0161.wav",
+ "render_seconds": 44.50766643101815,
+ "audio_seconds": 2.2291156462585033,
+ "rtf": 19.966512955809534,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THAT'S A KIND OF WILD AXILE",
+ "raw": {
+ "normalized_reference": "that's kind of wild actually.",
+ "normalized_hypothesis": "that's a kind of wild axile",
+ "errors": 2,
+ "words": 5,
+ "wer": 0.4
+ },
+ "semantic": {
+ "normalized_reference": "that is kind of wild actually",
+ "normalized_hypothesis": "that is a kind of wild axile",
+ "errors": 2,
+ "words": 6,
+ "wer": 0.3333333333333333
+ }
+ },
+ {
+ "id": "release500_0162",
+ "category": "technical",
+ "text": "Version pinning will save you a lot of pain later down the line.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0162.wav",
+ "render_seconds": 76.50037862698082,
+ "audio_seconds": 4.31891156462585,
+ "rtf": 17.71288378617405,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "VERSION PINNING WILL SAVE YOU A LOT OF PAIN LATER DOWN THE LINE",
+ "raw": {
+ "normalized_reference": "version pinning will save you a lot of pain later down the line.",
+ "normalized_hypothesis": "version pinning will save you a lot of pain later down the line",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "version pinning will save you a lot of pain later down the line",
+ "normalized_hypothesis": "version pinning will save you a lot of pain later down the line",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0163",
+ "category": "conversational",
+ "text": "At some point you just have to commit and see what happens.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0163.wav",
+ "render_seconds": 46.49331007502042,
+ "audio_seconds": 4.040272108843538,
+ "rtf": 11.507469997689926,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "AT SOME POINT YOU JUST HAVE TO COMMIT AND SEE WHAT HAPPENS",
+ "raw": {
+ "normalized_reference": "at some point you just have to commit and see what happens.",
+ "normalized_hypothesis": "at some point you just have to commit and see what happens",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "at some point you just have to commit and see what happens",
+ "normalized_hypothesis": "at some point you just have to commit and see what happens",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0164",
+ "category": "emotional",
+ "text": "Nobody told me it was going to be this hard, and honestly, I'm grateful.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0164.wav",
+ "render_seconds": 77.01046454103198,
+ "audio_seconds": 5.085170068027211,
+ "rtf": 15.144127632079009,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "NOBODY TOLD ME IT WAS GOING TO BE THIS HARD AND HONESTLY I'M GRATEFUL",
+ "raw": {
+ "normalized_reference": "nobody told me it was going to be this hard, and honestly, i'm grateful.",
+ "normalized_hypothesis": "nobody told me it was going to be this hard and honestly i'm grateful",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "nobody told me it was going to be this hard and honestly i am grateful",
+ "normalized_hypothesis": "nobody told me it was going to be this hard and honestly i am grateful",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0165",
+ "category": "narrative",
+ "text": "Chapter seven. In which everything that could go wrong did, and a few things that couldn't.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0165.wav",
+ "render_seconds": 97.00171974499244,
+ "audio_seconds": 6.060408163265306,
+ "rtf": 16.005806396499967,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "CHAPTER SEVEN IN WHICH EVERYTHING THAT COULD GO WRONG DID AND A FEW THINGS THAT COULDN'T",
+ "raw": {
+ "normalized_reference": "chapter seven. in which everything that could go wrong did, and a few things that couldn't.",
+ "normalized_hypothesis": "chapter seven in which everything that could go wrong did and a few things that couldn't",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "chapter 7 in which everything that could go wrong did and a few things that could not",
+ "normalized_hypothesis": "chapter 7 in which everything that could go wrong did and a few things that could not",
+ "errors": 0,
+ "words": 17,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0166",
+ "category": "short",
+ "text": "Wait, what?",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0166.wav",
+ "render_seconds": 36.80749314709101,
+ "audio_seconds": 1.3931972789115645,
+ "rtf": 26.41944088194521,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WAIT WHAT",
+ "raw": {
+ "normalized_reference": "wait, what?",
+ "normalized_hypothesis": "wait what",
+ "errors": 0,
+ "words": 2,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "wait what",
+ "normalized_hypothesis": "wait what",
+ "errors": 0,
+ "words": 2,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0167",
+ "category": "technical",
+ "text": "The graphics processor remained idle because the data loader exhausted every central processing core.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0167.wav",
+ "render_seconds": 116.69261429400649,
+ "audio_seconds": 7.174965986394557,
+ "rtf": 16.26385609566421,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE GRAPHICS PROCESSOR REMAINED IDLE BECAUSE THE DATA LODER EXHAUSTED EVERY CENTRAL PROCESSING CORPE",
+ "raw": {
+ "normalized_reference": "the graphics processor remained idle because the data loader exhausted every central processing core.",
+ "normalized_hypothesis": "the graphics processor remained idle because the data loder exhausted every central processing corpe",
+ "errors": 2,
+ "words": 14,
+ "wer": 0.14285714285714285
+ },
+ "semantic": {
+ "normalized_reference": "the graphics processor remained idle because the data loader exhausted every central processing core",
+ "normalized_hypothesis": "the graphics processor remained idle because the data loder exhausted every central processing corpe",
+ "errors": 2,
+ "words": 14,
+ "wer": 0.14285714285714285
+ }
+ },
+ {
+ "id": "release500_0168",
+ "category": "conversational",
+ "text": "Could you speak a little more slowly? I am trying to write every detail down.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0168.wav",
+ "render_seconds": 114.00417210010346,
+ "audio_seconds": 5.712108843537415,
+ "rtf": 19.958333292105575,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "COULD YOU SPEAK A LITTLE MORE SLOWLY I AM TRYING TO WRITE EVERY DETAIL DOWN",
+ "raw": {
+ "normalized_reference": "could you speak a little more slowly? i am trying to write every detail down.",
+ "normalized_hypothesis": "could you speak a little more slowly i am trying to write every detail down",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "could you speak a little more slowly i am trying to write every detail down",
+ "normalized_hypothesis": "could you speak a little more slowly i am trying to write every detail down",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0169",
+ "category": "emotional",
+ "text": "I was so excited I could barely form a complete sentence.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0169.wav",
+ "render_seconds": 33.42832995997742,
+ "audio_seconds": 4.31891156462585,
+ "rtf": 7.739989453308784,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I WAS SO EXCITED I COULD BARELY FORM A COMPLETE SENTENCE",
+ "raw": {
+ "normalized_reference": "i was so excited i could barely form a complete sentence.",
+ "normalized_hypothesis": "i was so excited i could barely form a complete sentence",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i was so excited i could barely form a complete sentence",
+ "normalized_hypothesis": "i was so excited i could barely form a complete sentence",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0170",
+ "category": "narrative",
+ "text": "He looked at me like I had said something in a language he didn't speak.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0170.wav",
+ "render_seconds": 78.58920265594497,
+ "audio_seconds": 4.667210884353741,
+ "rtf": 16.838579743505,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE LOOKED AT ME LIKE I HAD SAID SOMETHING IN THE LANGUAGE HE DIDN'T SPEAK",
+ "raw": {
+ "normalized_reference": "he looked at me like i had said something in a language he didn't speak.",
+ "normalized_hypothesis": "he looked at me like i had said something in the language he didn't speak",
+ "errors": 1,
+ "words": 15,
+ "wer": 0.06666666666666667
+ },
+ "semantic": {
+ "normalized_reference": "he looked at me like i had said something in a language he did not speak",
+ "normalized_hypothesis": "he looked at me like i had said something in the language he did not speak",
+ "errors": 1,
+ "words": 16,
+ "wer": 0.0625
+ }
+ },
+ {
+ "id": "release500_0171",
+ "category": "short",
+ "text": "That's actually hilarious.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0171.wav",
+ "render_seconds": 55.39334730489645,
+ "audio_seconds": 2.159455782312925,
+ "rtf": 25.651531167527025,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "TAT ACTARY ON TER YES",
+ "raw": {
+ "normalized_reference": "that's actually hilarious.",
+ "normalized_hypothesis": "tat actary on ter yes",
+ "errors": 5,
+ "words": 3,
+ "wer": 1.6666666666666667
+ },
+ "semantic": {
+ "normalized_reference": "that is actually hilarious",
+ "normalized_hypothesis": "tat actary on ter yes",
+ "errors": 5,
+ "words": 4,
+ "wer": 1.25
+ }
+ },
+ {
+ "id": "release500_0172",
+ "category": "technical",
+ "text": "The encoder maps phonemes, stress, and phrase boundaries into a shared hidden representation.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0172.wav",
+ "render_seconds": 96.20001360599417,
+ "audio_seconds": 6.617687074829932,
+ "rtf": 14.536803042846572,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE ENCOTER MAPS PHONIMES STRESS AND PHRASE BOUNDARIES INTO A SHARED HIDDEN REPRESENTATION",
+ "raw": {
+ "normalized_reference": "the encoder maps phonemes, stress, and phrase boundaries into a shared hidden representation.",
+ "normalized_hypothesis": "the encoter maps phonimes stress and phrase boundaries into a shared hidden representation",
+ "errors": 2,
+ "words": 13,
+ "wer": 0.15384615384615385
+ },
+ "semantic": {
+ "normalized_reference": "the encoder maps phonemes stress and phrase boundaries into a shared hidden representation",
+ "normalized_hypothesis": "the encoter maps phonimes stress and phrase boundaries into a shared hidden representation",
+ "errors": 2,
+ "words": 13,
+ "wer": 0.15384615384615385
+ }
+ },
+ {
+ "id": "release500_0173",
+ "category": "conversational",
+ "text": "She showed up three hours late and acted like nothing had happened.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0173.wav",
+ "render_seconds": 78.99925330001861,
+ "audio_seconds": 4.597551020408163,
+ "rtf": 17.182898666996234,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SHE SHOWED UP THREE HOURS LATE AND ACTED LIKE NOTHING HAD HAPPENED",
+ "raw": {
+ "normalized_reference": "she showed up three hours late and acted like nothing had happened.",
+ "normalized_hypothesis": "she showed up three hours late and acted like nothing had happened",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "she showed up 3 hours late and acted like nothing had happened",
+ "normalized_hypothesis": "she showed up 3 hours late and acted like nothing had happened",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0174",
+ "category": "emotional",
+ "text": "Oh my god. This is the best thing I've ever tasted.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0174.wav",
+ "render_seconds": 78.99571508297231,
+ "audio_seconds": 3.9706122448979593,
+ "rtf": 19.89509682907917,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "O MY GOD THIS IS THE BEST THING I'VE EVER TASTED",
+ "raw": {
+ "normalized_reference": "oh my god. this is the best thing i've ever tasted.",
+ "normalized_hypothesis": "o my god this is the best thing i've ever tasted",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ },
+ "semantic": {
+ "normalized_reference": "0 my god this is the best thing i have ever tasted",
+ "normalized_hypothesis": "0 my god this is the best thing i have ever tasted",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0175",
+ "category": "narrative",
+ "text": "The note was short, which was somehow more frightening than a long one would have been.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0175.wav",
+ "render_seconds": 46.504088663030416,
+ "audio_seconds": 5.572789115646258,
+ "rtf": 8.34484989436703,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE NOTE WAS SHORT WHICH WAS SOMEHOW MORE FRIGHTENING THAN A LONG ONE WOULD HAVE BEEN",
+ "raw": {
+ "normalized_reference": "the note was short, which was somehow more frightening than a long one would have been.",
+ "normalized_hypothesis": "the note was short which was somehow more frightening than a long one would have been",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the note was short which was somehow more frightening than a long one would have been",
+ "normalized_hypothesis": "the note was short which was somehow more frightening than a long one would have been",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0176",
+ "category": "short",
+ "text": "Noted.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0176.wav",
+ "render_seconds": 38.10715349006932,
+ "audio_seconds": 1.3235374149659864,
+ "rtf": 28.791897425165452,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "NOTED",
+ "raw": {
+ "normalized_reference": "noted.",
+ "normalized_hypothesis": "noted",
+ "errors": 0,
+ "words": 1,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "noted",
+ "normalized_hypothesis": "noted",
+ "errors": 0,
+ "words": 1,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0177",
+ "category": "technical",
+ "text": "It's a simple process once you understand the underlying structure.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0177.wav",
+ "render_seconds": 78.19265920203179,
+ "audio_seconds": 4.597551020408163,
+ "rtf": 17.007458722032837,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT'S A SIMPLE PROCESS ONCE YOU UNDERSTAND THE UNDERLYING STRUCTURE",
+ "raw": {
+ "normalized_reference": "it's a simple process once you understand the underlying structure.",
+ "normalized_hypothesis": "it's a simple process once you understand the underlying structure",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it is a simple process once you understand the underlying structure",
+ "normalized_hypothesis": "it is a simple process once you understand the underlying structure",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0178",
+ "category": "conversational",
+ "text": "You have to hear this story because it's almost too good to be true.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0178.wav",
+ "render_seconds": 80.99190462101251,
+ "audio_seconds": 4.458231292517007,
+ "rtf": 18.16682430921759,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "YOU HAVE TO HEAR THIS STORY BECAUSE IT'S ALMOST TOO GOOD TO BE TRUE",
+ "raw": {
+ "normalized_reference": "you have to hear this story because it's almost too good to be true.",
+ "normalized_hypothesis": "you have to hear this story because it's almost too good to be true",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "you have to hear this story because it is almost too good to be true",
+ "normalized_hypothesis": "you have to hear this story because it is almost too good to be true",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0179",
+ "category": "emotional",
+ "text": "I have never felt more seen in my entire life than in that moment.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0179.wav",
+ "render_seconds": 91.00180944101885,
+ "audio_seconds": 4.458231292517007,
+ "rtf": 20.412087994125017,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I HAVE NEVER FELT MORE SEEN IN MY ENTIRE LIFE THAN IN THAT MOMENT",
+ "raw": {
+ "normalized_reference": "i have never felt more seen in my entire life than in that moment.",
+ "normalized_hypothesis": "i have never felt more seen in my entire life than in that moment",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i have never felt more seen in my entire life than in that moment",
+ "normalized_hypothesis": "i have never felt more seen in my entire life than in that moment",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0180",
+ "category": "narrative",
+ "text": "She knew. She had always known. She just waited for the right moment.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0180.wav",
+ "render_seconds": 82.90699441102333,
+ "audio_seconds": 5.224489795918367,
+ "rtf": 15.868916898984937,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SHE KNEW SHE HAD ALWAYS KNOWN SHE JUST WAITED FOR THE RIGHT MOMENT",
+ "raw": {
+ "normalized_reference": "she knew. she had always known. she just waited for the right moment.",
+ "normalized_hypothesis": "she knew she had always known she just waited for the right moment",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "she knew she had always known she just waited for the right moment",
+ "normalized_hypothesis": "she knew she had always known she just waited for the right moment",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0181",
+ "category": "short",
+ "text": "Yeah, no. Hard pass.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0181.wav",
+ "render_seconds": 18.877496225060895,
+ "audio_seconds": 1.9504761904761905,
+ "rtf": 9.678403826325166,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "GE NI SHILD PASS",
+ "raw": {
+ "normalized_reference": "yeah, no. hard pass.",
+ "normalized_hypothesis": "ge ni shild pass",
+ "errors": 3,
+ "words": 4,
+ "wer": 0.75
+ },
+ "semantic": {
+ "normalized_reference": "yeah no hard pass",
+ "normalized_hypothesis": "ge ni shild pass",
+ "errors": 3,
+ "words": 4,
+ "wer": 0.75
+ }
+ },
+ {
+ "id": "release500_0182",
+ "category": "technical",
+ "text": "Please verify the sample rate, hop length, and mel filter configuration before inference.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0182.wav",
+ "render_seconds": 104.69275171903428,
+ "audio_seconds": 6.199727891156463,
+ "rtf": 16.886668827574216,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "PLEASE VERIFY THE SAMPLE RATE HOP LENGTH AND MAIL FILTER CONFIGURATION BEFORE INFERENCE",
+ "raw": {
+ "normalized_reference": "please verify the sample rate, hop length, and mel filter configuration before inference.",
+ "normalized_hypothesis": "please verify the sample rate hop length and mail filter configuration before inference",
+ "errors": 1,
+ "words": 13,
+ "wer": 0.07692307692307693
+ },
+ "semantic": {
+ "normalized_reference": "please verify the sample rate hop length and mel filter configuration before inference",
+ "normalized_hypothesis": "please verify the sample rate hop length and mail filter configuration before inference",
+ "errors": 1,
+ "words": 13,
+ "wer": 0.07692307692307693
+ }
+ },
+ {
+ "id": "release500_0183",
+ "category": "conversational",
+ "text": "That's the worst idea I've ever heard and I'm completely in.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0183.wav",
+ "render_seconds": 56.61120985401794,
+ "audio_seconds": 3.8312925170068026,
+ "rtf": 14.776008253800848,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THAT'S THE WORST IDEA I'VE EVER HEARD AND I'M COMPLETELY IN",
+ "raw": {
+ "normalized_reference": "that's the worst idea i've ever heard and i'm completely in.",
+ "normalized_hypothesis": "that's the worst idea i've ever heard and i'm completely in",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "that is the worst idea i have ever heard and i am completely in",
+ "normalized_hypothesis": "that is the worst idea i have ever heard and i am completely in",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0184",
+ "category": "emotional",
+ "text": "I remember thinking: this is it. This is the thing I've been waiting for.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0184.wav",
+ "render_seconds": 92.70035656006075,
+ "audio_seconds": 5.154829931972789,
+ "rtf": 17.98320367178121,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I REMEMBER THINKING THIS IS IT THIS IS THE THING I'VE BEEN WAITING FOYER",
+ "raw": {
+ "normalized_reference": "i remember thinking: this is it. this is the thing i've been waiting for.",
+ "normalized_hypothesis": "i remember thinking this is it this is the thing i've been waiting foyer",
+ "errors": 1,
+ "words": 14,
+ "wer": 0.07142857142857142
+ },
+ "semantic": {
+ "normalized_reference": "i remember thinking this is it this is the thing i have been waiting for",
+ "normalized_hypothesis": "i remember thinking this is it this is the thing i have been waiting foyer",
+ "errors": 1,
+ "words": 15,
+ "wer": 0.06666666666666667
+ }
+ },
+ {
+ "id": "release500_0185",
+ "category": "narrative",
+ "text": "We never talked about it again, which I think was the right call.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0185.wav",
+ "render_seconds": 87.39801979297772,
+ "audio_seconds": 4.458231292517007,
+ "rtf": 19.60374284296833,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WE NEVER TALKED ABOUT IT AGAIN WHICH I THINK WAS THE RIGHT CALL",
+ "raw": {
+ "normalized_reference": "we never talked about it again, which i think was the right call.",
+ "normalized_hypothesis": "we never talked about it again which i think was the right call",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "we never talked about it again which i think was the right call",
+ "normalized_hypothesis": "we never talked about it again which i think was the right call",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0186",
+ "category": "short",
+ "text": "In hindsight, obvious.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0186.wav",
+ "render_seconds": 46.29941581108142,
+ "audio_seconds": 1.9504761904761905,
+ "rtf": 23.737493457829828,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "AN HIGNSIGHT OBVIOUS",
+ "raw": {
+ "normalized_reference": "in hindsight, obvious.",
+ "normalized_hypothesis": "an hignsight obvious",
+ "errors": 2,
+ "words": 3,
+ "wer": 0.6666666666666666
+ },
+ "semantic": {
+ "normalized_reference": "in hindsight obvious",
+ "normalized_hypothesis": "an hignsight obvious",
+ "errors": 2,
+ "words": 3,
+ "wer": 0.6666666666666666
+ }
+ },
+ {
+ "id": "release500_0187",
+ "category": "technical",
+ "text": "Once it's configured correctly, it basically runs itself.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0187.wav",
+ "render_seconds": 30.989758559037,
+ "audio_seconds": 4.179591836734694,
+ "rtf": 7.414541842738344,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "ONCE IT CONSIGURED CORRECTLY IT BASICALLY RUNS ITSELF",
+ "raw": {
+ "normalized_reference": "once it's configured correctly, it basically runs itself.",
+ "normalized_hypothesis": "once it consigured correctly it basically runs itself",
+ "errors": 2,
+ "words": 8,
+ "wer": 0.25
+ },
+ "semantic": {
+ "normalized_reference": "once it is configured correctly it basically runs itself",
+ "normalized_hypothesis": "once it consigured correctly it basically runs itself",
+ "errors": 2,
+ "words": 9,
+ "wer": 0.2222222222222222
+ }
+ },
+ {
+ "id": "release500_0188",
+ "category": "conversational",
+ "text": "You got engaged? When? How? Tell me everything right now.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0188.wav",
+ "render_seconds": 73.0189430139726,
+ "audio_seconds": 4.597551020408163,
+ "rtf": 15.882138705986584,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IF THAT INOT URGED WHEL HOW TELL ME EVERYTHING LITME",
+ "raw": {
+ "normalized_reference": "you got engaged? when? how? tell me everything right now.",
+ "normalized_hypothesis": "if that inot urged whel how tell me everything litme",
+ "errors": 7,
+ "words": 10,
+ "wer": 0.7
+ },
+ "semantic": {
+ "normalized_reference": "you got engaged when how tell me everything right now",
+ "normalized_hypothesis": "if that inot urged whel how tell me everything litme",
+ "errors": 7,
+ "words": 10,
+ "wer": 0.7
+ }
+ },
+ {
+ "id": "release500_0189",
+ "category": "emotional",
+ "text": "It's hard to explain how much that moment meant to me.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0189.wav",
+ "render_seconds": 66.28985990094952,
+ "audio_seconds": 3.4829931972789114,
+ "rtf": 19.03244024499918,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT'S HARD TO EXPLAIN HOW MUCH THAT MOMENT MEANT TO ME",
+ "raw": {
+ "normalized_reference": "it's hard to explain how much that moment meant to me.",
+ "normalized_hypothesis": "it's hard to explain how much that moment meant to me",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it is hard to explain how much that moment meant to me",
+ "normalized_hypothesis": "it is hard to explain how much that moment meant to me",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0190",
+ "category": "narrative",
+ "text": "He told the story three different times that night and somehow it got better each time.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0190.wav",
+ "render_seconds": 107.10718399297912,
+ "audio_seconds": 5.642448979591837,
+ "rtf": 18.982392996440886,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE TOLD THE STORY THREE DIFFERENT TIMES THAT NIGHT AND SOMEHOW IT GOT BETTER EACH TIME",
+ "raw": {
+ "normalized_reference": "he told the story three different times that night and somehow it got better each time.",
+ "normalized_hypothesis": "he told the story three different times that night and somehow it got better each time",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he told the story 3 different times that night and somehow it got better each time",
+ "normalized_hypothesis": "he told the story 3 different times that night and somehow it got better each time",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0191",
+ "category": "short",
+ "text": "Right. Moving on.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0191.wav",
+ "render_seconds": 34.410585061996244,
+ "audio_seconds": 1.811156462585034,
+ "rtf": 18.99923378948861,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "LIGHT MISSING ON",
+ "raw": {
+ "normalized_reference": "right. moving on.",
+ "normalized_hypothesis": "light missing on",
+ "errors": 2,
+ "words": 3,
+ "wer": 0.6666666666666666
+ },
+ "semantic": {
+ "normalized_reference": "right moving on",
+ "normalized_hypothesis": "light missing on",
+ "errors": 2,
+ "words": 3,
+ "wer": 0.6666666666666666
+ }
+ },
+ {
+ "id": "release500_0192",
+ "category": "conversational",
+ "text": "No, I did not move your keys; check beside the small ceramic bowl.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0192.wav",
+ "render_seconds": 80.69222360302228,
+ "audio_seconds": 4.667210884353741,
+ "rtf": 17.28917454180894,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "NO I DID NOT MOVE YOUR TEYS CHECK BESIDE THE SMALL SERAMMIC BOWL",
+ "raw": {
+ "normalized_reference": "no, i did not move your keys; check beside the small ceramic bowl.",
+ "normalized_hypothesis": "no i did not move your teys check beside the small serammic bowl",
+ "errors": 2,
+ "words": 13,
+ "wer": 0.15384615384615385
+ },
+ "semantic": {
+ "normalized_reference": "no i did not move your keys check beside the small ceramic bowl",
+ "normalized_hypothesis": "no i did not move your teys check beside the small serammic bowl",
+ "errors": 2,
+ "words": 13,
+ "wer": 0.15384615384615385
+ }
+ },
+ {
+ "id": "release500_0193",
+ "category": "emotional",
+ "text": "I was devastated. I'm not going to pretend otherwise.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0193.wav",
+ "render_seconds": 26.12720021198038,
+ "audio_seconds": 4.1099319727891155,
+ "rtf": 6.35708824013691,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I WAS DEVASTATED IAM NOT GOING TO PRETEND OTHERWISE",
+ "raw": {
+ "normalized_reference": "i was devastated. i'm not going to pretend otherwise.",
+ "normalized_hypothesis": "i was devastated iam not going to pretend otherwise",
+ "errors": 1,
+ "words": 9,
+ "wer": 0.1111111111111111
+ },
+ "semantic": {
+ "normalized_reference": "i was devastated i am not going to pretend otherwise",
+ "normalized_hypothesis": "i was devastated iam not going to pretend otherwise",
+ "errors": 2,
+ "words": 10,
+ "wer": 0.2
+ }
+ },
+ {
+ "id": "release500_0194",
+ "category": "narrative",
+ "text": "That was the last summer before everything changed, though we didn't know it yet.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0194.wav",
+ "render_seconds": 64.37945774104446,
+ "audio_seconds": 5.015510204081632,
+ "rtf": 12.836073524215408,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THAT WAS THE LAST SUMMER BEFORE EVERYTHING CHANGED THOUGH WE DIDN'T KNOW IT YET",
+ "raw": {
+ "normalized_reference": "that was the last summer before everything changed, though we didn't know it yet.",
+ "normalized_hypothesis": "that was the last summer before everything changed though we didn't know it yet",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "that was the last summer before everything changed though we did not know it yet",
+ "normalized_hypothesis": "that was the last summer before everything changed though we did not know it yet",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0195",
+ "category": "short",
+ "text": "Bold of you to assume.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0195.wav",
+ "render_seconds": 49.709628745913506,
+ "audio_seconds": 1.8808163265306121,
+ "rtf": 26.429815630965297,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "BOTH OF HER TO MASAM",
+ "raw": {
+ "normalized_reference": "bold of you to assume.",
+ "normalized_hypothesis": "both of her to masam",
+ "errors": 3,
+ "words": 5,
+ "wer": 0.6
+ },
+ "semantic": {
+ "normalized_reference": "bold of you to assume",
+ "normalized_hypothesis": "both of her to masam",
+ "errors": 3,
+ "words": 5,
+ "wer": 0.6
+ }
+ },
+ {
+ "id": "release500_0196",
+ "category": "conversational",
+ "text": "It's one of those decisions that makes total sense in the moment.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0196.wav",
+ "render_seconds": 79.19231077795848,
+ "audio_seconds": 4.388571428571429,
+ "rtf": 18.04512289862335,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT'S ONE OF THOSE DECISIONS THAT MAKES TOTAL SENSE IN THE MOMENT",
+ "raw": {
+ "normalized_reference": "it's one of those decisions that makes total sense in the moment.",
+ "normalized_hypothesis": "it's one of those decisions that makes total sense in the moment",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it is one of those decisions that makes total sense in the moment",
+ "normalized_hypothesis": "it is one of those decisions that makes total sense in the moment",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0197",
+ "category": "emotional",
+ "text": "I was furious. Not the loud kind. The quiet kind that lasts for days.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0197.wav",
+ "render_seconds": 100.68786248005927,
+ "audio_seconds": 5.363809523809524,
+ "rtf": 18.771707316062187,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I WAS FARIOUS NOT THE LOUD KIND THE QUIET KIND THAT LASTS FOR DAYS",
+ "raw": {
+ "normalized_reference": "i was furious. not the loud kind. the quiet kind that lasts for days.",
+ "normalized_hypothesis": "i was farious not the loud kind the quiet kind that lasts for days",
+ "errors": 1,
+ "words": 14,
+ "wer": 0.07142857142857142
+ },
+ "semantic": {
+ "normalized_reference": "i was furious not the loud kind the quiet kind that lasts for days",
+ "normalized_hypothesis": "i was farious not the loud kind the quiet kind that lasts for days",
+ "errors": 1,
+ "words": 14,
+ "wer": 0.07142857142857142
+ }
+ },
+ {
+ "id": "release500_0198",
+ "category": "narrative",
+ "text": "That's the part of the story I usually leave out.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0198.wav",
+ "render_seconds": 51.69975427899044,
+ "audio_seconds": 3.3436734693877552,
+ "rtf": 15.46196264447346,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THAT'S THE PART OF THE STORY I USUALLY LEAVE OUT",
+ "raw": {
+ "normalized_reference": "that's the part of the story i usually leave out.",
+ "normalized_hypothesis": "that's the part of the story i usually leave out",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "that is the part of the story i usually leave out",
+ "normalized_hypothesis": "that is the part of the story i usually leave out",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0199",
+ "category": "short",
+ "text": "I've said too much.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0199.wav",
+ "render_seconds": 16.68404570000712,
+ "audio_seconds": 1.7414965986394557,
+ "rtf": 9.580291866800962,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I SAID TONIK",
+ "raw": {
+ "normalized_reference": "i've said too much.",
+ "normalized_hypothesis": "i said tonik",
+ "errors": 3,
+ "words": 4,
+ "wer": 0.75
+ },
+ "semantic": {
+ "normalized_reference": "i have said too much",
+ "normalized_hypothesis": "i said tonik",
+ "errors": 3,
+ "words": 5,
+ "wer": 0.6
+ }
+ },
+ {
+ "id": "release500_0200",
+ "category": "conversational",
+ "text": "I've been faking that I understood this the entire time.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0200.wav",
+ "render_seconds": 76.50434159603901,
+ "audio_seconds": 3.622312925170068,
+ "rtf": 21.12030163502429,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I'VE BEEN SAKING THAT I UNDERSTOOD THIS THE ENTIRE TIME",
+ "raw": {
+ "normalized_reference": "i've been faking that i understood this the entire time.",
+ "normalized_hypothesis": "i've been saking that i understood this the entire time",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ },
+ "semantic": {
+ "normalized_reference": "i have been faking that i understood this the entire time",
+ "normalized_hypothesis": "i have been saking that i understood this the entire time",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ }
+ },
+ {
+ "id": "release500_0201",
+ "category": "phonetic_balance",
+ "text": "Peace be unto you and grace before the Lord.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0201.wav",
+ "render_seconds": 40.68864465598017,
+ "audio_seconds": 3.0650340136054424,
+ "rtf": 13.275103791790404,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "PEACE BE UNTO YOU IN GRACE BEFORE THE LORD",
+ "raw": {
+ "normalized_reference": "peace be unto you and grace before the lord.",
+ "normalized_hypothesis": "peace be unto you in grace before the lord",
+ "errors": 1,
+ "words": 9,
+ "wer": 0.1111111111111111
+ },
+ "semantic": {
+ "normalized_reference": "peace be unto you and grace before the lord",
+ "normalized_hypothesis": "peace be unto you in grace before the lord",
+ "errors": 1,
+ "words": 9,
+ "wer": 0.1111111111111111
+ }
+ },
+ {
+ "id": "release500_0202",
+ "category": "phonetic_balance",
+ "text": "From now on we're pals.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0202.wav",
+ "render_seconds": 46.31092496099882,
+ "audio_seconds": 1.8808163265306121,
+ "rtf": 24.62277911337828,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "FROM NOW ON YOR PAT",
+ "raw": {
+ "normalized_reference": "from now on we're pals.",
+ "normalized_hypothesis": "from now on yor pat",
+ "errors": 2,
+ "words": 5,
+ "wer": 0.4
+ },
+ "semantic": {
+ "normalized_reference": "from now on we are pals",
+ "normalized_hypothesis": "from now on yor pat",
+ "errors": 3,
+ "words": 6,
+ "wer": 0.5
+ }
+ },
+ {
+ "id": "release500_0203",
+ "category": "phonetic_balance",
+ "text": "Here we allow our solicitors to look after our legal work.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0203.wav",
+ "render_seconds": 68.31000176095404,
+ "audio_seconds": 4.179591836734694,
+ "rtf": 16.34370159319701,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HERE WE ALLOW OUR SOLICITORS TO LOOK AFTER OUR LEGAL WORK",
+ "raw": {
+ "normalized_reference": "here we allow our solicitors to look after our legal work.",
+ "normalized_hypothesis": "here we allow our solicitors to look after our legal work",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "here we allow our solicitors to look after our legal work",
+ "normalized_hypothesis": "here we allow our solicitors to look after our legal work",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0204",
+ "category": "phonetic_balance",
+ "text": "Her words sent a strange chill through Philip.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0204.wav",
+ "render_seconds": 59.90024443494622,
+ "audio_seconds": 3.2043537414965986,
+ "rtf": 18.693393197896345,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HER WORDS SENT A STRANGE CHILL THROUGH PHILIP",
+ "raw": {
+ "normalized_reference": "her words sent a strange chill through philip.",
+ "normalized_hypothesis": "her words sent a strange chill through philip",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "her words sent a strange chill through philip",
+ "normalized_hypothesis": "her words sent a strange chill through philip",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0205",
+ "category": "phonetic_balance",
+ "text": "It was introduced by Representative Dick of Ohio.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0205.wav",
+ "render_seconds": 27.99281141499523,
+ "audio_seconds": 3.7616326530612243,
+ "rtf": 7.44166536097421,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT WAS INTRODUCED BY REPRESENTATIVE DICK OF OHIO",
+ "raw": {
+ "normalized_reference": "it was introduced by representative dick of ohio.",
+ "normalized_hypothesis": "it was introduced by representative dick of ohio",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it was introduced by representative dick of ohio",
+ "normalized_hypothesis": "it was introduced by representative dick of ohio",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0206",
+ "category": "phonetic_balance",
+ "text": "It's the aurora borealis.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0206.wav",
+ "render_seconds": 27.798951274948195,
+ "audio_seconds": 2.020136054421769,
+ "rtf": 13.760930217596258,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT'S THE AURORA BERIOLIN",
+ "raw": {
+ "normalized_reference": "it's the aurora borealis.",
+ "normalized_hypothesis": "it's the aurora beriolin",
+ "errors": 1,
+ "words": 4,
+ "wer": 0.25
+ },
+ "semantic": {
+ "normalized_reference": "it is the aurora borealis",
+ "normalized_hypothesis": "it is the aurora beriolin",
+ "errors": 1,
+ "words": 5,
+ "wer": 0.2
+ }
+ },
+ {
+ "id": "release500_0207",
+ "category": "phonetic_balance",
+ "text": "The boy grew and prospered.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0207.wav",
+ "render_seconds": 40.599264263990335,
+ "audio_seconds": 2.159455782312925,
+ "rtf": 18.80069256176468,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE BOY TALER TASSBOP",
+ "raw": {
+ "normalized_reference": "the boy grew and prospered.",
+ "normalized_hypothesis": "the boy taler tassbop",
+ "errors": 3,
+ "words": 5,
+ "wer": 0.6
+ },
+ "semantic": {
+ "normalized_reference": "the boy grew and prospered",
+ "normalized_hypothesis": "the boy taler tassbop",
+ "errors": 3,
+ "words": 5,
+ "wer": 0.6
+ }
+ },
+ {
+ "id": "release500_0208",
+ "category": "phonetic_balance",
+ "text": "His voice was passionately rebellious.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0208.wav",
+ "render_seconds": 64.88870784209576,
+ "audio_seconds": 2.995374149659864,
+ "rtf": 21.66297250360664,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HER VOICE WAS PASSIONATELY REBELLIOUS",
+ "raw": {
+ "normalized_reference": "his voice was passionately rebellious.",
+ "normalized_hypothesis": "her voice was passionately rebellious",
+ "errors": 1,
+ "words": 5,
+ "wer": 0.2
+ },
+ "semantic": {
+ "normalized_reference": "his voice was passionately rebellious",
+ "normalized_hypothesis": "her voice was passionately rebellious",
+ "errors": 1,
+ "words": 5,
+ "wer": 0.2
+ }
+ },
+ {
+ "id": "release500_0209",
+ "category": "phonetic_balance",
+ "text": "That's where they cut off the Scottish Chiefs and killed all hands.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0209.wav",
+ "render_seconds": 76.79028228507377,
+ "audio_seconds": 4.388571428571429,
+ "rtf": 17.497785676937383,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THAT'S WHERE THEY CUT OFF THE SCOTTISH CHIEFS AND KILLED ALL HANDS",
+ "raw": {
+ "normalized_reference": "that's where they cut off the scottish chiefs and killed all hands.",
+ "normalized_hypothesis": "that's where they cut off the scottish chiefs and killed all hands",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "that is where they cut off the scottish chiefs and killed all hands",
+ "normalized_hypothesis": "that is where they cut off the scottish chiefs and killed all hands",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0210",
+ "category": "phonetic_balance",
+ "text": "And Tom King patiently endured.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0210.wav",
+ "render_seconds": 65.00650330004282,
+ "audio_seconds": 2.4380952380952383,
+ "rtf": 26.662823619158186,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "AND TOM PATIENTLY ENDURED",
+ "raw": {
+ "normalized_reference": "and tom king patiently endured.",
+ "normalized_hypothesis": "and tom patiently endured",
+ "errors": 1,
+ "words": 5,
+ "wer": 0.2
+ },
+ "semantic": {
+ "normalized_reference": "and tom king patiently endured",
+ "normalized_hypothesis": "and tom patiently endured",
+ "errors": 1,
+ "words": 5,
+ "wer": 0.2
+ }
+ },
+ {
+ "id": "release500_0211",
+ "category": "phonetic_balance",
+ "text": "From that moment his friendship for Belize turns to hatred and jealousy.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0211.wav",
+ "render_seconds": 47.60291245498229,
+ "audio_seconds": 5.154829931972789,
+ "rtf": 9.234623272384919,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "FROM THAT MOMENT HIS FRIENDSHIP FOR BOLAYS TURNS TO HATRED AND JEALOUSY",
+ "raw": {
+ "normalized_reference": "from that moment his friendship for belize turns to hatred and jealousy.",
+ "normalized_hypothesis": "from that moment his friendship for bolays turns to hatred and jealousy",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "from that moment his friendship for belize turns to hatred and jealousy",
+ "normalized_hypothesis": "from that moment his friendship for bolays turns to hatred and jealousy",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ }
+ },
+ {
+ "id": "release500_0212",
+ "category": "phonetic_balance",
+ "text": "Then you don't believe in altruism.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0212.wav",
+ "render_seconds": 57.117999506997876,
+ "audio_seconds": 2.4380952380952383,
+ "rtf": 23.427304485292094,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THOUGH I DON'T BELIEVE AN E TRAINFOR",
+ "raw": {
+ "normalized_reference": "then you don't believe in altruism.",
+ "normalized_hypothesis": "though i don't believe an e trainfor",
+ "errors": 5,
+ "words": 6,
+ "wer": 0.8333333333333334
+ },
+ "semantic": {
+ "normalized_reference": "then you do not believe in altruism",
+ "normalized_hypothesis": "though i do not believe an e trainfor",
+ "errors": 5,
+ "words": 7,
+ "wer": 0.7142857142857143
+ }
+ },
+ {
+ "id": "release500_0213",
+ "category": "phonetic_balance",
+ "text": "What Pascal glimpsed with the vision of a seer, I have lived.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0213.wav",
+ "render_seconds": 64.60190822998993,
+ "audio_seconds": 4.31891156462585,
+ "rtf": 14.957914109451425,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WHAT WAT TASCO GLIMPSED WITH THE VISION OF A SEER I HAVE LIVED",
+ "raw": {
+ "normalized_reference": "what pascal glimpsed with the vision of a seer, i have lived.",
+ "normalized_hypothesis": "what wat tasco glimpsed with the vision of a seer i have lived",
+ "errors": 2,
+ "words": 12,
+ "wer": 0.16666666666666666
+ },
+ "semantic": {
+ "normalized_reference": "what pascal glimpsed with the vision of a seer i have lived",
+ "normalized_hypothesis": "what wat tasco glimpsed with the vision of a seer i have lived",
+ "errors": 2,
+ "words": 12,
+ "wer": 0.16666666666666666
+ }
+ },
+ {
+ "id": "release500_0214",
+ "category": "phonetic_balance",
+ "text": "You were making them talk shop, Ruth charged him.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0214.wav",
+ "render_seconds": 70.70681391202379,
+ "audio_seconds": 3.274013605442177,
+ "rtf": 21.596371436726017,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "YOU WERE MAKING THEM TALK SHAP RUS CHARGED EM",
+ "raw": {
+ "normalized_reference": "you were making them talk shop, ruth charged him.",
+ "normalized_hypothesis": "you were making them talk shap rus charged em",
+ "errors": 3,
+ "words": 9,
+ "wer": 0.3333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "you were making them talk shop ruth charged him",
+ "normalized_hypothesis": "you were making them talk shap rus charged em",
+ "errors": 3,
+ "words": 9,
+ "wer": 0.3333333333333333
+ }
+ },
+ {
+ "id": "release500_0215",
+ "category": "phonetic_balance",
+ "text": "Philip snatched at the letter which Gregson held out to him.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0215.wav",
+ "render_seconds": 73.00307296798564,
+ "audio_seconds": 3.9706122448979593,
+ "rtf": 18.385847941155923,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "PHILIP SNATCHED AT THE LETTER WHICH GREGSON HELD OUT TO HIM",
+ "raw": {
+ "normalized_reference": "philip snatched at the letter which gregson held out to him.",
+ "normalized_hypothesis": "philip snatched at the letter which gregson held out to him",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "philip snatched at the letter which gregson held out to him",
+ "normalized_hypothesis": "philip snatched at the letter which gregson held out to him",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0216",
+ "category": "phonetic_balance",
+ "text": "For a time the exciting thrill of his adventure was gone.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0216.wav",
+ "render_seconds": 80.90028269705363,
+ "audio_seconds": 4.040272108843538,
+ "rtf": 20.023473795236534,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "FOR A TIME THE EXCITING THRILL OF HIS ADVENTURE WAS GONE",
+ "raw": {
+ "normalized_reference": "for a time the exciting thrill of his adventure was gone.",
+ "normalized_hypothesis": "for a time the exciting thrill of his adventure was gone",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "for a time the exciting thrill of his adventure was gone",
+ "normalized_hypothesis": "for a time the exciting thrill of his adventure was gone",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0217",
+ "category": "phonetic_balance",
+ "text": "Obviously, it was a disease that could be contracted by contact.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0217.wav",
+ "render_seconds": 47.70699213095941,
+ "audio_seconds": 5.085170068027211,
+ "rtf": 9.38159225606142,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "OBVIOUSLY IT WAS A DISEASE THAT COULD BE CONTRACTED BY CONTACT",
+ "raw": {
+ "normalized_reference": "obviously, it was a disease that could be contracted by contact.",
+ "normalized_hypothesis": "obviously it was a disease that could be contracted by contact",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "obviously it was a disease that could be contracted by contact",
+ "normalized_hypothesis": "obviously it was a disease that could be contracted by contact",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0218",
+ "category": "phonetic_balance",
+ "text": "Blind with rage, he darted in.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0218.wav",
+ "render_seconds": 44.784837095066905,
+ "audio_seconds": 2.2291156462585033,
+ "rtf": 20.090854043502304,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "BLIND WITH RAGE HE DARTED IN",
+ "raw": {
+ "normalized_reference": "blind with rage, he darted in.",
+ "normalized_hypothesis": "blind with rage he darted in",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "blind with rage he darted in",
+ "normalized_hypothesis": "blind with rage he darted in",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0219",
+ "category": "phonetic_balance",
+ "text": "You live on an income which your father earned.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0219.wav",
+ "render_seconds": 61.30533464089967,
+ "audio_seconds": 3.1346938775510202,
+ "rtf": 19.55704034768284,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "NOW IVE AD AN ANTEON WITH YOUR BAD IRAND",
+ "raw": {
+ "normalized_reference": "you live on an income which your father earned.",
+ "normalized_hypothesis": "now ive ad an anteon with your bad irand",
+ "errors": 7,
+ "words": 9,
+ "wer": 0.7777777777777778
+ },
+ "semantic": {
+ "normalized_reference": "you live on an income which your father earned",
+ "normalized_hypothesis": "now ive ad an anteon with your bad irand",
+ "errors": 7,
+ "words": 9,
+ "wer": 0.7777777777777778
+ }
+ },
+ {
+ "id": "release500_0220",
+ "category": "phonetic_balance",
+ "text": "Something about it seemed to fascinate him, to challenge his presence.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0220.wav",
+ "render_seconds": 90.90409748698585,
+ "audio_seconds": 4.806530612244898,
+ "rtf": 18.91262218436781,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SOMETHING ABOUT IT SEEMED TO FASCINATE HIM TO CHALLENGE HIS PRESENCE",
+ "raw": {
+ "normalized_reference": "something about it seemed to fascinate him, to challenge his presence.",
+ "normalized_hypothesis": "something about it seemed to fascinate him to challenge his presence",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "something about it seemed to fascinate him to challenge his presence",
+ "normalized_hypothesis": "something about it seemed to fascinate him to challenge his presence",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0221",
+ "category": "phonetic_balance",
+ "text": "Her beautiful hair was done up in shining coils.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0221.wav",
+ "render_seconds": 48.39576102490537,
+ "audio_seconds": 3.4829931972789114,
+ "rtf": 13.89487670050994,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HER BEAUTIFUL HAIR WAS DONE UP IN SHINING COILS",
+ "raw": {
+ "normalized_reference": "her beautiful hair was done up in shining coils.",
+ "normalized_hypothesis": "her beautiful hair was done up in shining coils",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "her beautiful hair was done up in shining coils",
+ "normalized_hypothesis": "her beautiful hair was done up in shining coils",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0222",
+ "category": "phonetic_balance",
+ "text": "It does, was her audacious answer.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0222.wav",
+ "render_seconds": 64.49648476706352,
+ "audio_seconds": 2.716734693877551,
+ "rtf": 23.740442860472598,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT DARES WITH HER I DOS FOR SAMPIN",
+ "raw": {
+ "normalized_reference": "it does, was her audacious answer.",
+ "normalized_hypothesis": "it dares with her i dos for sampin",
+ "errors": 6,
+ "words": 6,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "it does was her audacious answer",
+ "normalized_hypothesis": "it dares with her i dos for sampin",
+ "errors": 6,
+ "words": 6,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0223",
+ "category": "phonetic_balance",
+ "text": "He caught himself with a jerk.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0223.wav",
+ "render_seconds": 16.895986302988604,
+ "audio_seconds": 2.2291156462585033,
+ "rtf": 7.579681355405654,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE CAUGHT HIMSELF WITH A JERK",
+ "raw": {
+ "normalized_reference": "he caught himself with a jerk.",
+ "normalized_hypothesis": "he caught himself with a jerk",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he caught himself with a jerk",
+ "normalized_hypothesis": "he caught himself with a jerk",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0224",
+ "category": "phonetic_balance",
+ "text": "He did not believe in the burning of daylight for such a luxury.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0224.wav",
+ "render_seconds": 84.80448635201901,
+ "audio_seconds": 4.388571428571429,
+ "rtf": 19.323938947400165,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE DID NOT BELIEVE IN THE BURNING OF DAYLIGHT FOR SUCH A LUXURY",
+ "raw": {
+ "normalized_reference": "he did not believe in the burning of daylight for such a luxury.",
+ "normalized_hypothesis": "he did not believe in the burning of daylight for such a luxury",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he did not believe in the burning of daylight for such a luxury",
+ "normalized_hypothesis": "he did not believe in the burning of daylight for such a luxury",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0225",
+ "category": "phonetic_balance",
+ "text": "He had comparatively no advantages at first.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0225.wav",
+ "render_seconds": 71.69164062803611,
+ "audio_seconds": 3.4133333333333336,
+ "rtf": 21.003410340244955,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HAD COMPARATIVELY NO ADVANTAGES AT FIRST",
+ "raw": {
+ "normalized_reference": "he had comparatively no advantages at first.",
+ "normalized_hypothesis": "had comparatively no advantages at first",
+ "errors": 1,
+ "words": 7,
+ "wer": 0.14285714285714285
+ },
+ "semantic": {
+ "normalized_reference": "he had comparatively no advantages at 1st",
+ "normalized_hypothesis": "had comparatively no advantages at 1st",
+ "errors": 1,
+ "words": 7,
+ "wer": 0.14285714285714285
+ }
+ },
+ {
+ "id": "release500_0226",
+ "category": "phonetic_balance",
+ "text": "Such men believe, when they come together.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0226.wav",
+ "render_seconds": 55.39808167202864,
+ "audio_seconds": 2.8560544217687074,
+ "rtf": 19.3967178110428,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SUCH MEN BELIEVE WHEN THEY COME TOGETHER",
+ "raw": {
+ "normalized_reference": "such men believe, when they come together.",
+ "normalized_hypothesis": "such men believe when they come together",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "such men believe when they come together",
+ "normalized_hypothesis": "such men believe when they come together",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0227",
+ "category": "phonetic_balance",
+ "text": "The ship should be in within a week or ten days.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0227.wav",
+ "render_seconds": 39.50758587103337,
+ "audio_seconds": 3.274013605442177,
+ "rtf": 12.067019454458746,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE SHIP SHOULD BE IN WITHIN A WEEK OR TEN DAYS",
+ "raw": {
+ "normalized_reference": "the ship should be in within a week or ten days.",
+ "normalized_hypothesis": "the ship should be in within a week or ten days",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the ship should be in within a week or 10 days",
+ "normalized_hypothesis": "the ship should be in within a week or 10 days",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0228",
+ "category": "phonetic_balance",
+ "text": "Such a dog the wise driver kills, or turns loose.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0228.wav",
+ "render_seconds": 72.1157341429498,
+ "audio_seconds": 3.622312925170068,
+ "rtf": 19.908753228315845,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THAT'S A DOG THE WISE DRIVER KILLS OR TURNS LOOSE",
+ "raw": {
+ "normalized_reference": "such a dog the wise driver kills, or turns loose.",
+ "normalized_hypothesis": "that's a dog the wise driver kills or turns loose",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ },
+ "semantic": {
+ "normalized_reference": "such a dog the wise driver kills or turns loose",
+ "normalized_hypothesis": "that is a dog the wise driver kills or turns loose",
+ "errors": 2,
+ "words": 10,
+ "wer": 0.2
+ }
+ },
+ {
+ "id": "release500_0229",
+ "category": "phonetic_balance",
+ "text": "You have heard always how he was the lover of the Princess Naomi.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0229.wav",
+ "render_seconds": 32.687658572918735,
+ "audio_seconds": 4.458231292517007,
+ "rtf": 7.331979080534445,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "YOU HAVE HEARD ALWAYS HOW HE WAS THE LOVER OF THE PRINCESS NAOMI",
+ "raw": {
+ "normalized_reference": "you have heard always how he was the lover of the princess naomi.",
+ "normalized_hypothesis": "you have heard always how he was the lover of the princess naomi",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "you have heard always how he was the lover of the princess naomi",
+ "normalized_hypothesis": "you have heard always how he was the lover of the princess naomi",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0230",
+ "category": "phonetic_balance",
+ "text": "I want to know how all this is possible.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0230.wav",
+ "render_seconds": 60.4882631029468,
+ "audio_seconds": 2.9257142857142857,
+ "rtf": 20.67469930276502,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "YEIN TO KNOW HEW I'LL BEFIS PARTABLE",
+ "raw": {
+ "normalized_reference": "i want to know how all this is possible.",
+ "normalized_hypothesis": "yein to know hew i'll befis partable",
+ "errors": 7,
+ "words": 9,
+ "wer": 0.7777777777777778
+ },
+ "semantic": {
+ "normalized_reference": "i want to know how all this is possible",
+ "normalized_hypothesis": "yein to know hew i will befis partable",
+ "errors": 7,
+ "words": 9,
+ "wer": 0.7777777777777778
+ }
+ },
+ {
+ "id": "release500_0231",
+ "category": "phonetic_balance",
+ "text": "It was my idea to a tee.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0231.wav",
+ "render_seconds": 44.600040287012234,
+ "audio_seconds": 2.159455782312925,
+ "rtf": 20.653370470611137,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT WAS MERLY DO TO MATEN",
+ "raw": {
+ "normalized_reference": "it was my idea to a tee.",
+ "normalized_hypothesis": "it was merly do to maten",
+ "errors": 4,
+ "words": 7,
+ "wer": 0.5714285714285714
+ },
+ "semantic": {
+ "normalized_reference": "it was my idea to a tee",
+ "normalized_hypothesis": "it was merly do to maten",
+ "errors": 4,
+ "words": 7,
+ "wer": 0.5714285714285714
+ }
+ },
+ {
+ "id": "release500_0232",
+ "category": "phonetic_balance",
+ "text": "They likewise are disinclined to being eaten.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0232.wav",
+ "render_seconds": 61.39736158796586,
+ "audio_seconds": 3.2043537414965986,
+ "rtf": 19.16060664366292,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THEY LIKEWISE ARE DISINCLINED TO BEING EATEN",
+ "raw": {
+ "normalized_reference": "they likewise are disinclined to being eaten.",
+ "normalized_hypothesis": "they likewise are disinclined to being eaten",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "they likewise are disinclined to being eaten",
+ "normalized_hypothesis": "they likewise are disinclined to being eaten",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0233",
+ "category": "phonetic_balance",
+ "text": "This state of mind comes of an undue prominence of the ego.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0233.wav",
+ "render_seconds": 76.19261164800264,
+ "audio_seconds": 4.1099319727891155,
+ "rtf": 18.53865517786081,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THIS STATE OF MIND COMES OF AN UNDUE PROMINENCE OF THE EGO",
+ "raw": {
+ "normalized_reference": "this state of mind comes of an undue prominence of the ego.",
+ "normalized_hypothesis": "this state of mind comes of an undue prominence of the ego",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "this state of mind comes of an undue prominence of the ego",
+ "normalized_hypothesis": "this state of mind comes of an undue prominence of the ego",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0234",
+ "category": "phonetic_balance",
+ "text": "Then it is as I said, Womble announced with finality.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0234.wav",
+ "render_seconds": 67.09139321104158,
+ "audio_seconds": 3.8312925170068026,
+ "rtf": 17.511425429728536,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THEN ATIT AS I SAT WUMBLE ANNOUNCED WITH FINALITY",
+ "raw": {
+ "normalized_reference": "then it is as i said, womble announced with finality.",
+ "normalized_hypothesis": "then atit as i sat wumble announced with finality",
+ "errors": 4,
+ "words": 10,
+ "wer": 0.4
+ },
+ "semantic": {
+ "normalized_reference": "then it is as i said womble announced with finality",
+ "normalized_hypothesis": "then atit as i sat wumble announced with finality",
+ "errors": 4,
+ "words": 10,
+ "wer": 0.4
+ }
+ },
+ {
+ "id": "release500_0235",
+ "category": "phonetic_balance",
+ "text": "I can see that knife now.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0235.wav",
+ "render_seconds": 32.19995296106208,
+ "audio_seconds": 2.020136054421769,
+ "rtf": 15.939497189103333,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I CON' FOND THAT NO IFE NO",
+ "raw": {
+ "normalized_reference": "i can see that knife now.",
+ "normalized_hypothesis": "i con' fond that no ife no",
+ "errors": 5,
+ "words": 6,
+ "wer": 0.8333333333333334
+ },
+ "semantic": {
+ "normalized_reference": "i can see that knife now",
+ "normalized_hypothesis": "i con fond that no ife no",
+ "errors": 5,
+ "words": 6,
+ "wer": 0.8333333333333334
+ }
+ },
+ {
+ "id": "release500_0236",
+ "category": "phonetic_balance",
+ "text": "If I meet her again I shall apologize, said Eileen.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0236.wav",
+ "render_seconds": 73.49967968300916,
+ "audio_seconds": 3.7616326530612243,
+ "rtf": 19.539302867119407,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IF I MEET HER AGAIN I SHALL APOLOGIZE SAID EILEN",
+ "raw": {
+ "normalized_reference": "if i meet her again i shall apologize, said eileen.",
+ "normalized_hypothesis": "if i meet her again i shall apologize said eilen",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ },
+ "semantic": {
+ "normalized_reference": "if i meet her again i shall apologize said eileen",
+ "normalized_hypothesis": "if i meet her again i shall apologize said eilen",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ }
+ },
+ {
+ "id": "release500_0237",
+ "category": "phonetic_balance",
+ "text": "A wildly exciting time was his during the week preceding Thursday the eighteenth.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0237.wav",
+ "render_seconds": 76.90761393296998,
+ "audio_seconds": 5.572789115646258,
+ "rtf": 13.800560605647691,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "A WILDLY EXCITING TIME WAS AS DURING THE WEEK PRECEDING THURSDAY THE EIGHTEENTH",
+ "raw": {
+ "normalized_reference": "a wildly exciting time was his during the week preceding thursday the eighteenth.",
+ "normalized_hypothesis": "a wildly exciting time was as during the week preceding thursday the eighteenth",
+ "errors": 1,
+ "words": 13,
+ "wer": 0.07692307692307693
+ },
+ "semantic": {
+ "normalized_reference": "a wildly exciting time was his during the week preceding thursday the 18th",
+ "normalized_hypothesis": "a wildly exciting time was as during the week preceding thursday the 18th",
+ "errors": 1,
+ "words": 13,
+ "wer": 0.07692307692307693
+ }
+ },
+ {
+ "id": "release500_0238",
+ "category": "phonetic_balance",
+ "text": "In what bucolic school of fence he had been taught was beyond imagining.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0238.wav",
+ "render_seconds": 74.69169386499561,
+ "audio_seconds": 4.876190476190477,
+ "rtf": 15.317632530907302,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IN WHAT BUCOLICK SCHOOL OF FENCE HE HAD BEEN TAUGHT WAS BEYOND IMAGINING",
+ "raw": {
+ "normalized_reference": "in what bucolic school of fence he had been taught was beyond imagining.",
+ "normalized_hypothesis": "in what bucolick school of fence he had been taught was beyond imagining",
+ "errors": 1,
+ "words": 13,
+ "wer": 0.07692307692307693
+ },
+ "semantic": {
+ "normalized_reference": "in what bucolic school of fence he had been taught was beyond imagining",
+ "normalized_hypothesis": "in what bucolick school of fence he had been taught was beyond imagining",
+ "errors": 1,
+ "words": 13,
+ "wer": 0.07692307692307693
+ }
+ },
+ {
+ "id": "release500_0239",
+ "category": "phonetic_balance",
+ "text": "So she said, the irate skipper dashed on.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0239.wav",
+ "render_seconds": 64.50826901104301,
+ "audio_seconds": 3.0650340136054424,
+ "rtf": 21.046509997832302,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SECHENSE D LAROT SKIPPER DAFTBAN",
+ "raw": {
+ "normalized_reference": "so she said, the irate skipper dashed on.",
+ "normalized_hypothesis": "sechense d larot skipper daftban",
+ "errors": 7,
+ "words": 8,
+ "wer": 0.875
+ },
+ "semantic": {
+ "normalized_reference": "so she said the irate skipper dashed on",
+ "normalized_hypothesis": "sechense d larot skipper daftban",
+ "errors": 7,
+ "words": 8,
+ "wer": 0.875
+ }
+ },
+ {
+ "id": "release500_0240",
+ "category": "phonetic_balance",
+ "text": "It is not an attempt to smash the market.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0240.wav",
+ "render_seconds": 66.30265488405712,
+ "audio_seconds": 3.0650340136054424,
+ "rtf": 21.631947505229927,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT IS NOT AN ATTEMPT TO SMASH THE MARITRED",
+ "raw": {
+ "normalized_reference": "it is not an attempt to smash the market.",
+ "normalized_hypothesis": "it is not an attempt to smash the maritred",
+ "errors": 1,
+ "words": 9,
+ "wer": 0.1111111111111111
+ },
+ "semantic": {
+ "normalized_reference": "it is not an attempt to smash the market",
+ "normalized_hypothesis": "it is not an attempt to smash the maritred",
+ "errors": 1,
+ "words": 9,
+ "wer": 0.1111111111111111
+ }
+ },
+ {
+ "id": "release500_0241",
+ "category": "phonetic_balance",
+ "text": "He destroyed everything that had belonged to the woman.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0241.wav",
+ "render_seconds": 42.7055435749935,
+ "audio_seconds": 3.622312925170068,
+ "rtf": 11.789578773895816,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE DESTROYED EVERYTHING THAT HAD BELONGED TO THE WOMAN",
+ "raw": {
+ "normalized_reference": "he destroyed everything that had belonged to the woman.",
+ "normalized_hypothesis": "he destroyed everything that had belonged to the woman",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he destroyed everything that had belonged to the woman",
+ "normalized_hypothesis": "he destroyed everything that had belonged to the woman",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0242",
+ "category": "phonetic_balance",
+ "text": "Did Thorpe go to see any one in Churchill.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0242.wav",
+ "render_seconds": 53.50194810505491,
+ "audio_seconds": 2.995374149659864,
+ "rtf": 17.86152428107529,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "DID THORP GO TO SEE ANYONE IN CHURCHILL",
+ "raw": {
+ "normalized_reference": "did thorpe go to see any one in churchill.",
+ "normalized_hypothesis": "did thorp go to see anyone in churchill",
+ "errors": 3,
+ "words": 9,
+ "wer": 0.3333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "did thorpe go to see any one in churchill",
+ "normalized_hypothesis": "did thorp go to see anyone in churchill",
+ "errors": 3,
+ "words": 9,
+ "wer": 0.3333333333333333
+ }
+ },
+ {
+ "id": "release500_0243",
+ "category": "phonetic_balance",
+ "text": "Many other similar disconcerting omissions will be noticed in the Manuscript.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0243.wav",
+ "render_seconds": 98.99978925101459,
+ "audio_seconds": 5.712108843537415,
+ "rtf": 17.33156562011617,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "MANY OTHER SIMILAR DISCONCERTING OMISSIONS WILL BE NOTICED IN THE MANUSCRIPT",
+ "raw": {
+ "normalized_reference": "many other similar disconcerting omissions will be noticed in the manuscript.",
+ "normalized_hypothesis": "many other similar disconcerting omissions will be noticed in the manuscript",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "many other similar disconcerting omissions will be noticed in the manuscript",
+ "normalized_hypothesis": "many other similar disconcerting omissions will be noticed in the manuscript",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0244",
+ "category": "phonetic_balance",
+ "text": "The ringing of the big bell aroused him.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0244.wav",
+ "render_seconds": 65.50034994003363,
+ "audio_seconds": 2.716734693877551,
+ "rtf": 24.109954530210693,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "DID WY NOT HIF THE BIG BROTHER RASED IM",
+ "raw": {
+ "normalized_reference": "the ringing of the big bell aroused him.",
+ "normalized_hypothesis": "did wy not hif the big brother rased im",
+ "errors": 7,
+ "words": 8,
+ "wer": 0.875
+ },
+ "semantic": {
+ "normalized_reference": "the ringing of the big bell aroused him",
+ "normalized_hypothesis": "did wy not hif the big brother rased im",
+ "errors": 7,
+ "words": 8,
+ "wer": 0.875
+ }
+ },
+ {
+ "id": "release500_0245",
+ "category": "phonetic_balance",
+ "text": "A sinewy hand, dripping with water, was clutching the rail.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0245.wav",
+ "render_seconds": 74.70232620602474,
+ "audio_seconds": 4.249251700680272,
+ "rtf": 17.580113268899904,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "A SINY HAND DRIPPING WITH WATER WAS CLUTCHING THE RAIL",
+ "raw": {
+ "normalized_reference": "a sinewy hand, dripping with water, was clutching the rail.",
+ "normalized_hypothesis": "a siny hand dripping with water was clutching the rail",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ },
+ "semantic": {
+ "normalized_reference": "a sinewy hand dripping with water was clutching the rail",
+ "normalized_hypothesis": "a siny hand dripping with water was clutching the rail",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ }
+ },
+ {
+ "id": "release500_0246",
+ "category": "phonetic_balance",
+ "text": "Enters now the psychology of the situation.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0246.wav",
+ "render_seconds": 56.38914462400135,
+ "audio_seconds": 3.1346938775510202,
+ "rtf": 17.9887245219796,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "ENTERS NOW THE PSYCHOLOGY OF THE SITUATION",
+ "raw": {
+ "normalized_reference": "enters now the psychology of the situation.",
+ "normalized_hypothesis": "enters now the psychology of the situation",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "enters now the psychology of the situation",
+ "normalized_hypothesis": "enters now the psychology of the situation",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0247",
+ "category": "phonetic_balance",
+ "text": "Wash your hands of me.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0247.wav",
+ "render_seconds": 54.69748455809895,
+ "audio_seconds": 1.8808163265306121,
+ "rtf": 29.081778899162856,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "YES YE IM DIVER",
+ "raw": {
+ "normalized_reference": "wash your hands of me.",
+ "normalized_hypothesis": "yes ye im diver",
+ "errors": 5,
+ "words": 5,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "wash your hands of me",
+ "normalized_hypothesis": "yes ye im diver",
+ "errors": 5,
+ "words": 5,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0248",
+ "category": "phonetic_balance",
+ "text": "He confessed that the sketch had startled him.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0248.wav",
+ "render_seconds": 62.39778457500506,
+ "audio_seconds": 3.1346938775510202,
+ "rtf": 19.905543256349272,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE CONFESSED THAT THE SKETCH HAD STARTLED HIM",
+ "raw": {
+ "normalized_reference": "he confessed that the sketch had startled him.",
+ "normalized_hypothesis": "he confessed that the sketch had startled him",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he confessed that the sketch had startled him",
+ "normalized_hypothesis": "he confessed that the sketch had startled him",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0249",
+ "category": "phonetic_balance",
+ "text": "Her face was against his breast.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0249.wav",
+ "render_seconds": 54.099360143067315,
+ "audio_seconds": 2.507755102040816,
+ "rtf": 21.572824275799956,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "YOUR FAITH WAS AGAINST YOU DERTHED",
+ "raw": {
+ "normalized_reference": "her face was against his breast.",
+ "normalized_hypothesis": "your faith was against you derthed",
+ "errors": 4,
+ "words": 6,
+ "wer": 0.6666666666666666
+ },
+ "semantic": {
+ "normalized_reference": "her face was against his breast",
+ "normalized_hypothesis": "your faith was against you derthed",
+ "errors": 4,
+ "words": 6,
+ "wer": 0.6666666666666666
+ }
+ },
+ {
+ "id": "release500_0250",
+ "category": "phonetic_balance",
+ "text": "He was a pariah; a wanderer without a friend or a home.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0250.wav",
+ "render_seconds": 60.31028752902057,
+ "audio_seconds": 3.900952380952381,
+ "rtf": 15.460400855828027,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THERE WAS A PERIOV A GANDERER WITHOUT A FRIEND OR AN HIM",
+ "raw": {
+ "normalized_reference": "he was a pariah; a wanderer without a friend or a home.",
+ "normalized_hypothesis": "there was a periov a ganderer without a friend or an him",
+ "errors": 5,
+ "words": 12,
+ "wer": 0.4166666666666667
+ },
+ "semantic": {
+ "normalized_reference": "he was a pariah a wanderer without a friend or a home",
+ "normalized_hypothesis": "there was a periov a ganderer without a friend or an him",
+ "errors": 5,
+ "words": 12,
+ "wer": 0.4166666666666667
+ }
+ },
+ {
+ "id": "release500_0251",
+ "category": "phonetic_balance",
+ "text": "A cry of joy burst from Philip's lips.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0251.wav",
+ "render_seconds": 52.593778323964216,
+ "audio_seconds": 2.786394557823129,
+ "rtf": 18.875208529352392,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "A CRY OF JOY BURST FROM PHILIP'S LIPS",
+ "raw": {
+ "normalized_reference": "a cry of joy burst from philip's lips.",
+ "normalized_hypothesis": "a cry of joy burst from philip's lips",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "a cry of joy burst from philip is lips",
+ "normalized_hypothesis": "a cry of joy burst from philip is lips",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0252",
+ "category": "phonetic_balance",
+ "text": "And the big Persian knew of his existence before he did of hers.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0252.wav",
+ "render_seconds": 74.51021319197025,
+ "audio_seconds": 4.31891156462585,
+ "rtf": 17.25208124247043,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "AND THE BIG PERSIAN KNEW OF HIS EXISTENCE BEFORE HE DID OF HERS",
+ "raw": {
+ "normalized_reference": "and the big persian knew of his existence before he did of hers.",
+ "normalized_hypothesis": "and the big persian knew of his existence before he did of hers",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "and the big persian knew of his existence before he did of hers",
+ "normalized_hypothesis": "and the big persian knew of his existence before he did of hers",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0253",
+ "category": "phonetic_balance",
+ "text": "The men stared into each other's face.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0253.wav",
+ "render_seconds": 21.394839835935272,
+ "audio_seconds": 2.5774149659863945,
+ "rtf": 8.300890666919566,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE MEN STARED INTO EACH OTHER'S FACE",
+ "raw": {
+ "normalized_reference": "the men stared into each other's face.",
+ "normalized_hypothesis": "the men stared into each other's face",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the men stared into each other is face",
+ "normalized_hypothesis": "the men stared into each other is face",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0254",
+ "category": "phonetic_balance",
+ "text": "It is the fire, partly, she said.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0254.wav",
+ "render_seconds": 48.41107422008645,
+ "audio_seconds": 2.716734693877551,
+ "rtf": 17.81958110565081,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "AND AS THE FIRE PARTLY SHE SAID",
+ "raw": {
+ "normalized_reference": "it is the fire, partly, she said.",
+ "normalized_hypothesis": "and as the fire partly she said",
+ "errors": 2,
+ "words": 7,
+ "wer": 0.2857142857142857
+ },
+ "semantic": {
+ "normalized_reference": "it is the fire partly she said",
+ "normalized_hypothesis": "and as the fire partly she said",
+ "errors": 2,
+ "words": 7,
+ "wer": 0.2857142857142857
+ }
+ },
+ {
+ "id": "release500_0255",
+ "category": "phonetic_balance",
+ "text": "It seemed nearer to him since he had seen and talked with Gregson.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0255.wav",
+ "render_seconds": 81.30002614797559,
+ "audio_seconds": 4.597551020408163,
+ "rtf": 17.683333102143127,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT SEEMED NEARER TO HIM SINCE HE HAD SEEN AND TALKED WITH GREGSON",
+ "raw": {
+ "normalized_reference": "it seemed nearer to him since he had seen and talked with gregson.",
+ "normalized_hypothesis": "it seemed nearer to him since he had seen and talked with gregson",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it seemed nearer to him since he had seen and talked with gregson",
+ "normalized_hypothesis": "it seemed nearer to him since he had seen and talked with gregson",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0256",
+ "category": "phonetic_balance",
+ "text": "Among my minor afflictions, I may mention a new and mysterious one.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0256.wav",
+ "render_seconds": 70.59623351204209,
+ "audio_seconds": 4.9458503401360545,
+ "rtf": 14.273831324645395,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "AMONG MY MINOR AFFLICTIONS I MAY MENTION A NEW AND MYSTERIOUS ONE",
+ "raw": {
+ "normalized_reference": "among my minor afflictions, i may mention a new and mysterious one.",
+ "normalized_hypothesis": "among my minor afflictions i may mention a new and mysterious one",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "among my minor afflictions i may mention a new and mysterious one",
+ "normalized_hypothesis": "among my minor afflictions i may mention a new and mysterious one",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0257",
+ "category": "phonetic_balance",
+ "text": "Between him and all domestic animals there must be no hostilities.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0257.wav",
+ "render_seconds": 83.69677852105815,
+ "audio_seconds": 4.667210884353741,
+ "rtf": 17.932932664697336,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "BETWEEN HIM AND ALL DOMESTIC ANIMALS THERE MUST BE NO HOSTILITIES",
+ "raw": {
+ "normalized_reference": "between him and all domestic animals there must be no hostilities.",
+ "normalized_hypothesis": "between him and all domestic animals there must be no hostilities",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "between him and all domestic animals there must be no hostilities",
+ "normalized_hypothesis": "between him and all domestic animals there must be no hostilities",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0258",
+ "category": "phonetic_balance",
+ "text": "With these arguments he convinced himself that he should go on alone.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0258.wav",
+ "render_seconds": 74.49195452709682,
+ "audio_seconds": 4.597551020408163,
+ "rtf": 16.202529171820597,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WITH THESE ARGUMENTS HE CONVINCED HIMSELF THAT HE SHOULD GO ON ALONE",
+ "raw": {
+ "normalized_reference": "with these arguments he convinced himself that he should go on alone.",
+ "normalized_hypothesis": "with these arguments he convinced himself that he should go on alone",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "with these arguments he convinced himself that he should go on alone",
+ "normalized_hypothesis": "with these arguments he convinced himself that he should go on alone",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0259",
+ "category": "phonetic_balance",
+ "text": "We are consumed in our own flesh-pots.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0259.wav",
+ "render_seconds": 34.19644083199091,
+ "audio_seconds": 2.8560544217687074,
+ "rtf": 11.973315554265112,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WE ARE CONCERNED IN OUR OWN FLESH POTS",
+ "raw": {
+ "normalized_reference": "we are consumed in our own flesh-pots.",
+ "normalized_hypothesis": "we are concerned in our own flesh pots",
+ "errors": 1,
+ "words": 8,
+ "wer": 0.125
+ },
+ "semantic": {
+ "normalized_reference": "we are consumed in our own flesh pots",
+ "normalized_hypothesis": "we are concerned in our own flesh pots",
+ "errors": 1,
+ "words": 8,
+ "wer": 0.125
+ }
+ },
+ {
+ "id": "release500_0260",
+ "category": "phonetic_balance",
+ "text": "Mercedes screamed, cried, laughed, and manifested the chaotic abandonment of hysteria.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0260.wav",
+ "render_seconds": 91.18836701591499,
+ "audio_seconds": 6.199727891156463,
+ "rtf": 14.708446663601105,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "MARCEDES SCREAMED CRIED LAUGHED AND MANIFESTED THE CHAOTIC ABANDONMENT OF HYSTERIA",
+ "raw": {
+ "normalized_reference": "mercedes screamed, cried, laughed, and manifested the chaotic abandonment of hysteria.",
+ "normalized_hypothesis": "marcedes screamed cried laughed and manifested the chaotic abandonment of hysteria",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ },
+ "semantic": {
+ "normalized_reference": "mercedes screamed cried laughed and manifested the chaotic abandonment of hysteria",
+ "normalized_hypothesis": "marcedes screamed cried laughed and manifested the chaotic abandonment of hysteria",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ }
+ },
+ {
+ "id": "release500_0261",
+ "category": "phonetic_balance",
+ "text": "Philip made no effort to follow.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0261.wav",
+ "render_seconds": 63.59963294793852,
+ "audio_seconds": 2.2987755102040817,
+ "rtf": 27.666743736230355,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "FILL IT BY THE NOWN EFFORT TO FAIL",
+ "raw": {
+ "normalized_reference": "philip made no effort to follow.",
+ "normalized_hypothesis": "fill it by the nown effort to fail",
+ "errors": 6,
+ "words": 6,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "philip made no effort to follow",
+ "normalized_hypothesis": "fill it by the nown effort to fail",
+ "errors": 6,
+ "words": 6,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0262",
+ "category": "phonetic_balance",
+ "text": "She said with chattering teeth.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0262.wav",
+ "render_seconds": 51.599916751962155,
+ "audio_seconds": 2.2987755102040817,
+ "rtf": 22.446696740466493,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SHE SAID WITH CHATTERING ATTENT",
+ "raw": {
+ "normalized_reference": "she said with chattering teeth.",
+ "normalized_hypothesis": "she said with chattering attent",
+ "errors": 1,
+ "words": 5,
+ "wer": 0.2
+ },
+ "semantic": {
+ "normalized_reference": "she said with chattering teeth",
+ "normalized_hypothesis": "she said with chattering attent",
+ "errors": 1,
+ "words": 5,
+ "wer": 0.2
+ }
+ },
+ {
+ "id": "release500_0263",
+ "category": "phonetic_balance",
+ "text": "Halfway around the track one donkey got into an argument with its rider.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0263.wav",
+ "render_seconds": 91.50479103601538,
+ "audio_seconds": 5.015510204081632,
+ "rtf": 18.244363447122208,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HALF WAY AROUND THE TRACK ONE DONKEY GOT INTO AN ARGUMENT WITH ITS RIDER",
+ "raw": {
+ "normalized_reference": "halfway around the track one donkey got into an argument with its rider.",
+ "normalized_hypothesis": "half way around the track one donkey got into an argument with its rider",
+ "errors": 2,
+ "words": 13,
+ "wer": 0.15384615384615385
+ },
+ "semantic": {
+ "normalized_reference": "halfway around the track one donkey got into an argument with its rider",
+ "normalized_hypothesis": "half way around the track one donkey got into an argument with its rider",
+ "errors": 2,
+ "words": 13,
+ "wer": 0.15384615384615385
+ }
+ },
+ {
+ "id": "release500_0264",
+ "category": "phonetic_balance",
+ "text": "Earth and gravel seemed to fill the pan.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0264.wav",
+ "render_seconds": 53.298670324031264,
+ "audio_seconds": 2.8560544217687074,
+ "rtf": 18.66164381105325,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "EARSTH AND GAVEL SEEMED TO FILL THE PAN",
+ "raw": {
+ "normalized_reference": "earth and gravel seemed to fill the pan.",
+ "normalized_hypothesis": "earsth and gavel seemed to fill the pan",
+ "errors": 2,
+ "words": 8,
+ "wer": 0.25
+ },
+ "semantic": {
+ "normalized_reference": "earth and gravel seemed to fill the pan",
+ "normalized_hypothesis": "earsth and gavel seemed to fill the pan",
+ "errors": 2,
+ "words": 8,
+ "wer": 0.25
+ }
+ },
+ {
+ "id": "release500_0265",
+ "category": "phonetic_balance",
+ "text": "Her achievements with cocoanuts were a revelation.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0265.wav",
+ "render_seconds": 40.108290101983584,
+ "audio_seconds": 3.5526530612244898,
+ "rtf": 11.289672650489406,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HER ACHIEVEMENTS WITH COCOA NUTS WERE A REVELATION",
+ "raw": {
+ "normalized_reference": "her achievements with cocoanuts were a revelation.",
+ "normalized_hypothesis": "her achievements with cocoa nuts were a revelation",
+ "errors": 2,
+ "words": 7,
+ "wer": 0.2857142857142857
+ },
+ "semantic": {
+ "normalized_reference": "her achievements with cocoanuts were a revelation",
+ "normalized_hypothesis": "her achievements with cocoa nuts were a revelation",
+ "errors": 2,
+ "words": 7,
+ "wer": 0.2857142857142857
+ }
+ },
+ {
+ "id": "release500_0266",
+ "category": "phonetic_balance",
+ "text": "For a full minute he crouched and listened.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0266.wav",
+ "render_seconds": 55.399573264992796,
+ "audio_seconds": 3.1346938775510202,
+ "rtf": 17.67304095042088,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "FOR A FULL MINUTE HE CROUCHED AND LISTENED",
+ "raw": {
+ "normalized_reference": "for a full minute he crouched and listened.",
+ "normalized_hypothesis": "for a full minute he crouched and listened",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "for a full minute he crouched and listened",
+ "normalized_hypothesis": "for a full minute he crouched and listened",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0267",
+ "category": "phonetic_balance",
+ "text": "It was edged with ice.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0267.wav",
+ "render_seconds": 46.69210641994141,
+ "audio_seconds": 1.9504761904761905,
+ "rtf": 23.938824092255118,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT WAS ANSERMGATYPE",
+ "raw": {
+ "normalized_reference": "it was edged with ice.",
+ "normalized_hypothesis": "it was ansermgatype",
+ "errors": 3,
+ "words": 5,
+ "wer": 0.6
+ },
+ "semantic": {
+ "normalized_reference": "it was edged with ice",
+ "normalized_hypothesis": "it was ansermgatype",
+ "errors": 3,
+ "words": 5,
+ "wer": 0.6
+ }
+ },
+ {
+ "id": "release500_0268",
+ "category": "phonetic_balance",
+ "text": "The Portuguese boy passed the Hawaiian.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0268.wav",
+ "render_seconds": 52.094459216925316,
+ "audio_seconds": 2.8560544217687074,
+ "rtf": 18.240009300895633,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE PORTUGUESE OY PASSED THE HAWAYAN",
+ "raw": {
+ "normalized_reference": "the portuguese boy passed the hawaiian.",
+ "normalized_hypothesis": "the portuguese oy passed the hawayan",
+ "errors": 2,
+ "words": 6,
+ "wer": 0.3333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "the portuguese boy passed the hawaiian",
+ "normalized_hypothesis": "the portuguese oy passed the hawayan",
+ "errors": 2,
+ "words": 6,
+ "wer": 0.3333333333333333
+ }
+ },
+ {
+ "id": "release500_0269",
+ "category": "phonetic_balance",
+ "text": "But he did not broach it, preferring to mature it carefully.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0269.wav",
+ "render_seconds": 64.79198672692291,
+ "audio_seconds": 4.1099319727891155,
+ "rtf": 15.764734588284012,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "BUT HE DID NOT BROACH IT PREFERRING TO MATURE IT CAREFULLY",
+ "raw": {
+ "normalized_reference": "but he did not broach it, preferring to mature it carefully.",
+ "normalized_hypothesis": "but he did not broach it preferring to mature it carefully",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "but he did not broach it preferring to mature it carefully",
+ "normalized_hypothesis": "but he did not broach it preferring to mature it carefully",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0270",
+ "category": "phonetic_balance",
+ "text": "Come on, Del Mar challenged.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0270.wav",
+ "render_seconds": 39.909629823989235,
+ "audio_seconds": 2.159455782312925,
+ "rtf": 18.481336895559533,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "GOD MAN DIL MARE TRIUMPH",
+ "raw": {
+ "normalized_reference": "come on, del mar challenged.",
+ "normalized_hypothesis": "god man dil mare triumph",
+ "errors": 5,
+ "words": 5,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "come on del mar challenged",
+ "normalized_hypothesis": "god man dil mare triumph",
+ "errors": 5,
+ "words": 5,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0271",
+ "category": "phonetic_balance",
+ "text": "For a full half minute Jeanne looked at him without speaking.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0271.wav",
+ "render_seconds": 58.31023634294979,
+ "audio_seconds": 4.179591836734694,
+ "rtf": 13.951179593772165,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "FOR A FULL HALF MINUTE JAN LOOKED AT HIM WITHOUT SPEAKING",
+ "raw": {
+ "normalized_reference": "for a full half minute jeanne looked at him without speaking.",
+ "normalized_hypothesis": "for a full half minute jan looked at him without speaking",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ },
+ "semantic": {
+ "normalized_reference": "for a full half minute jeanne looked at him without speaking",
+ "normalized_hypothesis": "for a full half minute jan looked at him without speaking",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ }
+ },
+ {
+ "id": "release500_0272",
+ "category": "phonetic_balance",
+ "text": "He bore no grudges and had few enemies.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0272.wav",
+ "render_seconds": 29.90095866797492,
+ "audio_seconds": 2.786394557823129,
+ "rtf": 10.73105694382889,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE BORE NO GRUDGES AND HAD FEW ENEMIES",
+ "raw": {
+ "normalized_reference": "he bore no grudges and had few enemies.",
+ "normalized_hypothesis": "he bore no grudges and had few enemies",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he bore no grudges and had few enemies",
+ "normalized_hypothesis": "he bore no grudges and had few enemies",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0273",
+ "category": "phonetic_balance",
+ "text": "I'll go over tomorrow afternoon.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0273.wav",
+ "render_seconds": 53.60737494705245,
+ "audio_seconds": 2.2987755102040817,
+ "rtf": 23.31996957036195,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "TE GAFU TO MARRY AFTER MON",
+ "raw": {
+ "normalized_reference": "i'll go over tomorrow afternoon.",
+ "normalized_hypothesis": "te gafu to marry after mon",
+ "errors": 6,
+ "words": 5,
+ "wer": 1.2
+ },
+ "semantic": {
+ "normalized_reference": "i will go over tomorrow afternoon",
+ "normalized_hypothesis": "te gafu to marry after mon",
+ "errors": 6,
+ "words": 6,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0274",
+ "category": "phonetic_balance",
+ "text": "Well, there are better men in Hawaii, that's all.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0274.wav",
+ "render_seconds": 56.00135361694265,
+ "audio_seconds": 3.274013605442177,
+ "rtf": 17.104801740547227,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WELL THERE ARE BETTER MEN IN HAWAII THAT'S ALL",
+ "raw": {
+ "normalized_reference": "well, there are better men in hawaii, that's all.",
+ "normalized_hypothesis": "well there are better men in hawaii that's all",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "well there are better men in hawaii that is all",
+ "normalized_hypothesis": "well there are better men in hawaii that is all",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0275",
+ "category": "phonetic_balance",
+ "text": "He was an athlete and a giant.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0275.wav",
+ "render_seconds": 52.600108951097354,
+ "audio_seconds": 2.36843537414966,
+ "rtf": 22.208800596884508,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THERE WAS A NESLATE AND A GIANT",
+ "raw": {
+ "normalized_reference": "he was an athlete and a giant.",
+ "normalized_hypothesis": "there was a neslate and a giant",
+ "errors": 3,
+ "words": 7,
+ "wer": 0.42857142857142855
+ },
+ "semantic": {
+ "normalized_reference": "he was an athlete and a giant",
+ "normalized_hypothesis": "there was a neslate and a giant",
+ "errors": 3,
+ "words": 7,
+ "wer": 0.42857142857142855
+ }
+ },
+ {
+ "id": "release500_0276",
+ "category": "phonetic_balance",
+ "text": "McVeigh when he returns from a trip to Honolulu.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0276.wav",
+ "render_seconds": 55.38810772099532,
+ "audio_seconds": 3.3436734693877552,
+ "rtf": 16.565046864799626,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "MAKVAY WHEN HE RETURNS FROM A TRIP TO VONNOLULU",
+ "raw": {
+ "normalized_reference": "mcveigh when he returns from a trip to honolulu.",
+ "normalized_hypothesis": "makvay when he returns from a trip to vonnolulu",
+ "errors": 2,
+ "words": 9,
+ "wer": 0.2222222222222222
+ },
+ "semantic": {
+ "normalized_reference": "mcveigh when he returns from a trip to honolulu",
+ "normalized_hypothesis": "makvay when he returns from a trip to vonnolulu",
+ "errors": 2,
+ "words": 9,
+ "wer": 0.2222222222222222
+ }
+ },
+ {
+ "id": "release500_0277",
+ "category": "phonetic_balance",
+ "text": "You're a devil for fighting, and will surely win.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0277.wav",
+ "render_seconds": 32.88942808797583,
+ "audio_seconds": 3.274013605442177,
+ "rtf": 10.045599087708707,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "YOU'RE A DEVIL FOR FIGHTING AND WILL SURELY WIN",
+ "raw": {
+ "normalized_reference": "you're a devil for fighting, and will surely win.",
+ "normalized_hypothesis": "you're a devil for fighting and will surely win",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "you are a devil for fighting and will surely win",
+ "normalized_hypothesis": "you are a devil for fighting and will surely win",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0278",
+ "category": "phonetic_balance",
+ "text": "All an appearance can know is mirage.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0278.wav",
+ "render_seconds": 31.810034321970306,
+ "audio_seconds": 2.8560544217687074,
+ "rtf": 11.137754966962737,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "ALL IN APPEARANCE CAN KNOW HIS MIRAGE",
+ "raw": {
+ "normalized_reference": "all an appearance can know is mirage.",
+ "normalized_hypothesis": "all in appearance can know his mirage",
+ "errors": 2,
+ "words": 7,
+ "wer": 0.2857142857142857
+ },
+ "semantic": {
+ "normalized_reference": "all an appearance can know is mirage",
+ "normalized_hypothesis": "all in appearance can know his mirage",
+ "errors": 2,
+ "words": 7,
+ "wer": 0.2857142857142857
+ }
+ },
+ {
+ "id": "release500_0279",
+ "category": "phonetic_balance",
+ "text": "You fired me out of your house, in short.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0279.wav",
+ "render_seconds": 66.29974531300832,
+ "audio_seconds": 2.9257142857142857,
+ "rtf": 22.66104576128214,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "YOU FIRED ME OUT OF YOUR HOUSE IN SHORT",
+ "raw": {
+ "normalized_reference": "you fired me out of your house, in short.",
+ "normalized_hypothesis": "you fired me out of your house in short",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "you fired me out of your house in short",
+ "normalized_hypothesis": "you fired me out of your house in short",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0280",
+ "category": "phonetic_balance",
+ "text": "He made sure that the magazine was loaded, and resumed his paddling.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0280.wav",
+ "render_seconds": 70.5980139960302,
+ "audio_seconds": 4.876190476190477,
+ "rtf": 14.47810833902963,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE MADE SURE THAT THE MAGAZINE WAS LOADED AND RESUMED HIS PADDLING",
+ "raw": {
+ "normalized_reference": "he made sure that the magazine was loaded, and resumed his paddling.",
+ "normalized_hypothesis": "he made sure that the magazine was loaded and resumed his paddling",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he made sure that the magazine was loaded and resumed his paddling",
+ "normalized_hypothesis": "he made sure that the magazine was loaded and resumed his paddling",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0281",
+ "category": "phonetic_balance",
+ "text": "Change chairs, Daylight commanded.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0281.wav",
+ "render_seconds": 47.59949245606549,
+ "audio_seconds": 2.507755102040816,
+ "rtf": 18.980917401914137,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "TUNES CIL TERLIGHT COMMANDED",
+ "raw": {
+ "normalized_reference": "change chairs, daylight commanded.",
+ "normalized_hypothesis": "tunes cil terlight commanded",
+ "errors": 3,
+ "words": 4,
+ "wer": 0.75
+ },
+ "semantic": {
+ "normalized_reference": "change chairs daylight commanded",
+ "normalized_hypothesis": "tunes cil terlight commanded",
+ "errors": 3,
+ "words": 4,
+ "wer": 0.75
+ }
+ },
+ {
+ "id": "release500_0282",
+ "category": "phonetic_balance",
+ "text": "By that answer my professional medical prestige stood or fell.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0282.wav",
+ "render_seconds": 73.81132563599385,
+ "audio_seconds": 4.667210884353741,
+ "rtf": 15.814868336769907,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "BY THAT ANSWER MY PROFESSIONAL MEDICAL PRESTIGE STOOD OR FELL",
+ "raw": {
+ "normalized_reference": "by that answer my professional medical prestige stood or fell.",
+ "normalized_hypothesis": "by that answer my professional medical prestige stood or fell",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "by that answer my professional medical prestige stood or fell",
+ "normalized_hypothesis": "by that answer my professional medical prestige stood or fell",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0283",
+ "category": "phonetic_balance",
+ "text": "The moon had already begun its westward decline.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0283.wav",
+ "render_seconds": 33.006494734901935,
+ "audio_seconds": 3.4133333333333336,
+ "rtf": 9.669871504365801,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE MOON HAD ALREADY BEGUN ITS WESTWARD DECLINE",
+ "raw": {
+ "normalized_reference": "the moon had already begun its westward decline.",
+ "normalized_hypothesis": "the moon had already begun its westward decline",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the moon had already begun its westward decline",
+ "normalized_hypothesis": "the moon had already begun its westward decline",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0284",
+ "category": "phonetic_balance",
+ "text": "At sea, Monday, March 16, 1908.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0284.wav",
+ "render_seconds": 41.395895043970086,
+ "audio_seconds": 3.8312925170068026,
+ "rtf": 10.804681412399862,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "AT SEA MONDAY MARCH SIXTEENTH NINETEEN O EIGHT",
+ "raw": {
+ "normalized_reference": "at sea, monday, march 16, 1908.",
+ "normalized_hypothesis": "at sea monday march sixteenth nineteen o eight",
+ "errors": 4,
+ "words": 6,
+ "wer": 0.6666666666666666
+ },
+ "semantic": {
+ "normalized_reference": "at sea monday march 16 1908",
+ "normalized_hypothesis": "at sea monday march 16th 1908",
+ "errors": 1,
+ "words": 6,
+ "wer": 0.16666666666666666
+ }
+ },
+ {
+ "id": "release500_0285",
+ "category": "phonetic_balance",
+ "text": "And wherever I ranged, the way lay along alcohol-drenched roads.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0285.wav",
+ "render_seconds": 86.40386472502723,
+ "audio_seconds": 4.527891156462585,
+ "rtf": 19.08258430675932,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "AND WHEREVER I RANGED THE WAY LAY ALONG ALCHAL DRENCHED ROADS",
+ "raw": {
+ "normalized_reference": "and wherever i ranged, the way lay along alcohol-drenched roads.",
+ "normalized_hypothesis": "and wherever i ranged the way lay along alchal drenched roads",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ },
+ "semantic": {
+ "normalized_reference": "and wherever i ranged the way lay along alcohol drenched roads",
+ "normalized_hypothesis": "and wherever i ranged the way lay along alchal drenched roads",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ }
+ },
+ {
+ "id": "release500_0286",
+ "category": "phonetic_balance",
+ "text": "The date was nearly eighteen years old.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0286.wav",
+ "render_seconds": 45.90761814196594,
+ "audio_seconds": 2.8560544217687074,
+ "rtf": 16.073789698144516,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE DIGHT WAS NEARLY ABOUT TWON YEAROS OLD",
+ "raw": {
+ "normalized_reference": "the date was nearly eighteen years old.",
+ "normalized_hypothesis": "the dight was nearly about twon yearos old",
+ "errors": 4,
+ "words": 7,
+ "wer": 0.5714285714285714
+ },
+ "semantic": {
+ "normalized_reference": "the date was nearly 18 years old",
+ "normalized_hypothesis": "the dight was nearly about twon yearos old",
+ "errors": 4,
+ "words": 7,
+ "wer": 0.5714285714285714
+ }
+ },
+ {
+ "id": "release500_0287",
+ "category": "phonetic_balance",
+ "text": "Very early in my life, I separated from my mother.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0287.wav",
+ "render_seconds": 68.00331256096251,
+ "audio_seconds": 3.900952380952381,
+ "rtf": 17.432489792238925,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "VERY EARLY IN MY LIFE I SEPARATED FROM MY MOTHER",
+ "raw": {
+ "normalized_reference": "very early in my life, i separated from my mother.",
+ "normalized_hypothesis": "very early in my life i separated from my mother",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "very early in my life i separated from my mother",
+ "normalized_hypothesis": "very early in my life i separated from my mother",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0288",
+ "category": "phonetic_balance",
+ "text": "There was pride and strength, the ring of triumph in his voice.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0288.wav",
+ "render_seconds": 58.888243380934,
+ "audio_seconds": 4.458231292517007,
+ "rtf": 13.208880274959256,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THERE WAS PRIDE AND STRENGTH THE RING OF TRIUMPH IN HIS VOICE",
+ "raw": {
+ "normalized_reference": "there was pride and strength, the ring of triumph in his voice.",
+ "normalized_hypothesis": "there was pride and strength the ring of triumph in his voice",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "there was pride and strength the ring of triumph in his voice",
+ "normalized_hypothesis": "there was pride and strength the ring of triumph in his voice",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0289",
+ "category": "phonetic_balance",
+ "text": "From my earliest recollection my sleep was a period of terror.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0289.wav",
+ "render_seconds": 42.4981372730108,
+ "audio_seconds": 4.73687074829932,
+ "rtf": 8.971774728763481,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "TO MY EARLIEST RECOLLECTION MY SLEEP WAS A PERIOD OF TERROR",
+ "raw": {
+ "normalized_reference": "from my earliest recollection my sleep was a period of terror.",
+ "normalized_hypothesis": "to my earliest recollection my sleep was a period of terror",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ },
+ "semantic": {
+ "normalized_reference": "from my earliest recollection my sleep was a period of terror",
+ "normalized_hypothesis": "to my earliest recollection my sleep was a period of terror",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ }
+ },
+ {
+ "id": "release500_0290",
+ "category": "phonetic_balance",
+ "text": "We'll have to watch our chances.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0290.wav",
+ "render_seconds": 34.978232913068496,
+ "audio_seconds": 2.2987755102040817,
+ "rtf": 15.21602816708413,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WES HAVE TO MATCH OUR TRAMPEN",
+ "raw": {
+ "normalized_reference": "we'll have to watch our chances.",
+ "normalized_hypothesis": "wes have to match our trampen",
+ "errors": 3,
+ "words": 6,
+ "wer": 0.5
+ },
+ "semantic": {
+ "normalized_reference": "we will have to watch our chances",
+ "normalized_hypothesis": "wes have to match our trampen",
+ "errors": 4,
+ "words": 7,
+ "wer": 0.5714285714285714
+ }
+ },
+ {
+ "id": "release500_0291",
+ "category": "phonetic_balance",
+ "text": "Close beside him gleamed the white fangs of the wolf-dog.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0291.wav",
+ "render_seconds": 67.38768891594373,
+ "audio_seconds": 3.7616326530612243,
+ "rtf": 17.914478932732436,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "CLOSE BESIDE HIM GLEAMED THE WHITE FANGS OF THE WOLF DOG",
+ "raw": {
+ "normalized_reference": "close beside him gleamed the white fangs of the wolf-dog.",
+ "normalized_hypothesis": "close beside him gleamed the white fangs of the wolf dog",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "close beside him gleamed the white fangs of the wolf dog",
+ "normalized_hypothesis": "close beside him gleamed the white fangs of the wolf dog",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0292",
+ "category": "phonetic_balance",
+ "text": "Would you be satisfied with that one hundredth part of me.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0292.wav",
+ "render_seconds": 58.39204915298615,
+ "audio_seconds": 3.900952380952381,
+ "rtf": 14.968664944002796,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WOULD YOU BE SATISFIED WITH THAT ONE HUNDREDTH PART OF ME",
+ "raw": {
+ "normalized_reference": "would you be satisfied with that one hundredth part of me.",
+ "normalized_hypothesis": "would you be satisfied with that one hundredth part of me",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "would you be satisfied with that 100th part of me",
+ "normalized_hypothesis": "would you be satisfied with that 100th part of me",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0293",
+ "category": "phonetic_balance",
+ "text": "Two gigantic owls were tearing at the carcass.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0293.wav",
+ "render_seconds": 63.297486994066276,
+ "audio_seconds": 3.3436734693877552,
+ "rtf": 18.930522843684372,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "TWO GIGANTIC OWLS WERE TEARING AT THE CARCASE",
+ "raw": {
+ "normalized_reference": "two gigantic owls were tearing at the carcass.",
+ "normalized_hypothesis": "two gigantic owls were tearing at the carcase",
+ "errors": 1,
+ "words": 8,
+ "wer": 0.125
+ },
+ "semantic": {
+ "normalized_reference": "2 gigantic owls were tearing at the carcass",
+ "normalized_hypothesis": "2 gigantic owls were tearing at the carcase",
+ "errors": 1,
+ "words": 8,
+ "wer": 0.125
+ }
+ },
+ {
+ "id": "release500_0294",
+ "category": "phonetic_balance",
+ "text": "The lush vegetation of that sheltered spot make a natural shield.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0294.wav",
+ "render_seconds": 68.8004981239792,
+ "audio_seconds": 4.597551020408163,
+ "rtf": 14.964596981866924,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE LUSH VEGETATION OF THAT SHELTERED SPOT MAKE A NATURAL SHIELD",
+ "raw": {
+ "normalized_reference": "the lush vegetation of that sheltered spot make a natural shield.",
+ "normalized_hypothesis": "the lush vegetation of that sheltered spot make a natural shield",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the lush vegetation of that sheltered spot make a natural shield",
+ "normalized_hypothesis": "the lush vegetation of that sheltered spot make a natural shield",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0295",
+ "category": "phonetic_balance",
+ "text": "Below him the shadow was broken into a pool of rippling starlight.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0295.wav",
+ "render_seconds": 36.89564866293222,
+ "audio_seconds": 4.527891156462585,
+ "rtf": 8.148528175256965,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "BELOW HIM THE SHADOW WAS BREAKING INTO A POOL OF RIPPLING STARLIGHT",
+ "raw": {
+ "normalized_reference": "below him the shadow was broken into a pool of rippling starlight.",
+ "normalized_hypothesis": "below him the shadow was breaking into a pool of rippling starlight",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "below him the shadow was broken into a pool of rippling starlight",
+ "normalized_hypothesis": "below him the shadow was breaking into a pool of rippling starlight",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ }
+ },
+ {
+ "id": "release500_0296",
+ "category": "phonetic_balance",
+ "text": "Lots of men take women buggy riding.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0296.wav",
+ "render_seconds": 31.415752750006504,
+ "audio_seconds": 2.507755102040816,
+ "rtf": 12.527440468345693,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "LOTFIN MEN TUK WOMEN BUGGY RIDER",
+ "raw": {
+ "normalized_reference": "lots of men take women buggy riding.",
+ "normalized_hypothesis": "lotfin men tuk women buggy rider",
+ "errors": 4,
+ "words": 7,
+ "wer": 0.5714285714285714
+ },
+ "semantic": {
+ "normalized_reference": "lots of men take women buggy riding",
+ "normalized_hypothesis": "lotfin men tuk women buggy rider",
+ "errors": 4,
+ "words": 7,
+ "wer": 0.5714285714285714
+ }
+ },
+ {
+ "id": "release500_0297",
+ "category": "phonetic_balance",
+ "text": "But it contributed to the smash.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0297.wav",
+ "render_seconds": 44.90772803698201,
+ "audio_seconds": 2.4380952380952383,
+ "rtf": 18.4191853276684,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "BUT IT CONTRIBUTED TO THE SMASH",
+ "raw": {
+ "normalized_reference": "but it contributed to the smash.",
+ "normalized_hypothesis": "but it contributed to the smash",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "but it contributed to the smash",
+ "normalized_hypothesis": "but it contributed to the smash",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0298",
+ "category": "phonetic_balance",
+ "text": "No man ate of the seal meat or the oil.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0298.wav",
+ "render_seconds": 39.89926950400695,
+ "audio_seconds": 2.786394557823129,
+ "rtf": 14.319317912814997,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "NO MERE LAT OF THE SEAL MAT OR THE OIL",
+ "raw": {
+ "normalized_reference": "no man ate of the seal meat or the oil.",
+ "normalized_hypothesis": "no mere lat of the seal mat or the oil",
+ "errors": 3,
+ "words": 10,
+ "wer": 0.3
+ },
+ "semantic": {
+ "normalized_reference": "no man ate of the seal meat or the oil",
+ "normalized_hypothesis": "no mere lat of the seal mat or the oil",
+ "errors": 3,
+ "words": 10,
+ "wer": 0.3
+ }
+ },
+ {
+ "id": "release500_0299",
+ "category": "phonetic_balance",
+ "text": "You're going in for grab sharing.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0299.wav",
+ "render_seconds": 38.899090984952636,
+ "audio_seconds": 2.2291156462585033,
+ "rtf": 17.450458907434196,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "YOU'RE GOING IN FOR GRAB SHARING",
+ "raw": {
+ "normalized_reference": "you're going in for grab sharing.",
+ "normalized_hypothesis": "you're going in for grab sharing",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "you are going in for grab sharing",
+ "normalized_hypothesis": "you are going in for grab sharing",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0300",
+ "category": "phonetic_balance",
+ "text": "He it was that lived to found the family of the Patino.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0300.wav",
+ "render_seconds": 62.90078269294463,
+ "audio_seconds": 3.6919727891156464,
+ "rtf": 17.037173967907687,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT WAS THAT LIVED TO FOUND THE FAMILY OF THE PATINA",
+ "raw": {
+ "normalized_reference": "he it was that lived to found the family of the patino.",
+ "normalized_hypothesis": "it was that lived to found the family of the patina",
+ "errors": 2,
+ "words": 12,
+ "wer": 0.16666666666666666
+ },
+ "semantic": {
+ "normalized_reference": "he it was that lived to found the family of the patino",
+ "normalized_hypothesis": "it was that lived to found the family of the patina",
+ "errors": 2,
+ "words": 12,
+ "wer": 0.16666666666666666
+ }
+ },
+ {
+ "id": "release500_0301",
+ "category": "phonetic_balance",
+ "text": "For such countries nothing remained but reorganization.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0301.wav",
+ "render_seconds": 41.90868374996353,
+ "audio_seconds": 3.9706122448979593,
+ "rtf": 10.554715788179548,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "FOR SUCH COUNTRIES NOTHING REMAINED BUT REORGANIZATION",
+ "raw": {
+ "normalized_reference": "for such countries nothing remained but reorganization.",
+ "normalized_hypothesis": "for such countries nothing remained but reorganization",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "for such countries nothing remained but reorganization",
+ "normalized_hypothesis": "for such countries nothing remained but reorganization",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0302",
+ "category": "phonetic_balance",
+ "text": "Stand off butcher and baker and all the rest.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0302.wav",
+ "render_seconds": 42.398791291052476,
+ "audio_seconds": 3.3436734693877552,
+ "rtf": 12.680302571176583,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "STAND OFF BUTCHER AND BAKER AND ALL THE REST",
+ "raw": {
+ "normalized_reference": "stand off butcher and baker and all the rest.",
+ "normalized_hypothesis": "stand off butcher and baker and all the rest",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "stand off butcher and baker and all the rest",
+ "normalized_hypothesis": "stand off butcher and baker and all the rest",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0303",
+ "category": "phonetic_balance",
+ "text": "All eyes, however, were staring at him in certitude of expectancy.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0303.wav",
+ "render_seconds": 67.49231372890063,
+ "audio_seconds": 4.806530612244898,
+ "rtf": 14.04179421160042,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "ALL EYES HOWEVER WERE STARING AT HIM IN CERTITUDE OF EXPECTANCY",
+ "raw": {
+ "normalized_reference": "all eyes, however, were staring at him in certitude of expectancy.",
+ "normalized_hypothesis": "all eyes however were staring at him in certitude of expectancy",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "all eyes however were staring at him in certitude of expectancy",
+ "normalized_hypothesis": "all eyes however were staring at him in certitude of expectancy",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0304",
+ "category": "phonetic_balance",
+ "text": "Two days ago Jeanne learned where her father's men were hiding.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0304.wav",
+ "render_seconds": 67.70582575898152,
+ "audio_seconds": 3.9706122448979593,
+ "rtf": 17.051734489052706,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "TWO DAYS AGO CIANN LEARNED WHERE HER FATHER'S MEN WERE HIDING",
+ "raw": {
+ "normalized_reference": "two days ago jeanne learned where her father's men were hiding.",
+ "normalized_hypothesis": "two days ago ciann learned where her father's men were hiding",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ },
+ "semantic": {
+ "normalized_reference": "2 days ago jeanne learned where her father is men were hiding",
+ "normalized_hypothesis": "2 days ago ciann learned where her father is men were hiding",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ }
+ },
+ {
+ "id": "release500_0305",
+ "category": "phonetic_balance",
+ "text": "His previous wives had never lived long enough to bear him children.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0305.wav",
+ "render_seconds": 46.40942315000575,
+ "audio_seconds": 4.597551020408163,
+ "rtf": 10.094379147506578,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HIS PREVIOUS WIVES HAD NEVER LIVED LONG ENOUGH TO BEAR HIM CHILDREN",
+ "raw": {
+ "normalized_reference": "his previous wives had never lived long enough to bear him children.",
+ "normalized_hypothesis": "his previous wives had never lived long enough to bear him children",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "his previous wives had never lived long enough to bear him children",
+ "normalized_hypothesis": "his previous wives had never lived long enough to bear him children",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0306",
+ "category": "phonetic_balance",
+ "text": "But this little defect did not worry him.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0306.wav",
+ "render_seconds": 55.49938709696289,
+ "audio_seconds": 2.8560544217687074,
+ "rtf": 19.432188222307417,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THIS LITTLE DEFECT DID NOT WORRY HIM",
+ "raw": {
+ "normalized_reference": "but this little defect did not worry him.",
+ "normalized_hypothesis": "this little defect did not worry him",
+ "errors": 1,
+ "words": 8,
+ "wer": 0.125
+ },
+ "semantic": {
+ "normalized_reference": "but this little defect did not worry him",
+ "normalized_hypothesis": "this little defect did not worry him",
+ "errors": 1,
+ "words": 8,
+ "wer": 0.125
+ }
+ },
+ {
+ "id": "release500_0307",
+ "category": "phonetic_balance",
+ "text": "Eli Harding asked, as Shunk started to follow.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0307.wav",
+ "render_seconds": 29.992313339957036,
+ "audio_seconds": 3.4829931972789114,
+ "rtf": 8.611074337839227,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "YE'RE HARDING AFT AS SHUNK STARTED TO FOLLOW",
+ "raw": {
+ "normalized_reference": "eli harding asked, as shunk started to follow.",
+ "normalized_hypothesis": "ye're harding aft as shunk started to follow",
+ "errors": 2,
+ "words": 8,
+ "wer": 0.25
+ },
+ "semantic": {
+ "normalized_reference": "eli harding asked as shunk started to follow",
+ "normalized_hypothesis": "ye are harding aft as shunk started to follow",
+ "errors": 3,
+ "words": 8,
+ "wer": 0.375
+ }
+ },
+ {
+ "id": "release500_0308",
+ "category": "phonetic_balance",
+ "text": "And this was their sole conversation throughout the meal.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0308.wav",
+ "render_seconds": 65.61097859800793,
+ "audio_seconds": 3.900952380952381,
+ "rtf": 16.819220587868244,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "AND THIS WAS THEIR SOLE CONVERSATION THROUGHOUT THE MEAL",
+ "raw": {
+ "normalized_reference": "and this was their sole conversation throughout the meal.",
+ "normalized_hypothesis": "and this was their sole conversation throughout the meal",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "and this was their sole conversation throughout the meal",
+ "normalized_hypothesis": "and this was their sole conversation throughout the meal",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0309",
+ "category": "phonetic_balance",
+ "text": "Then, again, Tudor had such an irritating way about him.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0309.wav",
+ "render_seconds": 65.31061014102306,
+ "audio_seconds": 4.179591836734694,
+ "rtf": 15.626073715381493,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THEN AGAIN TUDOR HAD SUCH AN IRRITATING WAY ABOUT HIM",
+ "raw": {
+ "normalized_reference": "then, again, tudor had such an irritating way about him.",
+ "normalized_hypothesis": "then again tudor had such an irritating way about him",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "then again tudor had such an irritating way about him",
+ "normalized_hypothesis": "then again tudor had such an irritating way about him",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0310",
+ "category": "phonetic_balance",
+ "text": "In it there was something that was almost tragedy.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0310.wav",
+ "render_seconds": 32.71393072709907,
+ "audio_seconds": 3.4829931972789114,
+ "rtf": 9.392476204850709,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IN IT THERE WAS SOMETHING THAT WAS ALMOST TRAGEDY",
+ "raw": {
+ "normalized_reference": "in it there was something that was almost tragedy.",
+ "normalized_hypothesis": "in it there was something that was almost tragedy",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "in it there was something that was almost tragedy",
+ "normalized_hypothesis": "in it there was something that was almost tragedy",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0311",
+ "category": "phonetic_balance",
+ "text": "Thirty pounds, said the captain with finality.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0311.wav",
+ "render_seconds": 23.599484464037232,
+ "audio_seconds": 3.274013605442177,
+ "rtf": 7.208120462544617,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THIRTY POUNDS SAID THE CAPTAIN WITH FINALITY",
+ "raw": {
+ "normalized_reference": "thirty pounds, said the captain with finality.",
+ "normalized_hypothesis": "thirty pounds said the captain with finality",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "\u00a330 said the captain with finality",
+ "normalized_hypothesis": "\u00a330 said the captain with finality",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0312",
+ "category": "phonetic_balance",
+ "text": "It was over when he made his way through the ring of spectators.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0312.wav",
+ "render_seconds": 66.0076415190706,
+ "audio_seconds": 4.31891156462585,
+ "rtf": 15.283397340132591,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT WAS OVER WHEN HE MADE HIS WAY THROUGH THE RING OF SPECTATORS",
+ "raw": {
+ "normalized_reference": "it was over when he made his way through the ring of spectators.",
+ "normalized_hypothesis": "it was over when he made his way through the ring of spectators",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it was over when he made his way through the ring of spectators",
+ "normalized_hypothesis": "it was over when he made his way through the ring of spectators",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0313",
+ "category": "phonetic_balance",
+ "text": "She had died from cold and starvation.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0313.wav",
+ "render_seconds": 31.52514024404809,
+ "audio_seconds": 2.9257142857142857,
+ "rtf": 10.775194419352374,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SHE HAD DIED FROM COLD AND STARVATION",
+ "raw": {
+ "normalized_reference": "she had died from cold and starvation.",
+ "normalized_hypothesis": "she had died from cold and starvation",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "she had died from cold and starvation",
+ "normalized_hypothesis": "she had died from cold and starvation",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0314",
+ "category": "phonetic_balance",
+ "text": "We fished sharks on Niihau together.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0314.wav",
+ "render_seconds": 52.09936827199999,
+ "audio_seconds": 2.647074829931973,
+ "rtf": 19.68186455587993,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "REFISHE SHARK VON MEHALL TOGETHER",
+ "raw": {
+ "normalized_reference": "we fished sharks on niihau together.",
+ "normalized_hypothesis": "refishe shark von mehall together",
+ "errors": 5,
+ "words": 6,
+ "wer": 0.8333333333333334
+ },
+ "semantic": {
+ "normalized_reference": "we fished sharks on niihau together",
+ "normalized_hypothesis": "refishe shark von mehall together",
+ "errors": 5,
+ "words": 6,
+ "wer": 0.8333333333333334
+ }
+ },
+ {
+ "id": "release500_0315",
+ "category": "phonetic_balance",
+ "text": "His slim hands gripped the edges of the table.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0315.wav",
+ "render_seconds": 65.1922088349238,
+ "audio_seconds": 3.274013605442177,
+ "rtf": 19.912015248366437,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HIS SLIM HANDS GRIPPED THE EDGES OF THE TABLE",
+ "raw": {
+ "normalized_reference": "his slim hands gripped the edges of the table.",
+ "normalized_hypothesis": "his slim hands gripped the edges of the table",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "his slim hands gripped the edges of the table",
+ "normalized_hypothesis": "his slim hands gripped the edges of the table",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0316",
+ "category": "phonetic_balance",
+ "text": "We had been chased by them ourselves, more than once.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0316.wav",
+ "render_seconds": 51.9782966269413,
+ "audio_seconds": 3.7616326530612243,
+ "rtf": 13.818015053820117,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WE HAD BEEN CHASED BY THEM OURSELVES MORE THAN ONCE",
+ "raw": {
+ "normalized_reference": "we had been chased by them ourselves, more than once.",
+ "normalized_hypothesis": "we had been chased by them ourselves more than once",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "we had been chased by them ourselves more than once",
+ "normalized_hypothesis": "we had been chased by them ourselves more than once",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0317",
+ "category": "phonetic_balance",
+ "text": "Or have they already devised one.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0317.wav",
+ "render_seconds": 24.19049555505626,
+ "audio_seconds": 2.5774149659863945,
+ "rtf": 9.38556494561146,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "OR HAVE THEY ALREADY DEVISED ONE",
+ "raw": {
+ "normalized_reference": "or have they already devised one.",
+ "normalized_hypothesis": "or have they already devised one",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "or have they already devised one",
+ "normalized_hypothesis": "or have they already devised one",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0318",
+ "category": "phonetic_balance",
+ "text": "I cannot follow you, she said.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0318.wav",
+ "render_seconds": 48.49822791304905,
+ "audio_seconds": 2.2987755102040817,
+ "rtf": 21.097418037459192,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I CANNOT FOLLOW YOUR SUSOD",
+ "raw": {
+ "normalized_reference": "i cannot follow you, she said.",
+ "normalized_hypothesis": "i cannot follow your susod",
+ "errors": 3,
+ "words": 6,
+ "wer": 0.5
+ },
+ "semantic": {
+ "normalized_reference": "i cannot follow you she said",
+ "normalized_hypothesis": "i cannot follow your susod",
+ "errors": 3,
+ "words": 6,
+ "wer": 0.5
+ }
+ },
+ {
+ "id": "release500_0319",
+ "category": "phonetic_balance",
+ "text": "He went down in midstream, searching the shadows of both shores.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0319.wav",
+ "render_seconds": 51.67209360306151,
+ "audio_seconds": 4.388571428571429,
+ "rtf": 11.77424007882261,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE WENT DOWN IN MIDSTREAM SEARCHING THE SHADOWS OF BOTH SHORES",
+ "raw": {
+ "normalized_reference": "he went down in midstream, searching the shadows of both shores.",
+ "normalized_hypothesis": "he went down in midstream searching the shadows of both shores",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he went down in midstream searching the shadows of both shores",
+ "normalized_hypothesis": "he went down in midstream searching the shadows of both shores",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0320",
+ "category": "phonetic_balance",
+ "text": "I was sick once -- typhoid.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0320.wav",
+ "render_seconds": 30.190354880993254,
+ "audio_seconds": 2.2291156462585033,
+ "rtf": 13.543646751422145,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I WAS THICK ON TYPHOID",
+ "raw": {
+ "normalized_reference": "i was sick once -- typhoid.",
+ "normalized_hypothesis": "i was thick on typhoid",
+ "errors": 2,
+ "words": 5,
+ "wer": 0.4
+ },
+ "semantic": {
+ "normalized_reference": "i was sick once typhoid",
+ "normalized_hypothesis": "i was thick on typhoid",
+ "errors": 2,
+ "words": 5,
+ "wer": 0.4
+ }
+ },
+ {
+ "id": "release500_0321",
+ "category": "phonetic_balance",
+ "text": "Why not like any railroad station or ferry depot.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0321.wav",
+ "render_seconds": 59.707711667055264,
+ "audio_seconds": 3.622312925170068,
+ "rtf": 16.483311326354272,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WHY NOT LIKE ANY RAILROAD STATION OR FERRY DEPOT",
+ "raw": {
+ "normalized_reference": "why not like any railroad station or ferry depot.",
+ "normalized_hypothesis": "why not like any railroad station or ferry depot",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "why not like any railroad station or ferry depot",
+ "normalized_hypothesis": "why not like any railroad station or ferry depot",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0322",
+ "category": "phonetic_balance",
+ "text": "Red-Eye swung back and forth on the branch farther down.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0322.wav",
+ "render_seconds": 62.30721454601735,
+ "audio_seconds": 3.900952380952381,
+ "rtf": 15.972308416337455,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "RED EYE SWUNG BACK AND FORTH ON THE BRANCH FARTHER DOWN",
+ "raw": {
+ "normalized_reference": "red-eye swung back and forth on the branch farther down.",
+ "normalized_hypothesis": "red eye swung back and forth on the branch farther down",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "red eye swung back and forth on the branch farther down",
+ "normalized_hypothesis": "red eye swung back and forth on the branch farther down",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0323",
+ "category": "phonetic_balance",
+ "text": "Within himself he called it no longer his own.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0323.wav",
+ "render_seconds": 53.198633711086586,
+ "audio_seconds": 3.0650340136054424,
+ "rtf": 17.35662099505,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WITHIN HIMSELF HE CALLED IT NO LONGER HIS OWN",
+ "raw": {
+ "normalized_reference": "within himself he called it no longer his own.",
+ "normalized_hypothesis": "within himself he called it no longer his own",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "within himself he called it no longer his own",
+ "normalized_hypothesis": "within himself he called it no longer his own",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0324",
+ "category": "phonetic_balance",
+ "text": "It lasted as a deterrent for two days.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0324.wav",
+ "render_seconds": 40.49952131602913,
+ "audio_seconds": 2.8560544217687074,
+ "rtf": 14.180234454688174,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "TE LAST OF THE HADS ETURRIN FA CROTOLUS",
+ "raw": {
+ "normalized_reference": "it lasted as a deterrent for two days.",
+ "normalized_hypothesis": "te last of the hads eturrin fa crotolus",
+ "errors": 8,
+ "words": 8,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "it lasted as a deterrent for 2 days",
+ "normalized_hypothesis": "te last of the hads eturrin fa crotolus",
+ "errors": 8,
+ "words": 8,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0325",
+ "category": "phonetic_balance",
+ "text": "He loved to play Chinese lottery.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0325.wav",
+ "render_seconds": 35.21083924500272,
+ "audio_seconds": 2.4380952380952383,
+ "rtf": 14.441945784083144,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE LOVED TO PRY CHINESE WOTTERY",
+ "raw": {
+ "normalized_reference": "he loved to play chinese lottery.",
+ "normalized_hypothesis": "he loved to pry chinese wottery",
+ "errors": 2,
+ "words": 6,
+ "wer": 0.3333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "he loved to play chinese lottery",
+ "normalized_hypothesis": "he loved to pry chinese wottery",
+ "errors": 2,
+ "words": 6,
+ "wer": 0.3333333333333333
+ }
+ },
+ {
+ "id": "release500_0326",
+ "category": "phonetic_balance",
+ "text": "Lord, but I'm glad to see you again, Phil.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0326.wav",
+ "render_seconds": 51.59785965597257,
+ "audio_seconds": 2.9257142857142857,
+ "rtf": 17.6359871871,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "LORD BUT I'M GLAD TO SEE YOU AGAIN PHIL",
+ "raw": {
+ "normalized_reference": "lord, but i'm glad to see you again, phil.",
+ "normalized_hypothesis": "lord but i'm glad to see you again phil",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "lord but i am glad to see you again phil",
+ "normalized_hypothesis": "lord but i am glad to see you again phil",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0327",
+ "category": "phonetic_balance",
+ "text": "There is not an iota of truth in it, certainly not.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0327.wav",
+ "render_seconds": 66.69544376607519,
+ "audio_seconds": 3.6919727891156464,
+ "rtf": 18.064987900967445,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THERE IS NOT AN ILAND OF TRUTH IN IT CERTAINLY NOT",
+ "raw": {
+ "normalized_reference": "there is not an iota of truth in it, certainly not.",
+ "normalized_hypothesis": "there is not an iland of truth in it certainly not",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ },
+ "semantic": {
+ "normalized_reference": "there is not an iota of truth in it certainly not",
+ "normalized_hypothesis": "there is not an iland of truth in it certainly not",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ }
+ },
+ {
+ "id": "release500_0328",
+ "category": "phonetic_balance",
+ "text": "She added, with genuine sympathy in her face and voice.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0328.wav",
+ "render_seconds": 69.40363747405354,
+ "audio_seconds": 3.9706122448979593,
+ "rtf": 17.47932892798429,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SHE ADDED WITH GENUINE SYMPATHY IN HER FACE AND VOICE",
+ "raw": {
+ "normalized_reference": "she added, with genuine sympathy in her face and voice.",
+ "normalized_hypothesis": "she added with genuine sympathy in her face and voice",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "she added with genuine sympathy in her face and voice",
+ "normalized_hypothesis": "she added with genuine sympathy in her face and voice",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0329",
+ "category": "phonetic_balance",
+ "text": "We have plenty of capital ourselves, and yet we want more.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0329.wav",
+ "render_seconds": 61.16601952398196,
+ "audio_seconds": 4.179591836734694,
+ "rtf": 14.634448030640213,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WE HAVE PLENTY OF CAPITAL OURSELVES AND YET WE WANT MORE",
+ "raw": {
+ "normalized_reference": "we have plenty of capital ourselves, and yet we want more.",
+ "normalized_hypothesis": "we have plenty of capital ourselves and yet we want more",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "we have plenty of capital ourselves and yet we want more",
+ "normalized_hypothesis": "we have plenty of capital ourselves and yet we want more",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0330",
+ "category": "phonetic_balance",
+ "text": "And now put yourself in my place for a moment.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0330.wav",
+ "render_seconds": 57.69193570991047,
+ "audio_seconds": 3.274013605442177,
+ "rtf": 17.621165536396358,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "AND NOW PUT YOURSELF IN MY PLACE FOR A MOMENT",
+ "raw": {
+ "normalized_reference": "and now put yourself in my place for a moment.",
+ "normalized_hypothesis": "and now put yourself in my place for a moment",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "and now put yourself in my place for a moment",
+ "normalized_hypothesis": "and now put yourself in my place for a moment",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0331",
+ "category": "phonetic_balance",
+ "text": "Fast, but endure.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0331.wav",
+ "render_seconds": 18.198805535095744,
+ "audio_seconds": 1.6718367346938776,
+ "rtf": 10.88551600609975,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "FAR BUT NO DRAN",
+ "raw": {
+ "normalized_reference": "fast, but endure.",
+ "normalized_hypothesis": "far but no dran",
+ "errors": 3,
+ "words": 3,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "fast but endure",
+ "normalized_hypothesis": "far but no dran",
+ "errors": 3,
+ "words": 3,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0332",
+ "category": "phonetic_balance",
+ "text": "You were looking squeamish this afternoon, he began.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0332.wav",
+ "render_seconds": 53.79991964297369,
+ "audio_seconds": 3.4829931972789114,
+ "rtf": 15.4464613037444,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "YOU ARE LOOKING SQUEAMISH THIS AFTERNOON HE BEGAN",
+ "raw": {
+ "normalized_reference": "you were looking squeamish this afternoon, he began.",
+ "normalized_hypothesis": "you are looking squeamish this afternoon he began",
+ "errors": 1,
+ "words": 8,
+ "wer": 0.125
+ },
+ "semantic": {
+ "normalized_reference": "you were looking squeamish this afternoon he began",
+ "normalized_hypothesis": "you are looking squeamish this afternoon he began",
+ "errors": 1,
+ "words": 8,
+ "wer": 0.125
+ }
+ },
+ {
+ "id": "release500_0333",
+ "category": "phonetic_balance",
+ "text": "It is the nearest refuge.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0333.wav",
+ "render_seconds": 44.69639569998253,
+ "audio_seconds": 2.020136054421769,
+ "rtf": 22.125438334783915,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE MUT GOM OTER THROMH NESSERCH",
+ "raw": {
+ "normalized_reference": "it is the nearest refuge.",
+ "normalized_hypothesis": "the mut gom oter thromh nesserch",
+ "errors": 6,
+ "words": 5,
+ "wer": 1.2
+ },
+ "semantic": {
+ "normalized_reference": "it is the nearest refuge",
+ "normalized_hypothesis": "the mut gom oter thromh nesserch",
+ "errors": 6,
+ "words": 5,
+ "wer": 1.2
+ }
+ },
+ {
+ "id": "release500_0334",
+ "category": "phonetic_balance",
+ "text": "I want to die in it.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0334.wav",
+ "render_seconds": 36.895730805932544,
+ "audio_seconds": 1.7414965986394557,
+ "rtf": 21.186220423719078,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I WANT TO DIE N",
+ "raw": {
+ "normalized_reference": "i want to die in it.",
+ "normalized_hypothesis": "i want to die n",
+ "errors": 2,
+ "words": 6,
+ "wer": 0.3333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "i want to die in it",
+ "normalized_hypothesis": "i want to die n",
+ "errors": 2,
+ "words": 6,
+ "wer": 0.3333333333333333
+ }
+ },
+ {
+ "id": "release500_0335",
+ "category": "phonetic_balance",
+ "text": "He stopped, and Philip nodded at the horrified question in his eyes.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0335.wav",
+ "render_seconds": 64.33425731700845,
+ "audio_seconds": 4.73687074829932,
+ "rtf": 13.581594418658435,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE STOPPED AND PHILIP NODDED AT THE HORRIFIED QUESTION IN HIS EYES",
+ "raw": {
+ "normalized_reference": "he stopped, and philip nodded at the horrified question in his eyes.",
+ "normalized_hypothesis": "he stopped and philip nodded at the horrified question in his eyes",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he stopped and philip nodded at the horrified question in his eyes",
+ "normalized_hypothesis": "he stopped and philip nodded at the horrified question in his eyes",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0336",
+ "category": "phonetic_balance",
+ "text": "She turned, fearing that Jacques might see what was in her face.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0336.wav",
+ "render_seconds": 70.60388363199309,
+ "audio_seconds": 4.249251700680272,
+ "rtf": 16.61560401815923,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SHE TURNED FEARING THAT JACKS MIGHT SEE WHAT WAS IN HER FACE",
+ "raw": {
+ "normalized_reference": "she turned, fearing that jacques might see what was in her face.",
+ "normalized_hypothesis": "she turned fearing that jacks might see what was in her face",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "she turned fearing that jacques might see what was in her face",
+ "normalized_hypothesis": "she turned fearing that jacks might see what was in her face",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ }
+ },
+ {
+ "id": "release500_0337",
+ "category": "phonetic_balance",
+ "text": "It was a curious coincidence.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0337.wav",
+ "render_seconds": 26.40043653792236,
+ "audio_seconds": 2.4380952380952383,
+ "rtf": 10.828304048757216,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT WAS A CURIOUS COINCIDENCE",
+ "raw": {
+ "normalized_reference": "it was a curious coincidence.",
+ "normalized_hypothesis": "it was a curious coincidence",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it was a curious coincidence",
+ "normalized_hypothesis": "it was a curious coincidence",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0338",
+ "category": "phonetic_balance",
+ "text": "Daylight made no answer, and the door closed behind him.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0338.wav",
+ "render_seconds": 68.90685510099865,
+ "audio_seconds": 3.9706122448979593,
+ "rtf": 17.354214123915163,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "DAYLIGHT MADE NO ANSWER AND THE DOOR CLOSED BEHIND HIM",
+ "raw": {
+ "normalized_reference": "daylight made no answer, and the door closed behind him.",
+ "normalized_hypothesis": "daylight made no answer and the door closed behind him",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "daylight made no answer and the door closed behind him",
+ "normalized_hypothesis": "daylight made no answer and the door closed behind him",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0339",
+ "category": "phonetic_balance",
+ "text": "And he did hurt my arm.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0339.wav",
+ "render_seconds": 28.416230701026507,
+ "audio_seconds": 1.9504761904761905,
+ "rtf": 14.568868279334879,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I DID NOT HURT MY ARM",
+ "raw": {
+ "normalized_reference": "and he did hurt my arm.",
+ "normalized_hypothesis": "i did not hurt my arm",
+ "errors": 3,
+ "words": 6,
+ "wer": 0.5
+ },
+ "semantic": {
+ "normalized_reference": "and he did hurt my arm",
+ "normalized_hypothesis": "i did not hurt my arm",
+ "errors": 3,
+ "words": 6,
+ "wer": 0.5
+ }
+ },
+ {
+ "id": "release500_0340",
+ "category": "phonetic_balance",
+ "text": "Dunham, can your boy go along with Jesse.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0340.wav",
+ "render_seconds": 50.09988435602281,
+ "audio_seconds": 2.9257142857142857,
+ "rtf": 17.123983910749985,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "DUNHAM CAN YOUR BOY GO ALONG WITH JESSE",
+ "raw": {
+ "normalized_reference": "dunham, can your boy go along with jesse.",
+ "normalized_hypothesis": "dunham can your boy go along with jesse",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "dunham can your boy go along with jesse",
+ "normalized_hypothesis": "dunham can your boy go along with jesse",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0341",
+ "category": "phonetic_balance",
+ "text": "The creative joy, I murmured.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0341.wav",
+ "render_seconds": 48.50075435405597,
+ "audio_seconds": 2.36843537414966,
+ "rtf": 20.47797245532579,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT'S A SORT OF DRALY I KNON YER",
+ "raw": {
+ "normalized_reference": "the creative joy, i murmured.",
+ "normalized_hypothesis": "it's a sort of draly i knon yer",
+ "errors": 7,
+ "words": 5,
+ "wer": 1.4
+ },
+ "semantic": {
+ "normalized_reference": "the creative joy i murmured",
+ "normalized_hypothesis": "it is a sort of draly i knon yer",
+ "errors": 8,
+ "words": 5,
+ "wer": 1.6
+ }
+ },
+ {
+ "id": "release500_0342",
+ "category": "phonetic_balance",
+ "text": "He pulled, and the log crashed down to break his back.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0342.wav",
+ "render_seconds": 41.507601993042044,
+ "audio_seconds": 3.8312925170068026,
+ "rtf": 10.833837878155505,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE PULLED AND THE LOG CRASHED DOWN TO BREAK HIS BACK",
+ "raw": {
+ "normalized_reference": "he pulled, and the log crashed down to break his back.",
+ "normalized_hypothesis": "he pulled and the log crashed down to break his back",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he pulled and the log crashed down to break his back",
+ "normalized_hypothesis": "he pulled and the log crashed down to break his back",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0343",
+ "category": "phonetic_balance",
+ "text": "By golly, the boy wins.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0343.wav",
+ "render_seconds": 18.99812872102484,
+ "audio_seconds": 2.020136054421769,
+ "rtf": 9.404380798729296,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "BY GALIN THE BOY WIM",
+ "raw": {
+ "normalized_reference": "by golly, the boy wins.",
+ "normalized_hypothesis": "by galin the boy wim",
+ "errors": 2,
+ "words": 5,
+ "wer": 0.4
+ },
+ "semantic": {
+ "normalized_reference": "by golly the boy wins",
+ "normalized_hypothesis": "by galin the boy wim",
+ "errors": 2,
+ "words": 5,
+ "wer": 0.4
+ }
+ },
+ {
+ "id": "release500_0344",
+ "category": "phonetic_balance",
+ "text": "The boy threw back his head with pride.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0344.wav",
+ "render_seconds": 44.50350289302878,
+ "audio_seconds": 2.716734693877551,
+ "rtf": 16.381247308882287,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE BOY THREW BACK HIS HEAD WITH PRIDE",
+ "raw": {
+ "normalized_reference": "the boy threw back his head with pride.",
+ "normalized_hypothesis": "the boy threw back his head with pride",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the boy threw back his head with pride",
+ "normalized_hypothesis": "the boy threw back his head with pride",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0345",
+ "category": "phonetic_balance",
+ "text": "The last one I knew was an overseer.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0345.wav",
+ "render_seconds": 52.690776895033196,
+ "audio_seconds": 2.647074829931973,
+ "rtf": 19.905284240259764,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE LAST ONE I KNEW WAS AN OVERSENTER",
+ "raw": {
+ "normalized_reference": "the last one i knew was an overseer.",
+ "normalized_hypothesis": "the last one i knew was an oversenter",
+ "errors": 1,
+ "words": 8,
+ "wer": 0.125
+ },
+ "semantic": {
+ "normalized_reference": "the last one i knew was an overseer",
+ "normalized_hypothesis": "the last one i knew was an oversenter",
+ "errors": 1,
+ "words": 8,
+ "wer": 0.125
+ }
+ },
+ {
+ "id": "release500_0346",
+ "category": "phonetic_balance",
+ "text": "Two years ago I gave up civilization for this.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0346.wav",
+ "render_seconds": 57.59371170005761,
+ "audio_seconds": 3.4133333333333336,
+ "rtf": 16.873157724626253,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "TWO YEARS AGO I GAVE UP CIVILIZATION FOR THIS",
+ "raw": {
+ "normalized_reference": "two years ago i gave up civilization for this.",
+ "normalized_hypothesis": "two years ago i gave up civilization for this",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "2 years ago i gave up civilization for this",
+ "normalized_hypothesis": "2 years ago i gave up civilization for this",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0347",
+ "category": "phonetic_balance",
+ "text": "The river bared its bosom, and snorting steamboats challenged the wilderness.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0347.wav",
+ "render_seconds": 64.40615069901105,
+ "audio_seconds": 5.50312925170068,
+ "rtf": 11.703550426170175,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE RIVER BARED ITS BOSOM AND SNORTING STEAMBOATS CHALLENGED THE WILDERNESS",
+ "raw": {
+ "normalized_reference": "the river bared its bosom, and snorting steamboats challenged the wilderness.",
+ "normalized_hypothesis": "the river bared its bosom and snorting steamboats challenged the wilderness",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the river bared its bosom and snorting steamboats challenged the wilderness",
+ "normalized_hypothesis": "the river bared its bosom and snorting steamboats challenged the wilderness",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0348",
+ "category": "phonetic_balance",
+ "text": "Mab, she said.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0348.wav",
+ "render_seconds": 37.49199713708367,
+ "audio_seconds": 1.6021768707482993,
+ "rtf": 23.400660577238874,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "PERABSOCOD",
+ "raw": {
+ "normalized_reference": "mab, she said.",
+ "normalized_hypothesis": "perabsocod",
+ "errors": 3,
+ "words": 3,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "mab she said",
+ "normalized_hypothesis": "perabsocod",
+ "errors": 3,
+ "words": 3,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0349",
+ "category": "phonetic_balance",
+ "text": "The regret in his voice was provocative of a second burst of laughter.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0349.wav",
+ "render_seconds": 66.40094997501001,
+ "audio_seconds": 4.876190476190477,
+ "rtf": 13.617382319093847,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE REGRET IN HIS VOICE WAS PROVOCATIVE OF A SECOND BURST OF LAUGHTER",
+ "raw": {
+ "normalized_reference": "the regret in his voice was provocative of a second burst of laughter.",
+ "normalized_hypothesis": "the regret in his voice was provocative of a second burst of laughter",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the regret in his voice was provocative of a 2nd burst of laughter",
+ "normalized_hypothesis": "the regret in his voice was provocative of a 2nd burst of laughter",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0350",
+ "category": "phonetic_balance",
+ "text": "And each year something happened, and I did not go.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0350.wav",
+ "render_seconds": 54.688802125980146,
+ "audio_seconds": 3.622312925170068,
+ "rtf": 15.09775749797003,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THEN EACH HEAR FENTINGS HAPPENED AND I DID NOT GO",
+ "raw": {
+ "normalized_reference": "and each year something happened, and i did not go.",
+ "normalized_hypothesis": "then each hear fentings happened and i did not go",
+ "errors": 3,
+ "words": 10,
+ "wer": 0.3
+ },
+ "semantic": {
+ "normalized_reference": "and each year something happened and i did not go",
+ "normalized_hypothesis": "then each hear fentings happened and i did not go",
+ "errors": 3,
+ "words": 10,
+ "wer": 0.3
+ }
+ },
+ {
+ "id": "release500_0351",
+ "category": "phonetic_balance",
+ "text": "Gad, your letter came just in time.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0351.wav",
+ "render_seconds": 42.79335663700476,
+ "audio_seconds": 2.647074829931973,
+ "rtf": 16.16628141868755,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "DAD YOUR LETTER KIM JUSPIN TO HIM",
+ "raw": {
+ "normalized_reference": "gad, your letter came just in time.",
+ "normalized_hypothesis": "dad your letter kim juspin to him",
+ "errors": 5,
+ "words": 7,
+ "wer": 0.7142857142857143
+ },
+ "semantic": {
+ "normalized_reference": "gad your letter came just in time",
+ "normalized_hypothesis": "dad your letter kim juspin to him",
+ "errors": 5,
+ "words": 7,
+ "wer": 0.7142857142857143
+ }
+ },
+ {
+ "id": "release500_0352",
+ "category": "phonetic_balance",
+ "text": "But if Pierre did not return, until tomorrow.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0352.wav",
+ "render_seconds": 48.40563212300185,
+ "audio_seconds": 3.2043537414965986,
+ "rtf": 15.106207375342375,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "BUT IF PIERRE DID NOT RETURN UNTIL TENRY",
+ "raw": {
+ "normalized_reference": "but if pierre did not return, until tomorrow.",
+ "normalized_hypothesis": "but if pierre did not return until tenry",
+ "errors": 1,
+ "words": 8,
+ "wer": 0.125
+ },
+ "semantic": {
+ "normalized_reference": "but if pierre did not return until tomorrow",
+ "normalized_hypothesis": "but if pierre did not return until tenry",
+ "errors": 1,
+ "words": 8,
+ "wer": 0.125
+ }
+ },
+ {
+ "id": "release500_0353",
+ "category": "phonetic_balance",
+ "text": "Billinger may arrive in time.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0353.wav",
+ "render_seconds": 45.29558758705389,
+ "audio_seconds": 2.2987755102040817,
+ "rtf": 19.70422400360121,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "BILLANGER MAY ARRIVE IN TIME",
+ "raw": {
+ "normalized_reference": "billinger may arrive in time.",
+ "normalized_hypothesis": "billanger may arrive in time",
+ "errors": 1,
+ "words": 5,
+ "wer": 0.2
+ },
+ "semantic": {
+ "normalized_reference": "billinger may arrive in time",
+ "normalized_hypothesis": "billanger may arrive in time",
+ "errors": 1,
+ "words": 5,
+ "wer": 0.2
+ }
+ },
+ {
+ "id": "release500_0354",
+ "category": "phonetic_balance",
+ "text": "In such a tumbling of values was no time to sell.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0354.wav",
+ "render_seconds": 67.16993414401077,
+ "audio_seconds": 3.4829931972789114,
+ "rtf": 19.28511781087809,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IN SUCH A TUMBLING OF VALUES WAS NO TIME TO SELL",
+ "raw": {
+ "normalized_reference": "in such a tumbling of values was no time to sell.",
+ "normalized_hypothesis": "in such a tumbling of values was no time to sell",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "in such a tumbling of values was no time to sell",
+ "normalized_hypothesis": "in such a tumbling of values was no time to sell",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0355",
+ "category": "phonetic_balance",
+ "text": "There's not an iota of truth in it.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0355.wav",
+ "render_seconds": 37.18858742003795,
+ "audio_seconds": 2.4380952380952383,
+ "rtf": 15.253131558999938,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IM NOT AN IODA TRIFLE",
+ "raw": {
+ "normalized_reference": "there's not an iota of truth in it.",
+ "normalized_hypothesis": "im not an ioda trifle",
+ "errors": 6,
+ "words": 8,
+ "wer": 0.75
+ },
+ "semantic": {
+ "normalized_reference": "there is not an iota of truth in it",
+ "normalized_hypothesis": "im not an ioda trifle",
+ "errors": 7,
+ "words": 9,
+ "wer": 0.7777777777777778
+ }
+ },
+ {
+ "id": "release500_0356",
+ "category": "phonetic_balance",
+ "text": "It is dog eat dog, and you ate them up.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0356.wav",
+ "render_seconds": 48.39995520899538,
+ "audio_seconds": 2.8560544217687074,
+ "rtf": 16.9464401098569,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IF HIS DOG EAT DOG AND YOU RATE THEM UP",
+ "raw": {
+ "normalized_reference": "it is dog eat dog, and you ate them up.",
+ "normalized_hypothesis": "if his dog eat dog and you rate them up",
+ "errors": 3,
+ "words": 10,
+ "wer": 0.3
+ },
+ "semantic": {
+ "normalized_reference": "it is dog eat dog and you ate them up",
+ "normalized_hypothesis": "if his dog eat dog and you rate them up",
+ "errors": 3,
+ "words": 10,
+ "wer": 0.3
+ }
+ },
+ {
+ "id": "release500_0357",
+ "category": "phonetic_balance",
+ "text": "You can take a vacation on pay.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0357.wav",
+ "render_seconds": 47.408293746062554,
+ "audio_seconds": 2.4380952380952383,
+ "rtf": 19.44480798178347,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I GOT RIGHT A VACROTION AND PERL",
+ "raw": {
+ "normalized_reference": "you can take a vacation on pay.",
+ "normalized_hypothesis": "i got right a vacrotion and perl",
+ "errors": 6,
+ "words": 7,
+ "wer": 0.8571428571428571
+ },
+ "semantic": {
+ "normalized_reference": "you can take a vacation on pay",
+ "normalized_hypothesis": "i got right a vacrotion and perl",
+ "errors": 6,
+ "words": 7,
+ "wer": 0.8571428571428571
+ }
+ },
+ {
+ "id": "release500_0358",
+ "category": "phonetic_balance",
+ "text": "I have seen myself that one man contemplated by Pascal's philosophic eye.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0358.wav",
+ "render_seconds": 81.69244368898217,
+ "audio_seconds": 4.876190476190477,
+ "rtf": 16.753333178404546,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I HAVE SEEN MYSELF THAT ONE MAN CONTEMPLATED BY PASCAL'S PHILOSOPHIC EYE",
+ "raw": {
+ "normalized_reference": "i have seen myself that one man contemplated by pascal's philosophic eye.",
+ "normalized_hypothesis": "i have seen myself that one man contemplated by pascal's philosophic eye",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i have seen myself that one man contemplated by pascal is philosophic eye",
+ "normalized_hypothesis": "i have seen myself that one man contemplated by pascal is philosophic eye",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0359",
+ "category": "phonetic_balance",
+ "text": "Ah, I had forgotten, he exclaimed.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0359.wav",
+ "render_seconds": 44.70367567194626,
+ "audio_seconds": 2.5774149659863945,
+ "rtf": 17.34438430050702,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I'VEN'T FORGOTTEN HEXCAIMED",
+ "raw": {
+ "normalized_reference": "ah, i had forgotten, he exclaimed.",
+ "normalized_hypothesis": "i'ven't forgotten hexcaimed",
+ "errors": 5,
+ "words": 6,
+ "wer": 0.8333333333333334
+ },
+ "semantic": {
+ "normalized_reference": "ah i had forgotten he exclaimed",
+ "normalized_hypothesis": "i have not forgotten hexcaimed",
+ "errors": 5,
+ "words": 6,
+ "wer": 0.8333333333333334
+ }
+ },
+ {
+ "id": "release500_0360",
+ "category": "phonetic_balance",
+ "text": "Providence had delivered him through the maelstrom.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0360.wav",
+ "render_seconds": 47.60065295291133,
+ "audio_seconds": 3.4133333333333336,
+ "rtf": 13.94550379479824,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "PROVIDENCE HAD DELIVERED HIM ATHROUGH THE MAILSTRUM",
+ "raw": {
+ "normalized_reference": "providence had delivered him through the maelstrom.",
+ "normalized_hypothesis": "providence had delivered him athrough the mailstrum",
+ "errors": 2,
+ "words": 7,
+ "wer": 0.2857142857142857
+ },
+ "semantic": {
+ "normalized_reference": "providence had delivered him through the maelstrom",
+ "normalized_hypothesis": "providence had delivered him athrough the mailstrum",
+ "errors": 2,
+ "words": 7,
+ "wer": 0.2857142857142857
+ }
+ },
+ {
+ "id": "release500_0361",
+ "category": "phonetic_balance",
+ "text": "She was his now, forever.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0361.wav",
+ "render_seconds": 41.50763119303156,
+ "audio_seconds": 2.089795918367347,
+ "rtf": 19.862050082602988,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HUMAN HOS NEW A FOR OV HIM",
+ "raw": {
+ "normalized_reference": "she was his now, forever.",
+ "normalized_hypothesis": "human hos new a for ov him",
+ "errors": 7,
+ "words": 5,
+ "wer": 1.4
+ },
+ "semantic": {
+ "normalized_reference": "she was his now forever",
+ "normalized_hypothesis": "human hos new a for ov him",
+ "errors": 7,
+ "words": 5,
+ "wer": 1.4
+ }
+ },
+ {
+ "id": "release500_0362",
+ "category": "phonetic_balance",
+ "text": "It was unobtrusive, yet it was there.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0362.wav",
+ "render_seconds": 52.499631601036526,
+ "audio_seconds": 2.716734693877551,
+ "rtf": 19.324533867569034,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT WAS UNOBTRUSIVE YET IT WAS THERE",
+ "raw": {
+ "normalized_reference": "it was unobtrusive, yet it was there.",
+ "normalized_hypothesis": "it was unobtrusive yet it was there",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it was unobtrusive yet it was there",
+ "normalized_hypothesis": "it was unobtrusive yet it was there",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0363",
+ "category": "phonetic_balance",
+ "text": "Men like Joe Goose dated existence from drunk to drunk.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0363.wav",
+ "render_seconds": 58.39979052299168,
+ "audio_seconds": 3.9706122448979593,
+ "rtf": 14.708006453672864,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "MEN LIKE JOE GOOSE DATED EXISTENCE FROM DRUNK TO DRINK",
+ "raw": {
+ "normalized_reference": "men like joe goose dated existence from drunk to drunk.",
+ "normalized_hypothesis": "men like joe goose dated existence from drunk to drink",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ },
+ "semantic": {
+ "normalized_reference": "men like joe goose dated existence from drunk to drunk",
+ "normalized_hypothesis": "men like joe goose dated existence from drunk to drink",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ }
+ },
+ {
+ "id": "release500_0364",
+ "category": "phonetic_balance",
+ "text": "I was completely lost in my work.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0364.wav",
+ "render_seconds": 47.81108432204928,
+ "audio_seconds": 2.5774149659863945,
+ "rtf": 18.55001424023766,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I WAS COMPLETELY LOST IN MY WORK",
+ "raw": {
+ "normalized_reference": "i was completely lost in my work.",
+ "normalized_hypothesis": "i was completely lost in my work",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i was completely lost in my work",
+ "normalized_hypothesis": "i was completely lost in my work",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0365",
+ "category": "phonetic_balance",
+ "text": "The very thing, Ernest agreed.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0365.wav",
+ "render_seconds": 51.60790924797766,
+ "audio_seconds": 2.2987755102040817,
+ "rtf": 22.450173589763008,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE VERY THING ERENEUT TO GRANT",
+ "raw": {
+ "normalized_reference": "the very thing, ernest agreed.",
+ "normalized_hypothesis": "the very thing ereneut to grant",
+ "errors": 3,
+ "words": 5,
+ "wer": 0.6
+ },
+ "semantic": {
+ "normalized_reference": "the very thing ernest agreed",
+ "normalized_hypothesis": "the very thing ereneut to grant",
+ "errors": 3,
+ "words": 5,
+ "wer": 0.6
+ }
+ },
+ {
+ "id": "release500_0366",
+ "category": "phonetic_balance",
+ "text": "He was the leader, and Tudor was his lieutenant.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0366.wav",
+ "render_seconds": 58.503436186001636,
+ "audio_seconds": 3.622312925170068,
+ "rtf": 16.150850960303185,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE WAS THE LEADER AND TUTOR WAS HIS LIEUTENANT",
+ "raw": {
+ "normalized_reference": "he was the leader, and tudor was his lieutenant.",
+ "normalized_hypothesis": "he was the leader and tutor was his lieutenant",
+ "errors": 1,
+ "words": 9,
+ "wer": 0.1111111111111111
+ },
+ "semantic": {
+ "normalized_reference": "he was the leader and tudor was his lieutenant",
+ "normalized_hypothesis": "he was the leader and tutor was his lieutenant",
+ "errors": 1,
+ "words": 9,
+ "wer": 0.1111111111111111
+ }
+ },
+ {
+ "id": "release500_0367",
+ "category": "phonetic_balance",
+ "text": "The Fire-Men wore animal skins around their waists and across their shoulders.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0367.wav",
+ "render_seconds": 59.00359232502524,
+ "audio_seconds": 5.712108843537415,
+ "rtf": 10.329563728776094,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE FIREMEN WORE ANIMAL SKINS AROUND THEIR WAISTS AND ACROSS THEIR SHOULDERS",
+ "raw": {
+ "normalized_reference": "the fire-men wore animal skins around their waists and across their shoulders.",
+ "normalized_hypothesis": "the firemen wore animal skins around their waists and across their shoulders",
+ "errors": 2,
+ "words": 13,
+ "wer": 0.15384615384615385
+ },
+ "semantic": {
+ "normalized_reference": "the fire men wore animal skins around their waists and across their shoulders",
+ "normalized_hypothesis": "the firemen wore animal skins around their waists and across their shoulders",
+ "errors": 2,
+ "words": 13,
+ "wer": 0.15384615384615385
+ }
+ },
+ {
+ "id": "release500_0368",
+ "category": "phonetic_balance",
+ "text": "We could throw stones with our feet.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0368.wav",
+ "render_seconds": 43.21032852400094,
+ "audio_seconds": 2.5774149659863945,
+ "rtf": 16.764987048744032,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WE CUT DLAY STONES WITH OUR SHOT",
+ "raw": {
+ "normalized_reference": "we could throw stones with our feet.",
+ "normalized_hypothesis": "we cut dlay stones with our shot",
+ "errors": 3,
+ "words": 7,
+ "wer": 0.42857142857142855
+ },
+ "semantic": {
+ "normalized_reference": "we could throw stones with our feet",
+ "normalized_hypothesis": "we cut dlay stones with our shot",
+ "errors": 3,
+ "words": 7,
+ "wer": 0.42857142857142855
+ }
+ },
+ {
+ "id": "release500_0369",
+ "category": "phonetic_balance",
+ "text": "He turned sharply, and faced Gregson across the table.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0369.wav",
+ "render_seconds": 42.790321383974515,
+ "audio_seconds": 4.249251700680272,
+ "rtf": 10.070083957870539,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "LE TURNED SHARPLY AND FACED GREXON ACROSS THE TABLE",
+ "raw": {
+ "normalized_reference": "he turned sharply, and faced gregson across the table.",
+ "normalized_hypothesis": "le turned sharply and faced grexon across the table",
+ "errors": 2,
+ "words": 9,
+ "wer": 0.2222222222222222
+ },
+ "semantic": {
+ "normalized_reference": "he turned sharply and faced gregson across the table",
+ "normalized_hypothesis": "le turned sharply and faced grexon across the table",
+ "errors": 2,
+ "words": 9,
+ "wer": 0.2222222222222222
+ }
+ },
+ {
+ "id": "release500_0370",
+ "category": "phonetic_balance",
+ "text": "They are to attack your camp tomorrow night.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0370.wav",
+ "render_seconds": 49.991705862106755,
+ "audio_seconds": 3.0650340136054424,
+ "rtf": 16.31032661960603,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THEY ARE TO ATTACK YOUR CAMP TOMORAY A NIGHT",
+ "raw": {
+ "normalized_reference": "they are to attack your camp tomorrow night.",
+ "normalized_hypothesis": "they are to attack your camp tomoray a night",
+ "errors": 2,
+ "words": 8,
+ "wer": 0.25
+ },
+ "semantic": {
+ "normalized_reference": "they are to attack your camp tomorrow night",
+ "normalized_hypothesis": "they are to attack your camp tomoray a night",
+ "errors": 2,
+ "words": 8,
+ "wer": 0.25
+ }
+ },
+ {
+ "id": "release500_0371",
+ "category": "phonetic_balance",
+ "text": "And watch out for wet feet, was his parting advice.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0371.wav",
+ "render_seconds": 58.3955981049221,
+ "audio_seconds": 3.5526530612244898,
+ "rtf": 16.437180073191538,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "AND WATCH OUT FOR WET FEET WAS HIS PARTING ADVICE",
+ "raw": {
+ "normalized_reference": "and watch out for wet feet, was his parting advice.",
+ "normalized_hypothesis": "and watch out for wet feet was his parting advice",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "and watch out for wet feet was his parting advice",
+ "normalized_hypothesis": "and watch out for wet feet was his parting advice",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0372",
+ "category": "phonetic_balance",
+ "text": "Two weeks passed, and in that time Thorpe left camp three times.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0372.wav",
+ "render_seconds": 71.31071821902879,
+ "audio_seconds": 4.527891156462585,
+ "rtf": 15.749212106666512,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "TWO WEEKS PASSED AND IN THAT TIME THORPE LEFT CAMP THREE TIMES",
+ "raw": {
+ "normalized_reference": "two weeks passed, and in that time thorpe left camp three times.",
+ "normalized_hypothesis": "two weeks passed and in that time thorpe left camp three times",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "2 weeks passed and in that time thorpe left camp 3 times",
+ "normalized_hypothesis": "2 weeks passed and in that time thorpe left camp 3 times",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0373",
+ "category": "phonetic_balance",
+ "text": "But ever his gaze returned to that Crouched Venus on the piano.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0373.wav",
+ "render_seconds": 80.58846646803431,
+ "audio_seconds": 4.388571428571429,
+ "rtf": 18.363257333210942,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "BUT EVER HIS GAZE RETURNED TO THAT CROUCHED VENUS ON THE PIANO",
+ "raw": {
+ "normalized_reference": "but ever his gaze returned to that crouched venus on the piano.",
+ "normalized_hypothesis": "but ever his gaze returned to that crouched venus on the piano",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "but ever his gaze returned to that crouched venus on the piano",
+ "normalized_hypothesis": "but ever his gaze returned to that crouched venus on the piano",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0374",
+ "category": "phonetic_balance",
+ "text": "But she had become an automaton.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0374.wav",
+ "render_seconds": 46.59230939403642,
+ "audio_seconds": 2.36843537414966,
+ "rtf": 19.672189455777094,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "BUT SHOU HAD BECOME AN AUTOMOTUM",
+ "raw": {
+ "normalized_reference": "but she had become an automaton.",
+ "normalized_hypothesis": "but shou had become an automotum",
+ "errors": 2,
+ "words": 6,
+ "wer": 0.3333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "but she had become an automaton",
+ "normalized_hypothesis": "but shou had become an automotum",
+ "errors": 2,
+ "words": 6,
+ "wer": 0.3333333333333333
+ }
+ },
+ {
+ "id": "release500_0375",
+ "category": "phonetic_balance",
+ "text": "I do not blame you for anything; remember that.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0375.wav",
+ "render_seconds": 44.501929983031005,
+ "audio_seconds": 3.1346938775510202,
+ "rtf": 14.196579226357548,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I DON AT BLAME YOU FLENNICROM REMEMBER THAT",
+ "raw": {
+ "normalized_reference": "i do not blame you for anything; remember that.",
+ "normalized_hypothesis": "i don at blame you flennicrom remember that",
+ "errors": 4,
+ "words": 9,
+ "wer": 0.4444444444444444
+ },
+ "semantic": {
+ "normalized_reference": "i do not blame you for anything remember that",
+ "normalized_hypothesis": "i don at blame you flennicrom remember that",
+ "errors": 4,
+ "words": 9,
+ "wer": 0.4444444444444444
+ }
+ },
+ {
+ "id": "release500_0376",
+ "category": "phonetic_balance",
+ "text": "The next thing to watch out for is bed sores.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0376.wav",
+ "render_seconds": 50.298806952894665,
+ "audio_seconds": 3.0650340136054424,
+ "rtf": 16.410521622149137,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "ANEXT THING TO WATCH OUT FOR HIS BED SORES",
+ "raw": {
+ "normalized_reference": "the next thing to watch out for is bed sores.",
+ "normalized_hypothesis": "anext thing to watch out for his bed sores",
+ "errors": 3,
+ "words": 10,
+ "wer": 0.3
+ },
+ "semantic": {
+ "normalized_reference": "the next thing to watch out for is bed sores",
+ "normalized_hypothesis": "anext thing to watch out for his bed sores",
+ "errors": 3,
+ "words": 10,
+ "wer": 0.3
+ }
+ },
+ {
+ "id": "release500_0377",
+ "category": "phonetic_balance",
+ "text": "He will follow us soon.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0377.wav",
+ "render_seconds": 37.39577233802993,
+ "audio_seconds": 1.9504761904761905,
+ "rtf": 19.17263718502511,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE WILL FOLLOW US SOON",
+ "raw": {
+ "normalized_reference": "he will follow us soon.",
+ "normalized_hypothesis": "he will follow us soon",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he will follow us soon",
+ "normalized_hypothesis": "he will follow us soon",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0378",
+ "category": "phonetic_balance",
+ "text": "His beady black eyes saw bargains where other men saw bankruptcy.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0378.wav",
+ "render_seconds": 35.784436937072314,
+ "audio_seconds": 4.527891156462585,
+ "rtf": 7.9031133259459585,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HIS BETTY BLACK EYES SAW BARGAINS WHERE OTHER MEN SAW BANKRUPTCY",
+ "raw": {
+ "normalized_reference": "his beady black eyes saw bargains where other men saw bankruptcy.",
+ "normalized_hypothesis": "his betty black eyes saw bargains where other men saw bankruptcy",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ },
+ "semantic": {
+ "normalized_reference": "his beady black eyes saw bargains where other men saw bankruptcy",
+ "normalized_hypothesis": "his betty black eyes saw bargains where other men saw bankruptcy",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ }
+ },
+ {
+ "id": "release500_0379",
+ "category": "phonetic_balance",
+ "text": "They robbed me a few years later.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0379.wav",
+ "render_seconds": 30.30690105794929,
+ "audio_seconds": 2.36843537414966,
+ "rtf": 12.796169736668617,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "BEY WAG MUN A FULARS NITM",
+ "raw": {
+ "normalized_reference": "they robbed me a few years later.",
+ "normalized_hypothesis": "bey wag mun a fulars nitm",
+ "errors": 6,
+ "words": 7,
+ "wer": 0.8571428571428571
+ },
+ "semantic": {
+ "normalized_reference": "they robbed me a few years later",
+ "normalized_hypothesis": "bey wag mun a fulars nitm",
+ "errors": 6,
+ "words": 7,
+ "wer": 0.8571428571428571
+ }
+ },
+ {
+ "id": "release500_0380",
+ "category": "phonetic_balance",
+ "text": "Gregson held a lighted match until it burnt his fingertips.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0380.wav",
+ "render_seconds": 63.697357670986094,
+ "audio_seconds": 4.249251700680272,
+ "rtf": 14.990252909892027,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "JACKSON HELD A LIGHTED MATCH UNTIL IT BURNT HIS FINGERTIPS",
+ "raw": {
+ "normalized_reference": "gregson held a lighted match until it burnt his fingertips.",
+ "normalized_hypothesis": "jackson held a lighted match until it burnt his fingertips",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ },
+ "semantic": {
+ "normalized_reference": "gregson held a lighted match until it burnt his fingertips",
+ "normalized_hypothesis": "jackson held a lighted match until it burnt his fingertips",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ }
+ },
+ {
+ "id": "release500_0381",
+ "category": "phonetic_balance",
+ "text": "A burst of laughter was his reward.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0381.wav",
+ "render_seconds": 47.196240690071136,
+ "audio_seconds": 2.647074829931973,
+ "rtf": 17.829583114310385,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "A BURST OF LAUGHTER WAS HIS REWARD",
+ "raw": {
+ "normalized_reference": "a burst of laughter was his reward.",
+ "normalized_hypothesis": "a burst of laughter was his reward",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "a burst of laughter was his reward",
+ "normalized_hypothesis": "a burst of laughter was his reward",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0382",
+ "category": "phonetic_balance",
+ "text": "He cried, and swung the club wildly.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0382.wav",
+ "render_seconds": 45.60102565505076,
+ "audio_seconds": 2.716734693877551,
+ "rtf": 16.785233301513575,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE TRIBE AN THRATER A CLUB WILD ONE",
+ "raw": {
+ "normalized_reference": "he cried, and swung the club wildly.",
+ "normalized_hypothesis": "the tribe an thrater a club wild one",
+ "errors": 7,
+ "words": 7,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "he cried and swung the club wildly",
+ "normalized_hypothesis": "the tribe an thrater a club wild one",
+ "errors": 7,
+ "words": 7,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0383",
+ "category": "phonetic_balance",
+ "text": "Sandel would never become a world champion.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0383.wav",
+ "render_seconds": 48.79456333303824,
+ "audio_seconds": 3.2043537414965986,
+ "rtf": 15.227583241246224,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "VANDEL WOULD NEVER BECOME A WORRIED CHAMPION",
+ "raw": {
+ "normalized_reference": "sandel would never become a world champion.",
+ "normalized_hypothesis": "vandel would never become a worried champion",
+ "errors": 2,
+ "words": 7,
+ "wer": 0.2857142857142857
+ },
+ "semantic": {
+ "normalized_reference": "sandel would never become a world champion",
+ "normalized_hypothesis": "vandel would never become a worried champion",
+ "errors": 2,
+ "words": 7,
+ "wer": 0.2857142857142857
+ }
+ },
+ {
+ "id": "release500_0384",
+ "category": "phonetic_balance",
+ "text": "And after the bath a shave would not be bad.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0384.wav",
+ "render_seconds": 41.10922727303114,
+ "audio_seconds": 3.0650340136054424,
+ "rtf": 13.412323351241959,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "AND AFTER THE BATH TA SHAVE WOULD NOT BE BAD",
+ "raw": {
+ "normalized_reference": "and after the bath a shave would not be bad.",
+ "normalized_hypothesis": "and after the bath ta shave would not be bad",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ },
+ "semantic": {
+ "normalized_reference": "and after the bath a shave would not be bad",
+ "normalized_hypothesis": "and after the bath ta shave would not be bad",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ }
+ },
+ {
+ "id": "release500_0385",
+ "category": "phonetic_balance",
+ "text": "They saw each other for the first time in Boston.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0385.wav",
+ "render_seconds": 42.49437020299956,
+ "audio_seconds": 3.3436734693877552,
+ "rtf": 12.70888757291857,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THEY SAW EACH OTHER FOR THE FIRST TIME IN BOSTON",
+ "raw": {
+ "normalized_reference": "they saw each other for the first time in boston.",
+ "normalized_hypothesis": "they saw each other for the first time in boston",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "they saw each other for the 1st time in boston",
+ "normalized_hypothesis": "they saw each other for the 1st time in boston",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0386",
+ "category": "phonetic_balance",
+ "text": "I had been born with no organic, chemical predisposition toward alcohol.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0386.wav",
+ "render_seconds": 70.00925771601032,
+ "audio_seconds": 5.224489795918367,
+ "rtf": 13.400209484705103,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I HAD BEEN BORN WITH NO ORGANIC TYPICAL PREDISPOSITION TOWARD ALCOHOL",
+ "raw": {
+ "normalized_reference": "i had been born with no organic, chemical predisposition toward alcohol.",
+ "normalized_hypothesis": "i had been born with no organic typical predisposition toward alcohol",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ },
+ "semantic": {
+ "normalized_reference": "i had been born with no organic chemical predisposition toward alcohol",
+ "normalized_hypothesis": "i had been born with no organic typical predisposition toward alcohol",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ }
+ },
+ {
+ "id": "release500_0387",
+ "category": "phonetic_balance",
+ "text": "They had no fixed values, to be altered by adjectives and adverbs.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0387.wav",
+ "render_seconds": 72.50669521500822,
+ "audio_seconds": 4.9458503401360545,
+ "rtf": 14.660107004575,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THEY HAD NO FIXED VALUES TO BE ALTERED BY ADJECTIVES AND ADVERBS",
+ "raw": {
+ "normalized_reference": "they had no fixed values, to be altered by adjectives and adverbs.",
+ "normalized_hypothesis": "they had no fixed values to be altered by adjectives and adverbs",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "they had no fixed values to be altered by adjectives and adverbs",
+ "normalized_hypothesis": "they had no fixed values to be altered by adjectives and adverbs",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0388",
+ "category": "phonetic_balance",
+ "text": "Nor was Elam Harnish an exception.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0388.wav",
+ "render_seconds": 50.79827866400592,
+ "audio_seconds": 2.716734693877551,
+ "rtf": 18.69828466448535,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "NOR WAS ELAM HERNESS AN EXCEPTION",
+ "raw": {
+ "normalized_reference": "nor was elam harnish an exception.",
+ "normalized_hypothesis": "nor was elam herness an exception",
+ "errors": 1,
+ "words": 6,
+ "wer": 0.16666666666666666
+ },
+ "semantic": {
+ "normalized_reference": "nor was elam harnish an exception",
+ "normalized_hypothesis": "nor was elam herness an exception",
+ "errors": 1,
+ "words": 6,
+ "wer": 0.16666666666666666
+ }
+ },
+ {
+ "id": "release500_0389",
+ "category": "phonetic_balance",
+ "text": "He was a wise hyena.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0389.wav",
+ "render_seconds": 21.19734215491917,
+ "audio_seconds": 1.8808163265306121,
+ "rtf": 11.270288255111105,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I'VE A WISE WHOANNY",
+ "raw": {
+ "normalized_reference": "he was a wise hyena.",
+ "normalized_hypothesis": "i've a wise whoanny",
+ "errors": 3,
+ "words": 5,
+ "wer": 0.6
+ },
+ "semantic": {
+ "normalized_reference": "he was a wise hyena",
+ "normalized_hypothesis": "i have a wise whoanny",
+ "errors": 3,
+ "words": 5,
+ "wer": 0.6
+ }
+ },
+ {
+ "id": "release500_0390",
+ "category": "phonetic_balance",
+ "text": "And now behold the perversity of things.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0390.wav",
+ "render_seconds": 42.79965169506613,
+ "audio_seconds": 2.8560544217687074,
+ "rtf": 14.985586888278204,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "AND NOW BEHOLD THE PERVERSITY OF THINGS",
+ "raw": {
+ "normalized_reference": "and now behold the perversity of things.",
+ "normalized_hypothesis": "and now behold the perversity of things",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "and now behold the perversity of things",
+ "normalized_hypothesis": "and now behold the perversity of things",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0391",
+ "category": "phonetic_balance",
+ "text": "The Fire People, like ourselves, lived in caves.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0391.wav",
+ "render_seconds": 68.71404180303216,
+ "audio_seconds": 3.4829931972789114,
+ "rtf": 19.728445595792437,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE FIRE PEOPLE LIKE OURSELVES LIVED IN CAVES",
+ "raw": {
+ "normalized_reference": "the fire people, like ourselves, lived in caves.",
+ "normalized_hypothesis": "the fire people like ourselves lived in caves",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the fire people like ourselves lived in caves",
+ "normalized_hypothesis": "the fire people like ourselves lived in caves",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0392",
+ "category": "phonetic_balance",
+ "text": "To her the bridge was tambo, which is the native for taboo.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0392.wav",
+ "render_seconds": 60.69530165498145,
+ "audio_seconds": 4.040272108843538,
+ "rtf": 15.022577692757059,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "TO HER THE BRIDGE WAS TAMBO WHICH IS THE NATIVE FOR TABOO",
+ "raw": {
+ "normalized_reference": "to her the bridge was tambo, which is the native for taboo.",
+ "normalized_hypothesis": "to her the bridge was tambo which is the native for taboo",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "to her the bridge was tambo which is the native for taboo",
+ "normalized_hypothesis": "to her the bridge was tambo which is the native for taboo",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0393",
+ "category": "phonetic_balance",
+ "text": "The voyage was our idea of a good time.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0393.wav",
+ "render_seconds": 50.50033521710429,
+ "audio_seconds": 2.8560544217687074,
+ "rtf": 17.681853270089395,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE VOYAGE WAS OUR IDEA OF A GOOD TIME",
+ "raw": {
+ "normalized_reference": "the voyage was our idea of a good time.",
+ "normalized_hypothesis": "the voyage was our idea of a good time",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the voyage was our idea of a good time",
+ "normalized_hypothesis": "the voyage was our idea of a good time",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0394",
+ "category": "phonetic_balance",
+ "text": "Of course much grumbling went on, and little outbursts were continually occurring.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0394.wav",
+ "render_seconds": 95.90916807902977,
+ "audio_seconds": 5.712108843537415,
+ "rtf": 16.79050079508548,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "OF COURSE MUCH GRUMBLING WENT ON AND LITTLE OUTBURSTS WERE CONTINUALLY OCCURRING",
+ "raw": {
+ "normalized_reference": "of course much grumbling went on, and little outbursts were continually occurring.",
+ "normalized_hypothesis": "of course much grumbling went on and little outbursts were continually occurring",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "of course much grumbling went on and little outbursts were continually occurring",
+ "normalized_hypothesis": "of course much grumbling went on and little outbursts were continually occurring",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0395",
+ "category": "phonetic_balance",
+ "text": "Get down and dig in.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0395.wav",
+ "render_seconds": 25.104691768996418,
+ "audio_seconds": 1.811156462585034,
+ "rtf": 13.86113916031578,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "GOT DOWN AND GERDEM",
+ "raw": {
+ "normalized_reference": "get down and dig in.",
+ "normalized_hypothesis": "got down and gerdem",
+ "errors": 3,
+ "words": 5,
+ "wer": 0.6
+ },
+ "semantic": {
+ "normalized_reference": "get down and dig in",
+ "normalized_hypothesis": "got down and gerdem",
+ "errors": 3,
+ "words": 5,
+ "wer": 0.6
+ }
+ },
+ {
+ "id": "release500_0396",
+ "category": "phonetic_balance",
+ "text": "It was put together in a casual, helter-skelter sort of way.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0396.wav",
+ "render_seconds": 59.49858675501309,
+ "audio_seconds": 4.31891156462585,
+ "rtf": 13.7762919811412,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT WAS PUT TOGETHER IN A CASUAL HELTER SKELTER SORT OF WAY",
+ "raw": {
+ "normalized_reference": "it was put together in a casual, helter-skelter sort of way.",
+ "normalized_hypothesis": "it was put together in a casual helter skelter sort of way",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it was put together in a casual helter skelter sort of way",
+ "normalized_hypothesis": "it was put together in a casual helter skelter sort of way",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0397",
+ "category": "phonetic_balance",
+ "text": "He deluged me, overwhelmed me with argument.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0397.wav",
+ "render_seconds": 34.39524943998549,
+ "audio_seconds": 3.0650340136054424,
+ "rtf": 11.221816556458334,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE DILUGED ME OVERWHELMED ME WITH AN ARGUMENT",
+ "raw": {
+ "normalized_reference": "he deluged me, overwhelmed me with argument.",
+ "normalized_hypothesis": "he diluged me overwhelmed me with an argument",
+ "errors": 2,
+ "words": 7,
+ "wer": 0.2857142857142857
+ },
+ "semantic": {
+ "normalized_reference": "he deluged me overwhelmed me with argument",
+ "normalized_hypothesis": "he diluged me overwhelmed me with an argument",
+ "errors": 2,
+ "words": 7,
+ "wer": 0.2857142857142857
+ }
+ },
+ {
+ "id": "release500_0398",
+ "category": "phonetic_balance",
+ "text": "The girl faced him, her eyes shining with sudden fear.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0398.wav",
+ "render_seconds": 48.99578789505176,
+ "audio_seconds": 3.6919727891156464,
+ "rtf": 13.270896264321582,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE GIRL FACED HIM HER EYES SHINING WITH SUDDEN FEAR",
+ "raw": {
+ "normalized_reference": "the girl faced him, her eyes shining with sudden fear.",
+ "normalized_hypothesis": "the girl faced him her eyes shining with sudden fear",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the girl faced him her eyes shining with sudden fear",
+ "normalized_hypothesis": "the girl faced him her eyes shining with sudden fear",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0399",
+ "category": "phonetic_balance",
+ "text": "It will break our hearts and our backs to hoist anchor by hand.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0399.wav",
+ "render_seconds": 67.49203227902763,
+ "audio_seconds": 4.597551020408163,
+ "rtf": 14.679996367508672,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT WILL BREAK OUR HEARTS AND OUR BACKS TO HOIST ANCHOR BY HAND",
+ "raw": {
+ "normalized_reference": "it will break our hearts and our backs to hoist anchor by hand.",
+ "normalized_hypothesis": "it will break our hearts and our backs to hoist anchor by hand",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it will break our hearts and our backs to hoist anchor by hand",
+ "normalized_hypothesis": "it will break our hearts and our backs to hoist anchor by hand",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0400",
+ "category": "phonetic_balance",
+ "text": "He made no reply as he waited for Whittemore to continue.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0400.wav",
+ "render_seconds": 73.58779493893962,
+ "audio_seconds": 4.040272108843538,
+ "rtf": 18.213573976333723,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE MADE NO REPLY AS HE WAITED FOR WHITMORE TO CONTINUE",
+ "raw": {
+ "normalized_reference": "he made no reply as he waited for whittemore to continue.",
+ "normalized_hypothesis": "he made no reply as he waited for whitmore to continue",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ },
+ "semantic": {
+ "normalized_reference": "he made no reply as he waited for whittemore to continue",
+ "normalized_hypothesis": "he made no reply as he waited for whitmore to continue",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ }
+ },
+ {
+ "id": "release500_0401",
+ "category": "phonetic_balance",
+ "text": "Who the devil gave it to you to be judge and jury.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0401.wav",
+ "render_seconds": 45.79701575706713,
+ "audio_seconds": 3.3436734693877552,
+ "rtf": 13.69661726132989,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WHO THE DEVIL GAVE IT TO YOU TO BE JUDGE AND JURY",
+ "raw": {
+ "normalized_reference": "who the devil gave it to you to be judge and jury.",
+ "normalized_hypothesis": "who the devil gave it to you to be judge and jury",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "who the devil gave it to you to be judge and jury",
+ "normalized_hypothesis": "who the devil gave it to you to be judge and jury",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0402",
+ "category": "phonetic_balance",
+ "text": "Daylight found himself charmed and made curious by the little man.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0402.wav",
+ "render_seconds": 70.59723766800016,
+ "audio_seconds": 4.527891156462585,
+ "rtf": 15.59163752583537,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "DAYLIGHT FOUND HIMSELF CHARMED AND MADE CURIOUS BY THE LITTLE MAN",
+ "raw": {
+ "normalized_reference": "daylight found himself charmed and made curious by the little man.",
+ "normalized_hypothesis": "daylight found himself charmed and made curious by the little man",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "daylight found himself charmed and made curious by the little man",
+ "normalized_hypothesis": "daylight found himself charmed and made curious by the little man",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0403",
+ "category": "phonetic_balance",
+ "text": "It's a Yankee, Joan cried.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0403.wav",
+ "render_seconds": 23.088385968934745,
+ "audio_seconds": 2.089795918367347,
+ "rtf": 11.048153442166038,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT PROFOUNDLY GLAGATIE",
+ "raw": {
+ "normalized_reference": "it's a yankee, joan cried.",
+ "normalized_hypothesis": "it profoundly glagatie",
+ "errors": 5,
+ "words": 5,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "it is a yankee joan cried",
+ "normalized_hypothesis": "it profoundly glagatie",
+ "errors": 5,
+ "words": 6,
+ "wer": 0.8333333333333334
+ }
+ },
+ {
+ "id": "release500_0404",
+ "category": "phonetic_balance",
+ "text": "But what they want with your toothbrush is more than I can imagine.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0404.wav",
+ "render_seconds": 71.50919498398434,
+ "audio_seconds": 4.388571428571429,
+ "rtf": 16.29441291952768,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "BUT WHAT THEY WANT WITH YOUR TOOTH BRUSH IS MORE THAN I CULD IMAGINE",
+ "raw": {
+ "normalized_reference": "but what they want with your toothbrush is more than i can imagine.",
+ "normalized_hypothesis": "but what they want with your tooth brush is more than i culd imagine",
+ "errors": 3,
+ "words": 13,
+ "wer": 0.23076923076923078
+ },
+ "semantic": {
+ "normalized_reference": "but what they want with your toothbrush is more than i can imagine",
+ "normalized_hypothesis": "but what they want with your tooth brush is more than i culd imagine",
+ "errors": 3,
+ "words": 13,
+ "wer": 0.23076923076923078
+ }
+ },
+ {
+ "id": "release500_0405",
+ "category": "phonetic_balance",
+ "text": "They were babbling and chattering all together.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0405.wav",
+ "render_seconds": 47.711288860067725,
+ "audio_seconds": 3.2043537414965986,
+ "rtf": 14.889519918541856,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THEY WERE BABBLING AND CHATTERING ALTOGETHER",
+ "raw": {
+ "normalized_reference": "they were babbling and chattering all together.",
+ "normalized_hypothesis": "they were babbling and chattering altogether",
+ "errors": 2,
+ "words": 7,
+ "wer": 0.2857142857142857
+ },
+ "semantic": {
+ "normalized_reference": "they were babbling and chattering all together",
+ "normalized_hypothesis": "they were babbling and chattering altogether",
+ "errors": 2,
+ "words": 7,
+ "wer": 0.2857142857142857
+ }
+ },
+ {
+ "id": "release500_0406",
+ "category": "phonetic_balance",
+ "text": "Men of Selden's stamp don't stop at women and children.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0406.wav",
+ "render_seconds": 65.10381821589544,
+ "audio_seconds": 3.4133333333333336,
+ "rtf": 19.07338424293812,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "MEN OF SELDOM STAMP DON'T STOP AT WOMEN AND CHILDREN",
+ "raw": {
+ "normalized_reference": "men of selden's stamp don't stop at women and children.",
+ "normalized_hypothesis": "men of seldom stamp don't stop at women and children",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ },
+ "semantic": {
+ "normalized_reference": "men of selden is stamp do not stop at women and children",
+ "normalized_hypothesis": "men of seldom stamp do not stop at women and children",
+ "errors": 2,
+ "words": 12,
+ "wer": 0.16666666666666666
+ }
+ },
+ {
+ "id": "release500_0407",
+ "category": "phonetic_balance",
+ "text": "Cameron looked at his hands with their long, sinewy fingers.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0407.wav",
+ "render_seconds": 60.201275527942926,
+ "audio_seconds": 4.179591836734694,
+ "rtf": 14.403625492525405,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "CAMERING LICKED UP HIS HANDS WITH THEIR LONG SIGNERY FINGERS",
+ "raw": {
+ "normalized_reference": "cameron looked at his hands with their long, sinewy fingers.",
+ "normalized_hypothesis": "camering licked up his hands with their long signery fingers",
+ "errors": 4,
+ "words": 10,
+ "wer": 0.4
+ },
+ "semantic": {
+ "normalized_reference": "cameron looked at his hands with their long sinewy fingers",
+ "normalized_hypothesis": "camering licked up his hands with their long signery fingers",
+ "errors": 4,
+ "words": 10,
+ "wer": 0.4
+ }
+ },
+ {
+ "id": "release500_0408",
+ "category": "phonetic_balance",
+ "text": "Saxon waited, for she knew a fresh idea had struck Billy.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0408.wav",
+ "render_seconds": 67.69975536607672,
+ "audio_seconds": 4.179591836734694,
+ "rtf": 16.19769537567265,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SAXON WAITED FOR SHE KNEW A FRESH IDEA HAD STRUCK BILLY",
+ "raw": {
+ "normalized_reference": "saxon waited, for she knew a fresh idea had struck billy.",
+ "normalized_hypothesis": "saxon waited for she knew a fresh idea had struck billy",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "saxon waited for she knew a fresh idea had struck billy",
+ "normalized_hypothesis": "saxon waited for she knew a fresh idea had struck billy",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0409",
+ "category": "phonetic_balance",
+ "text": "He saw only the effect in a general, sketchy way.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0409.wav",
+ "render_seconds": 30.415352317038924,
+ "audio_seconds": 3.4133333333333336,
+ "rtf": 8.910747749132497,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE SAW ONLY THE EFFECT IN THE GENERAL SKETCHY WAY",
+ "raw": {
+ "normalized_reference": "he saw only the effect in a general, sketchy way.",
+ "normalized_hypothesis": "he saw only the effect in the general sketchy way",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ },
+ "semantic": {
+ "normalized_reference": "he saw only the effect in a general sketchy way",
+ "normalized_hypothesis": "he saw only the effect in the general sketchy way",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ }
+ },
+ {
+ "id": "release500_0410",
+ "category": "phonetic_balance",
+ "text": "He began to follow the footprints of the dog.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0410.wav",
+ "render_seconds": 46.28845179104246,
+ "audio_seconds": 2.995374149659864,
+ "rtf": 15.453312166795152,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE BEGAN TO FOLLOW THE FOOTPRINTS OF THE DOG",
+ "raw": {
+ "normalized_reference": "he began to follow the footprints of the dog.",
+ "normalized_hypothesis": "he began to follow the footprints of the dog",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he began to follow the footprints of the dog",
+ "normalized_hypothesis": "he began to follow the footprints of the dog",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0411",
+ "category": "phonetic_balance",
+ "text": "Do you value your hide.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0411.wav",
+ "render_seconds": 38.296080961008556,
+ "audio_seconds": 1.811156462585034,
+ "rtf": 21.144545903201088,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "TONAV E LIKAD",
+ "raw": {
+ "normalized_reference": "do you value your hide.",
+ "normalized_hypothesis": "tonav e likad",
+ "errors": 5,
+ "words": 5,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "do you value your hide",
+ "normalized_hypothesis": "tonav e likad",
+ "errors": 5,
+ "words": 5,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0412",
+ "category": "phonetic_balance",
+ "text": "A rising tide of fat had submerged them.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0412.wav",
+ "render_seconds": 50.398992425063625,
+ "audio_seconds": 2.8560544217687074,
+ "rtf": 17.646369775353357,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "A RISING TIDE OF FAD HAD SUBMERGED THEM",
+ "raw": {
+ "normalized_reference": "a rising tide of fat had submerged them.",
+ "normalized_hypothesis": "a rising tide of fad had submerged them",
+ "errors": 1,
+ "words": 8,
+ "wer": 0.125
+ },
+ "semantic": {
+ "normalized_reference": "a rising tide of fat had submerged them",
+ "normalized_hypothesis": "a rising tide of fad had submerged them",
+ "errors": 1,
+ "words": 8,
+ "wer": 0.125
+ }
+ },
+ {
+ "id": "release500_0413",
+ "category": "phonetic_balance",
+ "text": "There was nothing on the rock.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0413.wav",
+ "render_seconds": 40.90144200401846,
+ "audio_seconds": 2.2291156462585033,
+ "rtf": 18.348730391207013,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "TRIK NOTA ANDE MONK",
+ "raw": {
+ "normalized_reference": "there was nothing on the rock.",
+ "normalized_hypothesis": "trik nota ande monk",
+ "errors": 6,
+ "words": 6,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "there was nothing on the rock",
+ "normalized_hypothesis": "trik nota ande monk",
+ "errors": 6,
+ "words": 6,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0414",
+ "category": "phonetic_balance",
+ "text": "Whatever he guessed he locked away in the taboo room of Naomi.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0414.wav",
+ "render_seconds": 67.39272928400896,
+ "audio_seconds": 4.458231292517007,
+ "rtf": 15.116472175215634,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WHENEVER HE GUESSED HE LOCKED AWAY IN THE TABOO ROOM OF NAOMI",
+ "raw": {
+ "normalized_reference": "whatever he guessed he locked away in the taboo room of naomi.",
+ "normalized_hypothesis": "whenever he guessed he locked away in the taboo room of naomi",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "whatever he guessed he locked away in the taboo room of naomi",
+ "normalized_hypothesis": "whenever he guessed he locked away in the taboo room of naomi",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ }
+ },
+ {
+ "id": "release500_0415",
+ "category": "phonetic_balance",
+ "text": "Will we ever forget it.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0415.wav",
+ "render_seconds": 16.89603050693404,
+ "audio_seconds": 1.9504761904761905,
+ "rtf": 8.66251564076208,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "YOA RE AT TE THRICK ABIT",
+ "raw": {
+ "normalized_reference": "will we ever forget it.",
+ "normalized_hypothesis": "yoa re at te thrick abit",
+ "errors": 6,
+ "words": 5,
+ "wer": 1.2
+ },
+ "semantic": {
+ "normalized_reference": "will we ever forget it",
+ "normalized_hypothesis": "yoa re at te thrick abit",
+ "errors": 6,
+ "words": 5,
+ "wer": 1.2
+ }
+ },
+ {
+ "id": "release500_0416",
+ "category": "phonetic_balance",
+ "text": "Jeanne and Pierre both gazed toward the great rock.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0416.wav",
+ "render_seconds": 54.9072054940043,
+ "audio_seconds": 3.622312925170068,
+ "rtf": 15.158051396519367,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "JE AND PIERRE BOTH GEZZED TOWARD THE GREAT ROCK",
+ "raw": {
+ "normalized_reference": "jeanne and pierre both gazed toward the great rock.",
+ "normalized_hypothesis": "je and pierre both gezzed toward the great rock",
+ "errors": 2,
+ "words": 9,
+ "wer": 0.2222222222222222
+ },
+ "semantic": {
+ "normalized_reference": "jeanne and pierre both gazed toward the great rock",
+ "normalized_hypothesis": "je and pierre both gezzed toward the great rock",
+ "errors": 2,
+ "words": 9,
+ "wer": 0.2222222222222222
+ }
+ },
+ {
+ "id": "release500_0417",
+ "category": "phonetic_balance",
+ "text": "Does the old boy often go off at half-cock that way.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0417.wav",
+ "render_seconds": 47.9935970089864,
+ "audio_seconds": 3.4829931972789114,
+ "rtf": 13.779411641251954,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "DOES THE OLD BOY OFTEN GO OFF AT HAUSE COCK THAT WAY",
+ "raw": {
+ "normalized_reference": "does the old boy often go off at half-cock that way.",
+ "normalized_hypothesis": "does the old boy often go off at hause cock that way",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "does the old boy often go off at half cock that way",
+ "normalized_hypothesis": "does the old boy often go off at hause cock that way",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ }
+ },
+ {
+ "id": "release500_0418",
+ "category": "phonetic_balance",
+ "text": "The temperature dropped to fifty below zero and remained there the whole trip.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0418.wav",
+ "render_seconds": 78.09716012491845,
+ "audio_seconds": 5.363809523809524,
+ "rtf": 14.56001742385731,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE TEMPERATURE DROPPED TO FIFTY BELOW ZERO AND REMAINED THERE THE WHOLE TRIP",
+ "raw": {
+ "normalized_reference": "the temperature dropped to fifty below zero and remained there the whole trip.",
+ "normalized_hypothesis": "the temperature dropped to fifty below zero and remained there the whole trip",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the temperature dropped to 50 below 0 and remained there the whole trip",
+ "normalized_hypothesis": "the temperature dropped to 50 below 0 and remained there the whole trip",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0419",
+ "category": "phonetic_balance",
+ "text": "It's worth eight dollars.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0419.wav",
+ "render_seconds": 40.10603851394262,
+ "audio_seconds": 2.020136054421769,
+ "rtf": 19.853137330110336,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT'S WORS ANT DYLER'S",
+ "raw": {
+ "normalized_reference": "it's worth eight dollars.",
+ "normalized_hypothesis": "it's wors ant dyler's",
+ "errors": 3,
+ "words": 4,
+ "wer": 0.75
+ },
+ "semantic": {
+ "normalized_reference": "it is worth $8",
+ "normalized_hypothesis": "it is wors ant dyler is",
+ "errors": 4,
+ "words": 4,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0420",
+ "category": "phonetic_balance",
+ "text": "But I did not enjoy it long.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0420.wav",
+ "render_seconds": 30.903302449034527,
+ "audio_seconds": 2.159455782312925,
+ "rtf": 14.310690083190764,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "BUT I DID NOT ENJOY IT LONG",
+ "raw": {
+ "normalized_reference": "but i did not enjoy it long.",
+ "normalized_hypothesis": "but i did not enjoy it long",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "but i did not enjoy it long",
+ "normalized_hypothesis": "but i did not enjoy it long",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0421",
+ "category": "phonetic_balance",
+ "text": "Shorty turned to their employers.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0421.wav",
+ "render_seconds": 50.088806048966944,
+ "audio_seconds": 2.4380952380952383,
+ "rtf": 20.544236856021595,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SHORTY TURNED TO THEIR EMPLOYERS",
+ "raw": {
+ "normalized_reference": "shorty turned to their employers.",
+ "normalized_hypothesis": "shorty turned to their employers",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "shorty turned to their employers",
+ "normalized_hypothesis": "shorty turned to their employers",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0422",
+ "category": "phonetic_balance",
+ "text": "It was the third or fourth time that Philip had heard MacDougall swear.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0422.wav",
+ "render_seconds": 75.50115589995403,
+ "audio_seconds": 5.015510204081632,
+ "rtf": 15.053534501537058,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT WAS THE THIRD OR FOURTH TIME THAT PHILIP HAD HEARD MACDUGAL SWEAR",
+ "raw": {
+ "normalized_reference": "it was the third or fourth time that philip had heard macdougall swear.",
+ "normalized_hypothesis": "it was the third or fourth time that philip had heard macdugal swear",
+ "errors": 1,
+ "words": 13,
+ "wer": 0.07692307692307693
+ },
+ "semantic": {
+ "normalized_reference": "it was the 3rd or 4th time that philip had heard macdougall swear",
+ "normalized_hypothesis": "it was the 3rd or 4th time that philip had heard macdugal swear",
+ "errors": 1,
+ "words": 13,
+ "wer": 0.07692307692307693
+ }
+ },
+ {
+ "id": "release500_0423",
+ "category": "phonetic_balance",
+ "text": "Ernest saw in the affair the most sinister import.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0423.wav",
+ "render_seconds": 64.29856270097662,
+ "audio_seconds": 3.6919727891156464,
+ "rtf": 17.415773726863875,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "ERNEST SAW IN THE AFFAIR THE MOST SINISTER IMPORT",
+ "raw": {
+ "normalized_reference": "ernest saw in the affair the most sinister import.",
+ "normalized_hypothesis": "ernest saw in the affair the most sinister import",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "ernest saw in the affair the most sinister import",
+ "normalized_hypothesis": "ernest saw in the affair the most sinister import",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0424",
+ "category": "phonetic_balance",
+ "text": "Also, at regular intervals, he would mutter.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0424.wav",
+ "render_seconds": 45.8996271060314,
+ "audio_seconds": 3.3436734693877552,
+ "rtf": 13.727305469943474,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "ALSO AT REGULAR INTERVALS HE WOULDN'T ETT HER",
+ "raw": {
+ "normalized_reference": "also, at regular intervals, he would mutter.",
+ "normalized_hypothesis": "also at regular intervals he wouldn't ett her",
+ "errors": 3,
+ "words": 7,
+ "wer": 0.42857142857142855
+ },
+ "semantic": {
+ "normalized_reference": "also at regular intervals he would mutter",
+ "normalized_hypothesis": "also at regular intervals he would not ett her",
+ "errors": 3,
+ "words": 7,
+ "wer": 0.42857142857142855
+ }
+ },
+ {
+ "id": "release500_0425",
+ "category": "phonetic_balance",
+ "text": "He may anticipate the day of his death.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0425.wav",
+ "render_seconds": 48.688397492980585,
+ "audio_seconds": 2.9257142857142857,
+ "rtf": 16.641542111858598,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE MAY ANTICIPATE THE DAY OF HIS DEATH",
+ "raw": {
+ "normalized_reference": "he may anticipate the day of his death.",
+ "normalized_hypothesis": "he may anticipate the day of his death",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he may anticipate the day of his death",
+ "normalized_hypothesis": "he may anticipate the day of his death",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0426",
+ "category": "phonetic_balance",
+ "text": "I defy any man to get a Solomon Island sore in California.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0426.wav",
+ "render_seconds": 50.70114224101417,
+ "audio_seconds": 4.31891156462585,
+ "rtf": 11.739333274680385,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I DEFY ANY MAN TO GET A SOLOMON ISLANDS OR IN CALIFORNIA",
+ "raw": {
+ "normalized_reference": "i defy any man to get a solomon island sore in california.",
+ "normalized_hypothesis": "i defy any man to get a solomon islands or in california",
+ "errors": 2,
+ "words": 12,
+ "wer": 0.16666666666666666
+ },
+ "semantic": {
+ "normalized_reference": "i defy any man to get a solomon island sore in california",
+ "normalized_hypothesis": "i defy any man to get a solomon islands or in california",
+ "errors": 2,
+ "words": 12,
+ "wer": 0.16666666666666666
+ }
+ },
+ {
+ "id": "release500_0427",
+ "category": "phonetic_balance",
+ "text": "Now our figuring was all right.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0427.wav",
+ "render_seconds": 21.79856337502133,
+ "audio_seconds": 2.4380952380952383,
+ "rtf": 8.940817009286093,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "NOW IL TURNTELING WAS I A RIGHT",
+ "raw": {
+ "normalized_reference": "now our figuring was all right.",
+ "normalized_hypothesis": "now il turnteling was i a right",
+ "errors": 4,
+ "words": 6,
+ "wer": 0.6666666666666666
+ },
+ "semantic": {
+ "normalized_reference": "now our figuring was all right",
+ "normalized_hypothesis": "now il turnteling was i a right",
+ "errors": 4,
+ "words": 6,
+ "wer": 0.6666666666666666
+ }
+ },
+ {
+ "id": "release500_0428",
+ "category": "phonetic_balance",
+ "text": "MacDougall tapped his forehead suspiciously with a stubby forefinger.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0428.wav",
+ "render_seconds": 71.49118694197387,
+ "audio_seconds": 5.085170068027211,
+ "rtf": 14.058760274601559,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "MACDUGALL TAPPED HIS FOREHEAD SUSPICIOUSLY WITH A STUBBY FOREFINGER",
+ "raw": {
+ "normalized_reference": "macdougall tapped his forehead suspiciously with a stubby forefinger.",
+ "normalized_hypothesis": "macdugall tapped his forehead suspiciously with a stubby forefinger",
+ "errors": 1,
+ "words": 9,
+ "wer": 0.1111111111111111
+ },
+ "semantic": {
+ "normalized_reference": "macdougall tapped his forehead suspiciously with a stubby forefinger",
+ "normalized_hypothesis": "macdugall tapped his forehead suspiciously with a stubby forefinger",
+ "errors": 1,
+ "words": 9,
+ "wer": 0.1111111111111111
+ }
+ },
+ {
+ "id": "release500_0429",
+ "category": "phonetic_balance",
+ "text": "O'Brien took off his coat and bared his right arm.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0429.wav",
+ "render_seconds": 60.09927179699298,
+ "audio_seconds": 3.4133333333333336,
+ "rtf": 17.607208534275287,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "OBRIAN TOOK OFF HIS COAT AND BARED HIS RIGHT ARM",
+ "raw": {
+ "normalized_reference": "o'brien took off his coat and bared his right arm.",
+ "normalized_hypothesis": "obrian took off his coat and bared his right arm",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ },
+ "semantic": {
+ "normalized_reference": "0 brien took off his coat and bared his right arm",
+ "normalized_hypothesis": "obrian took off his coat and bared his right arm",
+ "errors": 2,
+ "words": 11,
+ "wer": 0.18181818181818182
+ }
+ },
+ {
+ "id": "release500_0430",
+ "category": "phonetic_balance",
+ "text": "But it won't continue, she said with easy confidence.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0430.wav",
+ "render_seconds": 57.308990280958824,
+ "audio_seconds": 3.5526530612244898,
+ "rtf": 16.131321942595257,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "BUT IT WON'T CONTINUE SHE SAID WITH EASY CONFIDENCE",
+ "raw": {
+ "normalized_reference": "but it won't continue, she said with easy confidence.",
+ "normalized_hypothesis": "but it won't continue she said with easy confidence",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "but it will not continue she said with easy confidence",
+ "normalized_hypothesis": "but it will not continue she said with easy confidence",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0431",
+ "category": "phonetic_balance",
+ "text": "I was still weak from my prolonged immersion.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0431.wav",
+ "render_seconds": 36.81127409497276,
+ "audio_seconds": 3.3436734693877552,
+ "rtf": 11.009231144126375,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I WAS STILL WEAK FROM MY PRELANGD DIMERSION",
+ "raw": {
+ "normalized_reference": "i was still weak from my prolonged immersion.",
+ "normalized_hypothesis": "i was still weak from my prelangd dimersion",
+ "errors": 2,
+ "words": 8,
+ "wer": 0.25
+ },
+ "semantic": {
+ "normalized_reference": "i was still weak from my prolonged immersion",
+ "normalized_hypothesis": "i was still weak from my prelangd dimersion",
+ "errors": 2,
+ "words": 8,
+ "wer": 0.25
+ }
+ },
+ {
+ "id": "release500_0432",
+ "category": "phonetic_balance",
+ "text": "Oolong was two hundred and fifty miles from the nearest land.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0432.wav",
+ "render_seconds": 48.21655345603358,
+ "audio_seconds": 4.249251700680272,
+ "rtf": 11.347069284767123,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "OLONG WAS TWO HUNDRED AND FIFTY MILES FROM THE NEAREST LAND",
+ "raw": {
+ "normalized_reference": "oolong was two hundred and fifty miles from the nearest land.",
+ "normalized_hypothesis": "olong was two hundred and fifty miles from the nearest land",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ },
+ "semantic": {
+ "normalized_reference": "oolong was 250 miles from the nearest land",
+ "normalized_hypothesis": "olong was 250 miles from the nearest land",
+ "errors": 1,
+ "words": 8,
+ "wer": 0.125
+ }
+ },
+ {
+ "id": "release500_0433",
+ "category": "phonetic_balance",
+ "text": "I just do appreciate it without being able to express my feelings.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0433.wav",
+ "render_seconds": 37.79948809801135,
+ "audio_seconds": 4.806530612244898,
+ "rtf": 7.864193770391289,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I JUST DO APPRECIATE IT WITHOUT BEING ABLE TO EXPRESS MY FEELINGS",
+ "raw": {
+ "normalized_reference": "i just do appreciate it without being able to express my feelings.",
+ "normalized_hypothesis": "i just do appreciate it without being able to express my feelings",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i just do appreciate it without being able to express my feelings",
+ "normalized_hypothesis": "i just do appreciate it without being able to express my feelings",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0434",
+ "category": "phonetic_balance",
+ "text": "Now just what do you want to know.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0434.wav",
+ "render_seconds": 51.29997921502218,
+ "audio_seconds": 2.36843537414966,
+ "rtf": 21.659860249908835,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "NOW JUST WHAT DO YOU WANT TO KNOW",
+ "raw": {
+ "normalized_reference": "now just what do you want to know.",
+ "normalized_hypothesis": "now just what do you want to know",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "now just what do you want to know",
+ "normalized_hypothesis": "now just what do you want to know",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0435",
+ "category": "phonetic_balance",
+ "text": "But she swung obediently on her heel into the wind.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0435.wav",
+ "render_seconds": 60.00744204397779,
+ "audio_seconds": 3.4133333333333336,
+ "rtf": 17.580305286321618,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "BUT SHE SWUNG OBEDIENTLY ON HER HEEL INTO THE WIND",
+ "raw": {
+ "normalized_reference": "but she swung obediently on her heel into the wind.",
+ "normalized_hypothesis": "but she swung obediently on her heel into the wind",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "but she swung obediently on her heel into the wind",
+ "normalized_hypothesis": "but she swung obediently on her heel into the wind",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0436",
+ "category": "phonetic_balance",
+ "text": "Ask him, I laughed, then turned to Pasquini.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0436.wav",
+ "render_seconds": 40.29292599798646,
+ "audio_seconds": 3.0650340136054424,
+ "rtf": 13.145996363867207,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "NAFKIMI LAUGHED THEN TURNED TO PAFKENEL",
+ "raw": {
+ "normalized_reference": "ask him, i laughed, then turned to pasquini.",
+ "normalized_hypothesis": "nafkimi laughed then turned to pafkenel",
+ "errors": 4,
+ "words": 8,
+ "wer": 0.5
+ },
+ "semantic": {
+ "normalized_reference": "ask him i laughed then turned to pasquini",
+ "normalized_hypothesis": "nafkimi laughed then turned to pafkenel",
+ "errors": 4,
+ "words": 8,
+ "wer": 0.5
+ }
+ },
+ {
+ "id": "release500_0437",
+ "category": "phonetic_balance",
+ "text": "He did not rush in.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0437.wav",
+ "render_seconds": 38.699202345917,
+ "audio_seconds": 1.7414965986394557,
+ "rtf": 22.221807597069528,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I O NOT NOW F",
+ "raw": {
+ "normalized_reference": "he did not rush in.",
+ "normalized_hypothesis": "i o not now f",
+ "errors": 4,
+ "words": 5,
+ "wer": 0.8
+ },
+ "semantic": {
+ "normalized_reference": "he did not rush in",
+ "normalized_hypothesis": "i 0 not now f",
+ "errors": 4,
+ "words": 5,
+ "wer": 0.8
+ }
+ },
+ {
+ "id": "release500_0438",
+ "category": "phonetic_balance",
+ "text": "They obeyed him, and went here and there at his commands.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0438.wav",
+ "render_seconds": 48.70280516904313,
+ "audio_seconds": 3.900952380952381,
+ "rtf": 12.484849957884592,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THEY OBEYED HIM AND WENT HERE AND THERE AT HIS COMMANDS",
+ "raw": {
+ "normalized_reference": "they obeyed him, and went here and there at his commands.",
+ "normalized_hypothesis": "they obeyed him and went here and there at his commands",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "they obeyed him and went here and there at his commands",
+ "normalized_hypothesis": "they obeyed him and went here and there at his commands",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0439",
+ "category": "phonetic_balance",
+ "text": "Tom Spink has a harpoon.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0439.wav",
+ "render_seconds": 38.82369073200971,
+ "audio_seconds": 2.159455782312925,
+ "rtf": 17.978460614936452,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "TAMSPOK AVE YOU HO DINNER",
+ "raw": {
+ "normalized_reference": "tom spink has a harpoon.",
+ "normalized_hypothesis": "tamspok ave you ho dinner",
+ "errors": 5,
+ "words": 5,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "tom spink has a harpoon",
+ "normalized_hypothesis": "tamspok ave you ho dinner",
+ "errors": 5,
+ "words": 5,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0440",
+ "category": "phonetic_balance",
+ "text": "It was my reports from the north which chiefly induced people to buy.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0440.wav",
+ "render_seconds": 56.10623948997818,
+ "audio_seconds": 4.876190476190477,
+ "rtf": 11.50616239540568,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT WAS MY REPORTS FROM THE NORTH WHICH CHIEFLY INDUCED PEOPLE TO BUY",
+ "raw": {
+ "normalized_reference": "it was my reports from the north which chiefly induced people to buy.",
+ "normalized_hypothesis": "it was my reports from the north which chiefly induced people to buy",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it was my reports from the north which chiefly induced people to buy",
+ "normalized_hypothesis": "it was my reports from the north which chiefly induced people to buy",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0441",
+ "category": "phonetic_balance",
+ "text": "Cherokee identified himself with his instinct.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0441.wav",
+ "render_seconds": 59.99933752510697,
+ "audio_seconds": 3.2043537414965986,
+ "rtf": 18.724317714399465,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SHEROKY IDENTIFIED HIMSELF WITH HIS INSTINCT",
+ "raw": {
+ "normalized_reference": "cherokee identified himself with his instinct.",
+ "normalized_hypothesis": "sheroky identified himself with his instinct",
+ "errors": 1,
+ "words": 6,
+ "wer": 0.16666666666666666
+ },
+ "semantic": {
+ "normalized_reference": "cherokee identified himself with his instinct",
+ "normalized_hypothesis": "sheroky identified himself with his instinct",
+ "errors": 1,
+ "words": 6,
+ "wer": 0.16666666666666666
+ }
+ },
+ {
+ "id": "release500_0442",
+ "category": "phonetic_balance",
+ "text": "I graduated last of my class.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0442.wav",
+ "render_seconds": 35.50398636702448,
+ "audio_seconds": 2.4380952380952383,
+ "rtf": 14.562181908349883,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I GRADUATE HAVE NAUGH AT MYCON",
+ "raw": {
+ "normalized_reference": "i graduated last of my class.",
+ "normalized_hypothesis": "i graduate have naugh at mycon",
+ "errors": 5,
+ "words": 6,
+ "wer": 0.8333333333333334
+ },
+ "semantic": {
+ "normalized_reference": "i graduated last of my class",
+ "normalized_hypothesis": "i graduate have naugh at mycon",
+ "errors": 5,
+ "words": 6,
+ "wer": 0.8333333333333334
+ }
+ },
+ {
+ "id": "release500_0443",
+ "category": "phonetic_balance",
+ "text": "It was a gigantic inadequacy.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0443.wav",
+ "render_seconds": 33.90033615392167,
+ "audio_seconds": 2.2987755102040817,
+ "rtf": 14.74712776582175,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT WAS A GIGANTIC INADEQUACY",
+ "raw": {
+ "normalized_reference": "it was a gigantic inadequacy.",
+ "normalized_hypothesis": "it was a gigantic inadequacy",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it was a gigantic inadequacy",
+ "normalized_hypothesis": "it was a gigantic inadequacy",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0444",
+ "category": "phonetic_balance",
+ "text": "Did I possess too much vitality.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0444.wav",
+ "render_seconds": 41.29556576604955,
+ "audio_seconds": 2.507755102040816,
+ "rtf": 16.467144551891504,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "DID I POSSESS TOO MUCH VITALITY",
+ "raw": {
+ "normalized_reference": "did i possess too much vitality.",
+ "normalized_hypothesis": "did i possess too much vitality",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "did i possess too much vitality",
+ "normalized_hypothesis": "did i possess too much vitality",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0445",
+ "category": "phonetic_balance",
+ "text": "Therefore, hurrah for the game.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0445.wav",
+ "render_seconds": 21.084058632026426,
+ "audio_seconds": 2.2291156462585033,
+ "rtf": 9.458485775475722,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "MAROPHINA HELD OUT FOR THE GUM",
+ "raw": {
+ "normalized_reference": "therefore, hurrah for the game.",
+ "normalized_hypothesis": "marophina held out for the gum",
+ "errors": 4,
+ "words": 5,
+ "wer": 0.8
+ },
+ "semantic": {
+ "normalized_reference": "therefore hurrah for the game",
+ "normalized_hypothesis": "marophina held out for the gum",
+ "errors": 4,
+ "words": 5,
+ "wer": 0.8
+ }
+ },
+ {
+ "id": "release500_0446",
+ "category": "phonetic_balance",
+ "text": "He had fulfilled his duty and paid properly.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0446.wav",
+ "render_seconds": 47.69303379300982,
+ "audio_seconds": 3.1346938775510202,
+ "rtf": 15.214574582405476,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE HAD FULFILLED HIS DUTY AND PAID PROPERLY",
+ "raw": {
+ "normalized_reference": "he had fulfilled his duty and paid properly.",
+ "normalized_hypothesis": "he had fulfilled his duty and paid properly",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he had fulfilled his duty and paid properly",
+ "normalized_hypothesis": "he had fulfilled his duty and paid properly",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0447",
+ "category": "phonetic_balance",
+ "text": "How can you manage all alone, Mr Young.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0447.wav",
+ "render_seconds": 45.49568775296211,
+ "audio_seconds": 2.8560544217687074,
+ "rtf": 15.929559117009887,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HOW CAN I HAT A DRAWL A LITTLE MISTER YOUNG",
+ "raw": {
+ "normalized_reference": "how can you manage all alone, mr young.",
+ "normalized_hypothesis": "how can i hat a drawl a little mister young",
+ "errors": 7,
+ "words": 8,
+ "wer": 0.875
+ },
+ "semantic": {
+ "normalized_reference": "how can you manage all alone mister young",
+ "normalized_hypothesis": "how can i hat a drawl a little mister young",
+ "errors": 6,
+ "words": 8,
+ "wer": 0.75
+ }
+ },
+ {
+ "id": "release500_0448",
+ "category": "phonetic_balance",
+ "text": "Red-Eye never committed a more outrageous deed.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0448.wav",
+ "render_seconds": 43.70079360692762,
+ "audio_seconds": 3.5526530612244898,
+ "rtf": 12.300889744597043,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "RED EYE NEVER COMMITTED A MORE OUTRAGEOUS DEED",
+ "raw": {
+ "normalized_reference": "red-eye never committed a more outrageous deed.",
+ "normalized_hypothesis": "red eye never committed a more outrageous deed",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "red eye never committed a more outrageous deed",
+ "normalized_hypothesis": "red eye never committed a more outrageous deed",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0449",
+ "category": "phonetic_balance",
+ "text": "He would keep his faith with Gregson for the promised day or two.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0449.wav",
+ "render_seconds": 71.59551129001193,
+ "audio_seconds": 4.597551020408163,
+ "rtf": 15.572532196424824,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE WOULD KEEP HIS FAITH WITH GREGSON FOR THE PROMISED DAY OR TWO",
+ "raw": {
+ "normalized_reference": "he would keep his faith with gregson for the promised day or two.",
+ "normalized_hypothesis": "he would keep his faith with gregson for the promised day or two",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he would keep his faith with gregson for the promised day or 2",
+ "normalized_hypothesis": "he would keep his faith with gregson for the promised day or 2",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0450",
+ "category": "phonetic_balance",
+ "text": "There was one chance, and only one, of saving Jeanne.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0450.wav",
+ "render_seconds": 40.0976740430342,
+ "audio_seconds": 3.8312925170068026,
+ "rtf": 10.465834666772066,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THER WAS ONE CHANCE AND ONLY ONE OF SAVING JAN",
+ "raw": {
+ "normalized_reference": "there was one chance, and only one, of saving jeanne.",
+ "normalized_hypothesis": "ther was one chance and only one of saving jan",
+ "errors": 2,
+ "words": 10,
+ "wer": 0.2
+ },
+ "semantic": {
+ "normalized_reference": "there was one chance and only one of saving jeanne",
+ "normalized_hypothesis": "ther was one chance and only one of saving jan",
+ "errors": 2,
+ "words": 10,
+ "wer": 0.2
+ }
+ },
+ {
+ "id": "release500_0451",
+ "category": "phonetic_balance",
+ "text": "I was beneath the water, suffocating and drowning.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0451.wav",
+ "render_seconds": 24.89961629291065,
+ "audio_seconds": 3.4829931972789114,
+ "rtf": 7.148913271597394,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I WAS BENEATH THE WATER SUFFOCATING AND DROWNING",
+ "raw": {
+ "normalized_reference": "i was beneath the water, suffocating and drowning.",
+ "normalized_hypothesis": "i was beneath the water suffocating and drowning",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i was beneath the water suffocating and drowning",
+ "normalized_hypothesis": "i was beneath the water suffocating and drowning",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0452",
+ "category": "phonetic_balance",
+ "text": "Give them their choice between a fine or an official whipping.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0452.wav",
+ "render_seconds": 52.30043519998435,
+ "audio_seconds": 4.388571428571429,
+ "rtf": 11.917416874996434,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "GIVE THEM THEIR CHOICE BETWEEN A FINE OR AN OFFICIAL WHIPPING",
+ "raw": {
+ "normalized_reference": "give them their choice between a fine or an official whipping.",
+ "normalized_hypothesis": "give them their choice between a fine or an official whipping",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "give them their choice between a fine or an official whipping",
+ "normalized_hypothesis": "give them their choice between a fine or an official whipping",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0453",
+ "category": "phonetic_balance",
+ "text": "They must have been swept away by the chaotic currents.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0453.wav",
+ "render_seconds": 57.59676754404791,
+ "audio_seconds": 3.900952380952381,
+ "rtf": 14.76479636749275,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THEY MUST HAVE BEEN SWEPT AWAY BY THE CHAOTIC CURRENTS",
+ "raw": {
+ "normalized_reference": "they must have been swept away by the chaotic currents.",
+ "normalized_hypothesis": "they must have been swept away by the chaotic currents",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "they must have been swept away by the chaotic currents",
+ "normalized_hypothesis": "they must have been swept away by the chaotic currents",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0454",
+ "category": "phonetic_balance",
+ "text": "You yellow giant thing of the frost.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0454.wav",
+ "render_seconds": 42.49232076504268,
+ "audio_seconds": 2.4380952380952383,
+ "rtf": 17.428490938787036,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "TE LELLE JOIAN THING OF THE FAFT",
+ "raw": {
+ "normalized_reference": "you yellow giant thing of the frost.",
+ "normalized_hypothesis": "te lelle joian thing of the faft",
+ "errors": 4,
+ "words": 7,
+ "wer": 0.5714285714285714
+ },
+ "semantic": {
+ "normalized_reference": "you yellow giant thing of the frost",
+ "normalized_hypothesis": "te lelle joian thing of the faft",
+ "errors": 4,
+ "words": 7,
+ "wer": 0.5714285714285714
+ }
+ },
+ {
+ "id": "release500_0455",
+ "category": "phonetic_balance",
+ "text": "He had peeled off his shirt and was wildly waving it.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0455.wav",
+ "render_seconds": 66.29271442000754,
+ "audio_seconds": 3.6919727891156464,
+ "rtf": 17.955905475643256,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE HAD PEELED OFF HIS SHIRT AND WAS WILDLY WAVING IT",
+ "raw": {
+ "normalized_reference": "he had peeled off his shirt and was wildly waving it.",
+ "normalized_hypothesis": "he had peeled off his shirt and was wildly waving it",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he had peeled off his shirt and was wildly waving it",
+ "normalized_hypothesis": "he had peeled off his shirt and was wildly waving it",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0456",
+ "category": "phonetic_balance",
+ "text": "To these he gave castor oil.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0456.wav",
+ "render_seconds": 27.798200483899564,
+ "audio_seconds": 2.2291156462585033,
+ "rtf": 12.47050619852672,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "TO NEY FROM DETCAST AR WO",
+ "raw": {
+ "normalized_reference": "to these he gave castor oil.",
+ "normalized_hypothesis": "to ney from detcast ar wo",
+ "errors": 5,
+ "words": 6,
+ "wer": 0.8333333333333334
+ },
+ "semantic": {
+ "normalized_reference": "to these he gave castor oil",
+ "normalized_hypothesis": "to ney from detcast ar wo",
+ "errors": 5,
+ "words": 6,
+ "wer": 0.8333333333333334
+ }
+ },
+ {
+ "id": "release500_0457",
+ "category": "phonetic_balance",
+ "text": "Tudor surveyed him with withering disgust.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0457.wav",
+ "render_seconds": 25.503747343900613,
+ "audio_seconds": 2.9257142857142857,
+ "rtf": 8.71710114293478,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "TUTER SURVEYED HIM WITH WITHERING DISGUST",
+ "raw": {
+ "normalized_reference": "tudor surveyed him with withering disgust.",
+ "normalized_hypothesis": "tuter surveyed him with withering disgust",
+ "errors": 1,
+ "words": 6,
+ "wer": 0.16666666666666666
+ },
+ "semantic": {
+ "normalized_reference": "tudor surveyed him with withering disgust",
+ "normalized_hypothesis": "tuter surveyed him with withering disgust",
+ "errors": 1,
+ "words": 6,
+ "wer": 0.16666666666666666
+ }
+ },
+ {
+ "id": "release500_0458",
+ "category": "phonetic_balance",
+ "text": "After all, it was simply a mistake in judgment.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0458.wav",
+ "render_seconds": 40.506310326978564,
+ "audio_seconds": 3.4829931972789114,
+ "rtf": 11.629741441534861,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "AFTER ALL IT WAS SIMPLY A MISTAKE IN JUDGMENT",
+ "raw": {
+ "normalized_reference": "after all, it was simply a mistake in judgment.",
+ "normalized_hypothesis": "after all it was simply a mistake in judgment",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "after all it was simply a mistake in judgment",
+ "normalized_hypothesis": "after all it was simply a mistake in judgment",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0459",
+ "category": "phonetic_balance",
+ "text": "That is the strange part of it.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0459.wav",
+ "render_seconds": 51.41124410997145,
+ "audio_seconds": 2.2987755102040817,
+ "rtf": 22.364621461191415,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THAT IS THE STRANGE PART OF IT",
+ "raw": {
+ "normalized_reference": "that is the strange part of it.",
+ "normalized_hypothesis": "that is the strange part of it",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "that is the strange part of it",
+ "normalized_hypothesis": "that is the strange part of it",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0460",
+ "category": "phonetic_balance",
+ "text": "Not a wheel moved in his empire.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0460.wav",
+ "render_seconds": 39.307403607992455,
+ "audio_seconds": 2.507755102040816,
+ "rtf": 15.674339003838138,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "NOT A RAAL MOONSTUN OF IMP",
+ "raw": {
+ "normalized_reference": "not a wheel moved in his empire.",
+ "normalized_hypothesis": "not a raal moonstun of imp",
+ "errors": 5,
+ "words": 7,
+ "wer": 0.7142857142857143
+ },
+ "semantic": {
+ "normalized_reference": "not a wheel moved in his empire",
+ "normalized_hypothesis": "not a raal moonstun of imp",
+ "errors": 5,
+ "words": 7,
+ "wer": 0.7142857142857143
+ }
+ },
+ {
+ "id": "release500_0461",
+ "category": "phonetic_balance",
+ "text": "Its diameter was not more than two hundred yards.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0461.wav",
+ "render_seconds": 51.40658588602673,
+ "audio_seconds": 3.4133333333333336,
+ "rtf": 15.060523208796893,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "ITS DIAMETER WAS NOT MORE THAN TWO HUNDRED YARDS",
+ "raw": {
+ "normalized_reference": "its diameter was not more than two hundred yards.",
+ "normalized_hypothesis": "its diameter was not more than two hundred yards",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "its diameter was not more than 200 yards",
+ "normalized_hypothesis": "its diameter was not more than 200 yards",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0462",
+ "category": "phonetic_balance",
+ "text": "Nimrod replied, with a slight manifestation of sensitiveness.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0462.wav",
+ "render_seconds": 42.29929517302662,
+ "audio_seconds": 4.667210884353741,
+ "rtf": 9.06307776124492,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "NIMROD REPLIED WITH A SLIGHT MANIFESTATION OF SENSITIVENESS",
+ "raw": {
+ "normalized_reference": "nimrod replied, with a slight manifestation of sensitiveness.",
+ "normalized_hypothesis": "nimrod replied with a slight manifestation of sensitiveness",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "nimrod replied with a slight manifestation of sensitiveness",
+ "normalized_hypothesis": "nimrod replied with a slight manifestation of sensitiveness",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0463",
+ "category": "phonetic_balance",
+ "text": "I tried to read George Moore last night, and was dreadfully bored.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0463.wav",
+ "render_seconds": 52.49574498098809,
+ "audio_seconds": 4.667210884353741,
+ "rtf": 11.247776516157373,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I TRIED TO READ GEORGE MOORE LAST NIGHT AND WAS DREADFULLY BORED",
+ "raw": {
+ "normalized_reference": "i tried to read george moore last night, and was dreadfully bored.",
+ "normalized_hypothesis": "i tried to read george moore last night and was dreadfully bored",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i tried to read george moore last night and was dreadfully bored",
+ "normalized_hypothesis": "i tried to read george moore last night and was dreadfully bored",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0464",
+ "category": "phonetic_balance",
+ "text": "Yes, it was a man who asked, a stranger.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0464.wav",
+ "render_seconds": 30.093770980020054,
+ "audio_seconds": 2.995374149659864,
+ "rtf": 10.046748578449645,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "YES IT WAS A MAN WHO ASKED A STRANGER",
+ "raw": {
+ "normalized_reference": "yes, it was a man who asked, a stranger.",
+ "normalized_hypothesis": "yes it was a man who asked a stranger",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "yes it was a man who asked a stranger",
+ "normalized_hypothesis": "yes it was a man who asked a stranger",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0465",
+ "category": "phonetic_balance",
+ "text": "We were now good friends.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0465.wav",
+ "render_seconds": 39.59023493004497,
+ "audio_seconds": 2.020136054421769,
+ "rtf": 19.597806218738587,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WE AR O GOD PLAMVR",
+ "raw": {
+ "normalized_reference": "we were now good friends.",
+ "normalized_hypothesis": "we ar o god plamvr",
+ "errors": 4,
+ "words": 5,
+ "wer": 0.8
+ },
+ "semantic": {
+ "normalized_reference": "we were now good friends",
+ "normalized_hypothesis": "we ar 0 god plamvr",
+ "errors": 4,
+ "words": 5,
+ "wer": 0.8
+ }
+ },
+ {
+ "id": "release500_0466",
+ "category": "phonetic_balance",
+ "text": "Yet, in accordance with Ernest's test of truth, it worked.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0466.wav",
+ "render_seconds": 39.81952378200367,
+ "audio_seconds": 3.900952380952381,
+ "rtf": 10.207641594507777,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "YET IN ACCORDANCE WITH ERNEST'S TEST OF TRUTH IT WENT",
+ "raw": {
+ "normalized_reference": "yet, in accordance with ernest's test of truth, it worked.",
+ "normalized_hypothesis": "yet in accordance with ernest's test of truth it went",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ },
+ "semantic": {
+ "normalized_reference": "yet in accordance with ernest is test of truth it worked",
+ "normalized_hypothesis": "yet in accordance with ernest is test of truth it went",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ }
+ },
+ {
+ "id": "release500_0467",
+ "category": "phonetic_balance",
+ "text": "It's the strap hangers that'll keep us from going under.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0467.wav",
+ "render_seconds": 66.60312551597599,
+ "audio_seconds": 3.4829931972789114,
+ "rtf": 19.12238173993842,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT'S THE STRAP HANGERS THAT'LL KEEP US FROM GOING UNDER",
+ "raw": {
+ "normalized_reference": "it's the strap hangers that'll keep us from going under.",
+ "normalized_hypothesis": "it's the strap hangers that'll keep us from going under",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it is the strap hangers that will keep us from going under",
+ "normalized_hypothesis": "it is the strap hangers that will keep us from going under",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0468",
+ "category": "phonetic_balance",
+ "text": "I never allow what can't be changed to annoy me.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0468.wav",
+ "render_seconds": 40.40542334411293,
+ "audio_seconds": 3.0650340136054424,
+ "rtf": 13.18269982152122,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I NEVER ALLOW WHAT CAN'T BE CHANGED TO ANNOY ME",
+ "raw": {
+ "normalized_reference": "i never allow what can't be changed to annoy me.",
+ "normalized_hypothesis": "i never allow what can't be changed to annoy me",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i never allow what can not be changed to annoy me",
+ "normalized_hypothesis": "i never allow what can not be changed to annoy me",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0469",
+ "category": "phonetic_balance",
+ "text": "After all superfluous flesh is gone what is left is stringy and resistant.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0469.wav",
+ "render_seconds": 41.10213894897606,
+ "audio_seconds": 5.224489795918367,
+ "rtf": 7.86720628320245,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "AFTER ALL SUPERFLUOUS FLESH IS GONE WHAT IS LEFT IS STRINGY AND RESISTANT",
+ "raw": {
+ "normalized_reference": "after all superfluous flesh is gone what is left is stringy and resistant.",
+ "normalized_hypothesis": "after all superfluous flesh is gone what is left is stringy and resistant",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "after all superfluous flesh is gone what is left is stringy and resistant",
+ "normalized_hypothesis": "after all superfluous flesh is gone what is left is stringy and resistant",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0470",
+ "category": "phonetic_balance",
+ "text": "Then he shouted, Shut up.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0470.wav",
+ "render_seconds": 33.61183233605698,
+ "audio_seconds": 2.089795918367347,
+ "rtf": 16.083786957683515,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "NONO SHOUTED SHOT UP",
+ "raw": {
+ "normalized_reference": "then he shouted, shut up.",
+ "normalized_hypothesis": "nono shouted shot up",
+ "errors": 3,
+ "words": 5,
+ "wer": 0.6
+ },
+ "semantic": {
+ "normalized_reference": "then he shouted shut up",
+ "normalized_hypothesis": "nono shouted shot up",
+ "errors": 3,
+ "words": 5,
+ "wer": 0.6
+ }
+ },
+ {
+ "id": "release500_0471",
+ "category": "phonetic_balance",
+ "text": "He had worshiped her, as Dante might have worshiped Beatrice.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0471.wav",
+ "render_seconds": 62.199409024906345,
+ "audio_seconds": 4.458231292517007,
+ "rtf": 13.95158863321111,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE HAD WORSHIPPED HER AS DANTI MIGHT HAVE WORSHIPPED BEATRICE",
+ "raw": {
+ "normalized_reference": "he had worshiped her, as dante might have worshiped beatrice.",
+ "normalized_hypothesis": "he had worshipped her as danti might have worshipped beatrice",
+ "errors": 3,
+ "words": 10,
+ "wer": 0.3
+ },
+ "semantic": {
+ "normalized_reference": "he had worshiped her as dante might have worshiped beatrice",
+ "normalized_hypothesis": "he had worshipped her as danti might have worshipped beatrice",
+ "errors": 3,
+ "words": 10,
+ "wer": 0.3
+ }
+ },
+ {
+ "id": "release500_0472",
+ "category": "phonetic_balance",
+ "text": "He looked like one who had passed through an uncomfortable hour or two.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0472.wav",
+ "render_seconds": 30.779865423915908,
+ "audio_seconds": 4.876190476190477,
+ "rtf": 6.312277088889004,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE LOOKED LIKE ONE WHO HAD PASSED THROUGH AN UNCOMFORTABLE HOUR OR TWO",
+ "raw": {
+ "normalized_reference": "he looked like one who had passed through an uncomfortable hour or two.",
+ "normalized_hypothesis": "he looked like one who had passed through an uncomfortable hour or two",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he looked like one who had passed through an uncomfortable hour or 2",
+ "normalized_hypothesis": "he looked like one who had passed through an uncomfortable hour or 2",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0473",
+ "category": "phonetic_balance",
+ "text": "The history of our westward-faring race is written in it.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0473.wav",
+ "render_seconds": 56.89556074899156,
+ "audio_seconds": 4.179591836734694,
+ "rtf": 13.61270740576458,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE HISTORY OF OUR WESTWARD FEARING RACE IS WRITTEN IN IT",
+ "raw": {
+ "normalized_reference": "the history of our westward-faring race is written in it.",
+ "normalized_hypothesis": "the history of our westward fearing race is written in it",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ },
+ "semantic": {
+ "normalized_reference": "the history of our westward faring race is written in it",
+ "normalized_hypothesis": "the history of our westward fearing race is written in it",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ }
+ },
+ {
+ "id": "release500_0474",
+ "category": "phonetic_balance",
+ "text": "He may desire to escape pain, or to enjoy pleasure.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0474.wav",
+ "render_seconds": 44.003337262081914,
+ "audio_seconds": 3.8312925170068026,
+ "rtf": 11.485246053845954,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "YOU MAY DESIRE TO ESCAPE PAIN OR TO ENJOY PLEASURE",
+ "raw": {
+ "normalized_reference": "he may desire to escape pain, or to enjoy pleasure.",
+ "normalized_hypothesis": "you may desire to escape pain or to enjoy pleasure",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ },
+ "semantic": {
+ "normalized_reference": "he may desire to escape pain or to enjoy pleasure",
+ "normalized_hypothesis": "you may desire to escape pain or to enjoy pleasure",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ }
+ },
+ {
+ "id": "release500_0475",
+ "category": "phonetic_balance",
+ "text": "And the air was growing chilly.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0475.wav",
+ "render_seconds": 25.10143750894349,
+ "audio_seconds": 2.2987755102040817,
+ "rtf": 10.919481870900489,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "AND THE ARM IS GROWING CHILLY",
+ "raw": {
+ "normalized_reference": "and the air was growing chilly.",
+ "normalized_hypothesis": "and the arm is growing chilly",
+ "errors": 2,
+ "words": 6,
+ "wer": 0.3333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "and the air was growing chilly",
+ "normalized_hypothesis": "and the arm is growing chilly",
+ "errors": 2,
+ "words": 6,
+ "wer": 0.3333333333333333
+ }
+ },
+ {
+ "id": "release500_0476",
+ "category": "phonetic_balance",
+ "text": "Dig in; you're sure good, was Daylight's answer.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0476.wav",
+ "render_seconds": 28.491644060006365,
+ "audio_seconds": 2.9257142857142857,
+ "rtf": 9.738354903322488,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "DIGGIN YOU'RE SURE GOOD WAS DAYLIGHT'S ANSWER",
+ "raw": {
+ "normalized_reference": "dig in; you're sure good, was daylight's answer.",
+ "normalized_hypothesis": "diggin you're sure good was daylight's answer",
+ "errors": 2,
+ "words": 8,
+ "wer": 0.25
+ },
+ "semantic": {
+ "normalized_reference": "dig in you are sure good was daylight is answer",
+ "normalized_hypothesis": "diggin you are sure good was daylight is answer",
+ "errors": 2,
+ "words": 10,
+ "wer": 0.2
+ }
+ },
+ {
+ "id": "release500_0477",
+ "category": "phonetic_balance",
+ "text": "My idea was that he would have more influence over the natives.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0477.wav",
+ "render_seconds": 56.70551932498347,
+ "audio_seconds": 4.458231292517007,
+ "rtf": 12.71928610347377,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "MY IDEA WAS THAT H E WOULD HAVE MORE INFLUENCE OVER THE NATIVES",
+ "raw": {
+ "normalized_reference": "my idea was that he would have more influence over the natives.",
+ "normalized_hypothesis": "my idea was that h e would have more influence over the natives",
+ "errors": 2,
+ "words": 12,
+ "wer": 0.16666666666666666
+ },
+ "semantic": {
+ "normalized_reference": "my idea was that he would have more influence over the natives",
+ "normalized_hypothesis": "my idea was that h e would have more influence over the natives",
+ "errors": 2,
+ "words": 12,
+ "wer": 0.16666666666666666
+ }
+ },
+ {
+ "id": "release500_0478",
+ "category": "phonetic_balance",
+ "text": "Now run along, and tell them to hurry.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0478.wav",
+ "render_seconds": 28.79980022890959,
+ "audio_seconds": 2.786394557823129,
+ "rtf": 10.335865804808861,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "NOW THEN ALOLE AND PRAY THEM TOHOY",
+ "raw": {
+ "normalized_reference": "now run along, and tell them to hurry.",
+ "normalized_hypothesis": "now then alole and pray them tohoy",
+ "errors": 5,
+ "words": 8,
+ "wer": 0.625
+ },
+ "semantic": {
+ "normalized_reference": "now run along and tell them to hurry",
+ "normalized_hypothesis": "now then alole and pray them tohoy",
+ "errors": 5,
+ "words": 8,
+ "wer": 0.625
+ }
+ },
+ {
+ "id": "release500_0479",
+ "category": "phonetic_balance",
+ "text": "I have no idea, replied Philip.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0479.wav",
+ "render_seconds": 35.29261709493585,
+ "audio_seconds": 2.36843537414966,
+ "rtf": 14.901237112119627,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I HAVE NO IDEA REPLIED FIP",
+ "raw": {
+ "normalized_reference": "i have no idea, replied philip.",
+ "normalized_hypothesis": "i have no idea replied fip",
+ "errors": 1,
+ "words": 6,
+ "wer": 0.16666666666666666
+ },
+ "semantic": {
+ "normalized_reference": "i have no idea replied philip",
+ "normalized_hypothesis": "i have no idea replied fip",
+ "errors": 1,
+ "words": 6,
+ "wer": 0.16666666666666666
+ }
+ },
+ {
+ "id": "release500_0480",
+ "category": "phonetic_balance",
+ "text": "For the rest, he was a mere automaton.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0480.wav",
+ "render_seconds": 30.688352441997267,
+ "audio_seconds": 2.786394557823129,
+ "rtf": 11.013642111751949,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "FOR THE REST HE WAS A MERE ADAMAIDEN",
+ "raw": {
+ "normalized_reference": "for the rest, he was a mere automaton.",
+ "normalized_hypothesis": "for the rest he was a mere adamaiden",
+ "errors": 1,
+ "words": 8,
+ "wer": 0.125
+ },
+ "semantic": {
+ "normalized_reference": "for the rest he was a mere automaton",
+ "normalized_hypothesis": "for the rest he was a mere adamaiden",
+ "errors": 1,
+ "words": 8,
+ "wer": 0.125
+ }
+ },
+ {
+ "id": "release500_0481",
+ "category": "phonetic_balance",
+ "text": "Why, the average review is more nauseating than cod liver oil.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0481.wav",
+ "render_seconds": 69.20198716490995,
+ "audio_seconds": 4.388571428571429,
+ "rtf": 15.768681971170887,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WHY THE AVERAGE REVIEW IS MORE NAUSEATING THAN COD LIVER OIL",
+ "raw": {
+ "normalized_reference": "why, the average review is more nauseating than cod liver oil.",
+ "normalized_hypothesis": "why the average review is more nauseating than cod liver oil",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "why the average review is more nauseating than cod liver oil",
+ "normalized_hypothesis": "why the average review is more nauseating than cod liver oil",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0482",
+ "category": "phonetic_balance",
+ "text": "I have been robbed, sir, I amended.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0482.wav",
+ "render_seconds": 31.50197906792164,
+ "audio_seconds": 2.716734693877551,
+ "rtf": 11.595530155710337,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WE HAVE BEEN RABBED FER LI AMMENTED",
+ "raw": {
+ "normalized_reference": "i have been robbed, sir, i amended.",
+ "normalized_hypothesis": "we have been rabbed fer li ammented",
+ "errors": 5,
+ "words": 7,
+ "wer": 0.7142857142857143
+ },
+ "semantic": {
+ "normalized_reference": "i have been robbed sir i amended",
+ "normalized_hypothesis": "we have been rabbed fer li ammented",
+ "errors": 5,
+ "words": 7,
+ "wer": 0.7142857142857143
+ }
+ },
+ {
+ "id": "release500_0483",
+ "category": "phonetic_balance",
+ "text": "He moved away as quietly as he had come.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0483.wav",
+ "render_seconds": 43.591895676916465,
+ "audio_seconds": 2.9257142857142857,
+ "rtf": 14.899573717696057,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE MOVED AWAY AS QUIETLY AS HE HAD COME",
+ "raw": {
+ "normalized_reference": "he moved away as quietly as he had come.",
+ "normalized_hypothesis": "he moved away as quietly as he had come",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he moved away as quietly as he had come",
+ "normalized_hypothesis": "he moved away as quietly as he had come",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0484",
+ "category": "phonetic_balance",
+ "text": "A trickle of fresh blood ran over his face.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0484.wav",
+ "render_seconds": 12.693803622038104,
+ "audio_seconds": 3.2043537414965986,
+ "rtf": 3.9614239394522786,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "A TRICKLE OF FRESH BLOOD RAN OVER HIS FACE",
+ "raw": {
+ "normalized_reference": "a trickle of fresh blood ran over his face.",
+ "normalized_hypothesis": "a trickle of fresh blood ran over his face",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "a trickle of fresh blood ran over his face",
+ "normalized_hypothesis": "a trickle of fresh blood ran over his face",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0485",
+ "category": "phonetic_balance",
+ "text": "Suppose you saw me at work through the window.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0485.wav",
+ "render_seconds": 53.307658123085275,
+ "audio_seconds": 3.0650340136054424,
+ "rtf": 17.392191370946232,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SUPPOSE YOU SAW ME AT WORK TRE THE WINDOW",
+ "raw": {
+ "normalized_reference": "suppose you saw me at work through the window.",
+ "normalized_hypothesis": "suppose you saw me at work tre the window",
+ "errors": 1,
+ "words": 9,
+ "wer": 0.1111111111111111
+ },
+ "semantic": {
+ "normalized_reference": "suppose you saw me at work through the window",
+ "normalized_hypothesis": "suppose you saw me at work tre the window",
+ "errors": 1,
+ "words": 9,
+ "wer": 0.1111111111111111
+ }
+ },
+ {
+ "id": "release500_0486",
+ "category": "phonetic_balance",
+ "text": "But who was Eileen's double.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0486.wav",
+ "render_seconds": 24.200655221007764,
+ "audio_seconds": 2.089795918367347,
+ "rtf": 11.580391658490042,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IL TELL ME EF THEY ARE AN AS DOUBLE",
+ "raw": {
+ "normalized_reference": "but who was eileen's double.",
+ "normalized_hypothesis": "il tell me ef they are an as double",
+ "errors": 8,
+ "words": 5,
+ "wer": 1.6
+ },
+ "semantic": {
+ "normalized_reference": "but who was eileen is double",
+ "normalized_hypothesis": "il tell me ef they are an as double",
+ "errors": 8,
+ "words": 6,
+ "wer": 1.3333333333333333
+ }
+ },
+ {
+ "id": "release500_0487",
+ "category": "phonetic_balance",
+ "text": "Darkness hid him from Jeanne.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0487.wav",
+ "render_seconds": 17.294839818030596,
+ "audio_seconds": 2.2291156462585033,
+ "rtf": 7.7586103919998095,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "DARKNESS LID HIM SANVRIEN",
+ "raw": {
+ "normalized_reference": "darkness hid him from jeanne.",
+ "normalized_hypothesis": "darkness lid him sanvrien",
+ "errors": 3,
+ "words": 5,
+ "wer": 0.6
+ },
+ "semantic": {
+ "normalized_reference": "darkness hid him from jeanne",
+ "normalized_hypothesis": "darkness lid him sanvrien",
+ "errors": 3,
+ "words": 5,
+ "wer": 0.6
+ }
+ },
+ {
+ "id": "release500_0488",
+ "category": "phonetic_balance",
+ "text": "By this time Charley was as enraged as the Greek.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0488.wav",
+ "render_seconds": 42.10388160403818,
+ "audio_seconds": 3.5526530612244898,
+ "rtf": 11.851391306283725,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "BY THIS TIME CHARLEY WAS AS ENRAGED AS THE GREEK",
+ "raw": {
+ "normalized_reference": "by this time charley was as enraged as the greek.",
+ "normalized_hypothesis": "by this time charley was as enraged as the greek",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "by this time charley was as enraged as the greek",
+ "normalized_hypothesis": "by this time charley was as enraged as the greek",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0489",
+ "category": "phonetic_balance",
+ "text": "I am writing these lines in Honolulu, Hawaii.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0489.wav",
+ "render_seconds": 46.409274923964404,
+ "audio_seconds": 3.274013605442177,
+ "rtf": 14.175040337896375,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I AM WRITING THESE LINES IN HALOLU HAWAI",
+ "raw": {
+ "normalized_reference": "i am writing these lines in honolulu, hawaii.",
+ "normalized_hypothesis": "i am writing these lines in halolu hawai",
+ "errors": 2,
+ "words": 8,
+ "wer": 0.25
+ },
+ "semantic": {
+ "normalized_reference": "i am writing these lines in honolulu hawaii",
+ "normalized_hypothesis": "i am writing these lines in halolu hawai",
+ "errors": 2,
+ "words": 8,
+ "wer": 0.25
+ }
+ },
+ {
+ "id": "release500_0490",
+ "category": "phonetic_balance",
+ "text": "Nobody knows how the natives got them.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0490.wav",
+ "render_seconds": 10.605359158013016,
+ "audio_seconds": 2.716734693877551,
+ "rtf": 3.903715435266209,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "TEY BUT ARE HOW THE MEROT GOT THEM",
+ "raw": {
+ "normalized_reference": "nobody knows how the natives got them.",
+ "normalized_hypothesis": "tey but are how the merot got them",
+ "errors": 4,
+ "words": 7,
+ "wer": 0.5714285714285714
+ },
+ "semantic": {
+ "normalized_reference": "nobody knows how the natives got them",
+ "normalized_hypothesis": "tey but are how the merot got them",
+ "errors": 4,
+ "words": 7,
+ "wer": 0.5714285714285714
+ }
+ },
+ {
+ "id": "release500_0491",
+ "category": "phonetic_balance",
+ "text": "And how would we ever find ourselves.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0491.wav",
+ "render_seconds": 49.29280564200599,
+ "audio_seconds": 2.716734693877551,
+ "rtf": 18.14413669214463,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "AND THOW WOULD E HAMPER FAIND OUR STRINS",
+ "raw": {
+ "normalized_reference": "and how would we ever find ourselves.",
+ "normalized_hypothesis": "and thow would e hamper faind our strins",
+ "errors": 6,
+ "words": 7,
+ "wer": 0.8571428571428571
+ },
+ "semantic": {
+ "normalized_reference": "and how would we ever find ourselves",
+ "normalized_hypothesis": "and thow would e hamper faind our strins",
+ "errors": 6,
+ "words": 7,
+ "wer": 0.8571428571428571
+ }
+ },
+ {
+ "id": "release500_0492",
+ "category": "phonetic_balance",
+ "text": "They are coming ashore, whoever they are.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0492.wav",
+ "render_seconds": 34.40752541599795,
+ "audio_seconds": 2.8560544217687074,
+ "rtf": 12.047223314004619,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THEY ARE COMING OF CHEN WHOREVER THEY ARE",
+ "raw": {
+ "normalized_reference": "they are coming ashore, whoever they are.",
+ "normalized_hypothesis": "they are coming of chen whorever they are",
+ "errors": 3,
+ "words": 7,
+ "wer": 0.42857142857142855
+ },
+ "semantic": {
+ "normalized_reference": "they are coming ashore whoever they are",
+ "normalized_hypothesis": "they are coming of chen whorever they are",
+ "errors": 3,
+ "words": 7,
+ "wer": 0.42857142857142855
+ }
+ },
+ {
+ "id": "release500_0493",
+ "category": "phonetic_balance",
+ "text": "This is my fifth voyage.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0493.wav",
+ "render_seconds": 19.597948632086627,
+ "audio_seconds": 2.159455782312925,
+ "rtf": 9.075410940387899,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "BU HAD MY SOT RAD",
+ "raw": {
+ "normalized_reference": "this is my fifth voyage.",
+ "normalized_hypothesis": "bu had my sot rad",
+ "errors": 4,
+ "words": 5,
+ "wer": 0.8
+ },
+ "semantic": {
+ "normalized_reference": "this is my 5th voyage",
+ "normalized_hypothesis": "bu had my sot rad",
+ "errors": 4,
+ "words": 5,
+ "wer": 0.8
+ }
+ },
+ {
+ "id": "release500_0494",
+ "category": "phonetic_balance",
+ "text": "And yet if she came he had no words to say.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0494.wav",
+ "render_seconds": 42.38802839803975,
+ "audio_seconds": 2.9257142857142857,
+ "rtf": 14.488095643861243,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "AND YET IF SHE CAME HE HAD NO WORDS TO SAY",
+ "raw": {
+ "normalized_reference": "and yet if she came he had no words to say.",
+ "normalized_hypothesis": "and yet if she came he had no words to say",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "and yet if she came he had no words to say",
+ "normalized_hypothesis": "and yet if she came he had no words to say",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0495",
+ "category": "phonetic_balance",
+ "text": "You must sleep, he urged.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0495.wav",
+ "render_seconds": 24.098106320016086,
+ "audio_seconds": 2.020136054421769,
+ "rtf": 11.928952145212703,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I MUST SRAP LARNED",
+ "raw": {
+ "normalized_reference": "you must sleep, he urged.",
+ "normalized_hypothesis": "i must srap larned",
+ "errors": 4,
+ "words": 5,
+ "wer": 0.8
+ },
+ "semantic": {
+ "normalized_reference": "you must sleep he urged",
+ "normalized_hypothesis": "i must srap larned",
+ "errors": 4,
+ "words": 5,
+ "wer": 0.8
+ }
+ },
+ {
+ "id": "release500_0496",
+ "category": "phonetic_balance",
+ "text": "Could the incident have anything to do with Jeanne and Pierre.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0496.wav",
+ "render_seconds": 14.69217577308882,
+ "audio_seconds": 4.179591836734694,
+ "rtf": 3.515217836334727,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "COULD THE INCIDENT HAVE ANYTHING TO DO WITH JAN AND PIERRE",
+ "raw": {
+ "normalized_reference": "could the incident have anything to do with jeanne and pierre.",
+ "normalized_hypothesis": "could the incident have anything to do with jan and pierre",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ },
+ "semantic": {
+ "normalized_reference": "could the incident have anything to do with jeanne and pierre",
+ "normalized_hypothesis": "could the incident have anything to do with jan and pierre",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ }
+ },
+ {
+ "id": "release500_0497",
+ "category": "phonetic_balance",
+ "text": "Each day she became a more vital part of him.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0497.wav",
+ "render_seconds": 58.90309573803097,
+ "audio_seconds": 3.274013605442177,
+ "rtf": 17.991096811607697,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "EACH DAY SHE BECAME A MORE VITAL PART OF HIM",
+ "raw": {
+ "normalized_reference": "each day she became a more vital part of him.",
+ "normalized_hypothesis": "each day she became a more vital part of him",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "each day she became a more vital part of him",
+ "normalized_hypothesis": "each day she became a more vital part of him",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0498",
+ "category": "phonetic_balance",
+ "text": "But why continue the tirade, for tirade it was.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0498.wav",
+ "render_seconds": 31.699932329007424,
+ "audio_seconds": 3.3436734693877552,
+ "rtf": 9.480570581795433,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "BUT WHY CONTINUE THE TERRID FOR TIRED IT WAS",
+ "raw": {
+ "normalized_reference": "but why continue the tirade, for tirade it was.",
+ "normalized_hypothesis": "but why continue the terrid for tired it was",
+ "errors": 2,
+ "words": 9,
+ "wer": 0.2222222222222222
+ },
+ "semantic": {
+ "normalized_reference": "but why continue the tirade for tirade it was",
+ "normalized_hypothesis": "but why continue the terrid for tired it was",
+ "errors": 2,
+ "words": 9,
+ "wer": 0.2222222222222222
+ }
+ },
+ {
+ "id": "release500_0499",
+ "category": "phonetic_balance",
+ "text": "That is what distinguishes all of us from the lower animals.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0499.wav",
+ "render_seconds": 55.103952111909166,
+ "audio_seconds": 4.31891156462585,
+ "rtf": 12.758759073290461,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THAT IS WHAT DISTINGUISHES ALL OF US FROM THE LOWER ANIMALS",
+ "raw": {
+ "normalized_reference": "that is what distinguishes all of us from the lower animals.",
+ "normalized_hypothesis": "that is what distinguishes all of us from the lower animals",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "that is what distinguishes all of us from the lower animals",
+ "normalized_hypothesis": "that is what distinguishes all of us from the lower animals",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0500",
+ "category": "phonetic_balance",
+ "text": "There was something pathetic in the girl's attitude now.",
+ "model": "supertonic3-james-3step",
+ "audio": "benchmark/audio/supertonic3-james-3step/release500_0500.wav",
+ "render_seconds": 39.00000892498065,
+ "audio_seconds": 3.622312925170068,
+ "rtf": 10.766604026390015,
+ "voice": "M2",
+ "steps": 3,
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THERE WAS SOMETHING PATHETIC IN THE GIRL'S ATTITUDE NOW",
+ "raw": {
+ "normalized_reference": "there was something pathetic in the girl's attitude now.",
+ "normalized_hypothesis": "there was something pathetic in the girl's attitude now",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "there was something pathetic in the girl is attitude now",
+ "normalized_hypothesis": "there was something pathetic in the girl is attitude now",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0001",
+ "category": "conversational",
+ "text": "It wasn't until later that I realized what had actually happened.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0001.wav",
+ "render_seconds": 144.48957702307962,
+ "audio_seconds": 4.31891156462585,
+ "rtf": 33.455090446057056,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT WASN'T UNTIL LATER THAT I REALIZED WHAT HAD ACTUALLY HAPPENED",
+ "raw": {
+ "normalized_reference": "it wasn't until later that i realized what had actually happened.",
+ "normalized_hypothesis": "it wasn't until later that i realized what had actually happened",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it was not until later that i realized what had actually happened",
+ "normalized_hypothesis": "it was not until later that i realized what had actually happened",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0002",
+ "category": "descriptive",
+ "text": "The light at that time of afternoon is perfect for about twenty minutes.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0002.wav",
+ "render_seconds": 170.39593176799826,
+ "audio_seconds": 4.806530612244898,
+ "rtf": 35.45091990757437,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE LIGHT AT THAT TIME OF AFTERNOON IS PERFECT FOR ABOUT TWENTY MINUTES",
+ "raw": {
+ "normalized_reference": "the light at that time of afternoon is perfect for about twenty minutes.",
+ "normalized_hypothesis": "the light at that time of afternoon is perfect for about twenty minutes",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the light at that time of afternoon is perfect for about 20 minutes",
+ "normalized_hypothesis": "the light at that time of afternoon is perfect for about 20 minutes",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0003",
+ "category": "emotional",
+ "text": "That was the scariest thing I've ever done, and also maybe the best.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0003.wav",
+ "render_seconds": 179.49796744505875,
+ "audio_seconds": 4.527891156462585,
+ "rtf": 39.64273018993935,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THAT WAS THE SCARIEST THING I'VE EVER DONE AND ALSO MAYBE THE BEST",
+ "raw": {
+ "normalized_reference": "that was the scariest thing i've ever done, and also maybe the best.",
+ "normalized_hypothesis": "that was the scariest thing i've ever done and also maybe the best",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "that was the scariest thing i have ever done and also maybe the best",
+ "normalized_hypothesis": "that was the scariest thing i have ever done and also maybe the best",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0004",
+ "category": "homographs",
+ "text": "After I record the record, please present the present to the bass player.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0004.wav",
+ "render_seconds": 157.99379248300102,
+ "audio_seconds": 5.015510204081632,
+ "rtf": 31.50104098171814,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "AFTER I RECORD THE RECORD PLEASE PRESENT THE PRESENT TO THE BASE PLAYER",
+ "raw": {
+ "normalized_reference": "after i record the record, please present the present to the bass player.",
+ "normalized_hypothesis": "after i record the record please present the present to the base player",
+ "errors": 1,
+ "words": 13,
+ "wer": 0.07692307692307693
+ },
+ "semantic": {
+ "normalized_reference": "after i record the record please present the present to the bass player",
+ "normalized_hypothesis": "after i record the record please present the present to the base player",
+ "errors": 1,
+ "words": 13,
+ "wer": 0.07692307692307693
+ }
+ },
+ {
+ "id": "release500_0005",
+ "category": "long",
+ "text": "If I'm being completely honest, I had no idea what I was doing for most of it, and I think that might be the most normal thing about the whole experience.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0005.wav",
+ "render_seconds": 302.5043903089827,
+ "audio_seconds": 9.47374149659864,
+ "rtf": 31.930825895723558,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IF I'M BEING COMPLETELY HONEST I HAD NO IDEA WHAT I WAS DOING FOR MOST OF IT AND I THINK THAT MIGHT BE THE MOST NORMAL THING ABOUT THE WHOLE EXPERIENCE",
+ "raw": {
+ "normalized_reference": "if i'm being completely honest, i had no idea what i was doing for most of it, and i think that might be the most normal thing about the whole experience.",
+ "normalized_hypothesis": "if i'm being completely honest i had no idea what i was doing for most of it and i think that might be the most normal thing about the whole experience",
+ "errors": 0,
+ "words": 31,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "if i am being completely honest i had no idea what i was doing for most of it and i think that might be the most normal thing about the whole experience",
+ "normalized_hypothesis": "if i am being completely honest i had no idea what i was doing for most of it and i think that might be the most normal thing about the whole experience",
+ "errors": 0,
+ "words": 32,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0006",
+ "category": "long_range",
+ "text": "The committee postponed its decision after discovering that three pages of the report had been printed in the wrong order.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0006.wav",
+ "render_seconds": 257.10046293900814,
+ "audio_seconds": 7.9412244897959186,
+ "rtf": 32.37541808185495,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE COMMITTEE POSTPONED ITS DECISION AFTER DISCOVERING THAT THREE PAGES OF THE REPORT HAD BEEN PRINTED IN THE WRONG ORDER",
+ "raw": {
+ "normalized_reference": "the committee postponed its decision after discovering that three pages of the report had been printed in the wrong order.",
+ "normalized_hypothesis": "the committee postponed its decision after discovering that three pages of the report had been printed in the wrong order",
+ "errors": 0,
+ "words": 20,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the committee postponed its decision after discovering that 3 pages of the report had been printed in the wrong order",
+ "normalized_hypothesis": "the committee postponed its decision after discovering that 3 pages of the report had been printed in the wrong order",
+ "errors": 0,
+ "words": 20,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0007",
+ "category": "names_places",
+ "text": "Gwendolyn photographed the eucalyptus trees outside Ljubljana.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0007.wav",
+ "render_seconds": 175.80659322696738,
+ "audio_seconds": 4.527891156462585,
+ "rtf": 38.82747777097987,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "GWENDOLEN PHOTOGRAPHED THE EUCALYPTUS TREES OUTSIDE JUBAJANA",
+ "raw": {
+ "normalized_reference": "gwendolyn photographed the eucalyptus trees outside ljubljana.",
+ "normalized_hypothesis": "gwendolen photographed the eucalyptus trees outside jubajana",
+ "errors": 2,
+ "words": 7,
+ "wer": 0.2857142857142857
+ },
+ "semantic": {
+ "normalized_reference": "gwendolyn photographed the eucalyptus trees outside ljubljana",
+ "normalized_hypothesis": "gwendolen photographed the eucalyptus trees outside jubajana",
+ "errors": 2,
+ "words": 7,
+ "wer": 0.2857142857142857
+ }
+ },
+ {
+ "id": "release500_0008",
+ "category": "narrative",
+ "text": "He had this theory, and the more time passes, the more I think he was right.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0008.wav",
+ "render_seconds": 173.3916494640289,
+ "audio_seconds": 5.294149659863946,
+ "rtf": 32.751557965681855,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE HAD THIS THEORY AND THE MORE TIME PASSES THE MORE I THINK HE WAS RIGHT",
+ "raw": {
+ "normalized_reference": "he had this theory, and the more time passes, the more i think he was right.",
+ "normalized_hypothesis": "he had this theory and the more time passes the more i think he was right",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he had this theory and the more time passes the more i think he was right",
+ "normalized_hypothesis": "he had this theory and the more time passes the more i think he was right",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0009",
+ "category": "numbers",
+ "text": "The package weighs twelve point six kilograms and arrived on July twenty-first.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0009.wav",
+ "render_seconds": 205.40336752007715,
+ "audio_seconds": 5.7817687074829935,
+ "rtf": 35.52604365758111,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE PACKAGE WEIGHS TWELVE POINT SIX KILOGRAMS AND ARRIVED ON JULY TWENTY FIRST",
+ "raw": {
+ "normalized_reference": "the package weighs twelve point six kilograms and arrived on july twenty-first.",
+ "normalized_hypothesis": "the package weighs twelve point six kilograms and arrived on july twenty first",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the package weighs 12.6 kilograms and arrived on july 21st",
+ "normalized_hypothesis": "the package weighs 12.6 kilograms and arrived on july 21st",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0010",
+ "category": "punctuation",
+ "text": "First, close the window; second, turn off the lamp; finally, lock the door.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0010.wav",
+ "render_seconds": 188.7054866560502,
+ "audio_seconds": 5.224489795918367,
+ "rtf": 36.11940955525961,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "FIRST CLOSE THE WINDOW SECOND TURN OFF THE LAMP FINALLY LOCK THE DOOR",
+ "raw": {
+ "normalized_reference": "first, close the window; second, turn off the lamp; finally, lock the door.",
+ "normalized_hypothesis": "first close the window second turn off the lamp finally lock the door",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "1st close the window 2nd turn off the lamp finally lock the door",
+ "normalized_hypothesis": "1st close the window 2nd turn off the lamp finally lock the door",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0011",
+ "category": "question",
+ "text": "When did you first realize something was wrong?",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0011.wav",
+ "render_seconds": 126.30049753002822,
+ "audio_seconds": 3.1346938775510202,
+ "rtf": 40.29117434226161,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WHEN DID YOU FIRST REALIZE SOMETHING WAS WRONG",
+ "raw": {
+ "normalized_reference": "when did you first realize something was wrong?",
+ "normalized_hypothesis": "when did you first realize something was wrong",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "when did you 1st realize something was wrong",
+ "normalized_hypothesis": "when did you 1st realize something was wrong",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0012",
+ "category": "rare_words",
+ "text": "The archaeologist catalogued a Byzantine reliquary and several anachronistic fragments.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0012.wav",
+ "render_seconds": 207.30320841202047,
+ "audio_seconds": 6.478367346938776,
+ "rtf": 31.99929817347846,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE ARCHEOLOGIST CATALOGUED A BISANTINE RELIQUARY AND SEVERAL ANACRONISTIC FRAGMENTS",
+ "raw": {
+ "normalized_reference": "the archaeologist catalogued a byzantine reliquary and several anachronistic fragments.",
+ "normalized_hypothesis": "the archeologist catalogued a bisantine reliquary and several anacronistic fragments",
+ "errors": 3,
+ "words": 10,
+ "wer": 0.3
+ },
+ "semantic": {
+ "normalized_reference": "the archaeologist catalogued a byzantine reliquary and several anachronistic fragments",
+ "normalized_hypothesis": "the archeologist catalogued a bisantine reliquary and several anacronistic fragments",
+ "errors": 3,
+ "words": 10,
+ "wer": 0.3
+ }
+ },
+ {
+ "id": "release500_0013",
+ "category": "short",
+ "text": "I have no idea.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0013.wav",
+ "render_seconds": 82.4928001590306,
+ "audio_seconds": 1.5325170068027212,
+ "rtf": 53.82830976286176,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I HAVE NO IDEA",
+ "raw": {
+ "normalized_reference": "i have no idea.",
+ "normalized_hypothesis": "i have no idea",
+ "errors": 0,
+ "words": 4,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i have no idea",
+ "normalized_hypothesis": "i have no idea",
+ "errors": 0,
+ "words": 4,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0014",
+ "category": "technical",
+ "text": "The system runs on three core components that all have to stay in sync.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0014.wav",
+ "render_seconds": 144.5139470390277,
+ "audio_seconds": 4.9458503401360545,
+ "rtf": 29.219231699407285,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE SYSTEM RUNS ON THREE CORPS COMPONENTS THAT ALL HAVE TO STAY AND SINK",
+ "raw": {
+ "normalized_reference": "the system runs on three core components that all have to stay in sync.",
+ "normalized_hypothesis": "the system runs on three corps components that all have to stay and sink",
+ "errors": 3,
+ "words": 14,
+ "wer": 0.21428571428571427
+ },
+ "semantic": {
+ "normalized_reference": "the system runs on 3 core components that all have to stay in sync",
+ "normalized_hypothesis": "the system runs on 3 corps components that all have to stay and sink",
+ "errors": 3,
+ "words": 14,
+ "wer": 0.21428571428571427
+ }
+ },
+ {
+ "id": "release500_0015",
+ "category": "conversational",
+ "text": "We improvised the entire second half and somehow nobody noticed.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0015.wav",
+ "render_seconds": 153.60720988898538,
+ "audio_seconds": 4.597551020408163,
+ "rtf": 33.41065911115183,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WE IMPROVISED THE ENTIRE SECOND HALF AND SOMEHOW NOBODY NOTICED",
+ "raw": {
+ "normalized_reference": "we improvised the entire second half and somehow nobody noticed.",
+ "normalized_hypothesis": "we improvised the entire second half and somehow nobody noticed",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "we improvised the entire 2nd half and somehow nobody noticed",
+ "normalized_hypothesis": "we improvised the entire 2nd half and somehow nobody noticed",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0016",
+ "category": "descriptive",
+ "text": "There's a smell in old buildings that's impossible to describe but immediately familiar.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0016.wav",
+ "render_seconds": 205.70407292200252,
+ "audio_seconds": 5.851428571428571,
+ "rtf": 35.15450464975629,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THERE'S A SMELL IN OLD BUILDINGS THAT'S IMPOSSIBLE TO DESCRIBE BUT IMMEDIATELY FAMILIAR",
+ "raw": {
+ "normalized_reference": "there's a smell in old buildings that's impossible to describe but immediately familiar.",
+ "normalized_hypothesis": "there's a smell in old buildings that's impossible to describe but immediately familiar",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "there is a smell in old buildings that is impossible to describe but immediately familiar",
+ "normalized_hypothesis": "there is a smell in old buildings that is impossible to describe but immediately familiar",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0017",
+ "category": "emotional",
+ "text": "I specifically asked for one thing. One thing. And it still didn't happen.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0017.wav",
+ "render_seconds": 191.3965175850317,
+ "audio_seconds": 5.712108843537415,
+ "rtf": 33.50715520793595,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I SPECIFICALLY ASKED FOR ONE THING ONE THING AND IT STILL DIDN'T HAPPEN",
+ "raw": {
+ "normalized_reference": "i specifically asked for one thing. one thing. and it still didn't happen.",
+ "normalized_hypothesis": "i specifically asked for one thing one thing and it still didn't happen",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i specifically asked for one thing one thing and it still did not happen",
+ "normalized_hypothesis": "i specifically asked for one thing one thing and it still did not happen",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0018",
+ "category": "homographs",
+ "text": "They refuse to place the refuse beside the produce aisle.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0018.wav",
+ "render_seconds": 133.39925820799544,
+ "audio_seconds": 3.9706122448979593,
+ "rtf": 33.59664706101859,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THEY REFUSED TO PLACE THE REFUSE BESIDE THE PRODUCE ILE",
+ "raw": {
+ "normalized_reference": "they refuse to place the refuse beside the produce aisle.",
+ "normalized_hypothesis": "they refused to place the refuse beside the produce ile",
+ "errors": 2,
+ "words": 10,
+ "wer": 0.2
+ },
+ "semantic": {
+ "normalized_reference": "they refuse to place the refuse beside the produce aisle",
+ "normalized_hypothesis": "they refused to place the refuse beside the produce ile",
+ "errors": 2,
+ "words": 10,
+ "wer": 0.2
+ }
+ },
+ {
+ "id": "release500_0019",
+ "category": "long",
+ "text": "There were days last year when I genuinely didn't know how I was going to get through it, and somehow I did, and I still don't fully understand how.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0019.wav",
+ "render_seconds": 256.6061690719798,
+ "audio_seconds": 8.916462585034013,
+ "rtf": 28.778920634140803,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THERE WERE DAYS LAST YEAR WHEN I GENUINELY DIDN'T KNOW HOW I WAS GOING TO GET THROUGH IT AND SOMEHOW I DID AND I STILL DON'T FULLY UNDERSTAND HOW",
+ "raw": {
+ "normalized_reference": "there were days last year when i genuinely didn't know how i was going to get through it, and somehow i did, and i still don't fully understand how.",
+ "normalized_hypothesis": "there were days last year when i genuinely didn't know how i was going to get through it and somehow i did and i still don't fully understand how",
+ "errors": 0,
+ "words": 29,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "there were days last year when i genuinely did not know how i was going to get through it and somehow i did and i still do not fully understand how",
+ "normalized_hypothesis": "there were days last year when i genuinely did not know how i was going to get through it and somehow i did and i still do not fully understand how",
+ "errors": 0,
+ "words": 31,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0020",
+ "category": "long_range",
+ "text": "Before announcing the winner, the judge thanked the volunteers who had worked through the storm to repair the damaged stage.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0020.wav",
+ "render_seconds": 195.6878811479546,
+ "audio_seconds": 7.801904761904762,
+ "rtf": 25.082064844403362,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "BEFORE ANNOUNCING THE WINNER THE JUDGE THANKED THE VOLUNTEERS WHO HAD WORKED THROUGH THE STORM TO REPAIR THE DAMAGED STAGE",
+ "raw": {
+ "normalized_reference": "before announcing the winner, the judge thanked the volunteers who had worked through the storm to repair the damaged stage.",
+ "normalized_hypothesis": "before announcing the winner the judge thanked the volunteers who had worked through the storm to repair the damaged stage",
+ "errors": 0,
+ "words": 20,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "before announcing the winner the judge thanked the volunteers who had worked through the storm to repair the damaged stage",
+ "normalized_hypothesis": "before announcing the winner the judge thanked the volunteers who had worked through the storm to repair the damaged stage",
+ "errors": 0,
+ "words": 20,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0021",
+ "category": "names_places",
+ "text": "Professor Nguyen mailed the fluorescent poster to Reykjavik on Thursday.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0021.wav",
+ "render_seconds": 169.68825406499673,
+ "audio_seconds": 5.154829931972789,
+ "rtf": 32.9183030874611,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "PROFESSOR GWYNN MAILED THE FLORESCENT POSTER TO RAIKHJEVIK ON THURSDAY",
+ "raw": {
+ "normalized_reference": "professor nguyen mailed the fluorescent poster to reykjavik on thursday.",
+ "normalized_hypothesis": "professor gwynn mailed the florescent poster to raikhjevik on thursday",
+ "errors": 3,
+ "words": 10,
+ "wer": 0.3
+ },
+ "semantic": {
+ "normalized_reference": "professor nguyen mailed the fluorescent poster to reykjavik on thursday",
+ "normalized_hypothesis": "professor gwynn mailed the florescent poster to raikhjevik on thursday",
+ "errors": 3,
+ "words": 10,
+ "wer": 0.3
+ }
+ },
+ {
+ "id": "release500_0022",
+ "category": "narrative",
+ "text": "The plan was simple. The execution was anything but.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0022.wav",
+ "render_seconds": 151.58793161890935,
+ "audio_seconds": 3.9706122448979593,
+ "rtf": 38.17747044267351,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE PLAN WAS SIMPLE THE EXECUTION WAS ANYTHING BUT",
+ "raw": {
+ "normalized_reference": "the plan was simple. the execution was anything but.",
+ "normalized_hypothesis": "the plan was simple the execution was anything but",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the plan was simple the execution was anything but",
+ "normalized_hypothesis": "the plan was simple the execution was anything but",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0023",
+ "category": "numbers",
+ "text": "Her score improved from eighty-two percent to ninety-seven percent in two weeks.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0023.wav",
+ "render_seconds": 199.00377235002816,
+ "audio_seconds": 5.50312925170068,
+ "rtf": 36.16192955826511,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HER SCORE IMPROVED FROM EIGHTY TWO PER CENT TO NINETY SEVEN PER CENT IN TWO WEEKS",
+ "raw": {
+ "normalized_reference": "her score improved from eighty-two percent to ninety-seven percent in two weeks.",
+ "normalized_hypothesis": "her score improved from eighty two per cent to ninety seven per cent in two weeks",
+ "errors": 4,
+ "words": 14,
+ "wer": 0.2857142857142857
+ },
+ "semantic": {
+ "normalized_reference": "her score improved from 82% to 97% in 2 weeks",
+ "normalized_hypothesis": "her score improved from 82% to 97% in 2 weeks",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0024",
+ "category": "punctuation",
+ "text": "Why would anyone pack spoons, batteries, and a violin for one short trip?",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0024.wav",
+ "render_seconds": 189.39552384999115,
+ "audio_seconds": 5.294149659863946,
+ "rtf": 35.774493737084576,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WHY WOULD ANYONE PACK SPOONS BATTERIES AND A VIOLIN FOR ONE SHORT TRIP",
+ "raw": {
+ "normalized_reference": "why would anyone pack spoons, batteries, and a violin for one short trip?",
+ "normalized_hypothesis": "why would anyone pack spoons batteries and a violin for one short trip",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "why would anyone pack spoons batteries and a violin for one short trip",
+ "normalized_hypothesis": "why would anyone pack spoons batteries and a violin for one short trip",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0025",
+ "category": "question",
+ "text": "How long have you been sitting on this?",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0025.wav",
+ "render_seconds": 113.7015101199504,
+ "audio_seconds": 2.5774149659863945,
+ "rtf": 44.114553387966396,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HOW LONG HAVE YOU BEEN SITTING ON THIS",
+ "raw": {
+ "normalized_reference": "how long have you been sitting on this?",
+ "normalized_hypothesis": "how long have you been sitting on this",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "how long have you been sitting on this",
+ "normalized_hypothesis": "how long have you been sitting on this",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0026",
+ "category": "rare_words",
+ "text": "A mischievous chameleon rested beside the chrysanthemum and the bougainvillea.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0026.wav",
+ "render_seconds": 150.89934963197447,
+ "audio_seconds": 5.572789115646258,
+ "rtf": 27.077886225464173,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "A MISCHIEVOUS CHAMELION RESTED BESIDE THE CHRYSANTHEMUM AND THE BAUGIN VILIA",
+ "raw": {
+ "normalized_reference": "a mischievous chameleon rested beside the chrysanthemum and the bougainvillea.",
+ "normalized_hypothesis": "a mischievous chamelion rested beside the chrysanthemum and the baugin vilia",
+ "errors": 3,
+ "words": 10,
+ "wer": 0.3
+ },
+ "semantic": {
+ "normalized_reference": "a mischievous chameleon rested beside the chrysanthemum and the bougainvillea",
+ "normalized_hypothesis": "a mischievous chamelion rested beside the chrysanthemum and the baugin vilia",
+ "errors": 3,
+ "words": 10,
+ "wer": 0.3
+ }
+ },
+ {
+ "id": "release500_0027",
+ "category": "short",
+ "text": "We don't talk about that.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0027.wav",
+ "render_seconds": 87.49977569805924,
+ "audio_seconds": 1.9504761904761905,
+ "rtf": 44.86072484519639,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WE DON'T TALK ABOUT THAT",
+ "raw": {
+ "normalized_reference": "we don't talk about that.",
+ "normalized_hypothesis": "we don't talk about that",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "we do not talk about that",
+ "normalized_hypothesis": "we do not talk about that",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0028",
+ "category": "technical",
+ "text": "Hash the UTF eight payload with S H A two fifty-six before opening the checkpoint.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0028.wav",
+ "render_seconds": 188.11445014097262,
+ "audio_seconds": 5.7817687074829935,
+ "rtf": 32.535796511110426,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HASH THE U T F PAY LOAD WITH S H A TWO FIFTY SIX BEFORE OPENING THE CHECK POINT",
+ "raw": {
+ "normalized_reference": "hash the utf eight payload with s h a two fifty-six before opening the checkpoint.",
+ "normalized_hypothesis": "hash the u t f pay load with s h a two fifty six before opening the check point",
+ "errors": 7,
+ "words": 16,
+ "wer": 0.4375
+ },
+ "semantic": {
+ "normalized_reference": "hash the utf 8 payload with s h a 256 before opening the checkpoint",
+ "normalized_hypothesis": "hash the u t f pay load with s h a 256 before opening the check point",
+ "errors": 7,
+ "words": 14,
+ "wer": 0.5
+ }
+ },
+ {
+ "id": "release500_0029",
+ "category": "conversational",
+ "text": "I thought I had it all figured out, and then this happened.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0029.wav",
+ "render_seconds": 158.59538096003234,
+ "audio_seconds": 4.179591836734694,
+ "rtf": 37.94518392110149,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I THOUGHT I HAD IT ALL FIGURED OUT AND THEN THIS HAPPENED",
+ "raw": {
+ "normalized_reference": "i thought i had it all figured out, and then this happened.",
+ "normalized_hypothesis": "i thought i had it all figured out and then this happened",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i thought i had it all figured out and then this happened",
+ "normalized_hypothesis": "i thought i had it all figured out and then this happened",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0030",
+ "category": "descriptive",
+ "text": "The kind of quiet you only get at three in the morning in a city.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0030.wav",
+ "render_seconds": 166.09637280600145,
+ "audio_seconds": 4.179591836734694,
+ "rtf": 39.73985482174839,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE KIND OF QUIET YOU ONLY GET AT THREE IN THE MORNING IN A CITY",
+ "raw": {
+ "normalized_reference": "the kind of quiet you only get at three in the morning in a city.",
+ "normalized_hypothesis": "the kind of quiet you only get at three in the morning in a city",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the kind of quiet you only get at 3 in the morning in a city",
+ "normalized_hypothesis": "the kind of quiet you only get at 3 in the morning in a city",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0031",
+ "category": "emotional",
+ "text": "The joy of it was almost overwhelming, like too much good all at once.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0031.wav",
+ "render_seconds": 193.0979104069993,
+ "audio_seconds": 4.876190476190477,
+ "rtf": 39.600157407685394,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE JOY OF IT WAS ALMOST OVERWHELMING LIKE TOO MUCH GOOD ALL AT ONCE",
+ "raw": {
+ "normalized_reference": "the joy of it was almost overwhelming, like too much good all at once.",
+ "normalized_hypothesis": "the joy of it was almost overwhelming like too much good all at once",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the joy of it was almost overwhelming like too much good all at once",
+ "normalized_hypothesis": "the joy of it was almost overwhelming like too much good all at once",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0032",
+ "category": "homographs",
+ "text": "Please close the door while I stand close to the microphone.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0032.wav",
+ "render_seconds": 93.71202471398283,
+ "audio_seconds": 4.1099319727891155,
+ "rtf": 22.80135664882726,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "PLEASE CLOSE THE DOOR WHILE I STAND CLOSE TO THE MICROPHONE",
+ "raw": {
+ "normalized_reference": "please close the door while i stand close to the microphone.",
+ "normalized_hypothesis": "please close the door while i stand close to the microphone",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "please close the door while i stand close to the microphone",
+ "normalized_hypothesis": "please close the door while i stand close to the microphone",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0033",
+ "category": "long",
+ "text": "I think the most underrated skill in the world is knowing when to stop explaining yourself and just let people draw their own conclusions.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0033.wav",
+ "render_seconds": 270.5096920909127,
+ "audio_seconds": 8.916462585034013,
+ "rtf": 30.338229932681404,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I THINK THE MOST UNDERRATED SKILL IN THE WORLD IS KNOWING WHEN TO STOP EXPLAINING YOURSELF AND JUST LET PEOPLE DRAW THEIR OWN CONCLUSIONS",
+ "raw": {
+ "normalized_reference": "i think the most underrated skill in the world is knowing when to stop explaining yourself and just let people draw their own conclusions.",
+ "normalized_hypothesis": "i think the most underrated skill in the world is knowing when to stop explaining yourself and just let people draw their own conclusions",
+ "errors": 0,
+ "words": 24,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i think the most underrated skill in the world is knowing when to stop explaining yourself and just let people draw their own conclusions",
+ "normalized_hypothesis": "i think the most underrated skill in the world is knowing when to stop explaining yourself and just let people draw their own conclusions",
+ "errors": 0,
+ "words": 24,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0034",
+ "category": "long_range",
+ "text": "When the final bus disappeared around the corner, Nadia remembered that her umbrella, notebook, and lunch were still inside the quiet office.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0034.wav",
+ "render_seconds": 267.98907835106365,
+ "audio_seconds": 9.404081632653062,
+ "rtf": 28.497102515629596,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WHEN THE FINAL BUST DISAPPEARED AROUND THE CORNER NEDIA REMEMBERED THAT HER UMBRELLA NOTE BOOK AND LUNCH WERE STILL INSIDE THE QUIET OFFICE",
+ "raw": {
+ "normalized_reference": "when the final bus disappeared around the corner, nadia remembered that her umbrella, notebook, and lunch were still inside the quiet office.",
+ "normalized_hypothesis": "when the final bust disappeared around the corner nedia remembered that her umbrella note book and lunch were still inside the quiet office",
+ "errors": 4,
+ "words": 22,
+ "wer": 0.18181818181818182
+ },
+ "semantic": {
+ "normalized_reference": "when the final bus disappeared around the corner nadia remembered that her umbrella notebook and lunch were still inside the quiet office",
+ "normalized_hypothesis": "when the final bust disappeared around the corner nedia remembered that her umbrella note book and lunch were still inside the quiet office",
+ "errors": 4,
+ "words": 22,
+ "wer": 0.18181818181818182
+ }
+ },
+ {
+ "id": "release500_0035",
+ "category": "names_places",
+ "text": "Aurelia and Joaquin crossed the Schuylkill River before noon.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0035.wav",
+ "render_seconds": 144.99596785893664,
+ "audio_seconds": 4.31891156462585,
+ "rtf": 33.57234008830596,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "ARALIA AND JOQUIN CROSSED THE SHOELKILL RIVER BEFORE NOON",
+ "raw": {
+ "normalized_reference": "aurelia and joaquin crossed the schuylkill river before noon.",
+ "normalized_hypothesis": "aralia and joquin crossed the shoelkill river before noon",
+ "errors": 3,
+ "words": 9,
+ "wer": 0.3333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "aurelia and joaquin crossed the schuylkill river before noon",
+ "normalized_hypothesis": "aralia and joquin crossed the shoelkill river before noon",
+ "errors": 3,
+ "words": 9,
+ "wer": 0.3333333333333333
+ }
+ },
+ {
+ "id": "release500_0036",
+ "category": "narrative",
+ "text": "The village had a name, but nobody used it. Everyone just called it home.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0036.wav",
+ "render_seconds": 188.30701990006492,
+ "audio_seconds": 5.224489795918367,
+ "rtf": 36.0431405277468,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE VILLAGE HAD A NAME BUT NOBODY USED IT EVERYONE JUST CALLED IT HOME",
+ "raw": {
+ "normalized_reference": "the village had a name, but nobody used it. everyone just called it home.",
+ "normalized_hypothesis": "the village had a name but nobody used it everyone just called it home",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the village had a name but nobody used it everyone just called it home",
+ "normalized_hypothesis": "the village had a name but nobody used it everyone just called it home",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0037",
+ "category": "numbers",
+ "text": "Increase the gain by three decibels, wait five seconds, then reduce it by one.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0037.wav",
+ "render_seconds": 191.00252298009582,
+ "audio_seconds": 5.50312925170068,
+ "rtf": 34.70798417483446,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "INCREASE THE GAIN BY THREE DISSIPLES WAIT FIVE SECONDS THEN REDUCE IT BY ONE",
+ "raw": {
+ "normalized_reference": "increase the gain by three decibels, wait five seconds, then reduce it by one.",
+ "normalized_hypothesis": "increase the gain by three dissiples wait five seconds then reduce it by one",
+ "errors": 1,
+ "words": 14,
+ "wer": 0.07142857142857142
+ },
+ "semantic": {
+ "normalized_reference": "increase the gain by 3 decibels wait 5 seconds then reduce it by one",
+ "normalized_hypothesis": "increase the gain by 3 dissiples wait 5 seconds then reduce it by one",
+ "errors": 1,
+ "words": 14,
+ "wer": 0.07142857142857142
+ }
+ },
+ {
+ "id": "release500_0038",
+ "category": "punctuation",
+ "text": "The result was unexpected: clear, stable, and surprisingly natural.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0038.wav",
+ "render_seconds": 154.6857005050406,
+ "audio_seconds": 5.433469387755102,
+ "rtf": 28.469047944512425,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE RESULT WAS UNEXPECTED CLEAR STABLE AND SURPRISINGLY NATURAL",
+ "raw": {
+ "normalized_reference": "the result was unexpected: clear, stable, and surprisingly natural.",
+ "normalized_hypothesis": "the result was unexpected clear stable and surprisingly natural",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the result was unexpected clear stable and surprisingly natural",
+ "normalized_hypothesis": "the result was unexpected clear stable and surprisingly natural",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0039",
+ "category": "question",
+ "text": "Who told you that was the right way to do it?",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0039.wav",
+ "render_seconds": 126.19904938701075,
+ "audio_seconds": 2.9257142857142857,
+ "rtf": 43.13444070845094,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WHO TOLD YOU THAT WAS THE RIGHT WAY TO DO IT",
+ "raw": {
+ "normalized_reference": "who told you that was the right way to do it?",
+ "normalized_hypothesis": "who told you that was the right way to do it",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "who told you that was the right way to do it",
+ "normalized_hypothesis": "who told you that was the right way to do it",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0040",
+ "category": "rare_words",
+ "text": "The meticulous cartographer examined an iridescent archipelago through binoculars.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0040.wav",
+ "render_seconds": 180.29475947702304,
+ "audio_seconds": 5.851428571428571,
+ "rtf": 30.812092684061554,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE METICULOUS CARDOGRAPHER EXAMINED AN IRIDESCENT ARCHIPELAGO THROUGH BONOCULARS",
+ "raw": {
+ "normalized_reference": "the meticulous cartographer examined an iridescent archipelago through binoculars.",
+ "normalized_hypothesis": "the meticulous cardographer examined an iridescent archipelago through bonoculars",
+ "errors": 2,
+ "words": 9,
+ "wer": 0.2222222222222222
+ },
+ "semantic": {
+ "normalized_reference": "the meticulous cartographer examined an iridescent archipelago through binoculars",
+ "normalized_hypothesis": "the meticulous cardographer examined an iridescent archipelago through bonoculars",
+ "errors": 2,
+ "words": 9,
+ "wer": 0.2222222222222222
+ }
+ },
+ {
+ "id": "release500_0041",
+ "category": "short",
+ "text": "I mean. Yeah. Kind of.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0041.wav",
+ "render_seconds": 114.09826706699096,
+ "audio_seconds": 2.2987755102040817,
+ "rtf": 49.63436688816191,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I MEAN YE KIND OF",
+ "raw": {
+ "normalized_reference": "i mean. yeah. kind of.",
+ "normalized_hypothesis": "i mean ye kind of",
+ "errors": 1,
+ "words": 5,
+ "wer": 0.2
+ },
+ "semantic": {
+ "normalized_reference": "i mean yeah kind of",
+ "normalized_hypothesis": "i mean ye kind of",
+ "errors": 1,
+ "words": 5,
+ "wer": 0.2
+ }
+ },
+ {
+ "id": "release500_0042",
+ "category": "technical",
+ "text": "The logs are your best friend when something behaves unexpectedly.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0042.wav",
+ "render_seconds": 173.2955003969837,
+ "audio_seconds": 4.527891156462585,
+ "rtf": 38.2728944686848,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE LOGS ARE YOUR BEST FRIEND WHEN SOMETHING BEHAVES UNEXPECTEDLY",
+ "raw": {
+ "normalized_reference": "the logs are your best friend when something behaves unexpectedly.",
+ "normalized_hypothesis": "the logs are your best friend when something behaves unexpectedly",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the logs are your best friend when something behaves unexpectedly",
+ "normalized_hypothesis": "the logs are your best friend when something behaves unexpectedly",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0043",
+ "category": "conversational",
+ "text": "He kept talking and I kept nodding and neither of us was listening.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0043.wav",
+ "render_seconds": 166.1927850000793,
+ "audio_seconds": 4.527891156462585,
+ "rtf": 36.70423586990934,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE KEPT TALKING AND I KEPT NODDING AND NEITHER OF US WAS LISTENING",
+ "raw": {
+ "normalized_reference": "he kept talking and i kept nodding and neither of us was listening.",
+ "normalized_hypothesis": "he kept talking and i kept nodding and neither of us was listening",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he kept talking and i kept nodding and neither of us was listening",
+ "normalized_hypothesis": "he kept talking and i kept nodding and neither of us was listening",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0044",
+ "category": "descriptive",
+ "text": "It's one of those places where everyone seems to know each other already.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0044.wav",
+ "render_seconds": 114.20044160599355,
+ "audio_seconds": 4.876190476190477,
+ "rtf": 23.42001243872914,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT'S ONE OF THOSE PLACES WHERE EVERYONE SEEMS TO KNOW EACH OTHER ALREADY",
+ "raw": {
+ "normalized_reference": "it's one of those places where everyone seems to know each other already.",
+ "normalized_hypothesis": "it's one of those places where everyone seems to know each other already",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it is one of those places where everyone seems to know each other already",
+ "normalized_hypothesis": "it is one of those places where everyone seems to know each other already",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0045",
+ "category": "emotional",
+ "text": "He said he was proud of me and nobody had ever said that to me before.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0045.wav",
+ "render_seconds": 166.0905514130136,
+ "audio_seconds": 4.876190476190477,
+ "rtf": 34.06153886399693,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE SAID HE WAS PROUD OF ME AND NOBODY HAD EVER SAID THAT TO ME BEFORE",
+ "raw": {
+ "normalized_reference": "he said he was proud of me and nobody had ever said that to me before.",
+ "normalized_hypothesis": "he said he was proud of me and nobody had ever said that to me before",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he said he was proud of me and nobody had ever said that to me before",
+ "normalized_hypothesis": "he said he was proud of me and nobody had ever said that to me before",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0046",
+ "category": "homographs",
+ "text": "The minute detail took a minute to explain clearly.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0046.wav",
+ "render_seconds": 137.69410212896764,
+ "audio_seconds": 3.622312925170068,
+ "rtf": 38.012757310994296,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE MINUTE DETAIL TOOK A MINUTE TO EXPLAIN CLEARLY",
+ "raw": {
+ "normalized_reference": "the minute detail took a minute to explain clearly.",
+ "normalized_hypothesis": "the minute detail took a minute to explain clearly",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the minute detail took a minute to explain clearly",
+ "normalized_hypothesis": "the minute detail took a minute to explain clearly",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0047",
+ "category": "long",
+ "text": "The strange thing about really wanting something is that getting it never quite feels the way you imagined it would, and not getting it never quite feels as bad.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0047.wav",
+ "render_seconds": 318.20927972893696,
+ "audio_seconds": 9.961360544217687,
+ "rtf": 31.944359238522818,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE STRANGE THING ABOUT REALLY WANTING SOMETHING IS THAT GETTING IT NEVER QUITE FEELS THE WAY YOU IMAGINED IT WOULD AND NOT GETTING IT NEVER QUITE FEELS AS BAD",
+ "raw": {
+ "normalized_reference": "the strange thing about really wanting something is that getting it never quite feels the way you imagined it would, and not getting it never quite feels as bad.",
+ "normalized_hypothesis": "the strange thing about really wanting something is that getting it never quite feels the way you imagined it would and not getting it never quite feels as bad",
+ "errors": 0,
+ "words": 29,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the strange thing about really wanting something is that getting it never quite feels the way you imagined it would and not getting it never quite feels as bad",
+ "normalized_hypothesis": "the strange thing about really wanting something is that getting it never quite feels the way you imagined it would and not getting it never quite feels as bad",
+ "errors": 0,
+ "words": 29,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0048",
+ "category": "long_range",
+ "text": "Even though the recipe looked simple at first, preparing the sauce required patience, careful timing, and a much larger pan.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0048.wav",
+ "render_seconds": 271.20498704595957,
+ "audio_seconds": 8.289523809523809,
+ "rtf": 32.71659425531452,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "EVEN THOUGH THE RECIPE LOOKED SIMPLE AT FIRST PREPARING THE SAUCE REQUIRED PATIENCE CAREFUL TIMING AND A MUCH LARGER PAN",
+ "raw": {
+ "normalized_reference": "even though the recipe looked simple at first, preparing the sauce required patience, careful timing, and a much larger pan.",
+ "normalized_hypothesis": "even though the recipe looked simple at first preparing the sauce required patience careful timing and a much larger pan",
+ "errors": 0,
+ "words": 20,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "even though the recipe looked simple at 1st preparing the sauce required patience careful timing and a much larger pan",
+ "normalized_hypothesis": "even though the recipe looked simple at 1st preparing the sauce required patience careful timing and a much larger pan",
+ "errors": 0,
+ "words": 20,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0049",
+ "category": "names_places",
+ "text": "Xiomara met Siobhan near the old library in Worcestershire.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0049.wav",
+ "render_seconds": 157.49648706498556,
+ "audio_seconds": 4.667210884353741,
+ "rtf": 33.74531191486835,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SHAH MARA MET SIOPAN NEAR THE OLD LIBRARY IN WERCESTERSHIRE",
+ "raw": {
+ "normalized_reference": "xiomara met siobhan near the old library in worcestershire.",
+ "normalized_hypothesis": "shah mara met siopan near the old library in wercestershire",
+ "errors": 4,
+ "words": 9,
+ "wer": 0.4444444444444444
+ },
+ "semantic": {
+ "normalized_reference": "xiomara met siobhan near the old library in worcestershire",
+ "normalized_hypothesis": "shah mara met siopan near the old library in wercestershire",
+ "errors": 4,
+ "words": 9,
+ "wer": 0.4444444444444444
+ }
+ },
+ {
+ "id": "release500_0050",
+ "category": "narrative",
+ "text": "She walked into the room and everyone went quiet, not because she was loud, but because she wasn't.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0050.wav",
+ "render_seconds": 195.6048700939864,
+ "audio_seconds": 6.408707482993197,
+ "rtf": 30.521734782413382,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SHE WALKED INTO THE ROOM AND EVERYONE WENT QUIET NOT BECAUSE SHE WAS LOUD BUT BECAUSE SHE WASN'T",
+ "raw": {
+ "normalized_reference": "she walked into the room and everyone went quiet, not because she was loud, but because she wasn't.",
+ "normalized_hypothesis": "she walked into the room and everyone went quiet not because she was loud but because she wasn't",
+ "errors": 0,
+ "words": 18,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "she walked into the room and everyone went quiet not because she was loud but because she was not",
+ "normalized_hypothesis": "she walked into the room and everyone went quiet not because she was loud but because she was not",
+ "errors": 0,
+ "words": 19,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0051",
+ "category": "numbers",
+ "text": "At exactly seven forty-five, Morgan placed eighteen dollars and fifty cents beside the notebook.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0051.wav",
+ "render_seconds": 221.61156587593723,
+ "audio_seconds": 6.826666666666667,
+ "rtf": 32.46263172010799,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "AT EXACTLY SEVEN FORTY FIVE MORGAN PLACED EIGHTEEN DOLLARS AND FIFTY CENTS BESIDE THE NOTE BOOK",
+ "raw": {
+ "normalized_reference": "at exactly seven forty-five, morgan placed eighteen dollars and fifty cents beside the notebook.",
+ "normalized_hypothesis": "at exactly seven forty five morgan placed eighteen dollars and fifty cents beside the note book",
+ "errors": 2,
+ "words": 15,
+ "wer": 0.13333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "at exactly 745 morgan placed $18.50 beside the notebook",
+ "normalized_hypothesis": "at exactly 745 morgan placed $18.50 beside the note book",
+ "errors": 2,
+ "words": 10,
+ "wer": 0.2
+ }
+ },
+ {
+ "id": "release500_0052",
+ "category": "punctuation",
+ "text": "Does punctuation work? Yes: commas, semicolons, and periods should create sensible boundaries.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0052.wav",
+ "render_seconds": 234.1030401220778,
+ "audio_seconds": 7.453605442176871,
+ "rtf": 31.408026885537236,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "DOES PUNCTUATION WORK YES COMMAS SEMICALONS AND PERIODS SHOULD CREATE SENSIBLE BOUNDARIES",
+ "raw": {
+ "normalized_reference": "does punctuation work? yes: commas, semicolons, and periods should create sensible boundaries.",
+ "normalized_hypothesis": "does punctuation work yes commas semicalons and periods should create sensible boundaries",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "does punctuation work yes commas semicolons and periods should create sensible boundaries",
+ "normalized_hypothesis": "does punctuation work yes commas semicalons and periods should create sensible boundaries",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ }
+ },
+ {
+ "id": "release500_0053",
+ "category": "question",
+ "text": "Have you ever just stopped and thought about how strange all of this is?",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0053.wav",
+ "render_seconds": 190.98922071699053,
+ "audio_seconds": 4.876190476190477,
+ "rtf": 39.167711279851574,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HAVE YOU EVER JUST STOPPED AND THOUGHT ABOUT HOW STRANGE ALL OF THIS IS",
+ "raw": {
+ "normalized_reference": "have you ever just stopped and thought about how strange all of this is?",
+ "normalized_hypothesis": "have you ever just stopped and thought about how strange all of this is",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "have you ever just stopped and thought about how strange all of this is",
+ "normalized_hypothesis": "have you ever just stopped and thought about how strange all of this is",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0054",
+ "category": "rare_words",
+ "text": "Her idiosyncratic explanation sounded plausible but remained fundamentally ambiguous.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0054.wav",
+ "render_seconds": 220.29730763600674,
+ "audio_seconds": 6.339047619047619,
+ "rtf": 34.75242984041572,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HER IDIOSYNCRATIC EXPLANATION SOUNDED PLAUSIBLE BUT REMAINED FUNDAMENTALLY AMBIGUOUS",
+ "raw": {
+ "normalized_reference": "her idiosyncratic explanation sounded plausible but remained fundamentally ambiguous.",
+ "normalized_hypothesis": "her idiosyncratic explanation sounded plausible but remained fundamentally ambiguous",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "her idiosyncratic explanation sounded plausible but remained fundamentally ambiguous",
+ "normalized_hypothesis": "her idiosyncratic explanation sounded plausible but remained fundamentally ambiguous",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0055",
+ "category": "short",
+ "text": "That tracks.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0055.wav",
+ "render_seconds": 77.29931494104676,
+ "audio_seconds": 1.4628571428571429,
+ "rtf": 52.84132857298118,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THAT TRACKS",
+ "raw": {
+ "normalized_reference": "that tracks.",
+ "normalized_hypothesis": "that tracks",
+ "errors": 0,
+ "words": 2,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "that tracks",
+ "normalized_hypothesis": "that tracks",
+ "errors": 0,
+ "words": 2,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0056",
+ "category": "technical",
+ "text": "The key thing to remember is that order matters here more than you'd think.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0056.wav",
+ "render_seconds": 135.3948296570452,
+ "audio_seconds": 4.597551020408163,
+ "rtf": 29.449337061413416,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE KEY THING TO REMEMBER IS THAT ORDER MATTERS HERE MORE THAN YOU'D THINK",
+ "raw": {
+ "normalized_reference": "the key thing to remember is that order matters here more than you'd think.",
+ "normalized_hypothesis": "the key thing to remember is that order matters here more than you'd think",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the key thing to remember is that order matters here more than you would think",
+ "normalized_hypothesis": "the key thing to remember is that order matters here more than you would think",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0057",
+ "category": "conversational",
+ "text": "I would not say that again in this room if I were you.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0057.wav",
+ "render_seconds": 150.48785761697218,
+ "audio_seconds": 3.7616326530612243,
+ "rtf": 40.00599513472025,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I WOULD NOT SAY THAT AGAIN IN THIS ROOM IF I WERE YOU",
+ "raw": {
+ "normalized_reference": "i would not say that again in this room if i were you.",
+ "normalized_hypothesis": "i would not say that again in this room if i were you",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i would not say that again in this room if i were you",
+ "normalized_hypothesis": "i would not say that again in this room if i were you",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0058",
+ "category": "descriptive",
+ "text": "The room was filled with the quiet that means something is about to change.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0058.wav",
+ "render_seconds": 172.1077914451016,
+ "audio_seconds": 4.73687074829932,
+ "rtf": 36.333647378259904,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE ROOM WAS FILLED WITH THE QUIET THAT MEANS SOMETHING IS ABOUT TO CHANGE",
+ "raw": {
+ "normalized_reference": "the room was filled with the quiet that means something is about to change.",
+ "normalized_hypothesis": "the room was filled with the quiet that means something is about to change",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the room was filled with the quiet that means something is about to change",
+ "normalized_hypothesis": "the room was filled with the quiet that means something is about to change",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0059",
+ "category": "emotional",
+ "text": "I didn't expect to feel that way about it, but I did.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0059.wav",
+ "render_seconds": 150.90618067095056,
+ "audio_seconds": 3.4829931972789114,
+ "rtf": 43.326579216073696,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I DIDN'T EXPECT TO FEEL THAT WAY ABOUT IT BUT I DID",
+ "raw": {
+ "normalized_reference": "i didn't expect to feel that way about it, but i did.",
+ "normalized_hypothesis": "i didn't expect to feel that way about it but i did",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i did not expect to feel that way about it but i did",
+ "normalized_hypothesis": "i did not expect to feel that way about it but i did",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0060",
+ "category": "homographs",
+ "text": "I object to the strange object that was left near the entrance.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0060.wav",
+ "render_seconds": 172.6041172950063,
+ "audio_seconds": 4.1099319727891155,
+ "rtf": 41.996830711013516,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I OBJECT TO THE STRANGE OBJECT THAT WAS LEFT NEAR THE ENTRANCE",
+ "raw": {
+ "normalized_reference": "i object to the strange object that was left near the entrance.",
+ "normalized_hypothesis": "i object to the strange object that was left near the entrance",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i object to the strange object that was left near the entrance",
+ "normalized_hypothesis": "i object to the strange object that was left near the entrance",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0061",
+ "category": "long",
+ "text": "I've learned more from the things that went wrong than from the things that worked, which is a frustrating way to have to learn things but apparently the only way that stuck.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0061.wav",
+ "render_seconds": 321.70934425399173,
+ "audio_seconds": 10.65795918367347,
+ "rtf": 30.184891751772355,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I'VE LEARNED MORE FROM THE THINGS THAT WENT WRONG THAN FROM THE THINGS THAT WORKED WHICH IS A FRUSTRATING WAY TO HAVE TO LEARN THINGS BUT APPARENTLY THE ONLY WAY THAT STUCK",
+ "raw": {
+ "normalized_reference": "i've learned more from the things that went wrong than from the things that worked, which is a frustrating way to have to learn things but apparently the only way that stuck.",
+ "normalized_hypothesis": "i've learned more from the things that went wrong than from the things that worked which is a frustrating way to have to learn things but apparently the only way that stuck",
+ "errors": 0,
+ "words": 32,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i have learned more from the things that went wrong than from the things that worked which is a frustrating way to have to learn things but apparently the only way that stuck",
+ "normalized_hypothesis": "i have learned more from the things that went wrong than from the things that worked which is a frustrating way to have to learn things but apparently the only way that stuck",
+ "errors": 0,
+ "words": 33,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0062",
+ "category": "long_range",
+ "text": "Because every nearby hotel was full, the exhausted hikers followed the narrow road until they found a farmhouse with one empty room.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0062.wav",
+ "render_seconds": 193.3061724229483,
+ "audio_seconds": 8.916462585034013,
+ "rtf": 21.679693104685516,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "BECAUSE EVERY NEARBY HOTEL WAS FULL THE EXHAUSTED HYKERS FOLLOWED THE NARROW ROAD UNTIL THEY FOUND A FARMHOUSE WITH ONE EMPTY ROOM",
+ "raw": {
+ "normalized_reference": "because every nearby hotel was full, the exhausted hikers followed the narrow road until they found a farmhouse with one empty room.",
+ "normalized_hypothesis": "because every nearby hotel was full the exhausted hykers followed the narrow road until they found a farmhouse with one empty room",
+ "errors": 1,
+ "words": 22,
+ "wer": 0.045454545454545456
+ },
+ "semantic": {
+ "normalized_reference": "because every nearby hotel was full the exhausted hikers followed the narrow road until they found a farmhouse with one empty room",
+ "normalized_hypothesis": "because every nearby hotel was full the exhausted hykers followed the narrow road until they found a farmhouse with one empty room",
+ "errors": 1,
+ "words": 22,
+ "wer": 0.045454545454545456
+ }
+ },
+ {
+ "id": "release500_0063",
+ "category": "names_places",
+ "text": "The itinerary sends Beatrice from Albuquerque to Des Moines and then to Boise.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0063.wav",
+ "render_seconds": 180.29913160100114,
+ "audio_seconds": 5.363809523809524,
+ "rtf": 33.6140071344196,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE ITINERARY SENDS BEATRICE FROM ALBUKERKY TO DESMOINS AND THEN TO BOYS",
+ "raw": {
+ "normalized_reference": "the itinerary sends beatrice from albuquerque to des moines and then to boise.",
+ "normalized_hypothesis": "the itinerary sends beatrice from albukerky to desmoins and then to boys",
+ "errors": 4,
+ "words": 13,
+ "wer": 0.3076923076923077
+ },
+ "semantic": {
+ "normalized_reference": "the itinerary sends beatrice from albuquerque to des moines and then to boise",
+ "normalized_hypothesis": "the itinerary sends beatrice from albukerky to desmoins and then to boys",
+ "errors": 4,
+ "words": 13,
+ "wer": 0.3076923076923077
+ }
+ },
+ {
+ "id": "release500_0064",
+ "category": "narrative",
+ "text": "We had been driving for six hours when we finally saw the sign for the town.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0064.wav",
+ "render_seconds": 169.9961345680058,
+ "audio_seconds": 5.085170068027211,
+ "rtf": 33.42978352618907,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WE HAD BEEN DRIVING FOR SIX HOURS WHEN WE FINALLY SAW THE SIGN FOR THE TOWN",
+ "raw": {
+ "normalized_reference": "we had been driving for six hours when we finally saw the sign for the town.",
+ "normalized_hypothesis": "we had been driving for six hours when we finally saw the sign for the town",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "we had been driving for 6 hours when we finally saw the sign for the town",
+ "normalized_hypothesis": "we had been driving for 6 hours when we finally saw the sign for the town",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0065",
+ "category": "numbers",
+ "text": "Call extension zero four nine after eight thirty in the morning.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0065.wav",
+ "render_seconds": 179.89916955307126,
+ "audio_seconds": 4.458231292517007,
+ "rtf": 40.35213916671978,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "CALL EXTENSION ZERO FOUR NINE AFTER EIGHT THIRTY IN THE MORNING",
+ "raw": {
+ "normalized_reference": "call extension zero four nine after eight thirty in the morning.",
+ "normalized_hypothesis": "call extension zero four nine after eight thirty in the morning",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "call extension 049 after 830 in the morning",
+ "normalized_hypothesis": "call extension 049 after 830 in the morning",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0066",
+ "category": "punctuation",
+ "text": "Stop - listen carefully - then answer the question without guessing.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0066.wav",
+ "render_seconds": 177.09646569692995,
+ "audio_seconds": 4.597551020408163,
+ "rtf": 38.51973907648068,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "STOP LISTEN CAREFULLY THEN ANSWER THE QUESTION WITHOUT GUESSING",
+ "raw": {
+ "normalized_reference": "stop - listen carefully - then answer the question without guessing.",
+ "normalized_hypothesis": "stop listen carefully then answer the question without guessing",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "stop listen carefully then answer the question without guessing",
+ "normalized_hypothesis": "stop listen carefully then answer the question without guessing",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0067",
+ "category": "question",
+ "text": "Does it matter at this point, honestly?",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0067.wav",
+ "render_seconds": 122.09159150102641,
+ "audio_seconds": 2.716734693877551,
+ "rtf": 44.94056478027564,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "DOES IT MATTER AT THIS POINT HONESTLY",
+ "raw": {
+ "normalized_reference": "does it matter at this point, honestly?",
+ "normalized_hypothesis": "does it matter at this point honestly",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "does it matter at this point honestly",
+ "normalized_hypothesis": "does it matter at this point honestly",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0068",
+ "category": "rare_words",
+ "text": "An entrepreneurial meteorologist described the unprecedented atmospheric phenomenon.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0068.wav",
+ "render_seconds": 179.09063885593787,
+ "audio_seconds": 5.851428571428571,
+ "rtf": 30.60631035135657,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "AN ONTROPERNURIAL METEOROLOGIST DESCRIBED THE UNPRECEDENTED ATMOSPHERIC PHENOMENON",
+ "raw": {
+ "normalized_reference": "an entrepreneurial meteorologist described the unprecedented atmospheric phenomenon.",
+ "normalized_hypothesis": "an ontropernurial meteorologist described the unprecedented atmospheric phenomenon",
+ "errors": 1,
+ "words": 8,
+ "wer": 0.125
+ },
+ "semantic": {
+ "normalized_reference": "an entrepreneurial meteorologist described the unprecedented atmospheric phenomenon",
+ "normalized_hypothesis": "an ontropernurial meteorologist described the unprecedented atmospheric phenomenon",
+ "errors": 1,
+ "words": 8,
+ "wer": 0.125
+ }
+ },
+ {
+ "id": "release500_0069",
+ "category": "short",
+ "text": "That's not what I expected.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0069.wav",
+ "render_seconds": 94.20020881399978,
+ "audio_seconds": 2.089795918367347,
+ "rtf": 45.076271795761606,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THAT'S NOT WHAT I EXPECTED",
+ "raw": {
+ "normalized_reference": "that's not what i expected.",
+ "normalized_hypothesis": "that's not what i expected",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "that is not what i expected",
+ "normalized_hypothesis": "that is not what i expected",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0070",
+ "category": "technical",
+ "text": "There's a known issue with that version that causes failures under load.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0070.wav",
+ "render_seconds": 192.00061726605054,
+ "audio_seconds": 4.806530612244898,
+ "rtf": 39.945780596282596,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THERE'S A KNOWN ISSUE WITH THAT VERSION THAT CAUSES FAILURES UNDER LOAD",
+ "raw": {
+ "normalized_reference": "there's a known issue with that version that causes failures under load.",
+ "normalized_hypothesis": "there's a known issue with that version that causes failures under load",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "there is a known issue with that version that causes failures under load",
+ "normalized_hypothesis": "there is a known issue with that version that causes failures under load",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0071",
+ "category": "conversational",
+ "text": "Don't look now but I think that's him over there.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0071.wav",
+ "render_seconds": 125.60043001594022,
+ "audio_seconds": 2.8560544217687074,
+ "rtf": 43.97690361171688,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "DON'T LOOK NOW BUT I THINK THAT'S HIM OVER THERE",
+ "raw": {
+ "normalized_reference": "don't look now but i think that's him over there.",
+ "normalized_hypothesis": "don't look now but i think that's him over there",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "do not look now but i think that is him over there",
+ "normalized_hypothesis": "do not look now but i think that is him over there",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0072",
+ "category": "descriptive",
+ "text": "The city at night looks completely different from how it feels during the day.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0072.wav",
+ "render_seconds": 205.99246078100987,
+ "audio_seconds": 5.154829931972789,
+ "rtf": 39.96105856050524,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE CITY AT NIGHT LOOKS COMPLETELY DIFFERENT FROM HOW IT FEELS DURING THE DAY",
+ "raw": {
+ "normalized_reference": "the city at night looks completely different from how it feels during the day.",
+ "normalized_hypothesis": "the city at night looks completely different from how it feels during the day",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the city at night looks completely different from how it feels during the day",
+ "normalized_hypothesis": "the city at night looks completely different from how it feels during the day",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0073",
+ "category": "emotional",
+ "text": "It's the kind of happy that makes you nervous because you know it can't last.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0073.wav",
+ "render_seconds": 182.49806611204986,
+ "audio_seconds": 4.876190476190477,
+ "rtf": 37.426361214385224,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT'S THE KIND OF HAPPY THAT MAKES YOU NERVOUS BECAUSE YOU KNOW IT CAN'T LAST",
+ "raw": {
+ "normalized_reference": "it's the kind of happy that makes you nervous because you know it can't last.",
+ "normalized_hypothesis": "it's the kind of happy that makes you nervous because you know it can't last",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it is the kind of happy that makes you nervous because you know it can not last",
+ "normalized_hypothesis": "it is the kind of happy that makes you nervous because you know it can not last",
+ "errors": 0,
+ "words": 17,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0074",
+ "category": "homographs",
+ "text": "The dove dove beneath the branch while the wind began to wind the loose cable.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0074.wav",
+ "render_seconds": 163.8029365990078,
+ "audio_seconds": 5.085170068027211,
+ "rtf": 32.21188955486696,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE DOVE DOVED BENEATH THE BRANCH WHILE THE WIND BEGAN TO WIN THE LOOSE CABLE",
+ "raw": {
+ "normalized_reference": "the dove dove beneath the branch while the wind began to wind the loose cable.",
+ "normalized_hypothesis": "the dove doved beneath the branch while the wind began to win the loose cable",
+ "errors": 2,
+ "words": 15,
+ "wer": 0.13333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "the dove dove beneath the branch while the wind began to wind the loose cable",
+ "normalized_hypothesis": "the dove doved beneath the branch while the wind began to win the loose cable",
+ "errors": 2,
+ "words": 15,
+ "wer": 0.13333333333333333
+ }
+ },
+ {
+ "id": "release500_0075",
+ "category": "long",
+ "text": "I've thought a lot about what I would do differently if I could go back, and the honest answer is probably not as much as I'd like to think.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0075.wav",
+ "render_seconds": 267.306908751023,
+ "audio_seconds": 8.70748299319728,
+ "rtf": 30.698527801875294,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I'VE THOUGHT A LOT ABOUT WHAT I WOULD DO DIFFERENTLY IF I COULD GO BACK AND THE HONEST ANSWER IS PROBABLY NOT AS MUCH AS I'D LIKE TO THINK",
+ "raw": {
+ "normalized_reference": "i've thought a lot about what i would do differently if i could go back, and the honest answer is probably not as much as i'd like to think.",
+ "normalized_hypothesis": "i've thought a lot about what i would do differently if i could go back and the honest answer is probably not as much as i'd like to think",
+ "errors": 0,
+ "words": 29,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i have thought a lot about what i would do differently if i could go back and the honest answer is probably not as much as i would like to think",
+ "normalized_hypothesis": "i have thought a lot about what i would do differently if i could go back and the honest answer is probably not as much as i would like to think",
+ "errors": 0,
+ "words": 31,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0076",
+ "category": "long_range",
+ "text": "Although the weather changed without warning, the musicians finished packing before anyone opened the enormous glass door.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0076.wav",
+ "render_seconds": 250.19262322608847,
+ "audio_seconds": 7.801904761904762,
+ "rtf": 32.06814628752355,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "ALTHOUGH THE WEATHER CHANGED WITHOUT WARNING THE MUSICIANS FINISHED PACKING BEFORE ANYONE OPENED THE ENORMOUS GLASS DOOR",
+ "raw": {
+ "normalized_reference": "although the weather changed without warning, the musicians finished packing before anyone opened the enormous glass door.",
+ "normalized_hypothesis": "although the weather changed without warning the musicians finished packing before anyone opened the enormous glass door",
+ "errors": 0,
+ "words": 17,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "although the weather changed without warning the musicians finished packing before anyone opened the enormous glass door",
+ "normalized_hypothesis": "although the weather changed without warning the musicians finished packing before anyone opened the enormous glass door",
+ "errors": 0,
+ "words": 17,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0077",
+ "category": "names_places",
+ "text": "Saskatchewan looked unusually bright beneath the pale February sunrise.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0077.wav",
+ "render_seconds": 190.80327391496394,
+ "audio_seconds": 5.154829931972789,
+ "rtf": 37.01446535248588,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SUSKACHUON LOOKED UNUSUALLY BRIGHT BENEATH THE PALE FEBRUARY SUNRISE",
+ "raw": {
+ "normalized_reference": "saskatchewan looked unusually bright beneath the pale february sunrise.",
+ "normalized_hypothesis": "suskachuon looked unusually bright beneath the pale february sunrise",
+ "errors": 1,
+ "words": 9,
+ "wer": 0.1111111111111111
+ },
+ "semantic": {
+ "normalized_reference": "saskatchewan looked unusually bright beneath the pale february sunrise",
+ "normalized_hypothesis": "suskachuon looked unusually bright beneath the pale february sunrise",
+ "errors": 1,
+ "words": 9,
+ "wer": 0.1111111111111111
+ }
+ },
+ {
+ "id": "release500_0078",
+ "category": "narrative",
+ "text": "She handled it with a grace I genuinely did not expect.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0078.wav",
+ "render_seconds": 158.5992601349717,
+ "audio_seconds": 3.900952380952381,
+ "rtf": 40.656548618584054,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SHE HANDLED IT WITH A GRACE I GENUINELY DID NOT EXPECT",
+ "raw": {
+ "normalized_reference": "she handled it with a grace i genuinely did not expect.",
+ "normalized_hypothesis": "she handled it with a grace i genuinely did not expect",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "she handled it with a grace i genuinely did not expect",
+ "normalized_hypothesis": "she handled it with a grace i genuinely did not expect",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0079",
+ "category": "numbers",
+ "text": "Apartment four hundred seven is across from room twenty-nine on the third floor.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0079.wav",
+ "render_seconds": 186.9999477979727,
+ "audio_seconds": 5.642448979591837,
+ "rtf": 33.141628479820106,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "APARTMENT FOUR HUNDRED SEVEN IS ACROSS FROM ROOM TWENTY NINE ON THE THIRD FLOOR",
+ "raw": {
+ "normalized_reference": "apartment four hundred seven is across from room twenty-nine on the third floor.",
+ "normalized_hypothesis": "apartment four hundred seven is across from room twenty nine on the third floor",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "apartment 407 is across from room 29 on the 3rd floor",
+ "normalized_hypothesis": "apartment 407 is across from room 29 on the 3rd floor",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0080",
+ "category": "punctuation",
+ "text": "\"Not yet,\" Rowan whispered. \"Give me another minute... please.\"",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0080.wav",
+ "render_seconds": 132.82155725499615,
+ "audio_seconds": 5.50312925170068,
+ "rtf": 24.135641955701683,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "NOT YET ROWAN WHISPERED GIVE ME ANOTHER MINUTE PLEASE",
+ "raw": {
+ "normalized_reference": "\"not yet,\" rowan whispered. \"give me another minute... please.\"",
+ "normalized_hypothesis": "not yet rowan whispered give me another minute please",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "not yet rowan whispered give me another minute please",
+ "normalized_hypothesis": "not yet rowan whispered give me another minute please",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0081",
+ "category": "question",
+ "text": "What are we actually trying to accomplish here?",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0081.wav",
+ "render_seconds": 103.99980497697834,
+ "audio_seconds": 3.274013605442177,
+ "rtf": 31.76523298623632,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WHAT ARE WE ACTUALLY TRYING TO ACCOMPLISH HERE",
+ "raw": {
+ "normalized_reference": "what are we actually trying to accomplish here?",
+ "normalized_hypothesis": "what are we actually trying to accomplish here",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "what are we actually trying to accomplish here",
+ "normalized_hypothesis": "what are we actually trying to accomplish here",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0082",
+ "category": "rare_words",
+ "text": "The physician recommended a noninvasive procedure for the recurring neurological symptoms.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0082.wav",
+ "render_seconds": 227.49787733901758,
+ "audio_seconds": 6.408707482993197,
+ "rtf": 35.498246400343476,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE PHYSICIAN RECOMMENDED A NON ENVASIVE PROCEDURE FOR THE RECURRING NEROLOGICAL SYMPTOMS",
+ "raw": {
+ "normalized_reference": "the physician recommended a noninvasive procedure for the recurring neurological symptoms.",
+ "normalized_hypothesis": "the physician recommended a non envasive procedure for the recurring nerological symptoms",
+ "errors": 3,
+ "words": 11,
+ "wer": 0.2727272727272727
+ },
+ "semantic": {
+ "normalized_reference": "the physician recommended a noninvasive procedure for the recurring neurological symptoms",
+ "normalized_hypothesis": "the physician recommended a non envasive procedure for the recurring nerological symptoms",
+ "errors": 3,
+ "words": 11,
+ "wer": 0.2727272727272727
+ }
+ },
+ {
+ "id": "release500_0083",
+ "category": "short",
+ "text": "Are you serious right now?",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0083.wav",
+ "render_seconds": 80.69410763599444,
+ "audio_seconds": 2.020136054421769,
+ "rtf": 39.944887602677746,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "ARE YOU SERIOUS RIGHT NOW",
+ "raw": {
+ "normalized_reference": "are you serious right now?",
+ "normalized_hypothesis": "are you serious right now",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "are you serious right now",
+ "normalized_hypothesis": "are you serious right now",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0084",
+ "category": "technical",
+ "text": "The performance improvement is real, but it comes with tradeoffs.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0084.wav",
+ "render_seconds": 181.11119199392851,
+ "audio_seconds": 4.597551020408163,
+ "rtf": 39.39297055975895,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE PERFORMANCE IMPROVEMENT IS REAL BUT IT COMES WITH TRADE OFFS",
+ "raw": {
+ "normalized_reference": "the performance improvement is real, but it comes with tradeoffs.",
+ "normalized_hypothesis": "the performance improvement is real but it comes with trade offs",
+ "errors": 2,
+ "words": 10,
+ "wer": 0.2
+ },
+ "semantic": {
+ "normalized_reference": "the performance improvement is real but it comes with tradeoffs",
+ "normalized_hypothesis": "the performance improvement is real but it comes with trade offs",
+ "errors": 2,
+ "words": 10,
+ "wer": 0.2
+ }
+ },
+ {
+ "id": "release500_0085",
+ "category": "conversational",
+ "text": "I don't know. It's just been one of those weeks, you know?",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0085.wav",
+ "render_seconds": 163.90710327599663,
+ "audio_seconds": 3.900952380952381,
+ "rtf": 42.017201767528434,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I DON'T KNOW IT'S JUST BEEN ONE OF THOSE WEEKS YOU KNOW",
+ "raw": {
+ "normalized_reference": "i don't know. it's just been one of those weeks, you know?",
+ "normalized_hypothesis": "i don't know it's just been one of those weeks you know",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i do not know it is just been one of those weeks you know",
+ "normalized_hypothesis": "i do not know it is just been one of those weeks you know",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0086",
+ "category": "descriptive",
+ "text": "The moment you stop expecting perfection, things get a lot more interesting.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0086.wav",
+ "render_seconds": 161.07786938804202,
+ "audio_seconds": 5.154829931972789,
+ "rtf": 31.247950274548902,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE MOMENT YOU STOP EXPECTING PERFECTION THINGS GET A LOT MORE INTERESTING",
+ "raw": {
+ "normalized_reference": "the moment you stop expecting perfection, things get a lot more interesting.",
+ "normalized_hypothesis": "the moment you stop expecting perfection things get a lot more interesting",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the moment you stop expecting perfection things get a lot more interesting",
+ "normalized_hypothesis": "the moment you stop expecting perfection things get a lot more interesting",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0087",
+ "category": "emotional",
+ "text": "It's the kind of thing that stays with you even years later.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0087.wav",
+ "render_seconds": 131.09238334093243,
+ "audio_seconds": 3.8312925170068026,
+ "rtf": 34.21622931661411,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT'S THE KIND OF THING THAT STAYS WITH YOU EVEN YEARS LATER",
+ "raw": {
+ "normalized_reference": "it's the kind of thing that stays with you even years later.",
+ "normalized_hypothesis": "it's the kind of thing that stays with you even years later",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it is the kind of thing that stays with you even years later",
+ "normalized_hypothesis": "it is the kind of thing that stays with you even years later",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0088",
+ "category": "long",
+ "text": "She told me once that the most important skill she'd developed was learning to disagree with someone without making them feel they were wrong.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0088.wav",
+ "render_seconds": 274.00661964400206,
+ "audio_seconds": 8.70748299319728,
+ "rtf": 31.46794772474086,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SHE TOLD ME ONCE THAT THE MOST IMPORTANT SKILL SHE DEVELOPED WAS LEARNING TO DISAGREE WITH SOMEONE WITHOUT MAKING THEM FEEL THEY WERE WRONG",
+ "raw": {
+ "normalized_reference": "she told me once that the most important skill she'd developed was learning to disagree with someone without making them feel they were wrong.",
+ "normalized_hypothesis": "she told me once that the most important skill she developed was learning to disagree with someone without making them feel they were wrong",
+ "errors": 1,
+ "words": 24,
+ "wer": 0.041666666666666664
+ },
+ "semantic": {
+ "normalized_reference": "she told me once that the most important skill she would developed was learning to disagree with someone without making them feel they were wrong",
+ "normalized_hypothesis": "she told me once that the most important skill she developed was learning to disagree with someone without making them feel they were wrong",
+ "errors": 1,
+ "words": 25,
+ "wer": 0.04
+ }
+ },
+ {
+ "id": "release500_0089",
+ "category": "narrative",
+ "text": "He stood at the edge of the cliff and wondered, not for the first time, how he'd gotten here.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0089.wav",
+ "render_seconds": 235.8968853870174,
+ "audio_seconds": 5.7817687074829935,
+ "rtf": 40.80012489633325,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE STOOD AT THE EDGE OF THE CLIFF AND WONDERED NOT FOR THE FIRST TIME HOW HE'D GOTTEN HERE",
+ "raw": {
+ "normalized_reference": "he stood at the edge of the cliff and wondered, not for the first time, how he'd gotten here.",
+ "normalized_hypothesis": "he stood at the edge of the cliff and wondered not for the first time how he'd gotten here",
+ "errors": 0,
+ "words": 19,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he stood at the edge of the cliff and wondered not for the 1st time how he would gotten here",
+ "normalized_hypothesis": "he stood at the edge of the cliff and wondered not for the 1st time how he would gotten here",
+ "errors": 0,
+ "words": 20,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0090",
+ "category": "question",
+ "text": "Why does asking for help feel so hard when you actually need it?",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0090.wav",
+ "render_seconds": 170.8886543479748,
+ "audio_seconds": 4.249251700680272,
+ "rtf": 40.21617602003121,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WHY DOES ASKING FOR HELP FEEL SO HARD WHEN YOU ACTUALLY NEED IT",
+ "raw": {
+ "normalized_reference": "why does asking for help feel so hard when you actually need it?",
+ "normalized_hypothesis": "why does asking for help feel so hard when you actually need it",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "why does asking for help feel so hard when you actually need it",
+ "normalized_hypothesis": "why does asking for help feel so hard when you actually need it",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0091",
+ "category": "short",
+ "text": "That's fair.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0091.wav",
+ "render_seconds": 91.19974928500596,
+ "audio_seconds": 1.4628571428571429,
+ "rtf": 62.343578612797046,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THAT'S FAIR",
+ "raw": {
+ "normalized_reference": "that's fair.",
+ "normalized_hypothesis": "that's fair",
+ "errors": 0,
+ "words": 2,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "that is fair",
+ "normalized_hypothesis": "that is fair",
+ "errors": 0,
+ "words": 3,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0092",
+ "category": "technical",
+ "text": "If you run into that error, the fix is usually just restarting the service.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0092.wav",
+ "render_seconds": 150.22765503602568,
+ "audio_seconds": 5.363809523809524,
+ "rtf": 28.00764165266814,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IF YOU RUN INTO THAT ERROR THE FIX IS USUALLY JUST RESTARTING THE SERVICE",
+ "raw": {
+ "normalized_reference": "if you run into that error, the fix is usually just restarting the service.",
+ "normalized_hypothesis": "if you run into that error the fix is usually just restarting the service",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "if you run into that error the fix is usually just restarting the service",
+ "normalized_hypothesis": "if you run into that error the fix is usually just restarting the service",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0093",
+ "category": "conversational",
+ "text": "It was a disaster in the best possible way.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0093.wav",
+ "render_seconds": 122.3073402219452,
+ "audio_seconds": 3.2043537414965986,
+ "rtf": 38.169113053298965,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT WAS A DISASTER IN THE BEST POSSIBLE WAY",
+ "raw": {
+ "normalized_reference": "it was a disaster in the best possible way.",
+ "normalized_hypothesis": "it was a disaster in the best possible way",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it was a disaster in the best possible way",
+ "normalized_hypothesis": "it was a disaster in the best possible way",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0094",
+ "category": "descriptive",
+ "text": "It had the look of a place that used to mean something to someone.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0094.wav",
+ "render_seconds": 161.29286159400363,
+ "audio_seconds": 4.458231292517007,
+ "rtf": 36.17866615954366,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT HAD THE LOOK OF A PLACE THAT USED TO MEAN SOMETHING TO SOMEONE",
+ "raw": {
+ "normalized_reference": "it had the look of a place that used to mean something to someone.",
+ "normalized_hypothesis": "it had the look of a place that used to mean something to someone",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it had the look of a place that used to mean something to someone",
+ "normalized_hypothesis": "it had the look of a place that used to mean something to someone",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0095",
+ "category": "emotional",
+ "text": "There's this feeling you get when something finally works after weeks of failure.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0095.wav",
+ "render_seconds": 204.904479224002,
+ "audio_seconds": 5.015510204081632,
+ "rtf": 40.85416455882201,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THERE'S THIS FEELING YOU GET WHEN SOMETHING FINALLY WORKS AFTER WEEKS OF FAILURE",
+ "raw": {
+ "normalized_reference": "there's this feeling you get when something finally works after weeks of failure.",
+ "normalized_hypothesis": "there's this feeling you get when something finally works after weeks of failure",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "there is this feeling you get when something finally works after weeks of failure",
+ "normalized_hypothesis": "there is this feeling you get when something finally works after weeks of failure",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0096",
+ "category": "long",
+ "text": "He had a theory that most arguments aren't actually about what they seem to be about, and the longer I live the more I think he was onto something.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0096.wav",
+ "render_seconds": 337.59983496705536,
+ "audio_seconds": 9.125442176870749,
+ "rtf": 36.99544947232611,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE HAD A THEORY THAT MOST ARGUMENTS AREN'T ACTUALLY ABOUT WHAT THEY SEEM TO BE ABOUT AND THE LONGER I LIVE THE MORE I THINK HE WAS ON TO SOMETHING",
+ "raw": {
+ "normalized_reference": "he had a theory that most arguments aren't actually about what they seem to be about, and the longer i live the more i think he was onto something.",
+ "normalized_hypothesis": "he had a theory that most arguments aren't actually about what they seem to be about and the longer i live the more i think he was on to something",
+ "errors": 2,
+ "words": 29,
+ "wer": 0.06896551724137931
+ },
+ "semantic": {
+ "normalized_reference": "he had a theory that most arguments are not actually about what they seem to be about and the longer i live the more i think he was onto something",
+ "normalized_hypothesis": "he had a theory that most arguments are not actually about what they seem to be about and the longer i live the more i think he was on to something",
+ "errors": 2,
+ "words": 30,
+ "wer": 0.06666666666666667
+ }
+ },
+ {
+ "id": "release500_0097",
+ "category": "narrative",
+ "text": "The thing nobody tells you about starting over is how ordinary most of it feels.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0097.wav",
+ "render_seconds": 187.19245061906986,
+ "audio_seconds": 5.433469387755102,
+ "rtf": 34.45173557817917,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE THING NOBODY TELLS YOU ABOUT STARTING OVER IS HOW ORDINARY MOST OF IT FEELS",
+ "raw": {
+ "normalized_reference": "the thing nobody tells you about starting over is how ordinary most of it feels.",
+ "normalized_hypothesis": "the thing nobody tells you about starting over is how ordinary most of it feels",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the thing nobody tells you about starting over is how ordinary most of it feels",
+ "normalized_hypothesis": "the thing nobody tells you about starting over is how ordinary most of it feels",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0098",
+ "category": "question",
+ "text": "What exactly are we waiting for at this point?",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0098.wav",
+ "render_seconds": 56.579715312924236,
+ "audio_seconds": 3.2043537414965986,
+ "rtf": 17.65713771866479,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WHAT EXACTLY ARE WE WAITING FOR AT THIS POINT",
+ "raw": {
+ "normalized_reference": "what exactly are we waiting for at this point?",
+ "normalized_hypothesis": "what exactly are we waiting for at this point",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "what exactly are we waiting for at this point",
+ "normalized_hypothesis": "what exactly are we waiting for at this point",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0099",
+ "category": "short",
+ "text": "I stand by what I said.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0099.wav",
+ "render_seconds": 103.61576511000749,
+ "audio_seconds": 2.089795918367347,
+ "rtf": 49.58176260146842,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I STAND BY WHAT I SAID",
+ "raw": {
+ "normalized_reference": "i stand by what i said.",
+ "normalized_hypothesis": "i stand by what i said",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i stand by what i said",
+ "normalized_hypothesis": "i stand by what i said",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0100",
+ "category": "technical",
+ "text": "Double-check that value before you push anything to production.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0100.wav",
+ "render_seconds": 162.6109438110143,
+ "audio_seconds": 4.31891156462585,
+ "rtf": 37.65090842398422,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "DOUBLE CHECK THAT VALUE BEFORE YOU PUSH ANYTHING TO PRODUCTION",
+ "raw": {
+ "normalized_reference": "double-check that value before you push anything to production.",
+ "normalized_hypothesis": "double check that value before you push anything to production",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "double check that value before you push anything to production",
+ "normalized_hypothesis": "double check that value before you push anything to production",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0101",
+ "category": "conversational",
+ "text": "I understand your point; I just do not agree with the conclusion.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0101.wav",
+ "render_seconds": 168.29973123699892,
+ "audio_seconds": 4.458231292517007,
+ "rtf": 37.75033644384589,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I UNDERSTAND YOUR POINT I JUST DO NOT AGREE WITH THE CONCLUSION",
+ "raw": {
+ "normalized_reference": "i understand your point; i just do not agree with the conclusion.",
+ "normalized_hypothesis": "i understand your point i just do not agree with the conclusion",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i understand your point i just do not agree with the conclusion",
+ "normalized_hypothesis": "i understand your point i just do not agree with the conclusion",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0102",
+ "category": "descriptive",
+ "text": "The sky that day was a strange shade of orange I'd never seen before.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0102.wav",
+ "render_seconds": 171.00677004898898,
+ "audio_seconds": 4.597551020408163,
+ "rtf": 37.1951870223742,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE SKY THAT DAY WAS A STRANGE SHADE OF ORANGE IH'D NEVER SEEN BEFORE",
+ "raw": {
+ "normalized_reference": "the sky that day was a strange shade of orange i'd never seen before.",
+ "normalized_hypothesis": "the sky that day was a strange shade of orange ih'd never seen before",
+ "errors": 1,
+ "words": 14,
+ "wer": 0.07142857142857142
+ },
+ "semantic": {
+ "normalized_reference": "the sky that day was a strange shade of orange i would never seen before",
+ "normalized_hypothesis": "the sky that day was a strange shade of orange ih would never seen before",
+ "errors": 1,
+ "words": 15,
+ "wer": 0.06666666666666667
+ }
+ },
+ {
+ "id": "release500_0103",
+ "category": "emotional",
+ "text": "It hurt more than I expected, and I thought I was prepared.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0103.wav",
+ "render_seconds": 163.80737377295736,
+ "audio_seconds": 4.388571428571429,
+ "rtf": 37.32589897170252,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT HURT MORE THAN I EXPECTED AND I THOUGHT I WAS PREPARED",
+ "raw": {
+ "normalized_reference": "it hurt more than i expected, and i thought i was prepared.",
+ "normalized_hypothesis": "it hurt more than i expected and i thought i was prepared",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it hurt more than i expected and i thought i was prepared",
+ "normalized_hypothesis": "it hurt more than i expected and i thought i was prepared",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0104",
+ "category": "long",
+ "text": "The hardest part wasn't making the decision, it was sitting with the uncertainty of not knowing whether it was the right one for months afterward.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0104.wav",
+ "render_seconds": 190.2936620110413,
+ "audio_seconds": 8.986122448979591,
+ "rtf": 21.17639316529121,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE HARDEST PART WASN'T MAKING THE DECISION IT WAS SITTING WITH THE UNCERTAINTY OF NOT KNOWING WHETHER IT WAS THE RIGHT ONE FOR MONTHS AFTERWARD",
+ "raw": {
+ "normalized_reference": "the hardest part wasn't making the decision, it was sitting with the uncertainty of not knowing whether it was the right one for months afterward.",
+ "normalized_hypothesis": "the hardest part wasn't making the decision it was sitting with the uncertainty of not knowing whether it was the right one for months afterward",
+ "errors": 0,
+ "words": 25,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the hardest part was not making the decision it was sitting with the uncertainty of not knowing whether it was the right one for months afterward",
+ "normalized_hypothesis": "the hardest part was not making the decision it was sitting with the uncertainty of not knowing whether it was the right one for months afterward",
+ "errors": 0,
+ "words": 26,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0105",
+ "category": "narrative",
+ "text": "There was this one day last summer when everything just clicked into place.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0105.wav",
+ "render_seconds": 197.19412988598924,
+ "audio_seconds": 5.154829931972789,
+ "rtf": 38.25424553056432,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THERE WAS THIS ONE DAY LAST SUMMER WHEN EVERYTHING JUST CLICKED INTO PLACE",
+ "raw": {
+ "normalized_reference": "there was this one day last summer when everything just clicked into place.",
+ "normalized_hypothesis": "there was this one day last summer when everything just clicked into place",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "there was this one day last summer when everything just clicked into place",
+ "normalized_hypothesis": "there was this one day last summer when everything just clicked into place",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0106",
+ "category": "question",
+ "text": "Have you ever thought about why we keep doing things the same way when they stop working?",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0106.wav",
+ "render_seconds": 178.0883134501055,
+ "audio_seconds": 5.433469387755102,
+ "rtf": 32.77616946760505,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HAVE YOU EVER THOUGHT ABOUT WHY WE KEEP DOING THINGS THE SAME WAY WHEN THEY STOP WORKING",
+ "raw": {
+ "normalized_reference": "have you ever thought about why we keep doing things the same way when they stop working?",
+ "normalized_hypothesis": "have you ever thought about why we keep doing things the same way when they stop working",
+ "errors": 0,
+ "words": 17,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "have you ever thought about why we keep doing things the same way when they stop working",
+ "normalized_hypothesis": "have you ever thought about why we keep doing things the same way when they stop working",
+ "errors": 0,
+ "words": 17,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0107",
+ "category": "short",
+ "text": "I can't believe you just said that.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0107.wav",
+ "render_seconds": 117.0924708719831,
+ "audio_seconds": 2.4380952380952383,
+ "rtf": 48.02620875608681,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I CAN'T BELIEVE YOU JUST SAID THAT",
+ "raw": {
+ "normalized_reference": "i can't believe you just said that.",
+ "normalized_hypothesis": "i can't believe you just said that",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i can not believe you just said that",
+ "normalized_hypothesis": "i can not believe you just said that",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0108",
+ "category": "technical",
+ "text": "You'll want to isolate that variable before drawing any conclusions.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0108.wav",
+ "render_seconds": 195.0017470959574,
+ "audio_seconds": 4.597551020408163,
+ "rtf": 42.414264948960906,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "YOU'LL WANT ISOLATE THAT VARIABLE BEFORE DRAWING ANY CONCLUSIONS",
+ "raw": {
+ "normalized_reference": "you'll want to isolate that variable before drawing any conclusions.",
+ "normalized_hypothesis": "you'll want isolate that variable before drawing any conclusions",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ },
+ "semantic": {
+ "normalized_reference": "you will want to isolate that variable before drawing any conclusions",
+ "normalized_hypothesis": "you will want isolate that variable before drawing any conclusions",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ }
+ },
+ {
+ "id": "release500_0109",
+ "category": "conversational",
+ "text": "The meeting that could have been an email lasted two hours.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0109.wav",
+ "render_seconds": 130.09551886399277,
+ "audio_seconds": 3.9706122448979593,
+ "rtf": 32.76459922047515,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE MEETING THAT COULD HAVE BEEN AN EMALE LASTED TWO HOURS",
+ "raw": {
+ "normalized_reference": "the meeting that could have been an email lasted two hours.",
+ "normalized_hypothesis": "the meeting that could have been an emale lasted two hours",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ },
+ "semantic": {
+ "normalized_reference": "the meeting that could have been an email lasted 2 hours",
+ "normalized_hypothesis": "the meeting that could have been an emale lasted 2 hours",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ }
+ },
+ {
+ "id": "release500_0110",
+ "category": "descriptive",
+ "text": "There's a specific kind of tired that sleep doesn't fix.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0110.wav",
+ "render_seconds": 73.60374982096255,
+ "audio_seconds": 3.622312925170068,
+ "rtf": 20.319544816108575,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THERE'S A SPECIFIC KIND OF TIRED THAT SLEEP DOESN'T FIX",
+ "raw": {
+ "normalized_reference": "there's a specific kind of tired that sleep doesn't fix.",
+ "normalized_hypothesis": "there's a specific kind of tired that sleep doesn't fix",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "there is a specific kind of tired that sleep does not fix",
+ "normalized_hypothesis": "there is a specific kind of tired that sleep does not fix",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0111",
+ "category": "emotional",
+ "text": "Something about that moment broke something open in me that I hadn't realized was closed.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0111.wav",
+ "render_seconds": 191.18641630304046,
+ "audio_seconds": 5.642448979591837,
+ "rtf": 33.88358795879985,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SOMETHING ABOUT THAT MOMENT BROKE SOMETHING OPEN IN ME THAT I HADN'T REALIZED WAS CLOSED",
+ "raw": {
+ "normalized_reference": "something about that moment broke something open in me that i hadn't realized was closed.",
+ "normalized_hypothesis": "something about that moment broke something open in me that i hadn't realized was closed",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "something about that moment broke something open in me that i had not realized was closed",
+ "normalized_hypothesis": "something about that moment broke something open in me that i had not realized was closed",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0112",
+ "category": "long",
+ "text": "The thing about making something good is that you don't usually know it's good while you're making it. You just know you're not done yet.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0112.wav",
+ "render_seconds": 227.912969725905,
+ "audio_seconds": 7.662585034013605,
+ "rtf": 29.74361376927205,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE THING ABOUT MAKING SOMETHING GOOD IS THAT YOU DON'T USUALLY KNOW IT'S GOOD WHILE YOU'RE MAKING IT YOU JUST KNOW YOU'RE NOT DONE YET",
+ "raw": {
+ "normalized_reference": "the thing about making something good is that you don't usually know it's good while you're making it. you just know you're not done yet.",
+ "normalized_hypothesis": "the thing about making something good is that you don't usually know it's good while you're making it you just know you're not done yet",
+ "errors": 0,
+ "words": 25,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the thing about making something good is that you do not usually know it is good while you are making it you just know you are not done yet",
+ "normalized_hypothesis": "the thing about making something good is that you do not usually know it is good while you are making it you just know you are not done yet",
+ "errors": 0,
+ "words": 29,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0113",
+ "category": "narrative",
+ "text": "She opened the door slowly, as if she expected to find something she wasn't ready for.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0113.wav",
+ "render_seconds": 229.01116321899462,
+ "audio_seconds": 5.921088435374149,
+ "rtf": 38.67720702342855,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SHE OPENED THE DOOR SLOWLY AS IF SHE EXPECTED TO FIND SOMETHING SHE WASN'T READY FOR",
+ "raw": {
+ "normalized_reference": "she opened the door slowly, as if she expected to find something she wasn't ready for.",
+ "normalized_hypothesis": "she opened the door slowly as if she expected to find something she wasn't ready for",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "she opened the door slowly as if she expected to find something she was not ready for",
+ "normalized_hypothesis": "she opened the door slowly as if she expected to find something she was not ready for",
+ "errors": 0,
+ "words": 17,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0114",
+ "category": "question",
+ "text": "What would you do differently if you could go back?",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0114.wav",
+ "render_seconds": 135.2906975609949,
+ "audio_seconds": 3.3436734693877552,
+ "rtf": 40.4616954375534,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WHAT WOULD YOU DO DIFFERENTLY IF YOU COULD GO BACK",
+ "raw": {
+ "normalized_reference": "what would you do differently if you could go back?",
+ "normalized_hypothesis": "what would you do differently if you could go back",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "what would you do differently if you could go back",
+ "normalized_hypothesis": "what would you do differently if you could go back",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0115",
+ "category": "short",
+ "text": "Okay. Okay okay okay.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0115.wav",
+ "render_seconds": 126.49804918107111,
+ "audio_seconds": 2.4380952380952383,
+ "rtf": 51.88396548442369,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "OK OK OK OK",
+ "raw": {
+ "normalized_reference": "okay. okay okay okay.",
+ "normalized_hypothesis": "ok ok ok ok",
+ "errors": 4,
+ "words": 4,
+ "wer": 1.0
+ },
+ "semantic": {
+ "normalized_reference": "okay okay okay okay",
+ "normalized_hypothesis": "ok ok ok ok",
+ "errors": 4,
+ "words": 4,
+ "wer": 1.0
+ }
+ },
+ {
+ "id": "release500_0116",
+ "category": "technical",
+ "text": "The latency goes up significantly at scale, which is expected.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0116.wav",
+ "render_seconds": 94.58960826997645,
+ "audio_seconds": 4.597551020408163,
+ "rtf": 20.57391159991498,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE LATENCY GOES UP SIGNIFICANTLY AT SCALE WHICH IS EXPECTED",
+ "raw": {
+ "normalized_reference": "the latency goes up significantly at scale, which is expected.",
+ "normalized_hypothesis": "the latency goes up significantly at scale which is expected",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the latency goes up significantly at scale which is expected",
+ "normalized_hypothesis": "the latency goes up significantly at scale which is expected",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0117",
+ "category": "conversational",
+ "text": "We ended up staying until two in the morning talking about nothing.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0117.wav",
+ "render_seconds": 169.9065376300132,
+ "audio_seconds": 4.458231292517007,
+ "rtf": 38.11074986513052,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WE ENDED UP STAYING UNTIL TWO IN THE MORNING TALKING ABOUT NOTHING",
+ "raw": {
+ "normalized_reference": "we ended up staying until two in the morning talking about nothing.",
+ "normalized_hypothesis": "we ended up staying until two in the morning talking about nothing",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "we ended up staying until 2 in the morning talking about nothing",
+ "normalized_hypothesis": "we ended up staying until 2 in the morning talking about nothing",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0118",
+ "category": "descriptive",
+ "text": "The rain started slowly, and then all at once, the way it usually does.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0118.wav",
+ "render_seconds": 184.69438406801783,
+ "audio_seconds": 5.154829931972789,
+ "rtf": 35.829384578228755,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE RAIN STARTED SLOWLY AND THEN ALL AT ONCE THE WAY IT USUALLY DOES",
+ "raw": {
+ "normalized_reference": "the rain started slowly, and then all at once, the way it usually does.",
+ "normalized_hypothesis": "the rain started slowly and then all at once the way it usually does",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the rain started slowly and then all at once the way it usually does",
+ "normalized_hypothesis": "the rain started slowly and then all at once the way it usually does",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0119",
+ "category": "emotional",
+ "text": "I honestly did not think I was going to make it through that year.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0119.wav",
+ "render_seconds": 184.18769364000764,
+ "audio_seconds": 4.31891156462585,
+ "rtf": 42.64678516425328,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I HONESTLY DID NOT THINK I WAS GOING TO MAKE IT THROUGH THAT YEAR",
+ "raw": {
+ "normalized_reference": "i honestly did not think i was going to make it through that year.",
+ "normalized_hypothesis": "i honestly did not think i was going to make it through that year",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i honestly did not think i was going to make it through that year",
+ "normalized_hypothesis": "i honestly did not think i was going to make it through that year",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0120",
+ "category": "long",
+ "text": "There's an assumption that people who seem confident all the time actually feel confident, and in my experience that's almost never true.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0120.wav",
+ "render_seconds": 327.906903670053,
+ "audio_seconds": 8.986122448979591,
+ "rtf": 36.490366732904704,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THERE'S AN ASSUMPTION THAT PEOPLE WHO SEEM CONFIDENT ALL THE TIME ACTUALLY FEEL CONFIDENT AND IN MY EXPERIENCE THAT'S ALMOST NEVER TRUE",
+ "raw": {
+ "normalized_reference": "there's an assumption that people who seem confident all the time actually feel confident, and in my experience that's almost never true.",
+ "normalized_hypothesis": "there's an assumption that people who seem confident all the time actually feel confident and in my experience that's almost never true",
+ "errors": 0,
+ "words": 22,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "there is an assumption that people who seem confident all the time actually feel confident and in my experience that is almost never true",
+ "normalized_hypothesis": "there is an assumption that people who seem confident all the time actually feel confident and in my experience that is almost never true",
+ "errors": 0,
+ "words": 24,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0121",
+ "category": "narrative",
+ "text": "The moment I opened the door I knew something was different, even if I couldn't say what.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0121.wav",
+ "render_seconds": 177.3054225560045,
+ "audio_seconds": 5.572789115646258,
+ "rtf": 31.816280658853348,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE MOMENT I OPENED THE DOOR I KNEW SOMETHING WAS DIFFERENT EVEN IF I COULDN'T SAY WHAT",
+ "raw": {
+ "normalized_reference": "the moment i opened the door i knew something was different, even if i couldn't say what.",
+ "normalized_hypothesis": "the moment i opened the door i knew something was different even if i couldn't say what",
+ "errors": 0,
+ "words": 17,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the moment i opened the door i knew something was different even if i could not say what",
+ "normalized_hypothesis": "the moment i opened the door i knew something was different even if i could not say what",
+ "errors": 0,
+ "words": 18,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0122",
+ "category": "question",
+ "text": "How do you explain something that doesn't make sense even to you?",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0122.wav",
+ "render_seconds": 138.50993507902604,
+ "audio_seconds": 4.040272108843538,
+ "rtf": 34.28232835502564,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HOW DO YOU EXPLAIN SOMETHING THAT DOESN'T MAKE SENSE EVEN TO YOU",
+ "raw": {
+ "normalized_reference": "how do you explain something that doesn't make sense even to you?",
+ "normalized_hypothesis": "how do you explain something that doesn't make sense even to you",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "how do you explain something that does not make sense even to you",
+ "normalized_hypothesis": "how do you explain something that does not make sense even to you",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0123",
+ "category": "short",
+ "text": "Hold on a second.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0123.wav",
+ "render_seconds": 84.58820638398174,
+ "audio_seconds": 1.6718367346938776,
+ "rtf": 50.59597305682502,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HOLD ON A SECOND",
+ "raw": {
+ "normalized_reference": "hold on a second.",
+ "normalized_hypothesis": "hold on a second",
+ "errors": 0,
+ "words": 4,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "hold on a 2nd",
+ "normalized_hypothesis": "hold on a 2nd",
+ "errors": 0,
+ "words": 4,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0124",
+ "category": "technical",
+ "text": "A deterministic split prevents duplicate normalized text from leaking into evaluation.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0124.wav",
+ "render_seconds": 211.10324295493774,
+ "audio_seconds": 6.339047619047619,
+ "rtf": 33.30204403586007,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "A DETERMINISTIC SPLIT PREVENTS DUPLICATE NORMALIZED TEXT FROM LEAKING INTO EVALUATION",
+ "raw": {
+ "normalized_reference": "a deterministic split prevents duplicate normalized text from leaking into evaluation.",
+ "normalized_hypothesis": "a deterministic split prevents duplicate normalized text from leaking into evaluation",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "a deterministic split prevents duplicate normalized text from leaking into evaluation",
+ "normalized_hypothesis": "a deterministic split prevents duplicate normalized text from leaking into evaluation",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0125",
+ "category": "conversational",
+ "text": "Honestly, that was much better than I expected, but the ending still felt rushed.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0125.wav",
+ "render_seconds": 229.00751723907888,
+ "audio_seconds": 5.642448979591837,
+ "rtf": 40.58654638568745,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HONESTLY THAT WAS MUCH BETTER THAN I EXPECTED BUT THE ENDING STILL FELT RUSHED",
+ "raw": {
+ "normalized_reference": "honestly, that was much better than i expected, but the ending still felt rushed.",
+ "normalized_hypothesis": "honestly that was much better than i expected but the ending still felt rushed",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "honestly that was much better than i expected but the ending still felt rushed",
+ "normalized_hypothesis": "honestly that was much better than i expected but the ending still felt rushed",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0126",
+ "category": "descriptive",
+ "text": "She had a habit of tilting her head slightly when thinking hard about something.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0126.wav",
+ "render_seconds": 204.49203690805007,
+ "audio_seconds": 5.224489795918367,
+ "rtf": 39.14105393943146,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SHE HAD A HABIT OF TILTING HER HEAD SLIGHTLY WHEN THINKING HARD ABOUT SOMETHING",
+ "raw": {
+ "normalized_reference": "she had a habit of tilting her head slightly when thinking hard about something.",
+ "normalized_hypothesis": "she had a habit of tilting her head slightly when thinking hard about something",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "she had a habit of tilting her head slightly when thinking hard about something",
+ "normalized_hypothesis": "she had a habit of tilting her head slightly when thinking hard about something",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0127",
+ "category": "emotional",
+ "text": "I cannot stress enough how much I did not want to be right about this.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0127.wav",
+ "render_seconds": 182.69252880394924,
+ "audio_seconds": 4.667210884353741,
+ "rtf": 39.14383415079953,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I CANNOT STRESS ENOUGH HOW MUCH I DID NOT WANT TO BE RIGHT ABOUT THIS",
+ "raw": {
+ "normalized_reference": "i cannot stress enough how much i did not want to be right about this.",
+ "normalized_hypothesis": "i cannot stress enough how much i did not want to be right about this",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i cannot stress enough how much i did not want to be right about this",
+ "normalized_hypothesis": "i cannot stress enough how much i did not want to be right about this",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0128",
+ "category": "long",
+ "text": "I used to think the goal was to get to a point where nothing was hard anymore, and now I think that point doesn't exist and I'm not sure I'd want it to.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0128.wav",
+ "render_seconds": 181.790965070948,
+ "audio_seconds": 9.264761904761905,
+ "rtf": 19.621763294047636,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I USED TO THINK THE GOAL WAS TO GET TO A POINT WHERE NOTHING WAS HARD ANY MORE AND NOW I THINK THAT POINT DOESN'T EXIST AND I'M NOT SURE I'D WANT IT TO",
+ "raw": {
+ "normalized_reference": "i used to think the goal was to get to a point where nothing was hard anymore, and now i think that point doesn't exist and i'm not sure i'd want it to.",
+ "normalized_hypothesis": "i used to think the goal was to get to a point where nothing was hard any more and now i think that point doesn't exist and i'm not sure i'd want it to",
+ "errors": 2,
+ "words": 33,
+ "wer": 0.06060606060606061
+ },
+ "semantic": {
+ "normalized_reference": "i used to think the goal was to get to a point where nothing was hard anymore and now i think that point does not exist and i am not sure i would want it to",
+ "normalized_hypothesis": "i used to think the goal was to get to a point where nothing was hard any more and now i think that point does not exist and i am not sure i would want it to",
+ "errors": 2,
+ "words": 36,
+ "wer": 0.05555555555555555
+ }
+ },
+ {
+ "id": "release500_0129",
+ "category": "narrative",
+ "text": "So this is the part of the story where everything goes completely sideways.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0129.wav",
+ "render_seconds": 197.40703524800483,
+ "audio_seconds": 5.224489795918367,
+ "rtf": 37.784940340438425,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SO THIS IS THE PART OF THE STORY WHERE EVERYTHING GOES COMPLETELY SIDEWAYS",
+ "raw": {
+ "normalized_reference": "so this is the part of the story where everything goes completely sideways.",
+ "normalized_hypothesis": "so this is the part of the story where everything goes completely sideways",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "so this is the part of the story where everything goes completely sideways",
+ "normalized_hypothesis": "so this is the part of the story where everything goes completely sideways",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0130",
+ "category": "question",
+ "text": "Is there a version of this where it actually works out okay?",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0130.wav",
+ "render_seconds": 150.68803610792384,
+ "audio_seconds": 4.179591836734694,
+ "rtf": 36.05328988910287,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IS THERE A VERSION OF THIS WHERE IT ACTUALLY WORKS OUT O QU",
+ "raw": {
+ "normalized_reference": "is there a version of this where it actually works out okay?",
+ "normalized_hypothesis": "is there a version of this where it actually works out o qu",
+ "errors": 2,
+ "words": 12,
+ "wer": 0.16666666666666666
+ },
+ "semantic": {
+ "normalized_reference": "is there a version of this where it actually works out okay",
+ "normalized_hypothesis": "is there a version of this where it actually works out 0 qu",
+ "errors": 2,
+ "words": 12,
+ "wer": 0.16666666666666666
+ }
+ },
+ {
+ "id": "release500_0131",
+ "category": "short",
+ "text": "Oh no. Oh no no no.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0131.wav",
+ "render_seconds": 95.0924721530173,
+ "audio_seconds": 2.089795918367347,
+ "rtf": 45.50323374509616,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "OH NO OH NO NO NO",
+ "raw": {
+ "normalized_reference": "oh no. oh no no no.",
+ "normalized_hypothesis": "oh no oh no no no",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "0 no 0 no no no",
+ "normalized_hypothesis": "0 no 0 no no no",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0132",
+ "category": "technical",
+ "text": "Always test on a copy before modifying the original.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0132.wav",
+ "render_seconds": 155.70008645695634,
+ "audio_seconds": 3.7616326530612243,
+ "rtf": 41.39162454639139,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "ALWAYS TEST ON A COPY BEFORE MODIFYING THE ORIGINAL",
+ "raw": {
+ "normalized_reference": "always test on a copy before modifying the original.",
+ "normalized_hypothesis": "always test on a copy before modifying the original",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "always test on a copy before modifying the original",
+ "normalized_hypothesis": "always test on a copy before modifying the original",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0133",
+ "category": "conversational",
+ "text": "I wasn't even supposed to be there. That's the wild part.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0133.wav",
+ "render_seconds": 162.20817584195174,
+ "audio_seconds": 3.9706122448979593,
+ "rtf": 40.85218244374813,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I WASN'T EVEN SUPPOSED TO BE THERE THAT'S THE WILD PART",
+ "raw": {
+ "normalized_reference": "i wasn't even supposed to be there. that's the wild part.",
+ "normalized_hypothesis": "i wasn't even supposed to be there that's the wild part",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i was not even supposed to be there that is the wild part",
+ "normalized_hypothesis": "i was not even supposed to be there that is the wild part",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0134",
+ "category": "descriptive",
+ "text": "It's a small town. Not the kind you leave. The kind you leave and come back to.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0134.wav",
+ "render_seconds": 139.6144138079835,
+ "audio_seconds": 5.7817687074829935,
+ "rtf": 24.147353668314164,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT'S A SMALL TOWN NOT THE KIND YOU LEAVE THE KIND YOU LEAVE AND COME BACK TO",
+ "raw": {
+ "normalized_reference": "it's a small town. not the kind you leave. the kind you leave and come back to.",
+ "normalized_hypothesis": "it's a small town not the kind you leave the kind you leave and come back to",
+ "errors": 0,
+ "words": 17,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it is a small town not the kind you leave the kind you leave and come back to",
+ "normalized_hypothesis": "it is a small town not the kind you leave the kind you leave and come back to",
+ "errors": 0,
+ "words": 18,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0135",
+ "category": "emotional",
+ "text": "She cried. I cried. We all just sort of stood there crying.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0135.wav",
+ "render_seconds": 179.39809406595305,
+ "audio_seconds": 4.806530612244898,
+ "rtf": 37.32382221990361,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SHE CRIED I CRIED WE ALL JUST SORT OF STOOD THERE CRYING",
+ "raw": {
+ "normalized_reference": "she cried. i cried. we all just sort of stood there crying.",
+ "normalized_hypothesis": "she cried i cried we all just sort of stood there crying",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "she cried i cried we all just sort of stood there crying",
+ "normalized_hypothesis": "she cried i cried we all just sort of stood there crying",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0136",
+ "category": "long",
+ "text": "There's a version of being busy that's actually just a very elaborate way of avoiding the one thing you know you should be doing.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0136.wav",
+ "render_seconds": 213.9000673040282,
+ "audio_seconds": 8.150204081632653,
+ "rtf": 26.244749844494645,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THERE'S A VERSION OF BEING BUSY THAT'S ACTUALLY JUST A VERY ELABORATE WAY OF AVOIDING THE ONE THING YOU KNOW YOU SHOULD BE DOING",
+ "raw": {
+ "normalized_reference": "there's a version of being busy that's actually just a very elaborate way of avoiding the one thing you know you should be doing.",
+ "normalized_hypothesis": "there's a version of being busy that's actually just a very elaborate way of avoiding the one thing you know you should be doing",
+ "errors": 0,
+ "words": 24,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "there is a version of being busy that is actually just a very elaborate way of avoiding the one thing you know you should be doing",
+ "normalized_hypothesis": "there is a version of being busy that is actually just a very elaborate way of avoiding the one thing you know you should be doing",
+ "errors": 0,
+ "words": 26,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0137",
+ "category": "narrative",
+ "text": "The letter arrived on a Tuesday, which felt appropriate somehow.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0137.wav",
+ "render_seconds": 173.59912472194992,
+ "audio_seconds": 4.667210884353741,
+ "rtf": 37.19547477572097,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE LETTER ARRIVED ON A TUESDAY WHICH FELT APPROPRIATE SOMEHOW",
+ "raw": {
+ "normalized_reference": "the letter arrived on a tuesday, which felt appropriate somehow.",
+ "normalized_hypothesis": "the letter arrived on a tuesday which felt appropriate somehow",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the letter arrived on a tuesday which felt appropriate somehow",
+ "normalized_hypothesis": "the letter arrived on a tuesday which felt appropriate somehow",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0138",
+ "category": "question",
+ "text": "What exactly did she say when you told her?",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0138.wav",
+ "render_seconds": 119.71911585098132,
+ "audio_seconds": 2.9257142857142857,
+ "rtf": 40.91961967562838,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WHAT EXACTLY DID SHE SAY WHEN YOU TOLD HER",
+ "raw": {
+ "normalized_reference": "what exactly did she say when you told her?",
+ "normalized_hypothesis": "what exactly did she say when you told her",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "what exactly did she say when you told her",
+ "normalized_hypothesis": "what exactly did she say when you told her",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0139",
+ "category": "short",
+ "text": "No way. Absolutely not.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0139.wav",
+ "render_seconds": 92.69116011692677,
+ "audio_seconds": 2.5774149659863945,
+ "rtf": 35.96283925566996,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "NO WAY ABSOLUTELY NOT",
+ "raw": {
+ "normalized_reference": "no way. absolutely not.",
+ "normalized_hypothesis": "no way absolutely not",
+ "errors": 0,
+ "words": 4,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "no way absolutely not",
+ "normalized_hypothesis": "no way absolutely not",
+ "errors": 0,
+ "words": 4,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0140",
+ "category": "technical",
+ "text": "The first thing you want to do is make sure everything is backed up.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0140.wav",
+ "render_seconds": 121.99403522501234,
+ "audio_seconds": 4.597551020408163,
+ "rtf": 26.53456909634945,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE FIRST THING YOU WANT TO DO IS MAKE SURE EVERYTHING IS BACKED UP",
+ "raw": {
+ "normalized_reference": "the first thing you want to do is make sure everything is backed up.",
+ "normalized_hypothesis": "the first thing you want to do is make sure everything is backed up",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the 1st thing you want to do is make sure everything is backed up",
+ "normalized_hypothesis": "the 1st thing you want to do is make sure everything is backed up",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0141",
+ "category": "conversational",
+ "text": "I have the most insane idea and you need to just hear me out.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0141.wav",
+ "render_seconds": 144.2943132860819,
+ "audio_seconds": 4.179591836734694,
+ "rtf": 34.52354175301764,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I HAVE THE MOST INSANE IDEA AND YOU NEED TO JUST HEAR ME OUT",
+ "raw": {
+ "normalized_reference": "i have the most insane idea and you need to just hear me out.",
+ "normalized_hypothesis": "i have the most insane idea and you need to just hear me out",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i have the most insane idea and you need to just hear me out",
+ "normalized_hypothesis": "i have the most insane idea and you need to just hear me out",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0142",
+ "category": "descriptive",
+ "text": "There's a coffee shop on the corner that's been there longer than anyone can remember.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0142.wav",
+ "render_seconds": 179.10101063700859,
+ "audio_seconds": 5.085170068027211,
+ "rtf": 35.220259743739646,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THERE'S A COFFEE SHOP ON THE CORNER THAT'S BEEN THERE LONGER THAN ANY ONE CAN REMEMBER",
+ "raw": {
+ "normalized_reference": "there's a coffee shop on the corner that's been there longer than anyone can remember.",
+ "normalized_hypothesis": "there's a coffee shop on the corner that's been there longer than any one can remember",
+ "errors": 2,
+ "words": 15,
+ "wer": 0.13333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "there is a coffee shop on the corner that has been there longer than anyone can remember",
+ "normalized_hypothesis": "there is a coffee shop on the corner that has been there longer than any one can remember",
+ "errors": 2,
+ "words": 17,
+ "wer": 0.11764705882352941
+ }
+ },
+ {
+ "id": "release500_0143",
+ "category": "emotional",
+ "text": "I kept it together until I got to the car, and then I completely fell apart.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0143.wav",
+ "render_seconds": 206.52414280793164,
+ "audio_seconds": 5.294149659863946,
+ "rtf": 39.00987997631316,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I KEPT IT TOGETHER UNTIL I GOT TO THE CAR AND THEN I COMPLETELY FELL APART",
+ "raw": {
+ "normalized_reference": "i kept it together until i got to the car, and then i completely fell apart.",
+ "normalized_hypothesis": "i kept it together until i got to the car and then i completely fell apart",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i kept it together until i got to the car and then i completely fell apart",
+ "normalized_hypothesis": "i kept it together until i got to the car and then i completely fell apart",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0144",
+ "category": "long",
+ "text": "The older I get, the more I think that most of what we call wisdom is just pattern recognition built from having been wrong enough times to start noticing.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0144.wav",
+ "render_seconds": 313.5820713500725,
+ "audio_seconds": 9.682721088435374,
+ "rtf": 32.38573831529666,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE OLDER I GET THE MORE I THINK THAT MOST OF WHAT WE CALL WISDOM IS JUST PATTERN RECOGNITION BUILT FROM HAVING BEEN WRONG ENOUGH TIMES TO START NOTICING",
+ "raw": {
+ "normalized_reference": "the older i get, the more i think that most of what we call wisdom is just pattern recognition built from having been wrong enough times to start noticing.",
+ "normalized_hypothesis": "the older i get the more i think that most of what we call wisdom is just pattern recognition built from having been wrong enough times to start noticing",
+ "errors": 0,
+ "words": 29,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the older i get the more i think that most of what we call wisdom is just pattern recognition built from having been wrong enough times to start noticing",
+ "normalized_hypothesis": "the older i get the more i think that most of what we call wisdom is just pattern recognition built from having been wrong enough times to start noticing",
+ "errors": 0,
+ "words": 29,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0145",
+ "category": "narrative",
+ "text": "I remember the exact moment I realized I had made a huge mistake.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0145.wav",
+ "render_seconds": 187.80746719695162,
+ "audio_seconds": 4.527891156462585,
+ "rtf": 41.477911174807524,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I REMEMBER THE EXACT MOMENT I REALIZED I HAD MADE A HUGE MISTAKE",
+ "raw": {
+ "normalized_reference": "i remember the exact moment i realized i had made a huge mistake.",
+ "normalized_hypothesis": "i remember the exact moment i realized i had made a huge mistake",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i remember the exact moment i realized i had made a huge mistake",
+ "normalized_hypothesis": "i remember the exact moment i realized i had made a huge mistake",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0146",
+ "category": "question",
+ "text": "Do you think it would have gone differently if we'd left earlier?",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0146.wav",
+ "render_seconds": 103.29142461402807,
+ "audio_seconds": 3.9706122448979593,
+ "rtf": 26.013979266485276,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "DO YOU THINK IT WOULD HAVE GONE DIFFERENTLY IF WE'D LEFT EARLIER",
+ "raw": {
+ "normalized_reference": "do you think it would have gone differently if we'd left earlier?",
+ "normalized_hypothesis": "do you think it would have gone differently if we'd left earlier",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "do you think it would have gone differently if we would left earlier",
+ "normalized_hypothesis": "do you think it would have gone differently if we would left earlier",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0147",
+ "category": "short",
+ "text": "You have to be kidding me.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0147.wav",
+ "render_seconds": 100.99945445195772,
+ "audio_seconds": 1.9504761904761905,
+ "rtf": 51.78194686257598,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "YOU HAVE TO BE KITTING ME",
+ "raw": {
+ "normalized_reference": "you have to be kidding me.",
+ "normalized_hypothesis": "you have to be kitting me",
+ "errors": 1,
+ "words": 6,
+ "wer": 0.16666666666666666
+ },
+ "semantic": {
+ "normalized_reference": "you have to be kidding me",
+ "normalized_hypothesis": "you have to be kitting me",
+ "errors": 1,
+ "words": 6,
+ "wer": 0.16666666666666666
+ }
+ },
+ {
+ "id": "release500_0148",
+ "category": "technical",
+ "text": "If the checksum fails, restart the worker process without deleting the cached tensors.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0148.wav",
+ "render_seconds": 184.21122604899574,
+ "audio_seconds": 5.851428571428571,
+ "rtf": 31.48141070173267,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IF THE CHEKSUM FAILS RESTART THE WORKER PROCESS WITHOUT DELIDING THE CATCH TENSERS",
+ "raw": {
+ "normalized_reference": "if the checksum fails, restart the worker process without deleting the cached tensors.",
+ "normalized_hypothesis": "if the cheksum fails restart the worker process without deliding the catch tensers",
+ "errors": 4,
+ "words": 13,
+ "wer": 0.3076923076923077
+ },
+ "semantic": {
+ "normalized_reference": "if the checksum fails restart the worker process without deleting the cached tensors",
+ "normalized_hypothesis": "if the cheksum fails restart the worker process without deliding the catch tensers",
+ "errors": 4,
+ "words": 13,
+ "wer": 0.3076923076923077
+ }
+ },
+ {
+ "id": "release500_0149",
+ "category": "conversational",
+ "text": "I need you to understand how close that was to not working.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0149.wav",
+ "render_seconds": 128.47528654895723,
+ "audio_seconds": 3.900952380952381,
+ "rtf": 32.93433859287234,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I NEED YOU TO UNDERSTAND HOW CLOSE THAT WAS TO NOT WORKING",
+ "raw": {
+ "normalized_reference": "i need you to understand how close that was to not working.",
+ "normalized_hypothesis": "i need you to understand how close that was to not working",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i need you to understand how close that was to not working",
+ "normalized_hypothesis": "i need you to understand how close that was to not working",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0150",
+ "category": "descriptive",
+ "text": "It was the sort of morning where the whole world looks rinsed clean.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0150.wav",
+ "render_seconds": 174.10902754589915,
+ "audio_seconds": 4.527891156462585,
+ "rtf": 38.45256467735453,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT WAS THE SORT OF MORNING WHERE THE WHOLE WORLD LOOKS RINCED CLEAN",
+ "raw": {
+ "normalized_reference": "it was the sort of morning where the whole world looks rinsed clean.",
+ "normalized_hypothesis": "it was the sort of morning where the whole world looks rinced clean",
+ "errors": 1,
+ "words": 13,
+ "wer": 0.07692307692307693
+ },
+ "semantic": {
+ "normalized_reference": "it was the sort of morning where the whole world looks rinsed clean",
+ "normalized_hypothesis": "it was the sort of morning where the whole world looks rinced clean",
+ "errors": 1,
+ "words": 13,
+ "wer": 0.07692307692307693
+ }
+ },
+ {
+ "id": "release500_0151",
+ "category": "emotional",
+ "text": "She started laughing and I started laughing and we couldn't stop for a full minute.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0151.wav",
+ "render_seconds": 179.99561694508884,
+ "audio_seconds": 5.433469387755102,
+ "rtf": 33.12719813066914,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SHE STARTED LAUGHING AND I STARTED LAUGHING AND WE COULDN'T STOP FOR A FULL MINUTE",
+ "raw": {
+ "normalized_reference": "she started laughing and i started laughing and we couldn't stop for a full minute.",
+ "normalized_hypothesis": "she started laughing and i started laughing and we couldn't stop for a full minute",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "she started laughing and i started laughing and we could not stop for a full minute",
+ "normalized_hypothesis": "she started laughing and i started laughing and we could not stop for a full minute",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0152",
+ "category": "long",
+ "text": "We spent three years building something we were genuinely proud of, and then the market shifted and none of it mattered the way we thought it would.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0152.wav",
+ "render_seconds": 231.64165761403274,
+ "audio_seconds": 9.543401360544218,
+ "rtf": 24.272442168441216,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WE SPENT THREE YEARS BUILDING SOMETHING WE WERE GENUINELY PROUD OF AND THEN THE MARKET SHIFTED AND NONE OF IT MATTERED THE WAY WE THOUGHT IT WOULD",
+ "raw": {
+ "normalized_reference": "we spent three years building something we were genuinely proud of, and then the market shifted and none of it mattered the way we thought it would.",
+ "normalized_hypothesis": "we spent three years building something we were genuinely proud of and then the market shifted and none of it mattered the way we thought it would",
+ "errors": 0,
+ "words": 27,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "we spent 3 years building something we were genuinely proud of and then the market shifted and none of it mattered the way we thought it would",
+ "normalized_hypothesis": "we spent 3 years building something we were genuinely proud of and then the market shifted and none of it mattered the way we thought it would",
+ "errors": 0,
+ "words": 27,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0153",
+ "category": "narrative",
+ "text": "And somehow, despite all of that, it actually turned out okay.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0153.wav",
+ "render_seconds": 156.49953109992202,
+ "audio_seconds": 4.527891156462585,
+ "rtf": 34.56344812453205,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "AND SOMEHOW DESPITE ALL OF THAT IT ACTUALLY TURNED OUT O K",
+ "raw": {
+ "normalized_reference": "and somehow, despite all of that, it actually turned out okay.",
+ "normalized_hypothesis": "and somehow despite all of that it actually turned out o k",
+ "errors": 2,
+ "words": 11,
+ "wer": 0.18181818181818182
+ },
+ "semantic": {
+ "normalized_reference": "and somehow despite all of that it actually turned out okay",
+ "normalized_hypothesis": "and somehow despite all of that it actually turned out 0 k",
+ "errors": 2,
+ "words": 11,
+ "wer": 0.18181818181818182
+ }
+ },
+ {
+ "id": "release500_0154",
+ "category": "question",
+ "text": "Is that something you actually want, or something you think you should want?",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0154.wav",
+ "render_seconds": 173.49401189107448,
+ "audio_seconds": 4.876190476190477,
+ "rtf": 35.57982665734926,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IS THAT SOMETHING YOU ACTUALLY WANT OR SOMETHING YOU THINK YOU SHOULD WANT",
+ "raw": {
+ "normalized_reference": "is that something you actually want, or something you think you should want?",
+ "normalized_hypothesis": "is that something you actually want or something you think you should want",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "is that something you actually want or something you think you should want",
+ "normalized_hypothesis": "is that something you actually want or something you think you should want",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0155",
+ "category": "short",
+ "text": "This is fine.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0155.wav",
+ "render_seconds": 98.60779886704404,
+ "audio_seconds": 1.5325170068027212,
+ "rtf": 64.343689779188,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THIS IS FINE",
+ "raw": {
+ "normalized_reference": "this is fine.",
+ "normalized_hypothesis": "this is fine",
+ "errors": 0,
+ "words": 3,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "this is fine",
+ "normalized_hypothesis": "this is fine",
+ "errors": 0,
+ "words": 3,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0156",
+ "category": "technical",
+ "text": "The tricky part isn't the setup, it's making sure all dependencies line up.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0156.wav",
+ "render_seconds": 184.59595783997793,
+ "audio_seconds": 4.9458503401360545,
+ "rtf": 37.32340146687494,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE TRICKY PART ISN'T THE SET UP IT'S MAKING SURE ALL DEPENDENCIES LINE UP",
+ "raw": {
+ "normalized_reference": "the tricky part isn't the setup, it's making sure all dependencies line up.",
+ "normalized_hypothesis": "the tricky part isn't the set up it's making sure all dependencies line up",
+ "errors": 2,
+ "words": 13,
+ "wer": 0.15384615384615385
+ },
+ "semantic": {
+ "normalized_reference": "the tricky part is not the setup it is making sure all dependencies line up",
+ "normalized_hypothesis": "the tricky part is not the set up it is making sure all dependencies line up",
+ "errors": 2,
+ "words": 15,
+ "wer": 0.13333333333333333
+ }
+ },
+ {
+ "id": "release500_0157",
+ "category": "conversational",
+ "text": "He called it a minor setback. I called it a disaster.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0157.wav",
+ "render_seconds": 158.30804827006068,
+ "audio_seconds": 4.1099319727891155,
+ "rtf": 38.51841084431098,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE CALLED IT A MINOR SETBACK I CALLED IT A DISASTER",
+ "raw": {
+ "normalized_reference": "he called it a minor setback. i called it a disaster.",
+ "normalized_hypothesis": "he called it a minor setback i called it a disaster",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he called it a minor setback i called it a disaster",
+ "normalized_hypothesis": "he called it a minor setback i called it a disaster",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0158",
+ "category": "descriptive",
+ "text": "The whole place smelled like pine and old wood and cold air.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0158.wav",
+ "render_seconds": 107.55815625994,
+ "audio_seconds": 4.31891156462585,
+ "rtf": 24.903995983825578,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE WHOLE PLACE SMELLED LIKE PINE AND OLD WOOD IN COLD AIR",
+ "raw": {
+ "normalized_reference": "the whole place smelled like pine and old wood and cold air.",
+ "normalized_hypothesis": "the whole place smelled like pine and old wood in cold air",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "the whole place smelled like pine and old wood and cold air",
+ "normalized_hypothesis": "the whole place smelled like pine and old wood in cold air",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ }
+ },
+ {
+ "id": "release500_0159",
+ "category": "emotional",
+ "text": "I've never been so relieved in my entire life.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0159.wav",
+ "render_seconds": 103.70968468405772,
+ "audio_seconds": 3.2043537414965986,
+ "rtf": 32.365242120746615,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I HAVE NEVER BEEN SO RELIEVED IN MY ENTIRE LIFE",
+ "raw": {
+ "normalized_reference": "i've never been so relieved in my entire life.",
+ "normalized_hypothesis": "i have never been so relieved in my entire life",
+ "errors": 2,
+ "words": 9,
+ "wer": 0.2222222222222222
+ },
+ "semantic": {
+ "normalized_reference": "i have never been so relieved in my entire life",
+ "normalized_hypothesis": "i have never been so relieved in my entire life",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0160",
+ "category": "narrative",
+ "text": "I wrote it all down that night so I wouldn't forget, and I still haven't looked at it.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0160.wav",
+ "render_seconds": 182.30416422104463,
+ "audio_seconds": 5.363809523809524,
+ "rtf": 33.987814707403565,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I WROTE IT ALL DOWN THAT NIGHT SO I WOULDN'T FORGET AND I STILL HAVEN'T LOOKED AT IT",
+ "raw": {
+ "normalized_reference": "i wrote it all down that night so i wouldn't forget, and i still haven't looked at it.",
+ "normalized_hypothesis": "i wrote it all down that night so i wouldn't forget and i still haven't looked at it",
+ "errors": 0,
+ "words": 18,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i wrote it all down that night so i would not forget and i still have not looked at it",
+ "normalized_hypothesis": "i wrote it all down that night so i would not forget and i still have not looked at it",
+ "errors": 0,
+ "words": 20,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0161",
+ "category": "short",
+ "text": "That's kind of wild actually.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0161.wav",
+ "render_seconds": 86.89950804307591,
+ "audio_seconds": 2.2291156462585033,
+ "rtf": 38.983849128210935,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THAT'S KIND OF WILD ACTUALLY",
+ "raw": {
+ "normalized_reference": "that's kind of wild actually.",
+ "normalized_hypothesis": "that's kind of wild actually",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "that is kind of wild actually",
+ "normalized_hypothesis": "that is kind of wild actually",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0162",
+ "category": "technical",
+ "text": "Version pinning will save you a lot of pain later down the line.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0162.wav",
+ "render_seconds": 177.2036938010715,
+ "audio_seconds": 4.31891156462585,
+ "rtf": 41.029711108804044,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "VERSION PINNING WILL SAVE YOU A LOT OF PAIN LATER DOWN THE LINE",
+ "raw": {
+ "normalized_reference": "version pinning will save you a lot of pain later down the line.",
+ "normalized_hypothesis": "version pinning will save you a lot of pain later down the line",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "version pinning will save you a lot of pain later down the line",
+ "normalized_hypothesis": "version pinning will save you a lot of pain later down the line",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0163",
+ "category": "conversational",
+ "text": "At some point you just have to commit and see what happens.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0163.wav",
+ "render_seconds": 142.88773106306326,
+ "audio_seconds": 4.040272108843538,
+ "rtf": 35.36586824196912,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "AT SOME POINT YOU JUST HAVE TO COMMIT AND SEE WHAT HAPPENS",
+ "raw": {
+ "normalized_reference": "at some point you just have to commit and see what happens.",
+ "normalized_hypothesis": "at some point you just have to commit and see what happens",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "at some point you just have to commit and see what happens",
+ "normalized_hypothesis": "at some point you just have to commit and see what happens",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0164",
+ "category": "emotional",
+ "text": "Nobody told me it was going to be this hard, and honestly, I'm grateful.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0164.wav",
+ "render_seconds": 89.99885719595477,
+ "audio_seconds": 5.085170068027211,
+ "rtf": 17.698298383729334,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "NOBODY TOLD ME IT WAS GOING TO BE THIS HARD AND HONESTLY I'M GRATEFUL",
+ "raw": {
+ "normalized_reference": "nobody told me it was going to be this hard, and honestly, i'm grateful.",
+ "normalized_hypothesis": "nobody told me it was going to be this hard and honestly i'm grateful",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "nobody told me it was going to be this hard and honestly i am grateful",
+ "normalized_hypothesis": "nobody told me it was going to be this hard and honestly i am grateful",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0165",
+ "category": "narrative",
+ "text": "Chapter seven. In which everything that could go wrong did, and a few things that couldn't.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0165.wav",
+ "render_seconds": 196.1990482499823,
+ "audio_seconds": 6.060408163265306,
+ "rtf": 32.373900068188085,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "CHAPTER SEVEN IN WHICH EVERYTHING THAT COULD GO WRONG DID AND A FEW THINGS THAT COULDN'T",
+ "raw": {
+ "normalized_reference": "chapter seven. in which everything that could go wrong did, and a few things that couldn't.",
+ "normalized_hypothesis": "chapter seven in which everything that could go wrong did and a few things that couldn't",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "chapter 7 in which everything that could go wrong did and a few things that could not",
+ "normalized_hypothesis": "chapter 7 in which everything that could go wrong did and a few things that could not",
+ "errors": 0,
+ "words": 17,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0166",
+ "category": "short",
+ "text": "Wait, what?",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0166.wav",
+ "render_seconds": 85.48826819704846,
+ "audio_seconds": 1.3931972789115645,
+ "rtf": 61.361208129717404,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WAIT WHAT",
+ "raw": {
+ "normalized_reference": "wait, what?",
+ "normalized_hypothesis": "wait what",
+ "errors": 0,
+ "words": 2,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "wait what",
+ "normalized_hypothesis": "wait what",
+ "errors": 0,
+ "words": 2,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0167",
+ "category": "technical",
+ "text": "The graphics processor remained idle because the data loader exhausted every central processing core.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0167.wav",
+ "render_seconds": 252.69250211201143,
+ "audio_seconds": 7.174965986394557,
+ "rtf": 35.21863414978922,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE GRAPHIC'S PROCESSOR REMAINED IDLE BECAUSE THE DATA LODER EXHAUSTED EVERY CENTRAL PROCESSING CORPE",
+ "raw": {
+ "normalized_reference": "the graphics processor remained idle because the data loader exhausted every central processing core.",
+ "normalized_hypothesis": "the graphic's processor remained idle because the data loder exhausted every central processing corpe",
+ "errors": 3,
+ "words": 14,
+ "wer": 0.21428571428571427
+ },
+ "semantic": {
+ "normalized_reference": "the graphics processor remained idle because the data loader exhausted every central processing core",
+ "normalized_hypothesis": "the graphic is processor remained idle because the data loder exhausted every central processing corpe",
+ "errors": 4,
+ "words": 14,
+ "wer": 0.2857142857142857
+ }
+ },
+ {
+ "id": "release500_0168",
+ "category": "conversational",
+ "text": "Could you speak a little more slowly? I am trying to write every detail down.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0168.wav",
+ "render_seconds": 211.395826210035,
+ "audio_seconds": 5.712108843537415,
+ "rtf": 37.00836801266571,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "COULD YOU SPEAK A LITTLE MORE SLOWLY I AM TRYING TO WRITE EVERY DETAIL DOWN",
+ "raw": {
+ "normalized_reference": "could you speak a little more slowly? i am trying to write every detail down.",
+ "normalized_hypothesis": "could you speak a little more slowly i am trying to write every detail down",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "could you speak a little more slowly i am trying to write every detail down",
+ "normalized_hypothesis": "could you speak a little more slowly i am trying to write every detail down",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0169",
+ "category": "emotional",
+ "text": "I was so excited I could barely form a complete sentence.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0169.wav",
+ "render_seconds": 156.512929058983,
+ "audio_seconds": 4.31891156462585,
+ "rtf": 36.238975194793504,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I WAS SO EXCITED I COULD BARELY FORM A COMPLETE SENTENCE",
+ "raw": {
+ "normalized_reference": "i was so excited i could barely form a complete sentence.",
+ "normalized_hypothesis": "i was so excited i could barely form a complete sentence",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i was so excited i could barely form a complete sentence",
+ "normalized_hypothesis": "i was so excited i could barely form a complete sentence",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0170",
+ "category": "narrative",
+ "text": "He looked at me like I had said something in a language he didn't speak.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0170.wav",
+ "render_seconds": 131.19891132391058,
+ "audio_seconds": 4.667210884353741,
+ "rtf": 28.110774202155515,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE LOOKED AT ME LIKE I HAD SAID SOMETHING IN A LANGUAGE HE DIDN'T SPEAK",
+ "raw": {
+ "normalized_reference": "he looked at me like i had said something in a language he didn't speak.",
+ "normalized_hypothesis": "he looked at me like i had said something in a language he didn't speak",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he looked at me like i had said something in a language he did not speak",
+ "normalized_hypothesis": "he looked at me like i had said something in a language he did not speak",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0171",
+ "category": "short",
+ "text": "That's actually hilarious.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0171.wav",
+ "render_seconds": 85.59090774389915,
+ "audio_seconds": 2.159455782312925,
+ "rtf": 39.6354064968283,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THAT'S ACTUALLY HILARIOUS",
+ "raw": {
+ "normalized_reference": "that's actually hilarious.",
+ "normalized_hypothesis": "that's actually hilarious",
+ "errors": 0,
+ "words": 3,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "that is actually hilarious",
+ "normalized_hypothesis": "that is actually hilarious",
+ "errors": 0,
+ "words": 4,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0172",
+ "category": "technical",
+ "text": "The encoder maps phonemes, stress, and phrase boundaries into a shared hidden representation.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0172.wav",
+ "render_seconds": 210.00737861997914,
+ "audio_seconds": 6.617687074829932,
+ "rtf": 31.734256432089776,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE ENCOUTER MAPS PHONIMES STRESS AND PHRASE BOUNDARIES INTO A SHARED HIDDEN REPRESENTATION",
+ "raw": {
+ "normalized_reference": "the encoder maps phonemes, stress, and phrase boundaries into a shared hidden representation.",
+ "normalized_hypothesis": "the encouter maps phonimes stress and phrase boundaries into a shared hidden representation",
+ "errors": 2,
+ "words": 13,
+ "wer": 0.15384615384615385
+ },
+ "semantic": {
+ "normalized_reference": "the encoder maps phonemes stress and phrase boundaries into a shared hidden representation",
+ "normalized_hypothesis": "the encouter maps phonimes stress and phrase boundaries into a shared hidden representation",
+ "errors": 2,
+ "words": 13,
+ "wer": 0.15384615384615385
+ }
+ },
+ {
+ "id": "release500_0173",
+ "category": "conversational",
+ "text": "She showed up three hours late and acted like nothing had happened.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0173.wav",
+ "render_seconds": 195.00905901391525,
+ "audio_seconds": 4.597551020408163,
+ "rtf": 42.41585534304797,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SHE SHOWED UP THREE HOURS LATE AND ACTED LIKE NOTHING HAD HAPPENED",
+ "raw": {
+ "normalized_reference": "she showed up three hours late and acted like nothing had happened.",
+ "normalized_hypothesis": "she showed up three hours late and acted like nothing had happened",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "she showed up 3 hours late and acted like nothing had happened",
+ "normalized_hypothesis": "she showed up 3 hours late and acted like nothing had happened",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0174",
+ "category": "emotional",
+ "text": "Oh my god. This is the best thing I've ever tasted.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0174.wav",
+ "render_seconds": 158.50374704203568,
+ "audio_seconds": 3.9706122448979593,
+ "rtf": 39.91922083192716,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "OH MY GOD THIS IS THE BEST THING I'VE EVER TASTED",
+ "raw": {
+ "normalized_reference": "oh my god. this is the best thing i've ever tasted.",
+ "normalized_hypothesis": "oh my god this is the best thing i've ever tasted",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "0 my god this is the best thing i have ever tasted",
+ "normalized_hypothesis": "0 my god this is the best thing i have ever tasted",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0175",
+ "category": "narrative",
+ "text": "The note was short, which was somehow more frightening than a long one would have been.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0175.wav",
+ "render_seconds": 153.29321441799402,
+ "audio_seconds": 5.572789115646258,
+ "rtf": 27.507449364557036,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE NOTE WAS SHORT WHICH WAS SOMEHOW MORE FRIGHTENING THAN A LONG ONE WOULD HAVE BEEN",
+ "raw": {
+ "normalized_reference": "the note was short, which was somehow more frightening than a long one would have been.",
+ "normalized_hypothesis": "the note was short which was somehow more frightening than a long one would have been",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the note was short which was somehow more frightening than a long one would have been",
+ "normalized_hypothesis": "the note was short which was somehow more frightening than a long one would have been",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0176",
+ "category": "short",
+ "text": "Noted.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0176.wav",
+ "render_seconds": 84.70032263104804,
+ "audio_seconds": 1.3235374149659864,
+ "rtf": 63.995412349733044,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "NOTED",
+ "raw": {
+ "normalized_reference": "noted.",
+ "normalized_hypothesis": "noted",
+ "errors": 0,
+ "words": 1,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "noted",
+ "normalized_hypothesis": "noted",
+ "errors": 0,
+ "words": 1,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0177",
+ "category": "technical",
+ "text": "It's a simple process once you understand the underlying structure.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0177.wav",
+ "render_seconds": 157.00311071495526,
+ "audio_seconds": 4.597551020408163,
+ "rtf": 34.1492916594141,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT'S A SIMPLE PROCESS ONCE YOU UNDERSTAND THE UNDERLYING STRUCTURE",
+ "raw": {
+ "normalized_reference": "it's a simple process once you understand the underlying structure.",
+ "normalized_hypothesis": "it's a simple process once you understand the underlying structure",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it is a simple process once you understand the underlying structure",
+ "normalized_hypothesis": "it is a simple process once you understand the underlying structure",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0178",
+ "category": "conversational",
+ "text": "You have to hear this story because it's almost too good to be true.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0178.wav",
+ "render_seconds": 166.39960729691666,
+ "audio_seconds": 4.458231292517007,
+ "rtf": 37.32413066504936,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "YOU HAVE TO HEAR THIS STORY BECAUSE IT'S ALMOST TOO GOOD TO BE TRUE",
+ "raw": {
+ "normalized_reference": "you have to hear this story because it's almost too good to be true.",
+ "normalized_hypothesis": "you have to hear this story because it's almost too good to be true",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "you have to hear this story because it is almost too good to be true",
+ "normalized_hypothesis": "you have to hear this story because it is almost too good to be true",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0179",
+ "category": "emotional",
+ "text": "I have never felt more seen in my entire life than in that moment.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0179.wav",
+ "render_seconds": 129.49014355696272,
+ "audio_seconds": 4.458231292517007,
+ "rtf": 29.045182957265503,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I HAVE NEVER FELT MORE SEN IN MY ENTIRE LIFE THAN IN THAT MOMENT",
+ "raw": {
+ "normalized_reference": "i have never felt more seen in my entire life than in that moment.",
+ "normalized_hypothesis": "i have never felt more sen in my entire life than in that moment",
+ "errors": 1,
+ "words": 14,
+ "wer": 0.07142857142857142
+ },
+ "semantic": {
+ "normalized_reference": "i have never felt more seen in my entire life than in that moment",
+ "normalized_hypothesis": "i have never felt more senator in my entire life than in that moment",
+ "errors": 1,
+ "words": 14,
+ "wer": 0.07142857142857142
+ }
+ },
+ {
+ "id": "release500_0180",
+ "category": "narrative",
+ "text": "She knew. She had always known. She just waited for the right moment.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0180.wav",
+ "render_seconds": 184.3956411619438,
+ "audio_seconds": 5.224489795918367,
+ "rtf": 35.2944781911533,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SHE KNEW SHE HAD ALWAYS KNOWN SHE JUST WAITED FOR THE RIGHT MOMENT",
+ "raw": {
+ "normalized_reference": "she knew. she had always known. she just waited for the right moment.",
+ "normalized_hypothesis": "she knew she had always known she just waited for the right moment",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "she knew she had always known she just waited for the right moment",
+ "normalized_hypothesis": "she knew she had always known she just waited for the right moment",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0181",
+ "category": "short",
+ "text": "Yeah, no. Hard pass.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0181.wav",
+ "render_seconds": 95.00375633803196,
+ "audio_seconds": 1.9504761904761905,
+ "rtf": 48.70798054440115,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "YE NO HARD PASS",
+ "raw": {
+ "normalized_reference": "yeah, no. hard pass.",
+ "normalized_hypothesis": "ye no hard pass",
+ "errors": 1,
+ "words": 4,
+ "wer": 0.25
+ },
+ "semantic": {
+ "normalized_reference": "yeah no hard pass",
+ "normalized_hypothesis": "ye no hard pass",
+ "errors": 1,
+ "words": 4,
+ "wer": 0.25
+ }
+ },
+ {
+ "id": "release500_0182",
+ "category": "technical",
+ "text": "Please verify the sample rate, hop length, and mel filter configuration before inference.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0182.wav",
+ "render_seconds": 227.49867438699584,
+ "audio_seconds": 6.199727891156463,
+ "rtf": 36.694945065493755,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "PLEASE VERIFY THE SAMPLE RATE HOP LENGTH AND MELL FILTER CONFIGURATION BEFORE INFERENCE",
+ "raw": {
+ "normalized_reference": "please verify the sample rate, hop length, and mel filter configuration before inference.",
+ "normalized_hypothesis": "please verify the sample rate hop length and mell filter configuration before inference",
+ "errors": 1,
+ "words": 13,
+ "wer": 0.07692307692307693
+ },
+ "semantic": {
+ "normalized_reference": "please verify the sample rate hop length and mel filter configuration before inference",
+ "normalized_hypothesis": "please verify the sample rate hop length and mell filter configuration before inference",
+ "errors": 1,
+ "words": 13,
+ "wer": 0.07692307692307693
+ }
+ },
+ {
+ "id": "release500_0183",
+ "category": "conversational",
+ "text": "That's the worst idea I've ever heard and I'm completely in.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0183.wav",
+ "render_seconds": 140.69971582002472,
+ "audio_seconds": 3.8312925170068026,
+ "rtf": 36.723824974331734,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THAT'S THE WORST IDEA I'VE EVER HEARD AND I'M COMPLETELY IN",
+ "raw": {
+ "normalized_reference": "that's the worst idea i've ever heard and i'm completely in.",
+ "normalized_hypothesis": "that's the worst idea i've ever heard and i'm completely in",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "that is the worst idea i have ever heard and i am completely in",
+ "normalized_hypothesis": "that is the worst idea i have ever heard and i am completely in",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0184",
+ "category": "emotional",
+ "text": "I remember thinking: this is it. This is the thing I've been waiting for.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0184.wav",
+ "render_seconds": 167.0922068769578,
+ "audio_seconds": 5.154829931972789,
+ "rtf": 32.414688570144634,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I REMEMBER THINKING THIS IS IT THIS IS THE THING I 'VE BEEN WAITING FOR",
+ "raw": {
+ "normalized_reference": "i remember thinking: this is it. this is the thing i've been waiting for.",
+ "normalized_hypothesis": "i remember thinking this is it this is the thing i 've been waiting for",
+ "errors": 2,
+ "words": 14,
+ "wer": 0.14285714285714285
+ },
+ "semantic": {
+ "normalized_reference": "i remember thinking this is it this is the thing i have been waiting for",
+ "normalized_hypothesis": "i remember thinking this is it this is the thing i have been waiting for",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0185",
+ "category": "narrative",
+ "text": "We never talked about it again, which I think was the right call.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0185.wav",
+ "render_seconds": 169.90739757893607,
+ "audio_seconds": 4.458231292517007,
+ "rtf": 38.11094275528504,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WE NEVER TALKED ABOUT IT AGAIN WHICH I THINK WAS THE RIGHT CALL",
+ "raw": {
+ "normalized_reference": "we never talked about it again, which i think was the right call.",
+ "normalized_hypothesis": "we never talked about it again which i think was the right call",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "we never talked about it again which i think was the right call",
+ "normalized_hypothesis": "we never talked about it again which i think was the right call",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0186",
+ "category": "short",
+ "text": "In hindsight, obvious.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0186.wav",
+ "render_seconds": 55.21192568494007,
+ "audio_seconds": 1.9504761904761905,
+ "rtf": 28.306895492767126,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IN HINDSIGHT OBVIOUS",
+ "raw": {
+ "normalized_reference": "in hindsight, obvious.",
+ "normalized_hypothesis": "in hindsight obvious",
+ "errors": 0,
+ "words": 3,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "in hindsight obvious",
+ "normalized_hypothesis": "in hindsight obvious",
+ "errors": 0,
+ "words": 3,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0187",
+ "category": "technical",
+ "text": "Once it's configured correctly, it basically runs itself.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0187.wav",
+ "render_seconds": 167.79994737298694,
+ "audio_seconds": 4.179591836734694,
+ "rtf": 40.14744834607597,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "ONCE ITS CONFIGURED CORRECTLY IT BASICALLY RUNS ITSELF",
+ "raw": {
+ "normalized_reference": "once it's configured correctly, it basically runs itself.",
+ "normalized_hypothesis": "once its configured correctly it basically runs itself",
+ "errors": 1,
+ "words": 8,
+ "wer": 0.125
+ },
+ "semantic": {
+ "normalized_reference": "once it is configured correctly it basically runs itself",
+ "normalized_hypothesis": "once its configured correctly it basically runs itself",
+ "errors": 2,
+ "words": 9,
+ "wer": 0.2222222222222222
+ }
+ },
+ {
+ "id": "release500_0188",
+ "category": "conversational",
+ "text": "You got engaged? When? How? Tell me everything right now.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0188.wav",
+ "render_seconds": 167.60357385501266,
+ "audio_seconds": 4.597551020408163,
+ "rtf": 36.454967679756834,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "YOU GOT ENGAGED WHEN HOW TELL ME EVERYTHING RIGHT NOW",
+ "raw": {
+ "normalized_reference": "you got engaged? when? how? tell me everything right now.",
+ "normalized_hypothesis": "you got engaged when how tell me everything right now",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "you got engaged when how tell me everything right now",
+ "normalized_hypothesis": "you got engaged when how tell me everything right now",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0189",
+ "category": "emotional",
+ "text": "It's hard to explain how much that moment meant to me.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0189.wav",
+ "render_seconds": 140.10762640601024,
+ "audio_seconds": 3.4829931972789114,
+ "rtf": 40.226213050163096,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT'S HARD TO EXPLAIN HOW MUCH THAT MOMENT MEANT TO ME",
+ "raw": {
+ "normalized_reference": "it's hard to explain how much that moment meant to me.",
+ "normalized_hypothesis": "it's hard to explain how much that moment meant to me",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it is hard to explain how much that moment meant to me",
+ "normalized_hypothesis": "it is hard to explain how much that moment meant to me",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0190",
+ "category": "narrative",
+ "text": "He told the story three different times that night and somehow it got better each time.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0190.wav",
+ "render_seconds": 201.40748736192472,
+ "audio_seconds": 5.642448979591837,
+ "rtf": 35.69504803506334,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE TOLD THE STORY THREE DIFFERENT TIMES THAT NIGHT AND SOMEHOW IT GOT BETTER EACH TIME",
+ "raw": {
+ "normalized_reference": "he told the story three different times that night and somehow it got better each time.",
+ "normalized_hypothesis": "he told the story three different times that night and somehow it got better each time",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he told the story 3 different times that night and somehow it got better each time",
+ "normalized_hypothesis": "he told the story 3 different times that night and somehow it got better each time",
+ "errors": 0,
+ "words": 16,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0191",
+ "category": "short",
+ "text": "Right. Moving on.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0191.wav",
+ "render_seconds": 88.69991307798773,
+ "audio_seconds": 1.811156462585034,
+ "rtf": 48.97418578149112,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "RIGHT MOVING ON",
+ "raw": {
+ "normalized_reference": "right. moving on.",
+ "normalized_hypothesis": "right moving on",
+ "errors": 0,
+ "words": 3,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "right moving on",
+ "normalized_hypothesis": "right moving on",
+ "errors": 0,
+ "words": 3,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0192",
+ "category": "conversational",
+ "text": "No, I did not move your keys; check beside the small ceramic bowl.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0192.wav",
+ "render_seconds": 177.68764486000873,
+ "audio_seconds": 4.667210884353741,
+ "rtf": 38.07148407535752,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "NO I DID NOT MOVE YOUR KEYS CHECK BESIDE THE SMALL SERAMIC BOWL",
+ "raw": {
+ "normalized_reference": "no, i did not move your keys; check beside the small ceramic bowl.",
+ "normalized_hypothesis": "no i did not move your keys check beside the small seramic bowl",
+ "errors": 1,
+ "words": 13,
+ "wer": 0.07692307692307693
+ },
+ "semantic": {
+ "normalized_reference": "no i did not move your keys check beside the small ceramic bowl",
+ "normalized_hypothesis": "no i did not move your keys check beside the small seramic bowl",
+ "errors": 1,
+ "words": 13,
+ "wer": 0.07692307692307693
+ }
+ },
+ {
+ "id": "release500_0193",
+ "category": "emotional",
+ "text": "I was devastated. I'm not going to pretend otherwise.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0193.wav",
+ "render_seconds": 159.89793053700123,
+ "audio_seconds": 4.1099319727891155,
+ "rtf": 38.90524991548461,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I WAS DEVASTATED I'M NOT GOING TO PRETEND OTHERWISE",
+ "raw": {
+ "normalized_reference": "i was devastated. i'm not going to pretend otherwise.",
+ "normalized_hypothesis": "i was devastated i'm not going to pretend otherwise",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i was devastated i am not going to pretend otherwise",
+ "normalized_hypothesis": "i was devastated i am not going to pretend otherwise",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0194",
+ "category": "narrative",
+ "text": "That was the last summer before everything changed, though we didn't know it yet.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0194.wav",
+ "render_seconds": 151.09806479304098,
+ "audio_seconds": 5.015510204081632,
+ "rtf": 30.126160379471877,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THAT WAS THE LAST SUMMER BEFORE EVERYTHING CHANGED THOUGH WE DIDN'T KNOW IT YET",
+ "raw": {
+ "normalized_reference": "that was the last summer before everything changed, though we didn't know it yet.",
+ "normalized_hypothesis": "that was the last summer before everything changed though we didn't know it yet",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "that was the last summer before everything changed though we did not know it yet",
+ "normalized_hypothesis": "that was the last summer before everything changed though we did not know it yet",
+ "errors": 0,
+ "words": 15,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0195",
+ "category": "short",
+ "text": "Bold of you to assume.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0195.wav",
+ "render_seconds": 100.08805331797339,
+ "audio_seconds": 1.8808163265306121,
+ "rtf": 53.21521932053707,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "BOLD OF IOU TO ASSUME",
+ "raw": {
+ "normalized_reference": "bold of you to assume.",
+ "normalized_hypothesis": "bold of iou to assume",
+ "errors": 1,
+ "words": 5,
+ "wer": 0.2
+ },
+ "semantic": {
+ "normalized_reference": "bold of you to assume",
+ "normalized_hypothesis": "bold of iou to assume",
+ "errors": 1,
+ "words": 5,
+ "wer": 0.2
+ }
+ },
+ {
+ "id": "release500_0196",
+ "category": "conversational",
+ "text": "It's one of those decisions that makes total sense in the moment.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0196.wav",
+ "render_seconds": 154.00022177398205,
+ "audio_seconds": 4.388571428571429,
+ "rtf": 35.09119636776934,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT'S ONE OF THOSE DECISIONS THAT MAKES TOTAL SENSE IN THE MOMENT",
+ "raw": {
+ "normalized_reference": "it's one of those decisions that makes total sense in the moment.",
+ "normalized_hypothesis": "it's one of those decisions that makes total sense in the moment",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it is one of those decisions that makes total sense in the moment",
+ "normalized_hypothesis": "it is one of those decisions that makes total sense in the moment",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0197",
+ "category": "emotional",
+ "text": "I was furious. Not the loud kind. The quiet kind that lasts for days.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0197.wav",
+ "render_seconds": 203.4922595010139,
+ "audio_seconds": 5.363809523809524,
+ "rtf": 37.938010027710334,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I WAS FURIOUS NOT THE LOUD KIND THE QUIET KIND THAT LASTS FOR DAYS",
+ "raw": {
+ "normalized_reference": "i was furious. not the loud kind. the quiet kind that lasts for days.",
+ "normalized_hypothesis": "i was furious not the loud kind the quiet kind that lasts for days",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i was furious not the loud kind the quiet kind that lasts for days",
+ "normalized_hypothesis": "i was furious not the loud kind the quiet kind that lasts for days",
+ "errors": 0,
+ "words": 14,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0198",
+ "category": "narrative",
+ "text": "That's the part of the story I usually leave out.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0198.wav",
+ "render_seconds": 126.10402616497595,
+ "audio_seconds": 3.3436734693877552,
+ "rtf": 37.71421680959364,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THAT'S THE PART OF THE STORY I USUALLY LEAVE OUT",
+ "raw": {
+ "normalized_reference": "that's the part of the story i usually leave out.",
+ "normalized_hypothesis": "that's the part of the story i usually leave out",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "that is the part of the story i usually leave out",
+ "normalized_hypothesis": "that is the part of the story i usually leave out",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0199",
+ "category": "short",
+ "text": "I've said too much.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0199.wav",
+ "render_seconds": 79.7018426869763,
+ "audio_seconds": 1.7414965986394557,
+ "rtf": 45.766292480412176,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I'VE SAID TOO MUCH",
+ "raw": {
+ "normalized_reference": "i've said too much.",
+ "normalized_hypothesis": "i've said too much",
+ "errors": 0,
+ "words": 4,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i have said too much",
+ "normalized_hypothesis": "i have said too much",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0200",
+ "category": "conversational",
+ "text": "I've been faking that I understood this the entire time.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0200.wav",
+ "render_seconds": 132.69919514399953,
+ "audio_seconds": 3.622312925170068,
+ "rtf": 36.63382978922763,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I'VE BEEN FAKING THAT I UNDERSTOOD THIS THE ENTIRE TIME",
+ "raw": {
+ "normalized_reference": "i've been faking that i understood this the entire time.",
+ "normalized_hypothesis": "i've been faking that i understood this the entire time",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i have been faking that i understood this the entire time",
+ "normalized_hypothesis": "i have been faking that i understood this the entire time",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0201",
+ "category": "phonetic_balance",
+ "text": "Peace be unto you and grace before the Lord.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0201.wav",
+ "render_seconds": 141.20551825605799,
+ "audio_seconds": 3.0650340136054424,
+ "rtf": 46.06980465119079,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "PEACE BE UNTO YOU IN GRACE BEFORE THE LORD",
+ "raw": {
+ "normalized_reference": "peace be unto you and grace before the lord.",
+ "normalized_hypothesis": "peace be unto you in grace before the lord",
+ "errors": 1,
+ "words": 9,
+ "wer": 0.1111111111111111
+ },
+ "semantic": {
+ "normalized_reference": "peace be unto you and grace before the lord",
+ "normalized_hypothesis": "peace be unto you in grace before the lord",
+ "errors": 1,
+ "words": 9,
+ "wer": 0.1111111111111111
+ }
+ },
+ {
+ "id": "release500_0202",
+ "category": "phonetic_balance",
+ "text": "From now on we're pals.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0202.wav",
+ "render_seconds": 103.29508127400186,
+ "audio_seconds": 1.8808163265306121,
+ "rtf": 54.92034486139422,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "FROM NOW ON WERD PAULS",
+ "raw": {
+ "normalized_reference": "from now on we're pals.",
+ "normalized_hypothesis": "from now on werd pauls",
+ "errors": 2,
+ "words": 5,
+ "wer": 0.4
+ },
+ "semantic": {
+ "normalized_reference": "from now on we are pals",
+ "normalized_hypothesis": "from now on werd pauls",
+ "errors": 3,
+ "words": 6,
+ "wer": 0.5
+ }
+ },
+ {
+ "id": "release500_0203",
+ "category": "phonetic_balance",
+ "text": "Here we allow our solicitors to look after our legal work.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0203.wav",
+ "render_seconds": 151.49941342906095,
+ "audio_seconds": 4.179591836734694,
+ "rtf": 36.24741825207025,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HERE WE ALLOW OUR SOLICITORS TO LOOK AFTER OUR LEGAL WORK",
+ "raw": {
+ "normalized_reference": "here we allow our solicitors to look after our legal work.",
+ "normalized_hypothesis": "here we allow our solicitors to look after our legal work",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "here we allow our solicitors to look after our legal work",
+ "normalized_hypothesis": "here we allow our solicitors to look after our legal work",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0204",
+ "category": "phonetic_balance",
+ "text": "Her words sent a strange chill through Philip.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0204.wav",
+ "render_seconds": 122.99952331290115,
+ "audio_seconds": 3.2043537414965986,
+ "rtf": 38.38512637354889,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HER WORDS SENT A STRANGE CHILL THROUGH PHILOT",
+ "raw": {
+ "normalized_reference": "her words sent a strange chill through philip.",
+ "normalized_hypothesis": "her words sent a strange chill through philot",
+ "errors": 1,
+ "words": 8,
+ "wer": 0.125
+ },
+ "semantic": {
+ "normalized_reference": "her words sent a strange chill through philip",
+ "normalized_hypothesis": "her words sent a strange chill through philot",
+ "errors": 1,
+ "words": 8,
+ "wer": 0.125
+ }
+ },
+ {
+ "id": "release500_0205",
+ "category": "phonetic_balance",
+ "text": "It was introduced by Representative Dick of Ohio.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0205.wav",
+ "render_seconds": 156.49971805210225,
+ "audio_seconds": 3.7616326530612243,
+ "rtf": 41.604200220014164,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT WAS INTRODUCED BY REPRESENTATIVE DICK OF OHIO",
+ "raw": {
+ "normalized_reference": "it was introduced by representative dick of ohio.",
+ "normalized_hypothesis": "it was introduced by representative dick of ohio",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it was introduced by representative dick of ohio",
+ "normalized_hypothesis": "it was introduced by representative dick of ohio",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0206",
+ "category": "phonetic_balance",
+ "text": "It's the aurora borealis.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0206.wav",
+ "render_seconds": 44.29847009794321,
+ "audio_seconds": 2.020136054421769,
+ "rtf": 21.92845872978735,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT'S THE AURORA BOREALIS",
+ "raw": {
+ "normalized_reference": "it's the aurora borealis.",
+ "normalized_hypothesis": "it's the aurora borealis",
+ "errors": 0,
+ "words": 4,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it is the aurora borealis",
+ "normalized_hypothesis": "it is the aurora borealis",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0207",
+ "category": "phonetic_balance",
+ "text": "The boy grew and prospered.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0207.wav",
+ "render_seconds": 104.49678320402745,
+ "audio_seconds": 2.159455782312925,
+ "rtf": 48.39033244390132,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE BOY GREW AND PROSPERED",
+ "raw": {
+ "normalized_reference": "the boy grew and prospered.",
+ "normalized_hypothesis": "the boy grew and prospered",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the boy grew and prospered",
+ "normalized_hypothesis": "the boy grew and prospered",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0208",
+ "category": "phonetic_balance",
+ "text": "His voice was passionately rebellious.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0208.wav",
+ "render_seconds": 124.50408441293985,
+ "audio_seconds": 2.995374149659864,
+ "rtf": 41.5654533264493,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HIS VOICE WAS PASSIONATELY REBELLIOUS",
+ "raw": {
+ "normalized_reference": "his voice was passionately rebellious.",
+ "normalized_hypothesis": "his voice was passionately rebellious",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "his voice was passionately rebellious",
+ "normalized_hypothesis": "his voice was passionately rebellious",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0209",
+ "category": "phonetic_balance",
+ "text": "That's where they cut off the Scottish Chiefs and killed all hands.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0209.wav",
+ "render_seconds": 160.80715630296618,
+ "audio_seconds": 4.388571428571429,
+ "rtf": 36.642255667993595,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THAT'S WHERE THEY CUT OFF THE SCOTTISH CHIEFS AND KILLED ALL HANDS",
+ "raw": {
+ "normalized_reference": "that's where they cut off the scottish chiefs and killed all hands.",
+ "normalized_hypothesis": "that's where they cut off the scottish chiefs and killed all hands",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "that is where they cut off the scottish chiefs and killed all hands",
+ "normalized_hypothesis": "that is where they cut off the scottish chiefs and killed all hands",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0210",
+ "category": "phonetic_balance",
+ "text": "And Tom King patiently endured.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0210.wav",
+ "render_seconds": 99.49595871998463,
+ "audio_seconds": 2.4380952380952383,
+ "rtf": 40.80888931874369,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "AND TOM KING PATIENTLY ENDURED",
+ "raw": {
+ "normalized_reference": "and tom king patiently endured.",
+ "normalized_hypothesis": "and tom king patiently endured",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "and tom king patiently endured",
+ "normalized_hypothesis": "and tom king patiently endured",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0211",
+ "category": "phonetic_balance",
+ "text": "From that moment his friendship for Belize turns to hatred and jealousy.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0211.wav",
+ "render_seconds": 190.4977680540178,
+ "audio_seconds": 5.154829931972789,
+ "rtf": 36.95519940870542,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "FROM THAT MOMENT HIS FRIENDSHIP FOR BULLAYS TURNS TO HATRED AND JEALOUSY",
+ "raw": {
+ "normalized_reference": "from that moment his friendship for belize turns to hatred and jealousy.",
+ "normalized_hypothesis": "from that moment his friendship for bullays turns to hatred and jealousy",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "from that moment his friendship for belize turns to hatred and jealousy",
+ "normalized_hypothesis": "from that moment his friendship for bullays turns to hatred and jealousy",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ }
+ },
+ {
+ "id": "release500_0212",
+ "category": "phonetic_balance",
+ "text": "Then you don't believe in altruism.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0212.wav",
+ "render_seconds": 73.53096668096259,
+ "audio_seconds": 2.4380952380952383,
+ "rtf": 30.15918555273856,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THEN YOU DON'T BELIEVE IN ALTRUISM",
+ "raw": {
+ "normalized_reference": "then you don't believe in altruism.",
+ "normalized_hypothesis": "then you don't believe in altruism",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "then you do not believe in altruism",
+ "normalized_hypothesis": "then you do not believe in altruism",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0213",
+ "category": "phonetic_balance",
+ "text": "What Pascal glimpsed with the vision of a seer, I have lived.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0213.wav",
+ "render_seconds": 162.4048843840137,
+ "audio_seconds": 4.31891156462585,
+ "rtf": 37.603197461646324,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WHAT PASCOL GLIMPSED WITH THE VISION OF A SEER I HAVE LIVED",
+ "raw": {
+ "normalized_reference": "what pascal glimpsed with the vision of a seer, i have lived.",
+ "normalized_hypothesis": "what pascol glimpsed with the vision of a seer i have lived",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "what pascal glimpsed with the vision of a seer i have lived",
+ "normalized_hypothesis": "what pascol glimpsed with the vision of a seer i have lived",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ }
+ },
+ {
+ "id": "release500_0214",
+ "category": "phonetic_balance",
+ "text": "You were making them talk shop, Ruth charged him.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0214.wav",
+ "render_seconds": 111.89531640789937,
+ "audio_seconds": 3.274013605442177,
+ "rtf": 34.1768025098928,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "YOU WERE MAKING THEM TALK SHOP RUTH CHARGED HIM",
+ "raw": {
+ "normalized_reference": "you were making them talk shop, ruth charged him.",
+ "normalized_hypothesis": "you were making them talk shop ruth charged him",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "you were making them talk shop ruth charged him",
+ "normalized_hypothesis": "you were making them talk shop ruth charged him",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0215",
+ "category": "phonetic_balance",
+ "text": "Philip snatched at the letter which Gregson held out to him.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0215.wav",
+ "render_seconds": 151.79221775499173,
+ "audio_seconds": 3.9706122448979593,
+ "rtf": 38.22891997324524,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "PHILIP SNATCHED AT THE LETTER WHICH GREGSON HELD OUT TO HIM",
+ "raw": {
+ "normalized_reference": "philip snatched at the letter which gregson held out to him.",
+ "normalized_hypothesis": "philip snatched at the letter which gregson held out to him",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "philip snatched at the letter which gregson held out to him",
+ "normalized_hypothesis": "philip snatched at the letter which gregson held out to him",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0216",
+ "category": "phonetic_balance",
+ "text": "For a time the exciting thrill of his adventure was gone.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0216.wav",
+ "render_seconds": 145.6996357330354,
+ "audio_seconds": 4.040272108843538,
+ "rtf": 36.06183737330988,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "FOR A TIME THE EXCITING THRILL OF HIS ADVENTURE WAS GONE",
+ "raw": {
+ "normalized_reference": "for a time the exciting thrill of his adventure was gone.",
+ "normalized_hypothesis": "for a time the exciting thrill of his adventure was gone",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "for a time the exciting thrill of his adventure was gone",
+ "normalized_hypothesis": "for a time the exciting thrill of his adventure was gone",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0217",
+ "category": "phonetic_balance",
+ "text": "Obviously, it was a disease that could be contracted by contact.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0217.wav",
+ "render_seconds": 180.42634402099065,
+ "audio_seconds": 5.085170068027211,
+ "rtf": 35.48088689411069,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "OBVIOUSLY IT WAS A DISEASE THAT COULD BE CONTRACTED BY CONTACT",
+ "raw": {
+ "normalized_reference": "obviously, it was a disease that could be contracted by contact.",
+ "normalized_hypothesis": "obviously it was a disease that could be contracted by contact",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "obviously it was a disease that could be contracted by contact",
+ "normalized_hypothesis": "obviously it was a disease that could be contracted by contact",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0218",
+ "category": "phonetic_balance",
+ "text": "Blind with rage, he darted in.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0218.wav",
+ "render_seconds": 72.27195741294418,
+ "audio_seconds": 2.2291156462585033,
+ "rtf": 32.42180706696409,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "BLIND WITH RAGE HE DARTED IN",
+ "raw": {
+ "normalized_reference": "blind with rage, he darted in.",
+ "normalized_hypothesis": "blind with rage he darted in",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "blind with rage he darted in",
+ "normalized_hypothesis": "blind with rage he darted in",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0219",
+ "category": "phonetic_balance",
+ "text": "You live on an income which your father earned.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0219.wav",
+ "render_seconds": 118.79254577099346,
+ "audio_seconds": 3.1346938775510202,
+ "rtf": 37.896059523298696,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "YOU LIVE ON AN INCOME WHICH YOUR FATHER EARNED",
+ "raw": {
+ "normalized_reference": "you live on an income which your father earned.",
+ "normalized_hypothesis": "you live on an income which your father earned",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "you live on an income which your father earned",
+ "normalized_hypothesis": "you live on an income which your father earned",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0220",
+ "category": "phonetic_balance",
+ "text": "Something about it seemed to fascinate him, to challenge his presence.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0220.wav",
+ "render_seconds": 174.38811884389725,
+ "audio_seconds": 4.806530612244898,
+ "rtf": 36.2814955135486,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SOMETHING ABOUT IT SEEMED TO FASCINATE HIM TO CHALLENGE HIS PRESENCE",
+ "raw": {
+ "normalized_reference": "something about it seemed to fascinate him, to challenge his presence.",
+ "normalized_hypothesis": "something about it seemed to fascinate him to challenge his presence",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "something about it seemed to fascinate him to challenge his presence",
+ "normalized_hypothesis": "something about it seemed to fascinate him to challenge his presence",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0221",
+ "category": "phonetic_balance",
+ "text": "Her beautiful hair was done up in shining coils.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0221.wav",
+ "render_seconds": 152.71167974406853,
+ "audio_seconds": 3.4829931972789114,
+ "rtf": 43.84495492651968,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HER BEAUTIFUL HAIR WAS DONE UP IN SHINING COILS",
+ "raw": {
+ "normalized_reference": "her beautiful hair was done up in shining coils.",
+ "normalized_hypothesis": "her beautiful hair was done up in shining coils",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "her beautiful hair was done up in shining coils",
+ "normalized_hypothesis": "her beautiful hair was done up in shining coils",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0222",
+ "category": "phonetic_balance",
+ "text": "It does, was her audacious answer.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0222.wav",
+ "render_seconds": 138.70300833694637,
+ "audio_seconds": 2.716734693877551,
+ "rtf": 51.05504363364162,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT DOES WAS HER AUDACIOUS ANSWER",
+ "raw": {
+ "normalized_reference": "it does, was her audacious answer.",
+ "normalized_hypothesis": "it does was her audacious answer",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it does was her audacious answer",
+ "normalized_hypothesis": "it does was her audacious answer",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0223",
+ "category": "phonetic_balance",
+ "text": "He caught himself with a jerk.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0223.wav",
+ "render_seconds": 105.6640717900591,
+ "audio_seconds": 2.2291156462585033,
+ "rtf": 47.401790018123435,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE CAUGHT HIMSELF WITH A JERK",
+ "raw": {
+ "normalized_reference": "he caught himself with a jerk.",
+ "normalized_hypothesis": "he caught himself with a jerk",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he caught himself with a jerk",
+ "normalized_hypothesis": "he caught himself with a jerk",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0224",
+ "category": "phonetic_balance",
+ "text": "He did not believe in the burning of daylight for such a luxury.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0224.wav",
+ "render_seconds": 94.21396563295275,
+ "audio_seconds": 4.388571428571429,
+ "rtf": 21.468026023133763,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE DID NOT BELIEVE IN THE BURNING OF DAYLIGHT FOR SUCH A LUXURY",
+ "raw": {
+ "normalized_reference": "he did not believe in the burning of daylight for such a luxury.",
+ "normalized_hypothesis": "he did not believe in the burning of daylight for such a luxury",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he did not believe in the burning of daylight for such a luxury",
+ "normalized_hypothesis": "he did not believe in the burning of daylight for such a luxury",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0225",
+ "category": "phonetic_balance",
+ "text": "He had comparatively no advantages at first.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0225.wav",
+ "render_seconds": 150.89984683995135,
+ "audio_seconds": 3.4133333333333336,
+ "rtf": 44.208939503892,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE HAD COMPARATIVELY NO ADVANTAGES AT FIRST",
+ "raw": {
+ "normalized_reference": "he had comparatively no advantages at first.",
+ "normalized_hypothesis": "he had comparatively no advantages at first",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he had comparatively no advantages at 1st",
+ "normalized_hypothesis": "he had comparatively no advantages at 1st",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0226",
+ "category": "phonetic_balance",
+ "text": "Such men believe, when they come together.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0226.wav",
+ "render_seconds": 130.6986567300046,
+ "audio_seconds": 2.8560544217687074,
+ "rtf": 45.761962984257515,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SUCH MEN BELIEVE WHEN THEY COME TOGETHER",
+ "raw": {
+ "normalized_reference": "such men believe, when they come together.",
+ "normalized_hypothesis": "such men believe when they come together",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "such men believe when they come together",
+ "normalized_hypothesis": "such men believe when they come together",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0227",
+ "category": "phonetic_balance",
+ "text": "The ship should be in within a week or ten days.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0227.wav",
+ "render_seconds": 143.38940703007393,
+ "audio_seconds": 3.274013605442177,
+ "rtf": 43.796215993643756,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE SHIP SHOULD BE IN WITHIN A WEEK OR TEN DAYS",
+ "raw": {
+ "normalized_reference": "the ship should be in within a week or ten days.",
+ "normalized_hypothesis": "the ship should be in within a week or ten days",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the ship should be in within a week or 10 days",
+ "normalized_hypothesis": "the ship should be in within a week or 10 days",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0228",
+ "category": "phonetic_balance",
+ "text": "Such a dog the wise driver kills, or turns loose.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0228.wav",
+ "render_seconds": 155.70062283601146,
+ "audio_seconds": 3.622312925170068,
+ "rtf": 42.98375818226729,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SUCH A DOG THE WISE DRIVER KILLS OR TURNS LOOSE",
+ "raw": {
+ "normalized_reference": "such a dog the wise driver kills, or turns loose.",
+ "normalized_hypothesis": "such a dog the wise driver kills or turns loose",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "such a dog the wise driver kills or turns loose",
+ "normalized_hypothesis": "such a dog the wise driver kills or turns loose",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0229",
+ "category": "phonetic_balance",
+ "text": "You have heard always how he was the lover of the Princess Naomi.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0229.wav",
+ "render_seconds": 154.11101181805134,
+ "audio_seconds": 4.458231292517007,
+ "rtf": 34.56774709663932,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "YOU HAVE HEARD ALWAYS HOW HE WAS THE LOVER OF THE PRINCESS NAOMI",
+ "raw": {
+ "normalized_reference": "you have heard always how he was the lover of the princess naomi.",
+ "normalized_hypothesis": "you have heard always how he was the lover of the princess naomi",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "you have heard always how he was the lover of the princess naomi",
+ "normalized_hypothesis": "you have heard always how he was the lover of the princess naomi",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0230",
+ "category": "phonetic_balance",
+ "text": "I want to know how all this is possible.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0230.wav",
+ "render_seconds": 81.68238760298118,
+ "audio_seconds": 2.9257142857142857,
+ "rtf": 27.918784825237708,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I WANT TO KNOW HOW ALL THIS IS POSSIBLE",
+ "raw": {
+ "normalized_reference": "i want to know how all this is possible.",
+ "normalized_hypothesis": "i want to know how all this is possible",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i want to know how all this is possible",
+ "normalized_hypothesis": "i want to know how all this is possible",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0231",
+ "category": "phonetic_balance",
+ "text": "It was my idea to a tee.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0231.wav",
+ "render_seconds": 89.59984059398994,
+ "audio_seconds": 2.159455782312925,
+ "rtf": 41.49186166619368,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT WAS MY IDEA TO A TA",
+ "raw": {
+ "normalized_reference": "it was my idea to a tee.",
+ "normalized_hypothesis": "it was my idea to a ta",
+ "errors": 1,
+ "words": 7,
+ "wer": 0.14285714285714285
+ },
+ "semantic": {
+ "normalized_reference": "it was my idea to a tee",
+ "normalized_hypothesis": "it was my idea to a ta",
+ "errors": 1,
+ "words": 7,
+ "wer": 0.14285714285714285
+ }
+ },
+ {
+ "id": "release500_0232",
+ "category": "phonetic_balance",
+ "text": "They likewise are disinclined to being eaten.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0232.wav",
+ "render_seconds": 139.69522951799445,
+ "audio_seconds": 3.2043537414965986,
+ "rtf": 43.59544569281841,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THEY LIKEWISE ARE DISINCLINED TO BEING EATEN",
+ "raw": {
+ "normalized_reference": "they likewise are disinclined to being eaten.",
+ "normalized_hypothesis": "they likewise are disinclined to being eaten",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "they likewise are disinclined to being eaten",
+ "normalized_hypothesis": "they likewise are disinclined to being eaten",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0233",
+ "category": "phonetic_balance",
+ "text": "This state of mind comes of an undue prominence of the ego.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0233.wav",
+ "render_seconds": 165.50575402600225,
+ "audio_seconds": 4.1099319727891155,
+ "rtf": 40.26970643839766,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THIS STATE OF MIND COMES OF AN UNDUE PROMINENCE OF THE EGO",
+ "raw": {
+ "normalized_reference": "this state of mind comes of an undue prominence of the ego.",
+ "normalized_hypothesis": "this state of mind comes of an undue prominence of the ego",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "this state of mind comes of an undue prominence of the ego",
+ "normalized_hypothesis": "this state of mind comes of an undue prominence of the ego",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0234",
+ "category": "phonetic_balance",
+ "text": "Then it is as I said, Womble announced with finality.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0234.wav",
+ "render_seconds": 143.99591594492085,
+ "audio_seconds": 3.8312925170068026,
+ "rtf": 37.58416129954433,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THEN IT IS AS I SAID WUMBELL ANNOUNCED WITH FINALITY",
+ "raw": {
+ "normalized_reference": "then it is as i said, womble announced with finality.",
+ "normalized_hypothesis": "then it is as i said wumbell announced with finality",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ },
+ "semantic": {
+ "normalized_reference": "then it is as i said womble announced with finality",
+ "normalized_hypothesis": "then it is as i said wumbell announced with finality",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ }
+ },
+ {
+ "id": "release500_0235",
+ "category": "phonetic_balance",
+ "text": "I can see that knife now.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0235.wav",
+ "render_seconds": 94.4952214009827,
+ "audio_seconds": 2.020136054421769,
+ "rtf": 46.77666199469442,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I CAN SEE THAT KNIFE NOW",
+ "raw": {
+ "normalized_reference": "i can see that knife now.",
+ "normalized_hypothesis": "i can see that knife now",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i can see that knife now",
+ "normalized_hypothesis": "i can see that knife now",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0236",
+ "category": "phonetic_balance",
+ "text": "If I meet her again I shall apologize, said Eileen.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0236.wav",
+ "render_seconds": 153.71506068203598,
+ "audio_seconds": 3.7616326530612243,
+ "rtf": 40.86392129676521,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IF I MEET HER AGAIN I SHALL APOLOGIZE SAID ILEEN",
+ "raw": {
+ "normalized_reference": "if i meet her again i shall apologize, said eileen.",
+ "normalized_hypothesis": "if i meet her again i shall apologize said ileen",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ },
+ "semantic": {
+ "normalized_reference": "if i meet her again i shall apologize said eileen",
+ "normalized_hypothesis": "if i meet her again i shall apologize said ileen",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ }
+ },
+ {
+ "id": "release500_0237",
+ "category": "phonetic_balance",
+ "text": "A wildly exciting time was his during the week preceding Thursday the eighteenth.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0237.wav",
+ "render_seconds": 187.4109411360696,
+ "audio_seconds": 5.572789115646258,
+ "rtf": 33.62964886108671,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "A WILDLY EXCITING TIME WAS HIS DURING THE WEEK PRECEDING THURSDAY THE EIGHTEENTH",
+ "raw": {
+ "normalized_reference": "a wildly exciting time was his during the week preceding thursday the eighteenth.",
+ "normalized_hypothesis": "a wildly exciting time was his during the week preceding thursday the eighteenth",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "a wildly exciting time was his during the week preceding thursday the 18th",
+ "normalized_hypothesis": "a wildly exciting time was his during the week preceding thursday the 18th",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0238",
+ "category": "phonetic_balance",
+ "text": "In what bucolic school of fence he had been taught was beyond imagining.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0238.wav",
+ "render_seconds": 170.8976426590234,
+ "audio_seconds": 4.876190476190477,
+ "rtf": 35.047368123432534,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IN WHAT BUCALIC SCHOOL OF FENCE HE HAD BEEN TAUGHT WAS BEYOND IMAGINING",
+ "raw": {
+ "normalized_reference": "in what bucolic school of fence he had been taught was beyond imagining.",
+ "normalized_hypothesis": "in what bucalic school of fence he had been taught was beyond imagining",
+ "errors": 1,
+ "words": 13,
+ "wer": 0.07692307692307693
+ },
+ "semantic": {
+ "normalized_reference": "in what bucolic school of fence he had been taught was beyond imagining",
+ "normalized_hypothesis": "in what bucalic school of fence he had been taught was beyond imagining",
+ "errors": 1,
+ "words": 13,
+ "wer": 0.07692307692307693
+ }
+ },
+ {
+ "id": "release500_0239",
+ "category": "phonetic_balance",
+ "text": "So she said, the irate skipper dashed on.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0239.wav",
+ "render_seconds": 124.01602367509622,
+ "audio_seconds": 3.0650340136054424,
+ "rtf": 40.46154891743418,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SO SHE SAID THE IROT SKIPPER DASHED ON",
+ "raw": {
+ "normalized_reference": "so she said, the irate skipper dashed on.",
+ "normalized_hypothesis": "so she said the irot skipper dashed on",
+ "errors": 1,
+ "words": 8,
+ "wer": 0.125
+ },
+ "semantic": {
+ "normalized_reference": "so she said the irate skipper dashed on",
+ "normalized_hypothesis": "so she said the irot skipper dashed on",
+ "errors": 1,
+ "words": 8,
+ "wer": 0.125
+ }
+ },
+ {
+ "id": "release500_0240",
+ "category": "phonetic_balance",
+ "text": "It is not an attempt to smash the market.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0240.wav",
+ "render_seconds": 111.5925358888926,
+ "audio_seconds": 3.0650340136054424,
+ "rtf": 36.40825367468752,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT IS NOT AN ATTEMPT TO SMASH THE MARKET",
+ "raw": {
+ "normalized_reference": "it is not an attempt to smash the market.",
+ "normalized_hypothesis": "it is not an attempt to smash the market",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it is not an attempt to smash the market",
+ "normalized_hypothesis": "it is not an attempt to smash the market",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0241",
+ "category": "phonetic_balance",
+ "text": "He destroyed everything that had belonged to the woman.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0241.wav",
+ "render_seconds": 137.30032210005447,
+ "audio_seconds": 3.622312925170068,
+ "rtf": 37.904047755235894,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE DESTROYED EVERYTHING THAT HAD BELONGED TO THE WOMAN",
+ "raw": {
+ "normalized_reference": "he destroyed everything that had belonged to the woman.",
+ "normalized_hypothesis": "he destroyed everything that had belonged to the woman",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he destroyed everything that had belonged to the woman",
+ "normalized_hypothesis": "he destroyed everything that had belonged to the woman",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0242",
+ "category": "phonetic_balance",
+ "text": "Did Thorpe go to see any one in Churchill.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0242.wav",
+ "render_seconds": 65.49456997099333,
+ "audio_seconds": 2.995374149659864,
+ "rtf": 21.865238430541467,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "DID THORPE GO TO SEE ANY ONE IN CHURCHILL",
+ "raw": {
+ "normalized_reference": "did thorpe go to see any one in churchill.",
+ "normalized_hypothesis": "did thorpe go to see any one in churchill",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "did thorpe go to see any one in churchill",
+ "normalized_hypothesis": "did thorpe go to see any one in churchill",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0243",
+ "category": "phonetic_balance",
+ "text": "Many other similar disconcerting omissions will be noticed in the Manuscript.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0243.wav",
+ "render_seconds": 208.19875388999935,
+ "audio_seconds": 5.712108843537415,
+ "rtf": 36.44866713727838,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "MANY OTHER SIMILAR DISCONCERTING OMISSIONS WILL BE NOTICED IN THE MANUSCRIPT",
+ "raw": {
+ "normalized_reference": "many other similar disconcerting omissions will be noticed in the manuscript.",
+ "normalized_hypothesis": "many other similar disconcerting omissions will be noticed in the manuscript",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "many other similar disconcerting omissions will be noticed in the manuscript",
+ "normalized_hypothesis": "many other similar disconcerting omissions will be noticed in the manuscript",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0244",
+ "category": "phonetic_balance",
+ "text": "The ringing of the big bell aroused him.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0244.wav",
+ "render_seconds": 126.7997991410084,
+ "audio_seconds": 2.716734693877551,
+ "rtf": 46.67360395064161,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE RINGING OF THE BIG BELL AROUSED HIM",
+ "raw": {
+ "normalized_reference": "the ringing of the big bell aroused him.",
+ "normalized_hypothesis": "the ringing of the big bell aroused him",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the ringing of the big bell aroused him",
+ "normalized_hypothesis": "the ringing of the big bell aroused him",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0245",
+ "category": "phonetic_balance",
+ "text": "A sinewy hand, dripping with water, was clutching the rail.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0245.wav",
+ "render_seconds": 123.57655678398442,
+ "audio_seconds": 4.249251700680272,
+ "rtf": 29.081957363034242,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "A SINOY HAND DRIPPING WITH WATER WAS CLUTCHING THE RAIL",
+ "raw": {
+ "normalized_reference": "a sinewy hand, dripping with water, was clutching the rail.",
+ "normalized_hypothesis": "a sinoy hand dripping with water was clutching the rail",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ },
+ "semantic": {
+ "normalized_reference": "a sinewy hand dripping with water was clutching the rail",
+ "normalized_hypothesis": "a sinoy hand dripping with water was clutching the rail",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ }
+ },
+ {
+ "id": "release500_0246",
+ "category": "phonetic_balance",
+ "text": "Enters now the psychology of the situation.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0246.wav",
+ "render_seconds": 132.20554474997334,
+ "audio_seconds": 3.1346938775510202,
+ "rtf": 42.174945916332646,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "ENTERS NOW THE PSYCHOLOGY OF THE SITUATION",
+ "raw": {
+ "normalized_reference": "enters now the psychology of the situation.",
+ "normalized_hypothesis": "enters now the psychology of the situation",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "enters now the psychology of the situation",
+ "normalized_hypothesis": "enters now the psychology of the situation",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0247",
+ "category": "phonetic_balance",
+ "text": "Wash your hands of me.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0247.wav",
+ "render_seconds": 89.69241090700962,
+ "audio_seconds": 1.8808163265306121,
+ "rtf": 47.68802229213836,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WASH YOUR HANDS OF ME",
+ "raw": {
+ "normalized_reference": "wash your hands of me.",
+ "normalized_hypothesis": "wash your hands of me",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "wash your hands of me",
+ "normalized_hypothesis": "wash your hands of me",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0248",
+ "category": "phonetic_balance",
+ "text": "He confessed that the sketch had startled him.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0248.wav",
+ "render_seconds": 66.59408195200376,
+ "audio_seconds": 3.1346938775510202,
+ "rtf": 21.24420583104287,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE CONFESSED THAT THE SKETCH HAD STARTLED HIM",
+ "raw": {
+ "normalized_reference": "he confessed that the sketch had startled him.",
+ "normalized_hypothesis": "he confessed that the sketch had startled him",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he confessed that the sketch had startled him",
+ "normalized_hypothesis": "he confessed that the sketch had startled him",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0249",
+ "category": "phonetic_balance",
+ "text": "Her face was against his breast.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0249.wav",
+ "render_seconds": 110.28980051109102,
+ "audio_seconds": 2.507755102040816,
+ "rtf": 43.97949401890837,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HER FACE WAS AGAINST HIS BREAST",
+ "raw": {
+ "normalized_reference": "her face was against his breast.",
+ "normalized_hypothesis": "her face was against his breast",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "her face was against his breast",
+ "normalized_hypothesis": "her face was against his breast",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0250",
+ "category": "phonetic_balance",
+ "text": "He was a pariah; a wanderer without a friend or a home.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0250.wav",
+ "render_seconds": 156.49989560095128,
+ "audio_seconds": 3.900952380952381,
+ "rtf": 40.118381440673545,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE WAS A PARIA A WANDERER WITHOUT A FRIEND OR A HOME",
+ "raw": {
+ "normalized_reference": "he was a pariah; a wanderer without a friend or a home.",
+ "normalized_hypothesis": "he was a paria a wanderer without a friend or a home",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "he was a pariah a wanderer without a friend or a home",
+ "normalized_hypothesis": "he was a paria a wanderer without a friend or a home",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ }
+ },
+ {
+ "id": "release500_0251",
+ "category": "phonetic_balance",
+ "text": "A cry of joy burst from Philip's lips.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0251.wav",
+ "render_seconds": 116.60638577397913,
+ "audio_seconds": 2.786394557823129,
+ "rtf": 41.84848317572005,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "A CRY OF JOY BURST FROM PHILIP'S LIPS",
+ "raw": {
+ "normalized_reference": "a cry of joy burst from philip's lips.",
+ "normalized_hypothesis": "a cry of joy burst from philip's lips",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "a cry of joy burst from philip is lips",
+ "normalized_hypothesis": "a cry of joy burst from philip is lips",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0252",
+ "category": "phonetic_balance",
+ "text": "And the big Persian knew of his existence before he did of hers.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0252.wav",
+ "render_seconds": 156.19583395903464,
+ "audio_seconds": 4.31891156462585,
+ "rtf": 36.165555052888884,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "AND THE BIG PERSIAN KNEW OF HIS EXISTENCE BEFORE HE DID OF HERS",
+ "raw": {
+ "normalized_reference": "and the big persian knew of his existence before he did of hers.",
+ "normalized_hypothesis": "and the big persian knew of his existence before he did of hers",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "and the big persian knew of his existence before he did of hers",
+ "normalized_hypothesis": "and the big persian knew of his existence before he did of hers",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0253",
+ "category": "phonetic_balance",
+ "text": "The men stared into each other's face.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0253.wav",
+ "render_seconds": 99.39987088704947,
+ "audio_seconds": 2.5774149659863945,
+ "rtf": 38.565722710083065,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE MEN STARED INTO EACH OTHER'S FACE",
+ "raw": {
+ "normalized_reference": "the men stared into each other's face.",
+ "normalized_hypothesis": "the men stared into each other's face",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the men stared into each other is face",
+ "normalized_hypothesis": "the men stared into each other is face",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0254",
+ "category": "phonetic_balance",
+ "text": "It is the fire, partly, she said.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0254.wav",
+ "render_seconds": 76.80247374996543,
+ "audio_seconds": 2.716734693877551,
+ "rtf": 28.270141329239078,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT IS THE FIRE PARTLY SHE SAID",
+ "raw": {
+ "normalized_reference": "it is the fire, partly, she said.",
+ "normalized_hypothesis": "it is the fire partly she said",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it is the fire partly she said",
+ "normalized_hypothesis": "it is the fire partly she said",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0255",
+ "category": "phonetic_balance",
+ "text": "It seemed nearer to him since he had seen and talked with Gregson.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0255.wav",
+ "render_seconds": 173.5110007530311,
+ "audio_seconds": 4.597551020408163,
+ "rtf": 37.7398749862328,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT SEEMED NEARER TO HIM SINCE HE HAD SEEN AND TALKED WITH GREGSON",
+ "raw": {
+ "normalized_reference": "it seemed nearer to him since he had seen and talked with gregson.",
+ "normalized_hypothesis": "it seemed nearer to him since he had seen and talked with gregson",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it seemed nearer to him since he had seen and talked with gregson",
+ "normalized_hypothesis": "it seemed nearer to him since he had seen and talked with gregson",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0256",
+ "category": "phonetic_balance",
+ "text": "Among my minor afflictions, I may mention a new and mysterious one.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0256.wav",
+ "render_seconds": 181.49999442393892,
+ "audio_seconds": 4.9458503401360545,
+ "rtf": 36.69742955039478,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "AMONG MY MINOR AFFLICTIONS I MAY MENTION A NEW AND MYSTERIOUS ONE",
+ "raw": {
+ "normalized_reference": "among my minor afflictions, i may mention a new and mysterious one.",
+ "normalized_hypothesis": "among my minor afflictions i may mention a new and mysterious one",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "among my minor afflictions i may mention a new and mysterious one",
+ "normalized_hypothesis": "among my minor afflictions i may mention a new and mysterious one",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0257",
+ "category": "phonetic_balance",
+ "text": "Between him and all domestic animals there must be no hostilities.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0257.wav",
+ "render_seconds": 166.90159599797335,
+ "audio_seconds": 4.667210884353741,
+ "rtf": 35.760457398119875,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "BETWEEN HIM AND ALL DOMESTIC ANIMALS THERE MUST BE NO HOSTILITIES",
+ "raw": {
+ "normalized_reference": "between him and all domestic animals there must be no hostilities.",
+ "normalized_hypothesis": "between him and all domestic animals there must be no hostilities",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "between him and all domestic animals there must be no hostilities",
+ "normalized_hypothesis": "between him and all domestic animals there must be no hostilities",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0258",
+ "category": "phonetic_balance",
+ "text": "With these arguments he convinced himself that he should go on alone.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0258.wav",
+ "render_seconds": 171.09754963498563,
+ "audio_seconds": 4.597551020408163,
+ "rtf": 37.21493222706985,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WITH THESE ARGUMENTS HE CONVINCED HIMSELF THAT HE SHOULD GO ON ALONE",
+ "raw": {
+ "normalized_reference": "with these arguments he convinced himself that he should go on alone.",
+ "normalized_hypothesis": "with these arguments he convinced himself that he should go on alone",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "with these arguments he convinced himself that he should go on alone",
+ "normalized_hypothesis": "with these arguments he convinced himself that he should go on alone",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0259",
+ "category": "phonetic_balance",
+ "text": "We are consumed in our own flesh-pots.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0259.wav",
+ "render_seconds": 122.80917294404935,
+ "audio_seconds": 2.8560544217687074,
+ "rtf": 42.999591327113315,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WE ARE CONSUMED IN OUR OWN FLESH POTS",
+ "raw": {
+ "normalized_reference": "we are consumed in our own flesh-pots.",
+ "normalized_hypothesis": "we are consumed in our own flesh pots",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "we are consumed in our own flesh pots",
+ "normalized_hypothesis": "we are consumed in our own flesh pots",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0260",
+ "category": "phonetic_balance",
+ "text": "Mercedes screamed, cried, laughed, and manifested the chaotic abandonment of hysteria.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0260.wav",
+ "render_seconds": 159.50137514597736,
+ "audio_seconds": 6.199727891156463,
+ "rtf": 25.727157376293313,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "MERCEDES SCREAMED CRIED LAUGHED AND MANIFESTED THE CHAOTIC ABANDONMENT OF HYSTERIA",
+ "raw": {
+ "normalized_reference": "mercedes screamed, cried, laughed, and manifested the chaotic abandonment of hysteria.",
+ "normalized_hypothesis": "mercedes screamed cried laughed and manifested the chaotic abandonment of hysteria",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "mercedes screamed cried laughed and manifested the chaotic abandonment of hysteria",
+ "normalized_hypothesis": "mercedes screamed cried laughed and manifested the chaotic abandonment of hysteria",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0261",
+ "category": "phonetic_balance",
+ "text": "Philip made no effort to follow.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0261.wav",
+ "render_seconds": 117.49560932500754,
+ "audio_seconds": 2.2987755102040817,
+ "rtf": 51.11225902810165,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "PHILIP MADE NO EFFORT TO FOLLOW",
+ "raw": {
+ "normalized_reference": "philip made no effort to follow.",
+ "normalized_hypothesis": "philip made no effort to follow",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "philip made no effort to follow",
+ "normalized_hypothesis": "philip made no effort to follow",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0262",
+ "category": "phonetic_balance",
+ "text": "She said with chattering teeth.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0262.wav",
+ "render_seconds": 112.71082032192498,
+ "audio_seconds": 2.2987755102040817,
+ "rtf": 49.030807846007846,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SHE SAID WITH CHATTERING TEETH",
+ "raw": {
+ "normalized_reference": "she said with chattering teeth.",
+ "normalized_hypothesis": "she said with chattering teeth",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "she said with chattering teeth",
+ "normalized_hypothesis": "she said with chattering teeth",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0263",
+ "category": "phonetic_balance",
+ "text": "Halfway around the track one donkey got into an argument with its rider.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0263.wav",
+ "render_seconds": 165.69103769597132,
+ "audio_seconds": 5.015510204081632,
+ "rtf": 33.035729358327615,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HALF WAY AROUND THE TRACK ONE DONKEY GOT INTO AN ARGUMENT WITH ITS RIDER",
+ "raw": {
+ "normalized_reference": "halfway around the track one donkey got into an argument with its rider.",
+ "normalized_hypothesis": "half way around the track one donkey got into an argument with its rider",
+ "errors": 2,
+ "words": 13,
+ "wer": 0.15384615384615385
+ },
+ "semantic": {
+ "normalized_reference": "halfway around the track one donkey got into an argument with its rider",
+ "normalized_hypothesis": "half way around the track one donkey got into an argument with its rider",
+ "errors": 2,
+ "words": 13,
+ "wer": 0.15384615384615385
+ }
+ },
+ {
+ "id": "release500_0264",
+ "category": "phonetic_balance",
+ "text": "Earth and gravel seemed to fill the pan.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0264.wav",
+ "render_seconds": 128.8076142220525,
+ "audio_seconds": 2.8560544217687074,
+ "rtf": 45.09984587138367,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "EARTH AND GRAVEL SEEMED TO FILL THE PAN",
+ "raw": {
+ "normalized_reference": "earth and gravel seemed to fill the pan.",
+ "normalized_hypothesis": "earth and gravel seemed to fill the pan",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "earth and gravel seemed to fill the pan",
+ "normalized_hypothesis": "earth and gravel seemed to fill the pan",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0265",
+ "category": "phonetic_balance",
+ "text": "Her achievements with cocoanuts were a revelation.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0265.wav",
+ "render_seconds": 135.29178818303626,
+ "audio_seconds": 3.5526530612244898,
+ "rtf": 38.08190269398424,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HER ACHIEVEMENTS WITH COCOA NUTS WERE A REVELATION",
+ "raw": {
+ "normalized_reference": "her achievements with cocoanuts were a revelation.",
+ "normalized_hypothesis": "her achievements with cocoa nuts were a revelation",
+ "errors": 2,
+ "words": 7,
+ "wer": 0.2857142857142857
+ },
+ "semantic": {
+ "normalized_reference": "her achievements with cocoanuts were a revelation",
+ "normalized_hypothesis": "her achievements with cocoa nuts were a revelation",
+ "errors": 2,
+ "words": 7,
+ "wer": 0.2857142857142857
+ }
+ },
+ {
+ "id": "release500_0266",
+ "category": "phonetic_balance",
+ "text": "For a full minute he crouched and listened.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0266.wav",
+ "render_seconds": 94.99260276998393,
+ "audio_seconds": 3.1346938775510202,
+ "rtf": 30.30362978990373,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "FOR A FULL MINUTE HE CROUCHED AND LISTENED",
+ "raw": {
+ "normalized_reference": "for a full minute he crouched and listened.",
+ "normalized_hypothesis": "for a full minute he crouched and listened",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "for a full minute he crouched and listened",
+ "normalized_hypothesis": "for a full minute he crouched and listened",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0267",
+ "category": "phonetic_balance",
+ "text": "It was edged with ice.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0267.wav",
+ "render_seconds": 93.4998834610451,
+ "audio_seconds": 1.9504761904761905,
+ "rtf": 47.9369519697741,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT WAS EDGED WITH ICE",
+ "raw": {
+ "normalized_reference": "it was edged with ice.",
+ "normalized_hypothesis": "it was edged with ice",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it was edged with ice",
+ "normalized_hypothesis": "it was edged with ice",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0268",
+ "category": "phonetic_balance",
+ "text": "The Portuguese boy passed the Hawaiian.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0268.wav",
+ "render_seconds": 123.9883021019632,
+ "audio_seconds": 2.8560544217687074,
+ "rtf": 43.41244380951932,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE PORTUGUESE BOY PASSED THE HAWAIAN",
+ "raw": {
+ "normalized_reference": "the portuguese boy passed the hawaiian.",
+ "normalized_hypothesis": "the portuguese boy passed the hawaian",
+ "errors": 1,
+ "words": 6,
+ "wer": 0.16666666666666666
+ },
+ "semantic": {
+ "normalized_reference": "the portuguese boy passed the hawaiian",
+ "normalized_hypothesis": "the portuguese boy passed the hawaian",
+ "errors": 1,
+ "words": 6,
+ "wer": 0.16666666666666666
+ }
+ },
+ {
+ "id": "release500_0269",
+ "category": "phonetic_balance",
+ "text": "But he did not broach it, preferring to mature it carefully.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0269.wav",
+ "render_seconds": 162.41152951307595,
+ "audio_seconds": 4.1099319727891155,
+ "rtf": 39.51684129770618,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "BUT HE DID NOT BROACH IT PREFERRING TO MATURE IT CAREFULLY",
+ "raw": {
+ "normalized_reference": "but he did not broach it, preferring to mature it carefully.",
+ "normalized_hypothesis": "but he did not broach it preferring to mature it carefully",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "but he did not broach it preferring to mature it carefully",
+ "normalized_hypothesis": "but he did not broach it preferring to mature it carefully",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0270",
+ "category": "phonetic_balance",
+ "text": "Come on, Del Mar challenged.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0270.wav",
+ "render_seconds": 92.99759810091928,
+ "audio_seconds": 2.159455782312925,
+ "rtf": 43.06529397944536,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "COME ON DELMAR CHALLENGED",
+ "raw": {
+ "normalized_reference": "come on, del mar challenged.",
+ "normalized_hypothesis": "come on delmar challenged",
+ "errors": 2,
+ "words": 5,
+ "wer": 0.4
+ },
+ "semantic": {
+ "normalized_reference": "come on del mar challenged",
+ "normalized_hypothesis": "come on delmar challenged",
+ "errors": 2,
+ "words": 5,
+ "wer": 0.4
+ }
+ },
+ {
+ "id": "release500_0271",
+ "category": "phonetic_balance",
+ "text": "For a full half minute Jeanne looked at him without speaking.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0271.wav",
+ "render_seconds": 162.0087238959968,
+ "audio_seconds": 4.179591836734694,
+ "rtf": 38.761852885272674,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "FOR A FULL HALF MINUTE JAN LOOKED AT HIM WITHOUT SPEAKING",
+ "raw": {
+ "normalized_reference": "for a full half minute jeanne looked at him without speaking.",
+ "normalized_hypothesis": "for a full half minute jan looked at him without speaking",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ },
+ "semantic": {
+ "normalized_reference": "for a full half minute jeanne looked at him without speaking",
+ "normalized_hypothesis": "for a full half minute jan looked at him without speaking",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ }
+ },
+ {
+ "id": "release500_0272",
+ "category": "phonetic_balance",
+ "text": "He bore no grudges and had few enemies.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0272.wav",
+ "render_seconds": 87.70124634995591,
+ "audio_seconds": 2.786394557823129,
+ "rtf": 31.474812532821094,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE BORE NO GRUDGES AND HAD FEW ENEMIES",
+ "raw": {
+ "normalized_reference": "he bore no grudges and had few enemies.",
+ "normalized_hypothesis": "he bore no grudges and had few enemies",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he bore no grudges and had few enemies",
+ "normalized_hypothesis": "he bore no grudges and had few enemies",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0273",
+ "category": "phonetic_balance",
+ "text": "I'll go over tomorrow afternoon.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0273.wav",
+ "render_seconds": 127.19252713792957,
+ "audio_seconds": 2.2987755102040817,
+ "rtf": 55.33055601703257,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I'LL GO OVER TO MORROW AFTERNOON",
+ "raw": {
+ "normalized_reference": "i'll go over tomorrow afternoon.",
+ "normalized_hypothesis": "i'll go over to morrow afternoon",
+ "errors": 2,
+ "words": 5,
+ "wer": 0.4
+ },
+ "semantic": {
+ "normalized_reference": "i will go over tomorrow afternoon",
+ "normalized_hypothesis": "i will go over to morrow afternoon",
+ "errors": 2,
+ "words": 6,
+ "wer": 0.3333333333333333
+ }
+ },
+ {
+ "id": "release500_0274",
+ "category": "phonetic_balance",
+ "text": "Well, there are better men in Hawaii, that's all.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0274.wav",
+ "render_seconds": 125.60755661095027,
+ "audio_seconds": 3.274013605442177,
+ "rtf": 38.3650075253692,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WELL THERE ARE BETTER MEN IN HAWA'I THAT'S ALL",
+ "raw": {
+ "normalized_reference": "well, there are better men in hawaii, that's all.",
+ "normalized_hypothesis": "well there are better men in hawa'i that's all",
+ "errors": 1,
+ "words": 9,
+ "wer": 0.1111111111111111
+ },
+ "semantic": {
+ "normalized_reference": "well there are better men in hawaii that is all",
+ "normalized_hypothesis": "well there are better men in hawa i that is all",
+ "errors": 2,
+ "words": 10,
+ "wer": 0.2
+ }
+ },
+ {
+ "id": "release500_0275",
+ "category": "phonetic_balance",
+ "text": "He was an athlete and a giant.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0275.wav",
+ "render_seconds": 107.79519376193639,
+ "audio_seconds": 2.36843537414966,
+ "rtf": 45.513251042637435,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE WAS AN ATHLETE AND A GIANT",
+ "raw": {
+ "normalized_reference": "he was an athlete and a giant.",
+ "normalized_hypothesis": "he was an athlete and a giant",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he was an athlete and a giant",
+ "normalized_hypothesis": "he was an athlete and a giant",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0276",
+ "category": "phonetic_balance",
+ "text": "McVeigh when he returns from a trip to Honolulu.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0276.wav",
+ "render_seconds": 135.294383415021,
+ "audio_seconds": 3.3436734693877552,
+ "rtf": 40.462797774267756,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "MAC VAY WHEN HE RETURNS FROM A TRIP TO HONOLULU",
+ "raw": {
+ "normalized_reference": "mcveigh when he returns from a trip to honolulu.",
+ "normalized_hypothesis": "mac vay when he returns from a trip to honolulu",
+ "errors": 2,
+ "words": 9,
+ "wer": 0.2222222222222222
+ },
+ "semantic": {
+ "normalized_reference": "mcveigh when he returns from a trip to honolulu",
+ "normalized_hypothesis": "mac vay when he returns from a trip to honolulu",
+ "errors": 2,
+ "words": 9,
+ "wer": 0.2222222222222222
+ }
+ },
+ {
+ "id": "release500_0277",
+ "category": "phonetic_balance",
+ "text": "You're a devil for fighting, and will surely win.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0277.wav",
+ "render_seconds": 123.59655887796544,
+ "audio_seconds": 3.274013605442177,
+ "rtf": 37.75077741659931,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "YOU'RE A DEVIL FOR FIGHTING AND WILL SURELY WIN",
+ "raw": {
+ "normalized_reference": "you're a devil for fighting, and will surely win.",
+ "normalized_hypothesis": "you're a devil for fighting and will surely win",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "you are a devil for fighting and will surely win",
+ "normalized_hypothesis": "you are a devil for fighting and will surely win",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0278",
+ "category": "phonetic_balance",
+ "text": "All an appearance can know is mirage.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0278.wav",
+ "render_seconds": 71.3037364219781,
+ "audio_seconds": 2.8560544217687074,
+ "rtf": 24.96581853570594,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "ALL IN APPEARANCE CAN KNOW HIS MIRAGE",
+ "raw": {
+ "normalized_reference": "all an appearance can know is mirage.",
+ "normalized_hypothesis": "all in appearance can know his mirage",
+ "errors": 2,
+ "words": 7,
+ "wer": 0.2857142857142857
+ },
+ "semantic": {
+ "normalized_reference": "all an appearance can know is mirage",
+ "normalized_hypothesis": "all in appearance can know his mirage",
+ "errors": 2,
+ "words": 7,
+ "wer": 0.2857142857142857
+ }
+ },
+ {
+ "id": "release500_0279",
+ "category": "phonetic_balance",
+ "text": "You fired me out of your house, in short.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0279.wav",
+ "render_seconds": 110.00324166705832,
+ "audio_seconds": 2.9257142857142857,
+ "rtf": 37.598764241670324,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "YOU FIRED ME OUT OF YOUR HOUSE IN SHORT",
+ "raw": {
+ "normalized_reference": "you fired me out of your house, in short.",
+ "normalized_hypothesis": "you fired me out of your house in short",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "you fired me out of your house in short",
+ "normalized_hypothesis": "you fired me out of your house in short",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0280",
+ "category": "phonetic_balance",
+ "text": "He made sure that the magazine was loaded, and resumed his paddling.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0280.wav",
+ "render_seconds": 162.699455367052,
+ "audio_seconds": 4.876190476190477,
+ "rtf": 33.36609924519621,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE MADE SURE THAT THE MAGAZINE WAS LOADED AND RESUMED HIS PADDLING",
+ "raw": {
+ "normalized_reference": "he made sure that the magazine was loaded, and resumed his paddling.",
+ "normalized_hypothesis": "he made sure that the magazine was loaded and resumed his paddling",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he made sure that the magazine was loaded and resumed his paddling",
+ "normalized_hypothesis": "he made sure that the magazine was loaded and resumed his paddling",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0281",
+ "category": "phonetic_balance",
+ "text": "Change chairs, Daylight commanded.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0281.wav",
+ "render_seconds": 110.99231745500583,
+ "audio_seconds": 2.507755102040816,
+ "rtf": 44.25963179765044,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "CHANGE CHAIRS DAYLIGHT COMMANDED",
+ "raw": {
+ "normalized_reference": "change chairs, daylight commanded.",
+ "normalized_hypothesis": "change chairs daylight commanded",
+ "errors": 0,
+ "words": 4,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "change chairs daylight commanded",
+ "normalized_hypothesis": "change chairs daylight commanded",
+ "errors": 0,
+ "words": 4,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0282",
+ "category": "phonetic_balance",
+ "text": "By that answer my professional medical prestige stood or fell.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0282.wav",
+ "render_seconds": 176.0012555270223,
+ "audio_seconds": 4.667210884353741,
+ "rtf": 37.71015706983483,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "BY THAT ANSWER MY PROFESSIONAL MEDICAL PRESTIGE STOOD OR FELL",
+ "raw": {
+ "normalized_reference": "by that answer my professional medical prestige stood or fell.",
+ "normalized_hypothesis": "by that answer my professional medical prestige stood or fell",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "by that answer my professional medical prestige stood or fell",
+ "normalized_hypothesis": "by that answer my professional medical prestige stood or fell",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0283",
+ "category": "phonetic_balance",
+ "text": "The moon had already begun its westward decline.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0283.wav",
+ "render_seconds": 140.7004451659741,
+ "audio_seconds": 3.4133333333333336,
+ "rtf": 41.22083354471897,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE MOON HAD ALREADY BEGUN ITS WESTWARD DECLINE",
+ "raw": {
+ "normalized_reference": "the moon had already begun its westward decline.",
+ "normalized_hypothesis": "the moon had already begun its westward decline",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the moon had already begun its westward decline",
+ "normalized_hypothesis": "the moon had already begun its westward decline",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0284",
+ "category": "phonetic_balance",
+ "text": "At sea, Monday, March 16, 1908.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0284.wav",
+ "render_seconds": 110.49829378596041,
+ "audio_seconds": 3.8312925170068026,
+ "rtf": 28.84099642495771,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "AT SEA MONDAY MARCH SIXTEENTH NINETEEN O EIGHT",
+ "raw": {
+ "normalized_reference": "at sea, monday, march 16, 1908.",
+ "normalized_hypothesis": "at sea monday march sixteenth nineteen o eight",
+ "errors": 4,
+ "words": 6,
+ "wer": 0.6666666666666666
+ },
+ "semantic": {
+ "normalized_reference": "at sea monday march 16 1908",
+ "normalized_hypothesis": "at sea monday march 16th 1908",
+ "errors": 1,
+ "words": 6,
+ "wer": 0.16666666666666666
+ }
+ },
+ {
+ "id": "release500_0285",
+ "category": "phonetic_balance",
+ "text": "And wherever I ranged, the way lay along alcohol-drenched roads.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0285.wav",
+ "render_seconds": 181.09901139803696,
+ "audio_seconds": 4.527891156462585,
+ "rtf": 39.99632613508328,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "AND WHEREVER I RANGED THE WAY LAY ALONG ALCOHOL DRENCHED ROADS",
+ "raw": {
+ "normalized_reference": "and wherever i ranged, the way lay along alcohol-drenched roads.",
+ "normalized_hypothesis": "and wherever i ranged the way lay along alcohol drenched roads",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "and wherever i ranged the way lay along alcohol drenched roads",
+ "normalized_hypothesis": "and wherever i ranged the way lay along alcohol drenched roads",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0286",
+ "category": "phonetic_balance",
+ "text": "The date was nearly eighteen years old.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0286.wav",
+ "render_seconds": 128.10178916691802,
+ "audio_seconds": 2.8560544217687074,
+ "rtf": 44.852712956214155,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE DATE WAS NEARLY EIGHTEEN YEARS OLD",
+ "raw": {
+ "normalized_reference": "the date was nearly eighteen years old.",
+ "normalized_hypothesis": "the date was nearly eighteen years old",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the date was nearly 18 years old",
+ "normalized_hypothesis": "the date was nearly 18 years old",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0287",
+ "category": "phonetic_balance",
+ "text": "Very early in my life, I separated from my mother.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0287.wav",
+ "render_seconds": 151.10867715091445,
+ "audio_seconds": 3.900952380952381,
+ "rtf": 38.736355226674846,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "VERY EARLY IN MY LIFE I SEPARATED FROM MY MOTHER",
+ "raw": {
+ "normalized_reference": "very early in my life, i separated from my mother.",
+ "normalized_hypothesis": "very early in my life i separated from my mother",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "very early in my life i separated from my mother",
+ "normalized_hypothesis": "very early in my life i separated from my mother",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0288",
+ "category": "phonetic_balance",
+ "text": "There was pride and strength, the ring of triumph in his voice.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0288.wav",
+ "render_seconds": 165.89893930708058,
+ "audio_seconds": 4.458231292517007,
+ "rtf": 37.21182873251472,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THERE WAS PRIDE AND STRENGTH THE RING OF TRIUMPH IN HIS VOICE",
+ "raw": {
+ "normalized_reference": "there was pride and strength, the ring of triumph in his voice.",
+ "normalized_hypothesis": "there was pride and strength the ring of triumph in his voice",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "there was pride and strength the ring of triumph in his voice",
+ "normalized_hypothesis": "there was pride and strength the ring of triumph in his voice",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0289",
+ "category": "phonetic_balance",
+ "text": "From my earliest recollection my sleep was a period of terror.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0289.wav",
+ "render_seconds": 180.30714368098415,
+ "audio_seconds": 4.73687074829932,
+ "rtf": 38.06461127226659,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "FROM MY EARLIEST RECOLLECTION MY SLEEP WAS A PERIOD OF TERROR",
+ "raw": {
+ "normalized_reference": "from my earliest recollection my sleep was a period of terror.",
+ "normalized_hypothesis": "from my earliest recollection my sleep was a period of terror",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "from my earliest recollection my sleep was a period of terror",
+ "normalized_hypothesis": "from my earliest recollection my sleep was a period of terror",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0290",
+ "category": "phonetic_balance",
+ "text": "We'll have to watch our chances.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0290.wav",
+ "render_seconds": 78.39858104998711,
+ "audio_seconds": 2.2987755102040817,
+ "rtf": 34.10449637295249,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WE'LL HAVE TO WATCH OUR CHANCES",
+ "raw": {
+ "normalized_reference": "we'll have to watch our chances.",
+ "normalized_hypothesis": "we'll have to watch our chances",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "we will have to watch our chances",
+ "normalized_hypothesis": "we will have to watch our chances",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0291",
+ "category": "phonetic_balance",
+ "text": "Close beside him gleamed the white fangs of the wolf-dog.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0291.wav",
+ "render_seconds": 144.80133695702534,
+ "audio_seconds": 3.7616326530612243,
+ "rtf": 38.49427903045921,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "CLOSE BESIDE HIM GLEAMED THE WHITE FANGS OF THE WOLF DOG",
+ "raw": {
+ "normalized_reference": "close beside him gleamed the white fangs of the wolf-dog.",
+ "normalized_hypothesis": "close beside him gleamed the white fangs of the wolf dog",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "close beside him gleamed the white fangs of the wolf dog",
+ "normalized_hypothesis": "close beside him gleamed the white fangs of the wolf dog",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0292",
+ "category": "phonetic_balance",
+ "text": "Would you be satisfied with that one hundredth part of me.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0292.wav",
+ "render_seconds": 139.69537051802035,
+ "audio_seconds": 3.900952380952381,
+ "rtf": 35.810580821269866,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WOULD YOU BE SATISFIED WITH THAT ONE HUNDREDTH PART OF ME",
+ "raw": {
+ "normalized_reference": "would you be satisfied with that one hundredth part of me.",
+ "normalized_hypothesis": "would you be satisfied with that one hundredth part of me",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "would you be satisfied with that 100th part of me",
+ "normalized_hypothesis": "would you be satisfied with that 100th part of me",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0293",
+ "category": "phonetic_balance",
+ "text": "Two gigantic owls were tearing at the carcass.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0293.wav",
+ "render_seconds": 111.19883950799704,
+ "audio_seconds": 3.3436734693877552,
+ "rtf": 33.256488866527434,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "TWO GIGANTIC OWLS WERE TEARING AT THE CARCASE",
+ "raw": {
+ "normalized_reference": "two gigantic owls were tearing at the carcass.",
+ "normalized_hypothesis": "two gigantic owls were tearing at the carcase",
+ "errors": 1,
+ "words": 8,
+ "wer": 0.125
+ },
+ "semantic": {
+ "normalized_reference": "2 gigantic owls were tearing at the carcass",
+ "normalized_hypothesis": "2 gigantic owls were tearing at the carcase",
+ "errors": 1,
+ "words": 8,
+ "wer": 0.125
+ }
+ },
+ {
+ "id": "release500_0294",
+ "category": "phonetic_balance",
+ "text": "The lush vegetation of that sheltered spot make a natural shield.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0294.wav",
+ "render_seconds": 181.1006852670107,
+ "audio_seconds": 4.597551020408163,
+ "rtf": 39.39068527203269,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE LUSH VEGETATION OF THAT SHELTERED SPOT MAKE A NATURAL SHIELD",
+ "raw": {
+ "normalized_reference": "the lush vegetation of that sheltered spot make a natural shield.",
+ "normalized_hypothesis": "the lush vegetation of that sheltered spot make a natural shield",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the lush vegetation of that sheltered spot make a natural shield",
+ "normalized_hypothesis": "the lush vegetation of that sheltered spot make a natural shield",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0295",
+ "category": "phonetic_balance",
+ "text": "Below him the shadow was broken into a pool of rippling starlight.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0295.wav",
+ "render_seconds": 175.69179761398118,
+ "audio_seconds": 4.527891156462585,
+ "rtf": 38.80212477352048,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "BELOW HIM THE SHADOW WAS BROKEN INTO A POOL OF RIPPLING STARLIGHT",
+ "raw": {
+ "normalized_reference": "below him the shadow was broken into a pool of rippling starlight.",
+ "normalized_hypothesis": "below him the shadow was broken into a pool of rippling starlight",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "below him the shadow was broken into a pool of rippling starlight",
+ "normalized_hypothesis": "below him the shadow was broken into a pool of rippling starlight",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0296",
+ "category": "phonetic_balance",
+ "text": "Lots of men take women buggy riding.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0296.wav",
+ "render_seconds": 86.09542912605684,
+ "audio_seconds": 2.507755102040816,
+ "rtf": 34.33167339824858,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "LOTS OF MEN TAKE WOMEN BUGGY RIDING",
+ "raw": {
+ "normalized_reference": "lots of men take women buggy riding.",
+ "normalized_hypothesis": "lots of men take women buggy riding",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "lots of men take women buggy riding",
+ "normalized_hypothesis": "lots of men take women buggy riding",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0297",
+ "category": "phonetic_balance",
+ "text": "But it contributed to the smash.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0297.wav",
+ "render_seconds": 122.70294041000307,
+ "audio_seconds": 2.4380952380952383,
+ "rtf": 50.32737790254031,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "BUT IT CONTRIBUTED TO THE SMASH",
+ "raw": {
+ "normalized_reference": "but it contributed to the smash.",
+ "normalized_hypothesis": "but it contributed to the smash",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "but it contributed to the smash",
+ "normalized_hypothesis": "but it contributed to the smash",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0298",
+ "category": "phonetic_balance",
+ "text": "No man ate of the seal meat or the oil.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0298.wav",
+ "render_seconds": 76.19988947198726,
+ "audio_seconds": 2.786394557823129,
+ "rtf": 27.347128301714182,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "NO MAN ATE OF THE SEAL MEAT OR THE OIL",
+ "raw": {
+ "normalized_reference": "no man ate of the seal meat or the oil.",
+ "normalized_hypothesis": "no man ate of the seal meat or the oil",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "no man ate of the seal meat or the oil",
+ "normalized_hypothesis": "no man ate of the seal meat or the oil",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0299",
+ "category": "phonetic_balance",
+ "text": "You're going in for grab sharing.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0299.wav",
+ "render_seconds": 98.39253861107863,
+ "audio_seconds": 2.2291156462585033,
+ "rtf": 44.13971916451587,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "YOU'RE GOING IN FOR GRAB SHARING",
+ "raw": {
+ "normalized_reference": "you're going in for grab sharing.",
+ "normalized_hypothesis": "you're going in for grab sharing",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "you are going in for grab sharing",
+ "normalized_hypothesis": "you are going in for grab sharing",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0300",
+ "category": "phonetic_balance",
+ "text": "He it was that lived to found the family of the Patino.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0300.wav",
+ "render_seconds": 135.30550636490807,
+ "audio_seconds": 3.6919727891156464,
+ "rtf": 36.648565440082336,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE IT WAS THAT LIVED TO FOUND THE FAMILY OF THE PETTINO",
+ "raw": {
+ "normalized_reference": "he it was that lived to found the family of the patino.",
+ "normalized_hypothesis": "he it was that lived to found the family of the pettino",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "he it was that lived to found the family of the patino",
+ "normalized_hypothesis": "he it was that lived to found the family of the pettino",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ }
+ },
+ {
+ "id": "release500_0301",
+ "category": "phonetic_balance",
+ "text": "For such countries nothing remained but reorganization.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0301.wav",
+ "render_seconds": 149.49207354802638,
+ "audio_seconds": 3.9706122448979593,
+ "rtf": 37.649627898094636,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "FOR SUCH COUNTRIES NOTHING REMAINED BUT REORGANIZATION",
+ "raw": {
+ "normalized_reference": "for such countries nothing remained but reorganization.",
+ "normalized_hypothesis": "for such countries nothing remained but reorganization",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "for such countries nothing remained but reorganization",
+ "normalized_hypothesis": "for such countries nothing remained but reorganization",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0302",
+ "category": "phonetic_balance",
+ "text": "Stand off butcher and baker and all the rest.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0302.wav",
+ "render_seconds": 92.71865101600997,
+ "audio_seconds": 3.3436734693877552,
+ "rtf": 27.72957702505181,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "STAND OFF BUTCHER AND BAKER AND ALL THE REST",
+ "raw": {
+ "normalized_reference": "stand off butcher and baker and all the rest.",
+ "normalized_hypothesis": "stand off butcher and baker and all the rest",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "stand off butcher and baker and all the rest",
+ "normalized_hypothesis": "stand off butcher and baker and all the rest",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0303",
+ "category": "phonetic_balance",
+ "text": "All eyes, however, were staring at him in certitude of expectancy.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0303.wav",
+ "render_seconds": 176.4957405719906,
+ "audio_seconds": 4.806530612244898,
+ "rtf": 36.71998678680171,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "ALL EYES HOWEVER WERE STARING AT HIM IN CERTITUDE OF EXPECTANCY",
+ "raw": {
+ "normalized_reference": "all eyes, however, were staring at him in certitude of expectancy.",
+ "normalized_hypothesis": "all eyes however were staring at him in certitude of expectancy",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "all eyes however were staring at him in certitude of expectancy",
+ "normalized_hypothesis": "all eyes however were staring at him in certitude of expectancy",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0304",
+ "category": "phonetic_balance",
+ "text": "Two days ago Jeanne learned where her father's men were hiding.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0304.wav",
+ "render_seconds": 166.2022139140172,
+ "audio_seconds": 3.9706122448979593,
+ "rtf": 41.85808224602612,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "TWO DAYS AGO JAN LEARNED WHERE HER FATHER'S MEN WERE HIDING",
+ "raw": {
+ "normalized_reference": "two days ago jeanne learned where her father's men were hiding.",
+ "normalized_hypothesis": "two days ago jan learned where her father's men were hiding",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ },
+ "semantic": {
+ "normalized_reference": "2 days ago jeanne learned where her father is men were hiding",
+ "normalized_hypothesis": "2 days ago jan learned where her father is men were hiding",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ }
+ },
+ {
+ "id": "release500_0305",
+ "category": "phonetic_balance",
+ "text": "His previous wives had never lived long enough to bear him children.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0305.wav",
+ "render_seconds": 171.7069034690503,
+ "audio_seconds": 4.597551020408163,
+ "rtf": 37.34747101377603,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HIS PREVIOUS WIVES HAD NEVER LIVED LONG ENOUGH TO BEAR HIM CHILDREN",
+ "raw": {
+ "normalized_reference": "his previous wives had never lived long enough to bear him children.",
+ "normalized_hypothesis": "his previous wives had never lived long enough to bear him children",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "his previous wives had never lived long enough to bear him children",
+ "normalized_hypothesis": "his previous wives had never lived long enough to bear him children",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0306",
+ "category": "phonetic_balance",
+ "text": "But this little defect did not worry him.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0306.wav",
+ "render_seconds": 130.60353156202473,
+ "audio_seconds": 2.8560544217687074,
+ "rtf": 45.72865648727524,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "BUT THIS LITTLE DEFECT DID NOT WORRY HIM",
+ "raw": {
+ "normalized_reference": "but this little defect did not worry him.",
+ "normalized_hypothesis": "but this little defect did not worry him",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "but this little defect did not worry him",
+ "normalized_hypothesis": "but this little defect did not worry him",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0307",
+ "category": "phonetic_balance",
+ "text": "Eli Harding asked, as Shunk started to follow.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0307.wav",
+ "render_seconds": 144.3118335370673,
+ "audio_seconds": 3.4829931972789114,
+ "rtf": 41.43328033193143,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "ILE HARDING ASKED AS SHUNK STARTED TO FOLLOW",
+ "raw": {
+ "normalized_reference": "eli harding asked, as shunk started to follow.",
+ "normalized_hypothesis": "ile harding asked as shunk started to follow",
+ "errors": 1,
+ "words": 8,
+ "wer": 0.125
+ },
+ "semantic": {
+ "normalized_reference": "eli harding asked as shunk started to follow",
+ "normalized_hypothesis": "ile harding asked as shunk started to follow",
+ "errors": 1,
+ "words": 8,
+ "wer": 0.125
+ }
+ },
+ {
+ "id": "release500_0308",
+ "category": "phonetic_balance",
+ "text": "And this was their sole conversation throughout the meal.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0308.wav",
+ "render_seconds": 109.88795409700833,
+ "audio_seconds": 3.900952380952381,
+ "rtf": 28.16951948287567,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "AND THIS WAS THEIR SOLE CONVERSATION THROUGHOUT THE MEAL",
+ "raw": {
+ "normalized_reference": "and this was their sole conversation throughout the meal.",
+ "normalized_hypothesis": "and this was their sole conversation throughout the meal",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "and this was their sole conversation throughout the meal",
+ "normalized_hypothesis": "and this was their sole conversation throughout the meal",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0309",
+ "category": "phonetic_balance",
+ "text": "Then, again, Tudor had such an irritating way about him.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0309.wav",
+ "render_seconds": 137.29976294899825,
+ "audio_seconds": 4.179591836734694,
+ "rtf": 32.85004093994587,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THEN AGAIN TUDOR HAD SUCH AN IRRITATING WAY ABOUT HIM",
+ "raw": {
+ "normalized_reference": "then, again, tudor had such an irritating way about him.",
+ "normalized_hypothesis": "then again tudor had such an irritating way about him",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "then again tudor had such an irritating way about him",
+ "normalized_hypothesis": "then again tudor had such an irritating way about him",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0310",
+ "category": "phonetic_balance",
+ "text": "In it there was something that was almost tragedy.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0310.wav",
+ "render_seconds": 144.90152485005092,
+ "audio_seconds": 3.4829931972789114,
+ "rtf": 41.60258623624509,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IN IT THERE WAS SOMETHING THAT WAS ALMOST TRAGEDY",
+ "raw": {
+ "normalized_reference": "in it there was something that was almost tragedy.",
+ "normalized_hypothesis": "in it there was something that was almost tragedy",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "in it there was something that was almost tragedy",
+ "normalized_hypothesis": "in it there was something that was almost tragedy",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0311",
+ "category": "phonetic_balance",
+ "text": "Thirty pounds, said the captain with finality.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0311.wav",
+ "render_seconds": 128.39992312190589,
+ "audio_seconds": 3.274013605442177,
+ "rtf": 39.2178954016792,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THIRTY POUNDS SAID THE CAPTAIN WITH FINALITY",
+ "raw": {
+ "normalized_reference": "thirty pounds, said the captain with finality.",
+ "normalized_hypothesis": "thirty pounds said the captain with finality",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "\u00a330 said the captain with finality",
+ "normalized_hypothesis": "\u00a330 said the captain with finality",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0312",
+ "category": "phonetic_balance",
+ "text": "It was over when he made his way through the ring of spectators.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0312.wav",
+ "render_seconds": 160.9956856520148,
+ "audio_seconds": 4.31891156462585,
+ "rtf": 37.27691184294068,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT WAS OVER WHEN HE MADE HIS WAY THROUGH THE RING OF SPECTATORS",
+ "raw": {
+ "normalized_reference": "it was over when he made his way through the ring of spectators.",
+ "normalized_hypothesis": "it was over when he made his way through the ring of spectators",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it was over when he made his way through the ring of spectators",
+ "normalized_hypothesis": "it was over when he made his way through the ring of spectators",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0313",
+ "category": "phonetic_balance",
+ "text": "She had died from cold and starvation.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0313.wav",
+ "render_seconds": 138.3968471980188,
+ "audio_seconds": 2.9257142857142857,
+ "rtf": 47.30360988213533,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SHE HAD DIED FROM COLD AND STARVATION",
+ "raw": {
+ "normalized_reference": "she had died from cold and starvation.",
+ "normalized_hypothesis": "she had died from cold and starvation",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "she had died from cold and starvation",
+ "normalized_hypothesis": "she had died from cold and starvation",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0314",
+ "category": "phonetic_balance",
+ "text": "We fished sharks on Niihau together.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0314.wav",
+ "render_seconds": 84.799765813048,
+ "audio_seconds": 2.647074829931973,
+ "rtf": 32.03527337201392,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WE FISHED SHARKS ON NIHO TOGETHER",
+ "raw": {
+ "normalized_reference": "we fished sharks on niihau together.",
+ "normalized_hypothesis": "we fished sharks on niho together",
+ "errors": 1,
+ "words": 6,
+ "wer": 0.16666666666666666
+ },
+ "semantic": {
+ "normalized_reference": "we fished sharks on niihau together",
+ "normalized_hypothesis": "we fished sharks on niho together",
+ "errors": 1,
+ "words": 6,
+ "wer": 0.16666666666666666
+ }
+ },
+ {
+ "id": "release500_0315",
+ "category": "phonetic_balance",
+ "text": "His slim hands gripped the edges of the table.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0315.wav",
+ "render_seconds": 139.40758166799787,
+ "audio_seconds": 3.274013605442177,
+ "rtf": 42.580025152085454,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HIS SLIM HANDS GRIPPED THE EDGES OF THE TABLE",
+ "raw": {
+ "normalized_reference": "his slim hands gripped the edges of the table.",
+ "normalized_hypothesis": "his slim hands gripped the edges of the table",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "his slim hands gripped the edges of the table",
+ "normalized_hypothesis": "his slim hands gripped the edges of the table",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0316",
+ "category": "phonetic_balance",
+ "text": "We had been chased by them ourselves, more than once.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0316.wav",
+ "render_seconds": 160.19796254707035,
+ "audio_seconds": 3.7616326530612243,
+ "rtf": 42.58734898440998,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WE HAD BEEN CHASED BY THEM OURSELVES MORE THAN ONCE",
+ "raw": {
+ "normalized_reference": "we had been chased by them ourselves, more than once.",
+ "normalized_hypothesis": "we had been chased by them ourselves more than once",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "we had been chased by them ourselves more than once",
+ "normalized_hypothesis": "we had been chased by them ourselves more than once",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0317",
+ "category": "phonetic_balance",
+ "text": "Or have they already devised one.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0317.wav",
+ "render_seconds": 118.99141736305319,
+ "audio_seconds": 2.5774149659863945,
+ "rtf": 46.1669614452302,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "OR HAVE THEY ALREADY DEVISED ONE",
+ "raw": {
+ "normalized_reference": "or have they already devised one.",
+ "normalized_hypothesis": "or have they already devised one",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "or have they already devised one",
+ "normalized_hypothesis": "or have they already devised one",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0318",
+ "category": "phonetic_balance",
+ "text": "I cannot follow you, she said.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0318.wav",
+ "render_seconds": 110.09986936999485,
+ "audio_seconds": 2.2987755102040817,
+ "rtf": 47.89500709454676,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I CANNOT FOLLOW YOU SHE SAID",
+ "raw": {
+ "normalized_reference": "i cannot follow you, she said.",
+ "normalized_hypothesis": "i cannot follow you she said",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i cannot follow you she said",
+ "normalized_hypothesis": "i cannot follow you she said",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0319",
+ "category": "phonetic_balance",
+ "text": "He went down in midstream, searching the shadows of both shores.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0319.wav",
+ "render_seconds": 160.31019487290177,
+ "audio_seconds": 4.388571428571429,
+ "rtf": 36.52901575879923,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE WENT DOWN IN MID STREAM SEARCHING THE SHADOWS OF BOTH SHORES",
+ "raw": {
+ "normalized_reference": "he went down in midstream, searching the shadows of both shores.",
+ "normalized_hypothesis": "he went down in mid stream searching the shadows of both shores",
+ "errors": 2,
+ "words": 11,
+ "wer": 0.18181818181818182
+ },
+ "semantic": {
+ "normalized_reference": "he went down in midstream searching the shadows of both shores",
+ "normalized_hypothesis": "he went down in mid stream searching the shadows of both shores",
+ "errors": 2,
+ "words": 11,
+ "wer": 0.18181818181818182
+ }
+ },
+ {
+ "id": "release500_0320",
+ "category": "phonetic_balance",
+ "text": "I was sick once -- typhoid.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0320.wav",
+ "render_seconds": 98.59971724194475,
+ "audio_seconds": 2.2291156462585033,
+ "rtf": 44.23266123829919,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I WAS SICK ONCE TYPHOID",
+ "raw": {
+ "normalized_reference": "i was sick once -- typhoid.",
+ "normalized_hypothesis": "i was sick once typhoid",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i was sick once typhoid",
+ "normalized_hypothesis": "i was sick once typhoid",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0321",
+ "category": "phonetic_balance",
+ "text": "Why not like any railroad station or ferry depot.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0321.wav",
+ "render_seconds": 150.5955464920262,
+ "audio_seconds": 3.622312925170068,
+ "rtf": 41.57441656837412,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WHY NOT LIKE ANY RAILROAD STATION OR FERRY DEPOT",
+ "raw": {
+ "normalized_reference": "why not like any railroad station or ferry depot.",
+ "normalized_hypothesis": "why not like any railroad station or ferry depot",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "why not like any railroad station or ferry depot",
+ "normalized_hypothesis": "why not like any railroad station or ferry depot",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0322",
+ "category": "phonetic_balance",
+ "text": "Red-Eye swung back and forth on the branch farther down.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0322.wav",
+ "render_seconds": 145.4036583720008,
+ "audio_seconds": 3.900952380952381,
+ "rtf": 37.27388703383809,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "RED EYES SWUNG BACK AND FORTH ON THE BRANCH FARTHER DOWN",
+ "raw": {
+ "normalized_reference": "red-eye swung back and forth on the branch farther down.",
+ "normalized_hypothesis": "red eyes swung back and forth on the branch farther down",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ },
+ "semantic": {
+ "normalized_reference": "red eye swung back and forth on the branch farther down",
+ "normalized_hypothesis": "red eyes swung back and forth on the branch farther down",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ }
+ },
+ {
+ "id": "release500_0323",
+ "category": "phonetic_balance",
+ "text": "Within himself he called it no longer his own.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0323.wav",
+ "render_seconds": 123.80804807099048,
+ "audio_seconds": 3.0650340136054424,
+ "rtf": 40.39369466094549,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WITHIN HIMSELF HE CALLED IT NO LONGER HIS OWN",
+ "raw": {
+ "normalized_reference": "within himself he called it no longer his own.",
+ "normalized_hypothesis": "within himself he called it no longer his own",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "within himself he called it no longer his own",
+ "normalized_hypothesis": "within himself he called it no longer his own",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0324",
+ "category": "phonetic_balance",
+ "text": "It lasted as a deterrent for two days.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0324.wav",
+ "render_seconds": 131.70378227403853,
+ "audio_seconds": 2.8560544217687074,
+ "rtf": 46.113890992482055,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT LASTED AS A DETERRANT FOR TWO DAYS",
+ "raw": {
+ "normalized_reference": "it lasted as a deterrent for two days.",
+ "normalized_hypothesis": "it lasted as a deterrant for two days",
+ "errors": 1,
+ "words": 8,
+ "wer": 0.125
+ },
+ "semantic": {
+ "normalized_reference": "it lasted as a deterrent for 2 days",
+ "normalized_hypothesis": "it lasted as a deterrant for 2 days",
+ "errors": 1,
+ "words": 8,
+ "wer": 0.125
+ }
+ },
+ {
+ "id": "release500_0325",
+ "category": "phonetic_balance",
+ "text": "He loved to play Chinese lottery.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0325.wav",
+ "render_seconds": 119.28722140903119,
+ "audio_seconds": 2.4380952380952383,
+ "rtf": 48.92639940604794,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE LOVED TO PLAY CHINESE LOTTERY",
+ "raw": {
+ "normalized_reference": "he loved to play chinese lottery.",
+ "normalized_hypothesis": "he loved to play chinese lottery",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he loved to play chinese lottery",
+ "normalized_hypothesis": "he loved to play chinese lottery",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0326",
+ "category": "phonetic_balance",
+ "text": "Lord, but I'm glad to see you again, Phil.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0326.wav",
+ "render_seconds": 99.2118411361007,
+ "audio_seconds": 2.9257142857142857,
+ "rtf": 33.91029726331567,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "LORD BUT I'M GLAD TO SEE YOU AGAIN PHIL",
+ "raw": {
+ "normalized_reference": "lord, but i'm glad to see you again, phil.",
+ "normalized_hypothesis": "lord but i'm glad to see you again phil",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "lord but i am glad to see you again phil",
+ "normalized_hypothesis": "lord but i am glad to see you again phil",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0327",
+ "category": "phonetic_balance",
+ "text": "There is not an iota of truth in it, certainly not.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0327.wav",
+ "render_seconds": 145.19193411397282,
+ "audio_seconds": 3.6919727891156464,
+ "rtf": 39.32638250802256,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THERE IS NOT AN IODA OF TRUTH IN IT CERTAINLY NOT",
+ "raw": {
+ "normalized_reference": "there is not an iota of truth in it, certainly not.",
+ "normalized_hypothesis": "there is not an ioda of truth in it certainly not",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ },
+ "semantic": {
+ "normalized_reference": "there is not an iota of truth in it certainly not",
+ "normalized_hypothesis": "there is not an ioda of truth in it certainly not",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ }
+ },
+ {
+ "id": "release500_0328",
+ "category": "phonetic_balance",
+ "text": "She added, with genuine sympathy in her face and voice.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0328.wav",
+ "render_seconds": 145.19768194598146,
+ "audio_seconds": 3.9706122448979593,
+ "rtf": 36.5680839604908,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SHE ADDED WITH GENUINE SYMPATHY IN HER FACE AND VOICE",
+ "raw": {
+ "normalized_reference": "she added, with genuine sympathy in her face and voice.",
+ "normalized_hypothesis": "she added with genuine sympathy in her face and voice",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "she added with genuine sympathy in her face and voice",
+ "normalized_hypothesis": "she added with genuine sympathy in her face and voice",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0329",
+ "category": "phonetic_balance",
+ "text": "We have plenty of capital ourselves, and yet we want more.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0329.wav",
+ "render_seconds": 150.00380608905107,
+ "audio_seconds": 4.179591836734694,
+ "rtf": 35.88958251154054,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WE HAVE PLENTY OF CAPITAL OURSELVES AND YET WE WANT MORE",
+ "raw": {
+ "normalized_reference": "we have plenty of capital ourselves, and yet we want more.",
+ "normalized_hypothesis": "we have plenty of capital ourselves and yet we want more",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "we have plenty of capital ourselves and yet we want more",
+ "normalized_hypothesis": "we have plenty of capital ourselves and yet we want more",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0330",
+ "category": "phonetic_balance",
+ "text": "And now put yourself in my place for a moment.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0330.wav",
+ "render_seconds": 126.39591903507244,
+ "audio_seconds": 3.274013605442177,
+ "rtf": 38.60580140075559,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "AND NOW PUT YOURSELF IN MY PLACE FOR A MOMENT",
+ "raw": {
+ "normalized_reference": "and now put yourself in my place for a moment.",
+ "normalized_hypothesis": "and now put yourself in my place for a moment",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "and now put yourself in my place for a moment",
+ "normalized_hypothesis": "and now put yourself in my place for a moment",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0331",
+ "category": "phonetic_balance",
+ "text": "Fast, but endure.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0331.wav",
+ "render_seconds": 81.29378070903476,
+ "audio_seconds": 1.6718367346938776,
+ "rtf": 48.62543035574589,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "FAST BUT ENDURE",
+ "raw": {
+ "normalized_reference": "fast, but endure.",
+ "normalized_hypothesis": "fast but endure",
+ "errors": 0,
+ "words": 3,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "fast but endure",
+ "normalized_hypothesis": "fast but endure",
+ "errors": 0,
+ "words": 3,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0332",
+ "category": "phonetic_balance",
+ "text": "You were looking squeamish this afternoon, he began.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0332.wav",
+ "render_seconds": 74.608209892991,
+ "audio_seconds": 3.4829931972789114,
+ "rtf": 21.420716512245463,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "YOU ARE LOOKING SQUEAMISH THIS AFTERNOON HE BEGAN",
+ "raw": {
+ "normalized_reference": "you were looking squeamish this afternoon, he began.",
+ "normalized_hypothesis": "you are looking squeamish this afternoon he began",
+ "errors": 1,
+ "words": 8,
+ "wer": 0.125
+ },
+ "semantic": {
+ "normalized_reference": "you were looking squeamish this afternoon he began",
+ "normalized_hypothesis": "you are looking squeamish this afternoon he began",
+ "errors": 1,
+ "words": 8,
+ "wer": 0.125
+ }
+ },
+ {
+ "id": "release500_0333",
+ "category": "phonetic_balance",
+ "text": "It is the nearest refuge.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0333.wav",
+ "render_seconds": 67.5107208150439,
+ "audio_seconds": 2.020136054421769,
+ "rtf": 33.418898032770244,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT IS THE NEAREST REFUGE",
+ "raw": {
+ "normalized_reference": "it is the nearest refuge.",
+ "normalized_hypothesis": "it is the nearest refuge",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it is the nearest refuge",
+ "normalized_hypothesis": "it is the nearest refuge",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0334",
+ "category": "phonetic_balance",
+ "text": "I want to die in it.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0334.wav",
+ "render_seconds": 97.8909430099884,
+ "audio_seconds": 1.7414965986394557,
+ "rtf": 56.21081493151678,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I WANT TO DIE IN IT",
+ "raw": {
+ "normalized_reference": "i want to die in it.",
+ "normalized_hypothesis": "i want to die in it",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i want to die in it",
+ "normalized_hypothesis": "i want to die in it",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0335",
+ "category": "phonetic_balance",
+ "text": "He stopped, and Philip nodded at the horrified question in his eyes.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0335.wav",
+ "render_seconds": 160.99368157691788,
+ "audio_seconds": 4.73687074829932,
+ "rtf": 33.98734948271905,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE STOPPED AND PHILIP NODDED AT THE HORRIFIED QUESTION IN HIS EYES",
+ "raw": {
+ "normalized_reference": "he stopped, and philip nodded at the horrified question in his eyes.",
+ "normalized_hypothesis": "he stopped and philip nodded at the horrified question in his eyes",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he stopped and philip nodded at the horrified question in his eyes",
+ "normalized_hypothesis": "he stopped and philip nodded at the horrified question in his eyes",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0336",
+ "category": "phonetic_balance",
+ "text": "She turned, fearing that Jacques might see what was in her face.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0336.wav",
+ "render_seconds": 168.80214889801573,
+ "audio_seconds": 4.249251700680272,
+ "rtf": 39.72514710554609,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SHE TURNED FEARING THAT JACKS MIGHT SEE WHAT WAS IN HER FACE",
+ "raw": {
+ "normalized_reference": "she turned, fearing that jacques might see what was in her face.",
+ "normalized_hypothesis": "she turned fearing that jacks might see what was in her face",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "she turned fearing that jacques might see what was in her face",
+ "normalized_hypothesis": "she turned fearing that jacks might see what was in her face",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ }
+ },
+ {
+ "id": "release500_0337",
+ "category": "phonetic_balance",
+ "text": "It was a curious coincidence.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0337.wav",
+ "render_seconds": 115.40780701104086,
+ "audio_seconds": 2.4380952380952383,
+ "rtf": 47.33523334437222,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT WAS A CURIOUS COINCIDENCE",
+ "raw": {
+ "normalized_reference": "it was a curious coincidence.",
+ "normalized_hypothesis": "it was a curious coincidence",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it was a curious coincidence",
+ "normalized_hypothesis": "it was a curious coincidence",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0338",
+ "category": "phonetic_balance",
+ "text": "Daylight made no answer, and the door closed behind him.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0338.wav",
+ "render_seconds": 88.88503494800534,
+ "audio_seconds": 3.9706122448979593,
+ "rtf": 22.385725290153484,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "DAYLIGHT MADE NO ANSWER AND THE DOOR CLOSED BEHIND HIM",
+ "raw": {
+ "normalized_reference": "daylight made no answer, and the door closed behind him.",
+ "normalized_hypothesis": "daylight made no answer and the door closed behind him",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "daylight made no answer and the door closed behind him",
+ "normalized_hypothesis": "daylight made no answer and the door closed behind him",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0339",
+ "category": "phonetic_balance",
+ "text": "And he did hurt my arm.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0339.wav",
+ "render_seconds": 72.90102104295511,
+ "audio_seconds": 1.9504761904761905,
+ "rtf": 37.376011765186945,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "AND HE DID HURT MY ARM",
+ "raw": {
+ "normalized_reference": "and he did hurt my arm.",
+ "normalized_hypothesis": "and he did hurt my arm",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "and he did hurt my arm",
+ "normalized_hypothesis": "and he did hurt my arm",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0340",
+ "category": "phonetic_balance",
+ "text": "Dunham, can your boy go along with Jesse.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0340.wav",
+ "render_seconds": 123.90290090604685,
+ "audio_seconds": 2.9257142857142857,
+ "rtf": 42.34962433312148,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "DUNHAM CAN YOUR BOY GO ALONG WITH JESSE",
+ "raw": {
+ "normalized_reference": "dunham, can your boy go along with jesse.",
+ "normalized_hypothesis": "dunham can your boy go along with jesse",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "dunham can your boy go along with jesse",
+ "normalized_hypothesis": "dunham can your boy go along with jesse",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0341",
+ "category": "phonetic_balance",
+ "text": "The creative joy, I murmured.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0341.wav",
+ "render_seconds": 124.90152400592342,
+ "audio_seconds": 2.36843537414966,
+ "rtf": 52.73588013807083,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE CREATIVE JOY I MURMURED",
+ "raw": {
+ "normalized_reference": "the creative joy, i murmured.",
+ "normalized_hypothesis": "the creative joy i murmured",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the creative joy i murmured",
+ "normalized_hypothesis": "the creative joy i murmured",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0342",
+ "category": "phonetic_balance",
+ "text": "He pulled, and the log crashed down to break his back.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0342.wav",
+ "render_seconds": 145.49746312305797,
+ "audio_seconds": 3.8312925170068026,
+ "rtf": 37.976077910315205,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE PULLED AND THE LOG CRASHED DOWN TO BREAK HIS BACK",
+ "raw": {
+ "normalized_reference": "he pulled, and the log crashed down to break his back.",
+ "normalized_hypothesis": "he pulled and the log crashed down to break his back",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he pulled and the log crashed down to break his back",
+ "normalized_hypothesis": "he pulled and the log crashed down to break his back",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0343",
+ "category": "phonetic_balance",
+ "text": "By golly, the boy wins.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0343.wav",
+ "render_seconds": 99.89419081900269,
+ "audio_seconds": 2.020136054421769,
+ "rtf": 49.44923912443896,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "BY GOLLY THE BOY WINS",
+ "raw": {
+ "normalized_reference": "by golly, the boy wins.",
+ "normalized_hypothesis": "by golly the boy wins",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "by golly the boy wins",
+ "normalized_hypothesis": "by golly the boy wins",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0344",
+ "category": "phonetic_balance",
+ "text": "The boy threw back his head with pride.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0344.wav",
+ "render_seconds": 55.69395537499804,
+ "audio_seconds": 2.716734693877551,
+ "rtf": 20.500329126914842,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE BOY THREW BACK HIS HEAD WITH PRIDE",
+ "raw": {
+ "normalized_reference": "the boy threw back his head with pride.",
+ "normalized_hypothesis": "the boy threw back his head with pride",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the boy threw back his head with pride",
+ "normalized_hypothesis": "the boy threw back his head with pride",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0345",
+ "category": "phonetic_balance",
+ "text": "The last one I knew was an overseer.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0345.wav",
+ "render_seconds": 121.69537639594637,
+ "audio_seconds": 2.647074829931973,
+ "rtf": 45.97353086503936,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE LAST ONE I KNEW WAS AN OVERSEER",
+ "raw": {
+ "normalized_reference": "the last one i knew was an overseer.",
+ "normalized_hypothesis": "the last one i knew was an overseer",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the last one i knew was an overseer",
+ "normalized_hypothesis": "the last one i knew was an overseer",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0346",
+ "category": "phonetic_balance",
+ "text": "Two years ago I gave up civilization for this.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0346.wav",
+ "render_seconds": 146.50045349705033,
+ "audio_seconds": 3.4133333333333336,
+ "rtf": 42.920054735463964,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "TWO YEARS AGO I GAVE UP CIVILIZATION FOR THIS",
+ "raw": {
+ "normalized_reference": "two years ago i gave up civilization for this.",
+ "normalized_hypothesis": "two years ago i gave up civilization for this",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "2 years ago i gave up civilization for this",
+ "normalized_hypothesis": "2 years ago i gave up civilization for this",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0347",
+ "category": "phonetic_balance",
+ "text": "The river bared its bosom, and snorting steamboats challenged the wilderness.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0347.wav",
+ "render_seconds": 199.09577607398387,
+ "audio_seconds": 5.50312925170068,
+ "rtf": 36.1786479960389,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE RIVER BARED ITS BOSOM AND SNORTING STEAMBOATS CHALLENGED THE WILDERNESS",
+ "raw": {
+ "normalized_reference": "the river bared its bosom, and snorting steamboats challenged the wilderness.",
+ "normalized_hypothesis": "the river bared its bosom and snorting steamboats challenged the wilderness",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the river bared its bosom and snorting steamboats challenged the wilderness",
+ "normalized_hypothesis": "the river bared its bosom and snorting steamboats challenged the wilderness",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0348",
+ "category": "phonetic_balance",
+ "text": "Mab, she said.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0348.wav",
+ "render_seconds": 80.50041246297769,
+ "audio_seconds": 1.6021768707482993,
+ "rtf": 50.24439806410378,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "MABB SHE SAID",
+ "raw": {
+ "normalized_reference": "mab, she said.",
+ "normalized_hypothesis": "mabb she said",
+ "errors": 1,
+ "words": 3,
+ "wer": 0.3333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "mab she said",
+ "normalized_hypothesis": "mabb she said",
+ "errors": 1,
+ "words": 3,
+ "wer": 0.3333333333333333
+ }
+ },
+ {
+ "id": "release500_0349",
+ "category": "phonetic_balance",
+ "text": "The regret in his voice was provocative of a second burst of laughter.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0349.wav",
+ "render_seconds": 176.50368143501692,
+ "audio_seconds": 4.876190476190477,
+ "rtf": 36.19704404429058,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE REGRET IN HIS VOICE WAS PROVOCATIVE OF A SECOND BURST OF LAUGHTER",
+ "raw": {
+ "normalized_reference": "the regret in his voice was provocative of a second burst of laughter.",
+ "normalized_hypothesis": "the regret in his voice was provocative of a second burst of laughter",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the regret in his voice was provocative of a 2nd burst of laughter",
+ "normalized_hypothesis": "the regret in his voice was provocative of a 2nd burst of laughter",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0350",
+ "category": "phonetic_balance",
+ "text": "And each year something happened, and I did not go.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0350.wav",
+ "render_seconds": 103.49894928804133,
+ "audio_seconds": 3.622312925170068,
+ "rtf": 28.572614080044463,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "AND EACH YEAR SOMETHING HAPPENED AND I DID NOT GO",
+ "raw": {
+ "normalized_reference": "and each year something happened, and i did not go.",
+ "normalized_hypothesis": "and each year something happened and i did not go",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "and each year something happened and i did not go",
+ "normalized_hypothesis": "and each year something happened and i did not go",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0351",
+ "category": "phonetic_balance",
+ "text": "Gad, your letter came just in time.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0351.wav",
+ "render_seconds": 114.69290540600196,
+ "audio_seconds": 2.647074829931973,
+ "rtf": 43.3281689316465,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "GAD YOUR LETTER CAME JUST IN TIME",
+ "raw": {
+ "normalized_reference": "gad, your letter came just in time.",
+ "normalized_hypothesis": "gad your letter came just in time",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "gad your letter came just in time",
+ "normalized_hypothesis": "gad your letter came just in time",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0352",
+ "category": "phonetic_balance",
+ "text": "But if Pierre did not return, until tomorrow.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0352.wav",
+ "render_seconds": 120.09822258690838,
+ "audio_seconds": 3.2043537414965986,
+ "rtf": 37.47970176688929,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "BUT IF PIERRE DID NOT RETURN UNTIL TO MORROW",
+ "raw": {
+ "normalized_reference": "but if pierre did not return, until tomorrow.",
+ "normalized_hypothesis": "but if pierre did not return until to morrow",
+ "errors": 2,
+ "words": 8,
+ "wer": 0.25
+ },
+ "semantic": {
+ "normalized_reference": "but if pierre did not return until tomorrow",
+ "normalized_hypothesis": "but if pierre did not return until to morrow",
+ "errors": 2,
+ "words": 8,
+ "wer": 0.25
+ }
+ },
+ {
+ "id": "release500_0353",
+ "category": "phonetic_balance",
+ "text": "Billinger may arrive in time.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0353.wav",
+ "render_seconds": 91.80361258005723,
+ "audio_seconds": 2.2987755102040817,
+ "rtf": 39.93587550091268,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "BILLINGER MAY ARRIVE IN TIME",
+ "raw": {
+ "normalized_reference": "billinger may arrive in time.",
+ "normalized_hypothesis": "billinger may arrive in time",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "billinger may arrive in time",
+ "normalized_hypothesis": "billinger may arrive in time",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0354",
+ "category": "phonetic_balance",
+ "text": "In such a tumbling of values was no time to sell.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0354.wav",
+ "render_seconds": 124.19518921303097,
+ "audio_seconds": 3.4829931972789114,
+ "rtf": 35.657603152960064,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IN SUCH A TUMBLING OF VALUES WAS NO TIME TO SELL",
+ "raw": {
+ "normalized_reference": "in such a tumbling of values was no time to sell.",
+ "normalized_hypothesis": "in such a tumbling of values was no time to sell",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "in such a tumbling of values was no time to sell",
+ "normalized_hypothesis": "in such a tumbling of values was no time to sell",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0355",
+ "category": "phonetic_balance",
+ "text": "There's not an iota of truth in it.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0355.wav",
+ "render_seconds": 95.59949900896754,
+ "audio_seconds": 2.4380952380952383,
+ "rtf": 39.21073201539684,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THERE'S NOT AN IOTA OF TRUTH IN IT",
+ "raw": {
+ "normalized_reference": "there's not an iota of truth in it.",
+ "normalized_hypothesis": "there's not an iota of truth in it",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "there is not an iota of truth in it",
+ "normalized_hypothesis": "there is not an iota of truth in it",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0356",
+ "category": "phonetic_balance",
+ "text": "It is dog eat dog, and you ate them up.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0356.wav",
+ "render_seconds": 79.30646503996104,
+ "audio_seconds": 2.8560544217687074,
+ "rtf": 27.76784098912508,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT IS DOG EAT DOG AND YOU EAT THEM UF",
+ "raw": {
+ "normalized_reference": "it is dog eat dog, and you ate them up.",
+ "normalized_hypothesis": "it is dog eat dog and you eat them uf",
+ "errors": 2,
+ "words": 10,
+ "wer": 0.2
+ },
+ "semantic": {
+ "normalized_reference": "it is dog eat dog and you ate them up",
+ "normalized_hypothesis": "it is dog eat dog and you eat them uf",
+ "errors": 2,
+ "words": 10,
+ "wer": 0.2
+ }
+ },
+ {
+ "id": "release500_0357",
+ "category": "phonetic_balance",
+ "text": "You can take a vacation on pay.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0357.wav",
+ "render_seconds": 109.1108391280286,
+ "audio_seconds": 2.4380952380952383,
+ "rtf": 44.752492611105474,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "YOU CAN TAKE A VACATION ON PAY",
+ "raw": {
+ "normalized_reference": "you can take a vacation on pay.",
+ "normalized_hypothesis": "you can take a vacation on pay",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "you can take a vacation on pay",
+ "normalized_hypothesis": "you can take a vacation on pay",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0358",
+ "category": "phonetic_balance",
+ "text": "I have seen myself that one man contemplated by Pascal's philosophic eye.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0358.wav",
+ "render_seconds": 163.09745570202358,
+ "audio_seconds": 4.876190476190477,
+ "rtf": 33.447720407641555,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I HAVE SEEN MYSELF THAT ONE MAN CONTEMPLATED BY PASCALL'S PHILOSOPHIC EYE",
+ "raw": {
+ "normalized_reference": "i have seen myself that one man contemplated by pascal's philosophic eye.",
+ "normalized_hypothesis": "i have seen myself that one man contemplated by pascall's philosophic eye",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "i have seen myself that one man contemplated by pascal is philosophic eye",
+ "normalized_hypothesis": "i have seen myself that one man contemplated by pascall is philosophic eye",
+ "errors": 1,
+ "words": 13,
+ "wer": 0.07692307692307693
+ }
+ },
+ {
+ "id": "release500_0359",
+ "category": "phonetic_balance",
+ "text": "Ah, I had forgotten, he exclaimed.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0359.wav",
+ "render_seconds": 131.39737610006705,
+ "audio_seconds": 2.5774149659863945,
+ "rtf": 50.98029530909485,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "AH I HAD FORGOTTEN HE EXCLAIMED",
+ "raw": {
+ "normalized_reference": "ah, i had forgotten, he exclaimed.",
+ "normalized_hypothesis": "ah i had forgotten he exclaimed",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "ah i had forgotten he exclaimed",
+ "normalized_hypothesis": "ah i had forgotten he exclaimed",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0360",
+ "category": "phonetic_balance",
+ "text": "Providence had delivered him through the maelstrom.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0360.wav",
+ "render_seconds": 115.09659487800673,
+ "audio_seconds": 3.4133333333333336,
+ "rtf": 33.71970553066603,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "PROVIDENCE HAD DELIVERED HIM THROUGH THE MAILSTRUM",
+ "raw": {
+ "normalized_reference": "providence had delivered him through the maelstrom.",
+ "normalized_hypothesis": "providence had delivered him through the mailstrum",
+ "errors": 1,
+ "words": 7,
+ "wer": 0.14285714285714285
+ },
+ "semantic": {
+ "normalized_reference": "providence had delivered him through the maelstrom",
+ "normalized_hypothesis": "providence had delivered him through the mailstrum",
+ "errors": 1,
+ "words": 7,
+ "wer": 0.14285714285714285
+ }
+ },
+ {
+ "id": "release500_0361",
+ "category": "phonetic_balance",
+ "text": "She was his now, forever.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0361.wav",
+ "render_seconds": 98.50019540800713,
+ "audio_seconds": 2.089795918367347,
+ "rtf": 47.133882568284655,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SHE WAS HIS NOW FOR EVER",
+ "raw": {
+ "normalized_reference": "she was his now, forever.",
+ "normalized_hypothesis": "she was his now for ever",
+ "errors": 2,
+ "words": 5,
+ "wer": 0.4
+ },
+ "semantic": {
+ "normalized_reference": "she was his now forever",
+ "normalized_hypothesis": "she was his now for ever",
+ "errors": 2,
+ "words": 5,
+ "wer": 0.4
+ }
+ },
+ {
+ "id": "release500_0362",
+ "category": "phonetic_balance",
+ "text": "It was unobtrusive, yet it was there.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0362.wav",
+ "render_seconds": 54.78923688700888,
+ "audio_seconds": 2.716734693877551,
+ "rtf": 20.167312255584697,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT WAS UNOBTRUSIVE YET IT WAS THERE",
+ "raw": {
+ "normalized_reference": "it was unobtrusive, yet it was there.",
+ "normalized_hypothesis": "it was unobtrusive yet it was there",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it was unobtrusive yet it was there",
+ "normalized_hypothesis": "it was unobtrusive yet it was there",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0363",
+ "category": "phonetic_balance",
+ "text": "Men like Joe Goose dated existence from drunk to drunk.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0363.wav",
+ "render_seconds": 150.6957912640646,
+ "audio_seconds": 3.9706122448979593,
+ "rtf": 37.95278460083864,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "MEN LIKE JOE GOOSE DATED EXISTENCE FROM DRUNK TO DRUNK",
+ "raw": {
+ "normalized_reference": "men like joe goose dated existence from drunk to drunk.",
+ "normalized_hypothesis": "men like joe goose dated existence from drunk to drunk",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "men like joe goose dated existence from drunk to drunk",
+ "normalized_hypothesis": "men like joe goose dated existence from drunk to drunk",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0364",
+ "category": "phonetic_balance",
+ "text": "I was completely lost in my work.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0364.wav",
+ "render_seconds": 113.29968268994708,
+ "audio_seconds": 2.5774149659863945,
+ "rtf": 43.95865011460679,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I WAS COMPLETELY LOST IN MY WORK",
+ "raw": {
+ "normalized_reference": "i was completely lost in my work.",
+ "normalized_hypothesis": "i was completely lost in my work",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i was completely lost in my work",
+ "normalized_hypothesis": "i was completely lost in my work",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0365",
+ "category": "phonetic_balance",
+ "text": "The very thing, Ernest agreed.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0365.wav",
+ "render_seconds": 113.60242960799951,
+ "audio_seconds": 2.2987755102040817,
+ "rtf": 49.4186705503549,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE VERY THING ERNEST AGREED",
+ "raw": {
+ "normalized_reference": "the very thing, ernest agreed.",
+ "normalized_hypothesis": "the very thing ernest agreed",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the very thing ernest agreed",
+ "normalized_hypothesis": "the very thing ernest agreed",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0366",
+ "category": "phonetic_balance",
+ "text": "He was the leader, and Tudor was his lieutenant.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0366.wav",
+ "render_seconds": 144.5026921579847,
+ "audio_seconds": 3.622312925170068,
+ "rtf": 39.892382337784994,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE WAS THE LEADER AND TUTOR WAS HIS LIEUTENANT",
+ "raw": {
+ "normalized_reference": "he was the leader, and tudor was his lieutenant.",
+ "normalized_hypothesis": "he was the leader and tutor was his lieutenant",
+ "errors": 1,
+ "words": 9,
+ "wer": 0.1111111111111111
+ },
+ "semantic": {
+ "normalized_reference": "he was the leader and tudor was his lieutenant",
+ "normalized_hypothesis": "he was the leader and tutor was his lieutenant",
+ "errors": 1,
+ "words": 9,
+ "wer": 0.1111111111111111
+ }
+ },
+ {
+ "id": "release500_0367",
+ "category": "phonetic_balance",
+ "text": "The Fire-Men wore animal skins around their waists and across their shoulders.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0367.wav",
+ "render_seconds": 188.79228968708776,
+ "audio_seconds": 5.712108843537415,
+ "rtf": 33.05124164443824,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE FIRE MEN WORE ANIMAL SKINS AROUND THEIR WAISTS AND ACROSS THEIR SHOULDERS",
+ "raw": {
+ "normalized_reference": "the fire-men wore animal skins around their waists and across their shoulders.",
+ "normalized_hypothesis": "the fire men wore animal skins around their waists and across their shoulders",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the fire men wore animal skins around their waists and across their shoulders",
+ "normalized_hypothesis": "the fire men wore animal skins around their waists and across their shoulders",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0368",
+ "category": "phonetic_balance",
+ "text": "We could throw stones with our feet.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0368.wav",
+ "render_seconds": 70.40826302394271,
+ "audio_seconds": 2.5774149659863945,
+ "rtf": 27.31739512383757,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WE COULD THROW STONES WITH OUR FEET",
+ "raw": {
+ "normalized_reference": "we could throw stones with our feet.",
+ "normalized_hypothesis": "we could throw stones with our feet",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "we could throw stones with our feet",
+ "normalized_hypothesis": "we could throw stones with our feet",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0369",
+ "category": "phonetic_balance",
+ "text": "He turned sharply, and faced Gregson across the table.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0369.wav",
+ "render_seconds": 160.1924027170753,
+ "audio_seconds": 4.249251700680272,
+ "rtf": 37.698967724465405,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE TURNED SHARPLY AND FACED GREGSON ACROSS THE TABLE",
+ "raw": {
+ "normalized_reference": "he turned sharply, and faced gregson across the table.",
+ "normalized_hypothesis": "he turned sharply and faced gregson across the table",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he turned sharply and faced gregson across the table",
+ "normalized_hypothesis": "he turned sharply and faced gregson across the table",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0370",
+ "category": "phonetic_balance",
+ "text": "They are to attack your camp tomorrow night.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0370.wav",
+ "render_seconds": 120.7102004759945,
+ "audio_seconds": 3.0650340136054424,
+ "rtf": 39.382988880440315,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THEY ARE TO ATTACK YOUR CAMP TO MORROW NIGHT",
+ "raw": {
+ "normalized_reference": "they are to attack your camp tomorrow night.",
+ "normalized_hypothesis": "they are to attack your camp to morrow night",
+ "errors": 2,
+ "words": 8,
+ "wer": 0.25
+ },
+ "semantic": {
+ "normalized_reference": "they are to attack your camp tomorrow night",
+ "normalized_hypothesis": "they are to attack your camp to morrow night",
+ "errors": 2,
+ "words": 8,
+ "wer": 0.25
+ }
+ },
+ {
+ "id": "release500_0371",
+ "category": "phonetic_balance",
+ "text": "And watch out for wet feet, was his parting advice.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0371.wav",
+ "render_seconds": 139.09965348092373,
+ "audio_seconds": 3.5526530612244898,
+ "rtf": 39.15373977806332,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "AND WATCH OUT FOR WET FEET WAS HIS PARTING ADVICE",
+ "raw": {
+ "normalized_reference": "and watch out for wet feet, was his parting advice.",
+ "normalized_hypothesis": "and watch out for wet feet was his parting advice",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "and watch out for wet feet was his parting advice",
+ "normalized_hypothesis": "and watch out for wet feet was his parting advice",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0372",
+ "category": "phonetic_balance",
+ "text": "Two weeks passed, and in that time Thorpe left camp three times.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0372.wav",
+ "render_seconds": 164.39656222204212,
+ "audio_seconds": 4.527891156462585,
+ "rtf": 36.307534024399324,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "TWO WEEKS PASSED AND IN THAT TIME THORPE LEFT CAMP THREE TIMES",
+ "raw": {
+ "normalized_reference": "two weeks passed, and in that time thorpe left camp three times.",
+ "normalized_hypothesis": "two weeks passed and in that time thorpe left camp three times",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "2 weeks passed and in that time thorpe left camp 3 times",
+ "normalized_hypothesis": "2 weeks passed and in that time thorpe left camp 3 times",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0373",
+ "category": "phonetic_balance",
+ "text": "But ever his gaze returned to that Crouched Venus on the piano.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0373.wav",
+ "render_seconds": 155.61082126491237,
+ "audio_seconds": 4.388571428571429,
+ "rtf": 35.45819494968706,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "BUT EVER HIS GAZE RETURNED TO THAT CROUCHED VENUS ON THE PIANO",
+ "raw": {
+ "normalized_reference": "but ever his gaze returned to that crouched venus on the piano.",
+ "normalized_hypothesis": "but ever his gaze returned to that crouched venus on the piano",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "but ever his gaze returned to that crouched venus on the piano",
+ "normalized_hypothesis": "but ever his gaze returned to that crouched venus on the piano",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0374",
+ "category": "phonetic_balance",
+ "text": "But she had become an automaton.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0374.wav",
+ "render_seconds": 59.1937680329429,
+ "audio_seconds": 2.36843537414966,
+ "rtf": 24.992773152695904,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "BUT SHE HAD BECOME AN AUTOMATAN",
+ "raw": {
+ "normalized_reference": "but she had become an automaton.",
+ "normalized_hypothesis": "but she had become an automatan",
+ "errors": 1,
+ "words": 6,
+ "wer": 0.16666666666666666
+ },
+ "semantic": {
+ "normalized_reference": "but she had become an automaton",
+ "normalized_hypothesis": "but she had become an automatan",
+ "errors": 1,
+ "words": 6,
+ "wer": 0.16666666666666666
+ }
+ },
+ {
+ "id": "release500_0375",
+ "category": "phonetic_balance",
+ "text": "I do not blame you for anything; remember that.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0375.wav",
+ "render_seconds": 127.7072864070069,
+ "audio_seconds": 3.1346938775510202,
+ "rtf": 40.73995464806861,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I DO NOT BLAME YOU FOR ANYTHING REMEMBER THAT",
+ "raw": {
+ "normalized_reference": "i do not blame you for anything; remember that.",
+ "normalized_hypothesis": "i do not blame you for anything remember that",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i do not blame you for anything remember that",
+ "normalized_hypothesis": "i do not blame you for anything remember that",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0376",
+ "category": "phonetic_balance",
+ "text": "The next thing to watch out for is bed sores.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0376.wav",
+ "render_seconds": 110.20190875290427,
+ "audio_seconds": 3.0650340136054424,
+ "rtf": 35.95454675665156,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE NEXT THING TO WATCH OUT FOR HIS BEDSORES",
+ "raw": {
+ "normalized_reference": "the next thing to watch out for is bed sores.",
+ "normalized_hypothesis": "the next thing to watch out for his bedsores",
+ "errors": 3,
+ "words": 10,
+ "wer": 0.3
+ },
+ "semantic": {
+ "normalized_reference": "the next thing to watch out for is bed sores",
+ "normalized_hypothesis": "the next thing to watch out for his bedsores",
+ "errors": 3,
+ "words": 10,
+ "wer": 0.3
+ }
+ },
+ {
+ "id": "release500_0377",
+ "category": "phonetic_balance",
+ "text": "He will follow us soon.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0377.wav",
+ "render_seconds": 85.29273871390615,
+ "audio_seconds": 1.9504761904761905,
+ "rtf": 43.72918732890696,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE WILL FOLLOW US SOON",
+ "raw": {
+ "normalized_reference": "he will follow us soon.",
+ "normalized_hypothesis": "he will follow us soon",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he will follow us soon",
+ "normalized_hypothesis": "he will follow us soon",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0378",
+ "category": "phonetic_balance",
+ "text": "His beady black eyes saw bargains where other men saw bankruptcy.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0378.wav",
+ "render_seconds": 128.69972406700253,
+ "audio_seconds": 4.527891156462585,
+ "rtf": 28.423767184268886,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HIS BETTY BLACK EYES SAW BARGAINS WHERE OTHER MEN SAW BANKRUPTCY",
+ "raw": {
+ "normalized_reference": "his beady black eyes saw bargains where other men saw bankruptcy.",
+ "normalized_hypothesis": "his betty black eyes saw bargains where other men saw bankruptcy",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ },
+ "semantic": {
+ "normalized_reference": "his beady black eyes saw bargains where other men saw bankruptcy",
+ "normalized_hypothesis": "his betty black eyes saw bargains where other men saw bankruptcy",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ }
+ },
+ {
+ "id": "release500_0379",
+ "category": "phonetic_balance",
+ "text": "They robbed me a few years later.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0379.wav",
+ "render_seconds": 125.78808541595936,
+ "audio_seconds": 2.36843537414966,
+ "rtf": 53.11020380326868,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THEY ROBBED ME A FEW YEARS LATER",
+ "raw": {
+ "normalized_reference": "they robbed me a few years later.",
+ "normalized_hypothesis": "they robbed me a few years later",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "they robbed me a few years later",
+ "normalized_hypothesis": "they robbed me a few years later",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0380",
+ "category": "phonetic_balance",
+ "text": "Gregson held a lighted match until it burnt his fingertips.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0380.wav",
+ "render_seconds": 104.29416135197971,
+ "audio_seconds": 4.249251700680272,
+ "rtf": 24.544124165504957,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "GREGSON HELD A LIGHTED MATCH UNTIL IT BURNT HIS FINGER TIPS",
+ "raw": {
+ "normalized_reference": "gregson held a lighted match until it burnt his fingertips.",
+ "normalized_hypothesis": "gregson held a lighted match until it burnt his finger tips",
+ "errors": 2,
+ "words": 10,
+ "wer": 0.2
+ },
+ "semantic": {
+ "normalized_reference": "gregson held a lighted match until it burnt his fingertips",
+ "normalized_hypothesis": "gregson held a lighted match until it burnt his finger tips",
+ "errors": 2,
+ "words": 10,
+ "wer": 0.2
+ }
+ },
+ {
+ "id": "release500_0381",
+ "category": "phonetic_balance",
+ "text": "A burst of laughter was his reward.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0381.wav",
+ "render_seconds": 112.99254741205368,
+ "audio_seconds": 2.647074829931973,
+ "rtf": 42.68581535149026,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "A BURST OF LAUGHTER WAS HIS REWARD",
+ "raw": {
+ "normalized_reference": "a burst of laughter was his reward.",
+ "normalized_hypothesis": "a burst of laughter was his reward",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "a burst of laughter was his reward",
+ "normalized_hypothesis": "a burst of laughter was his reward",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0382",
+ "category": "phonetic_balance",
+ "text": "He cried, and swung the club wildly.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0382.wav",
+ "render_seconds": 109.38749186589848,
+ "audio_seconds": 2.716734693877551,
+ "rtf": 40.264326182609864,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE CRIED AND SWUNG THE CLUB WILDLY",
+ "raw": {
+ "normalized_reference": "he cried, and swung the club wildly.",
+ "normalized_hypothesis": "he cried and swung the club wildly",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he cried and swung the club wildly",
+ "normalized_hypothesis": "he cried and swung the club wildly",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0383",
+ "category": "phonetic_balance",
+ "text": "Sandel would never become a world champion.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0383.wav",
+ "render_seconds": 128.50690559204668,
+ "audio_seconds": 3.2043537414965986,
+ "rtf": 40.10384494317014,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SANDALL WOULD NEVER BECOME A WORLD CHAMPION",
+ "raw": {
+ "normalized_reference": "sandel would never become a world champion.",
+ "normalized_hypothesis": "sandall would never become a world champion",
+ "errors": 1,
+ "words": 7,
+ "wer": 0.14285714285714285
+ },
+ "semantic": {
+ "normalized_reference": "sandel would never become a world champion",
+ "normalized_hypothesis": "sandall would never become a world champion",
+ "errors": 1,
+ "words": 7,
+ "wer": 0.14285714285714285
+ }
+ },
+ {
+ "id": "release500_0384",
+ "category": "phonetic_balance",
+ "text": "And after the bath a shave would not be bad.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0384.wav",
+ "render_seconds": 125.00666798499878,
+ "audio_seconds": 3.0650340136054424,
+ "rtf": 40.78475717727898,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "AND AFTER BATH A SHAVE WOULD NOT BE BAD",
+ "raw": {
+ "normalized_reference": "and after the bath a shave would not be bad.",
+ "normalized_hypothesis": "and after bath a shave would not be bad",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ },
+ "semantic": {
+ "normalized_reference": "and after the bath a shave would not be bad",
+ "normalized_hypothesis": "and after bath a shave would not be bad",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ }
+ },
+ {
+ "id": "release500_0385",
+ "category": "phonetic_balance",
+ "text": "They saw each other for the first time in Boston.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0385.wav",
+ "render_seconds": 127.90721304703038,
+ "audio_seconds": 3.3436734693877552,
+ "rtf": 38.253499995754936,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THEY SAW EACH OTHER FOR THE FIRST TIME IN BOSTON",
+ "raw": {
+ "normalized_reference": "they saw each other for the first time in boston.",
+ "normalized_hypothesis": "they saw each other for the first time in boston",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "they saw each other for the 1st time in boston",
+ "normalized_hypothesis": "they saw each other for the 1st time in boston",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0386",
+ "category": "phonetic_balance",
+ "text": "I had been born with no organic, chemical predisposition toward alcohol.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0386.wav",
+ "render_seconds": 165.60716925095767,
+ "audio_seconds": 5.224489795918367,
+ "rtf": 31.69824723944112,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I HAD BEEN BORN WITH NO ORGANIC CHEMICAL PREDISPOSITION TOWARD ALCOHOL",
+ "raw": {
+ "normalized_reference": "i had been born with no organic, chemical predisposition toward alcohol.",
+ "normalized_hypothesis": "i had been born with no organic chemical predisposition toward alcohol",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i had been born with no organic chemical predisposition toward alcohol",
+ "normalized_hypothesis": "i had been born with no organic chemical predisposition toward alcohol",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0387",
+ "category": "phonetic_balance",
+ "text": "They had no fixed values, to be altered by adjectives and adverbs.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0387.wav",
+ "render_seconds": 192.80699968105182,
+ "audio_seconds": 4.9458503401360545,
+ "rtf": 38.98358955919154,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THEY HAD NO FIXED VALUES TO BE ALTERED BY ADJECTIVES AND ADVERBS",
+ "raw": {
+ "normalized_reference": "they had no fixed values, to be altered by adjectives and adverbs.",
+ "normalized_hypothesis": "they had no fixed values to be altered by adjectives and adverbs",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "they had no fixed values to be altered by adjectives and adverbs",
+ "normalized_hypothesis": "they had no fixed values to be altered by adjectives and adverbs",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0388",
+ "category": "phonetic_balance",
+ "text": "Nor was Elam Harnish an exception.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0388.wav",
+ "render_seconds": 111.40671193704475,
+ "audio_seconds": 2.716734693877551,
+ "rtf": 41.007578762884556,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "NOR WAS ELAM HARNISH AN EXCEPTION",
+ "raw": {
+ "normalized_reference": "nor was elam harnish an exception.",
+ "normalized_hypothesis": "nor was elam harnish an exception",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "nor was elam harnish an exception",
+ "normalized_hypothesis": "nor was elam harnish an exception",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0389",
+ "category": "phonetic_balance",
+ "text": "He was a wise hyena.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0389.wav",
+ "render_seconds": 77.29535092203878,
+ "audio_seconds": 1.8808163265306121,
+ "rtf": 41.09670350672635,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE WAS A WISE HYENA",
+ "raw": {
+ "normalized_reference": "he was a wise hyena.",
+ "normalized_hypothesis": "he was a wise hyena",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he was a wise hyena",
+ "normalized_hypothesis": "he was a wise hyena",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0390",
+ "category": "phonetic_balance",
+ "text": "And now behold the perversity of things.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0390.wav",
+ "render_seconds": 118.90010294201784,
+ "audio_seconds": 2.8560544217687074,
+ "rtf": 41.63089541843708,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "AND NOW BEHOLD THE PERVERSITY OF THINGS",
+ "raw": {
+ "normalized_reference": "and now behold the perversity of things.",
+ "normalized_hypothesis": "and now behold the perversity of things",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "and now behold the perversity of things",
+ "normalized_hypothesis": "and now behold the perversity of things",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0391",
+ "category": "phonetic_balance",
+ "text": "The Fire People, like ourselves, lived in caves.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0391.wav",
+ "render_seconds": 139.59252273198217,
+ "audio_seconds": 3.4829931972789114,
+ "rtf": 40.078321956252694,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE FIRE PEOPLE LIKE OURSELVES LIVED IN CAVES",
+ "raw": {
+ "normalized_reference": "the fire people, like ourselves, lived in caves.",
+ "normalized_hypothesis": "the fire people like ourselves lived in caves",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the fire people like ourselves lived in caves",
+ "normalized_hypothesis": "the fire people like ourselves lived in caves",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0392",
+ "category": "phonetic_balance",
+ "text": "To her the bridge was tambo, which is the native for taboo.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0392.wav",
+ "render_seconds": 148.99898482393473,
+ "audio_seconds": 4.040272108843538,
+ "rtf": 36.87845293830551,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "TO HER THE BRIDGE WAS TAMBO WHICH IS THE NATIVE FOR TABOU",
+ "raw": {
+ "normalized_reference": "to her the bridge was tambo, which is the native for taboo.",
+ "normalized_hypothesis": "to her the bridge was tambo which is the native for tabou",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ },
+ "semantic": {
+ "normalized_reference": "to her the bridge was tambo which is the native for taboo",
+ "normalized_hypothesis": "to her the bridge was tambo which is the native for tabou",
+ "errors": 1,
+ "words": 12,
+ "wer": 0.08333333333333333
+ }
+ },
+ {
+ "id": "release500_0393",
+ "category": "phonetic_balance",
+ "text": "The voyage was our idea of a good time.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0393.wav",
+ "render_seconds": 117.19434719404671,
+ "audio_seconds": 2.8560544217687074,
+ "rtf": 41.033653385872874,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE VOYAGE WAS OUR IDEA OF A GOOD TIME",
+ "raw": {
+ "normalized_reference": "the voyage was our idea of a good time.",
+ "normalized_hypothesis": "the voyage was our idea of a good time",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the voyage was our idea of a good time",
+ "normalized_hypothesis": "the voyage was our idea of a good time",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0394",
+ "category": "phonetic_balance",
+ "text": "Of course much grumbling went on, and little outbursts were continually occurring.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0394.wav",
+ "render_seconds": 186.702436683001,
+ "audio_seconds": 5.712108843537415,
+ "rtf": 32.68537799209359,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "OF COURSE MUCH GRUMBLING WENT ON AND LITTLE OUTBURSTS WERE CONTINUALLY OCCURRING",
+ "raw": {
+ "normalized_reference": "of course much grumbling went on, and little outbursts were continually occurring.",
+ "normalized_hypothesis": "of course much grumbling went on and little outbursts were continually occurring",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "of course much grumbling went on and little outbursts were continually occurring",
+ "normalized_hypothesis": "of course much grumbling went on and little outbursts were continually occurring",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0395",
+ "category": "phonetic_balance",
+ "text": "Get down and dig in.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0395.wav",
+ "render_seconds": 81.00418672303203,
+ "audio_seconds": 1.811156462585034,
+ "rtf": 44.72511812006351,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "GET DOWN AND DIG IN",
+ "raw": {
+ "normalized_reference": "get down and dig in.",
+ "normalized_hypothesis": "get down and dig in",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "get down and dig in",
+ "normalized_hypothesis": "get down and dig in",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0396",
+ "category": "phonetic_balance",
+ "text": "It was put together in a casual, helter-skelter sort of way.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0396.wav",
+ "render_seconds": 140.20012241892982,
+ "audio_seconds": 4.31891156462585,
+ "rtf": 32.46191090534067,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT WAS PUT TOGETHER IN A CASUAL HELTER SKELTER SORT OF WAY",
+ "raw": {
+ "normalized_reference": "it was put together in a casual, helter-skelter sort of way.",
+ "normalized_hypothesis": "it was put together in a casual helter skelter sort of way",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it was put together in a casual helter skelter sort of way",
+ "normalized_hypothesis": "it was put together in a casual helter skelter sort of way",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0397",
+ "category": "phonetic_balance",
+ "text": "He deluged me, overwhelmed me with argument.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0397.wav",
+ "render_seconds": 102.50742393604014,
+ "audio_seconds": 3.0650340136054424,
+ "rtf": 33.44413911265514,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE DELUGED ME OVERWHELMED ME WITH ARGUMENT",
+ "raw": {
+ "normalized_reference": "he deluged me, overwhelmed me with argument.",
+ "normalized_hypothesis": "he deluged me overwhelmed me with argument",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he deluged me overwhelmed me with argument",
+ "normalized_hypothesis": "he deluged me overwhelmed me with argument",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0398",
+ "category": "phonetic_balance",
+ "text": "The girl faced him, her eyes shining with sudden fear.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0398.wav",
+ "render_seconds": 87.59573250403628,
+ "audio_seconds": 3.6919727891156464,
+ "rtf": 23.72599623764249,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE GIRL FACED HIM HER EYES SHINING WITH SUDDEN FEAR",
+ "raw": {
+ "normalized_reference": "the girl faced him, her eyes shining with sudden fear.",
+ "normalized_hypothesis": "the girl faced him her eyes shining with sudden fear",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the girl faced him her eyes shining with sudden fear",
+ "normalized_hypothesis": "the girl faced him her eyes shining with sudden fear",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0399",
+ "category": "phonetic_balance",
+ "text": "It will break our hearts and our backs to hoist anchor by hand.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0399.wav",
+ "render_seconds": 168.71729675191455,
+ "audio_seconds": 4.597551020408163,
+ "rtf": 36.69721031979675,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT WILL BREAK OUR HEARTS AND OUR BACKS TO HOIST ANCHOR BY HAND",
+ "raw": {
+ "normalized_reference": "it will break our hearts and our backs to hoist anchor by hand.",
+ "normalized_hypothesis": "it will break our hearts and our backs to hoist anchor by hand",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it will break our hearts and our backs to hoist anchor by hand",
+ "normalized_hypothesis": "it will break our hearts and our backs to hoist anchor by hand",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0400",
+ "category": "phonetic_balance",
+ "text": "He made no reply as he waited for Whittemore to continue.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0400.wav",
+ "render_seconds": 142.1903632429894,
+ "audio_seconds": 4.040272108843538,
+ "rtf": 35.19326407044626,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE MADE NO REPLY AS HE WAITED FOR WHITMORE TO CONTINUE",
+ "raw": {
+ "normalized_reference": "he made no reply as he waited for whittemore to continue.",
+ "normalized_hypothesis": "he made no reply as he waited for whitmore to continue",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ },
+ "semantic": {
+ "normalized_reference": "he made no reply as he waited for whittemore to continue",
+ "normalized_hypothesis": "he made no reply as he waited for whitmore to continue",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ }
+ },
+ {
+ "id": "release500_0401",
+ "category": "phonetic_balance",
+ "text": "Who the devil gave it to you to be judge and jury.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0401.wav",
+ "render_seconds": 113.69965788710397,
+ "audio_seconds": 3.3436734693877552,
+ "rtf": 34.004414285083584,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WHO THE DEVIL GAVE IT TO YOU TO BE JUDGE AND JURY",
+ "raw": {
+ "normalized_reference": "who the devil gave it to you to be judge and jury.",
+ "normalized_hypothesis": "who the devil gave it to you to be judge and jury",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "who the devil gave it to you to be judge and jury",
+ "normalized_hypothesis": "who the devil gave it to you to be judge and jury",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0402",
+ "category": "phonetic_balance",
+ "text": "Daylight found himself charmed and made curious by the little man.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0402.wav",
+ "render_seconds": 166.39715103199705,
+ "audio_seconds": 4.527891156462585,
+ "rtf": 36.749370795828675,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "DAYLIGHT FOUND HIMSELF CHARMED AND MADE CURIOUS BY THE LITTLE MAN",
+ "raw": {
+ "normalized_reference": "daylight found himself charmed and made curious by the little man.",
+ "normalized_hypothesis": "daylight found himself charmed and made curious by the little man",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "daylight found himself charmed and made curious by the little man",
+ "normalized_hypothesis": "daylight found himself charmed and made curious by the little man",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0403",
+ "category": "phonetic_balance",
+ "text": "It's a Yankee, Joan cried.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0403.wav",
+ "render_seconds": 82.81564847100526,
+ "audio_seconds": 2.089795918367347,
+ "rtf": 39.62858178788337,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT'S A YANKEE JOANE CRIED",
+ "raw": {
+ "normalized_reference": "it's a yankee, joan cried.",
+ "normalized_hypothesis": "it's a yankee joane cried",
+ "errors": 1,
+ "words": 5,
+ "wer": 0.2
+ },
+ "semantic": {
+ "normalized_reference": "it is a yankee joan cried",
+ "normalized_hypothesis": "it is a yankee joane cried",
+ "errors": 1,
+ "words": 6,
+ "wer": 0.16666666666666666
+ }
+ },
+ {
+ "id": "release500_0404",
+ "category": "phonetic_balance",
+ "text": "But what they want with your toothbrush is more than I can imagine.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0404.wav",
+ "render_seconds": 135.29682943108492,
+ "audio_seconds": 4.388571428571429,
+ "rtf": 30.829355664635234,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "BUT WHAT THEY WANT WITH YOUR TOOTH BRUSH IS MORE THAN I CAN IMAGINE",
+ "raw": {
+ "normalized_reference": "but what they want with your toothbrush is more than i can imagine.",
+ "normalized_hypothesis": "but what they want with your tooth brush is more than i can imagine",
+ "errors": 2,
+ "words": 13,
+ "wer": 0.15384615384615385
+ },
+ "semantic": {
+ "normalized_reference": "but what they want with your toothbrush is more than i can imagine",
+ "normalized_hypothesis": "but what they want with your tooth brush is more than i can imagine",
+ "errors": 2,
+ "words": 13,
+ "wer": 0.15384615384615385
+ }
+ },
+ {
+ "id": "release500_0405",
+ "category": "phonetic_balance",
+ "text": "They were babbling and chattering all together.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0405.wav",
+ "render_seconds": 107.5880376159912,
+ "audio_seconds": 3.2043537414965986,
+ "rtf": 33.57558069283014,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THEY WERE BABBLING AND CHATTERING ALTOGETHER",
+ "raw": {
+ "normalized_reference": "they were babbling and chattering all together.",
+ "normalized_hypothesis": "they were babbling and chattering altogether",
+ "errors": 2,
+ "words": 7,
+ "wer": 0.2857142857142857
+ },
+ "semantic": {
+ "normalized_reference": "they were babbling and chattering all together",
+ "normalized_hypothesis": "they were babbling and chattering altogether",
+ "errors": 2,
+ "words": 7,
+ "wer": 0.2857142857142857
+ }
+ },
+ {
+ "id": "release500_0406",
+ "category": "phonetic_balance",
+ "text": "Men of Selden's stamp don't stop at women and children.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0406.wav",
+ "render_seconds": 131.00622504099738,
+ "audio_seconds": 3.4133333333333336,
+ "rtf": 38.3807299924797,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "MEN OF SELDOM STAMP DON'T STOP AT WOMEN AND CHILDREN",
+ "raw": {
+ "normalized_reference": "men of selden's stamp don't stop at women and children.",
+ "normalized_hypothesis": "men of seldom stamp don't stop at women and children",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ },
+ "semantic": {
+ "normalized_reference": "men of selden is stamp do not stop at women and children",
+ "normalized_hypothesis": "men of seldom stamp do not stop at women and children",
+ "errors": 2,
+ "words": 12,
+ "wer": 0.16666666666666666
+ }
+ },
+ {
+ "id": "release500_0407",
+ "category": "phonetic_balance",
+ "text": "Cameron looked at his hands with their long, sinewy fingers.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0407.wav",
+ "render_seconds": 137.7075223509455,
+ "audio_seconds": 4.179591836734694,
+ "rtf": 32.94760056248208,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "CAMERON LOOKED AT HIS HANDS WITH THEIR LONG SINWY FINGERS",
+ "raw": {
+ "normalized_reference": "cameron looked at his hands with their long, sinewy fingers.",
+ "normalized_hypothesis": "cameron looked at his hands with their long sinwy fingers",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ },
+ "semantic": {
+ "normalized_reference": "cameron looked at his hands with their long sinewy fingers",
+ "normalized_hypothesis": "cameron looked at his hands with their long sinwy fingers",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ }
+ },
+ {
+ "id": "release500_0408",
+ "category": "phonetic_balance",
+ "text": "Saxon waited, for she knew a fresh idea had struck Billy.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0408.wav",
+ "render_seconds": 137.90217854199,
+ "audio_seconds": 4.179591836734694,
+ "rtf": 32.994173576940966,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SAXON WAITED FOR SHE KNEW A FRESH IDEA HAD STRUCK BILLY",
+ "raw": {
+ "normalized_reference": "saxon waited, for she knew a fresh idea had struck billy.",
+ "normalized_hypothesis": "saxon waited for she knew a fresh idea had struck billy",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "saxon waited for she knew a fresh idea had struck billy",
+ "normalized_hypothesis": "saxon waited for she knew a fresh idea had struck billy",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0409",
+ "category": "phonetic_balance",
+ "text": "He saw only the effect in a general, sketchy way.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0409.wav",
+ "render_seconds": 128.47979144204874,
+ "audio_seconds": 3.4133333333333336,
+ "rtf": 37.640563899037716,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE SAW ONLY THE EFFECT IN A GENERAL SKETCHY WAY",
+ "raw": {
+ "normalized_reference": "he saw only the effect in a general, sketchy way.",
+ "normalized_hypothesis": "he saw only the effect in a general sketchy way",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he saw only the effect in a general sketchy way",
+ "normalized_hypothesis": "he saw only the effect in a general sketchy way",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0410",
+ "category": "phonetic_balance",
+ "text": "He began to follow the footprints of the dog.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0410.wav",
+ "render_seconds": 93.816970793996,
+ "audio_seconds": 2.995374149659864,
+ "rtf": 31.320618429136566,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE BEGAN TO FOLLOW THE FOOTPRINTS OF THE DOG",
+ "raw": {
+ "normalized_reference": "he began to follow the footprints of the dog.",
+ "normalized_hypothesis": "he began to follow the footprints of the dog",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he began to follow the footprints of the dog",
+ "normalized_hypothesis": "he began to follow the footprints of the dog",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0411",
+ "category": "phonetic_balance",
+ "text": "Do you value your hide.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0411.wav",
+ "render_seconds": 70.79543694294989,
+ "audio_seconds": 1.811156462585034,
+ "rtf": 39.088526256812024,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "DO YOU VALUE YOUR HIDE",
+ "raw": {
+ "normalized_reference": "do you value your hide.",
+ "normalized_hypothesis": "do you value your hide",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "do you value your hide",
+ "normalized_hypothesis": "do you value your hide",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0412",
+ "category": "phonetic_balance",
+ "text": "A rising tide of fat had submerged them.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0412.wav",
+ "render_seconds": 116.99265528493561,
+ "audio_seconds": 2.8560544217687074,
+ "rtf": 40.96303431518086,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "A RISING TIDE OF FAT AT SUBMERGED THEM",
+ "raw": {
+ "normalized_reference": "a rising tide of fat had submerged them.",
+ "normalized_hypothesis": "a rising tide of fat at submerged them",
+ "errors": 1,
+ "words": 8,
+ "wer": 0.125
+ },
+ "semantic": {
+ "normalized_reference": "a rising tide of fat had submerged them",
+ "normalized_hypothesis": "a rising tide of fat at submerged them",
+ "errors": 1,
+ "words": 8,
+ "wer": 0.125
+ }
+ },
+ {
+ "id": "release500_0413",
+ "category": "phonetic_balance",
+ "text": "There was nothing on the rock.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0413.wav",
+ "render_seconds": 86.40459188702516,
+ "audio_seconds": 2.2291156462585033,
+ "rtf": 38.761825584084164,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THERE WAS NOTHING ON THE ROCK",
+ "raw": {
+ "normalized_reference": "there was nothing on the rock.",
+ "normalized_hypothesis": "there was nothing on the rock",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "there was nothing on the rock",
+ "normalized_hypothesis": "there was nothing on the rock",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0414",
+ "category": "phonetic_balance",
+ "text": "Whatever he guessed he locked away in the taboo room of Naomi.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0414.wav",
+ "render_seconds": 146.0921727620298,
+ "audio_seconds": 4.458231292517007,
+ "rtf": 32.769087823514376,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WHATEVER HE GUESSED HE LOCKED AWAY IN THE TABOO ROOM OF NAOMI",
+ "raw": {
+ "normalized_reference": "whatever he guessed he locked away in the taboo room of naomi.",
+ "normalized_hypothesis": "whatever he guessed he locked away in the taboo room of naomi",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "whatever he guessed he locked away in the taboo room of naomi",
+ "normalized_hypothesis": "whatever he guessed he locked away in the taboo room of naomi",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0415",
+ "category": "phonetic_balance",
+ "text": "Will we ever forget it.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0415.wav",
+ "render_seconds": 78.00979957601521,
+ "audio_seconds": 1.9504761904761905,
+ "rtf": 39.995258571687486,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WILL WE EVER FORGET IT",
+ "raw": {
+ "normalized_reference": "will we ever forget it.",
+ "normalized_hypothesis": "will we ever forget it",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "will we ever forget it",
+ "normalized_hypothesis": "will we ever forget it",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0416",
+ "category": "phonetic_balance",
+ "text": "Jeanne and Pierre both gazed toward the great rock.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0416.wav",
+ "render_seconds": 141.89520880999044,
+ "audio_seconds": 3.622312925170068,
+ "rtf": 39.17254299704889,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "JAN AND PIERRE BOTH GAZED TOWARD THE GREAT ROCK",
+ "raw": {
+ "normalized_reference": "jeanne and pierre both gazed toward the great rock.",
+ "normalized_hypothesis": "jan and pierre both gazed toward the great rock",
+ "errors": 1,
+ "words": 9,
+ "wer": 0.1111111111111111
+ },
+ "semantic": {
+ "normalized_reference": "jeanne and pierre both gazed toward the great rock",
+ "normalized_hypothesis": "jan and pierre both gazed toward the great rock",
+ "errors": 1,
+ "words": 9,
+ "wer": 0.1111111111111111
+ }
+ },
+ {
+ "id": "release500_0417",
+ "category": "phonetic_balance",
+ "text": "Does the old boy often go off at half-cock that way.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0417.wav",
+ "render_seconds": 118.60370925499592,
+ "audio_seconds": 3.4829931972789114,
+ "rtf": 34.052236836883594,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "DOES THE OLD BOY OFTEN GO OFF AT HALF COCK THAT WAY",
+ "raw": {
+ "normalized_reference": "does the old boy often go off at half-cock that way.",
+ "normalized_hypothesis": "does the old boy often go off at half cock that way",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "does the old boy often go off at half cock that way",
+ "normalized_hypothesis": "does the old boy often go off at half cock that way",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0418",
+ "category": "phonetic_balance",
+ "text": "The temperature dropped to fifty below zero and remained there the whole trip.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0418.wav",
+ "render_seconds": 172.6113193150377,
+ "audio_seconds": 5.363809523809524,
+ "rtf": 32.180732471731105,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE TEMPERATURE DROPPED TO FIFTY BELOW ZERO AND REMAINED THERE THE WHOLE TRIP",
+ "raw": {
+ "normalized_reference": "the temperature dropped to fifty below zero and remained there the whole trip.",
+ "normalized_hypothesis": "the temperature dropped to fifty below zero and remained there the whole trip",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the temperature dropped to 50 below 0 and remained there the whole trip",
+ "normalized_hypothesis": "the temperature dropped to 50 below 0 and remained there the whole trip",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0419",
+ "category": "phonetic_balance",
+ "text": "It's worth eight dollars.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0419.wav",
+ "render_seconds": 82.20326712797396,
+ "audio_seconds": 2.020136054421769,
+ "rtf": 40.69194594494939,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT'S WORTH EIGHT DOLLARS",
+ "raw": {
+ "normalized_reference": "it's worth eight dollars.",
+ "normalized_hypothesis": "it's worth eight dollars",
+ "errors": 0,
+ "words": 4,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it is worth $8",
+ "normalized_hypothesis": "it is worth $8",
+ "errors": 0,
+ "words": 4,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0420",
+ "category": "phonetic_balance",
+ "text": "But I did not enjoy it long.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0420.wav",
+ "render_seconds": 87.99971777200699,
+ "audio_seconds": 2.159455782312925,
+ "rtf": 40.75087737047955,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "BUT I DID NOT ENJOY IT LONG",
+ "raw": {
+ "normalized_reference": "but i did not enjoy it long.",
+ "normalized_hypothesis": "but i did not enjoy it long",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "but i did not enjoy it long",
+ "normalized_hypothesis": "but i did not enjoy it long",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0421",
+ "category": "phonetic_balance",
+ "text": "Shorty turned to their employers.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0421.wav",
+ "render_seconds": 108.18734872399364,
+ "audio_seconds": 2.4380952380952383,
+ "rtf": 44.37371725007551,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SHORTY TURNED TO THEIR EMPLOYERS",
+ "raw": {
+ "normalized_reference": "shorty turned to their employers.",
+ "normalized_hypothesis": "shorty turned to their employers",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "shorty turned to their employers",
+ "normalized_hypothesis": "shorty turned to their employers",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0422",
+ "category": "phonetic_balance",
+ "text": "It was the third or fourth time that Philip had heard MacDougall swear.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0422.wav",
+ "render_seconds": 119.79499838501215,
+ "audio_seconds": 5.015510204081632,
+ "rtf": 23.884907718365866,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT WAS THE THIRD OR FOURTH TIME THAT PHILIP HAD HEARD MACDUGAL SWEAR",
+ "raw": {
+ "normalized_reference": "it was the third or fourth time that philip had heard macdougall swear.",
+ "normalized_hypothesis": "it was the third or fourth time that philip had heard macdugal swear",
+ "errors": 1,
+ "words": 13,
+ "wer": 0.07692307692307693
+ },
+ "semantic": {
+ "normalized_reference": "it was the 3rd or 4th time that philip had heard macdougall swear",
+ "normalized_hypothesis": "it was the 3rd or 4th time that philip had heard macdugal swear",
+ "errors": 1,
+ "words": 13,
+ "wer": 0.07692307692307693
+ }
+ },
+ {
+ "id": "release500_0423",
+ "category": "phonetic_balance",
+ "text": "Ernest saw in the affair the most sinister import.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0423.wav",
+ "render_seconds": 127.59987312403973,
+ "audio_seconds": 3.6919727891156464,
+ "rtf": 34.56143379502108,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "ERNEST SAW IN THE AFFAIR THE MOST SINISTER IMPORT",
+ "raw": {
+ "normalized_reference": "ernest saw in the affair the most sinister import.",
+ "normalized_hypothesis": "ernest saw in the affair the most sinister import",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "ernest saw in the affair the most sinister import",
+ "normalized_hypothesis": "ernest saw in the affair the most sinister import",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0424",
+ "category": "phonetic_balance",
+ "text": "Also, at regular intervals, he would mutter.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0424.wav",
+ "render_seconds": 104.59484435198829,
+ "audio_seconds": 3.3436734693877552,
+ "rtf": 31.28141707304337,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "ALSO AT REGULAR INTERVALS HE WOULD MUTTER",
+ "raw": {
+ "normalized_reference": "also, at regular intervals, he would mutter.",
+ "normalized_hypothesis": "also at regular intervals he would mutter",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "also at regular intervals he would mutter",
+ "normalized_hypothesis": "also at regular intervals he would mutter",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0425",
+ "category": "phonetic_balance",
+ "text": "He may anticipate the day of his death.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0425.wav",
+ "render_seconds": 57.284214230952784,
+ "audio_seconds": 2.9257142857142857,
+ "rtf": 19.57956541097019,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE MAY ANTICIPATE THE DAY OF HIS DEATH",
+ "raw": {
+ "normalized_reference": "he may anticipate the day of his death.",
+ "normalized_hypothesis": "he may anticipate the day of his death",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he may anticipate the day of his death",
+ "normalized_hypothesis": "he may anticipate the day of his death",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0426",
+ "category": "phonetic_balance",
+ "text": "I defy any man to get a Solomon Island sore in California.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0426.wav",
+ "render_seconds": 124.20933483995032,
+ "audio_seconds": 4.31891156462585,
+ "rtf": 28.759406850857953,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I DEFY ANY MAN TO GET A SOLOMON ISLAND SORE IN CALIFORNIA",
+ "raw": {
+ "normalized_reference": "i defy any man to get a solomon island sore in california.",
+ "normalized_hypothesis": "i defy any man to get a solomon island sore in california",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i defy any man to get a solomon island sore in california",
+ "normalized_hypothesis": "i defy any man to get a solomon island sore in california",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0427",
+ "category": "phonetic_balance",
+ "text": "Now our figuring was all right.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0427.wav",
+ "render_seconds": 106.39812796097249,
+ "audio_seconds": 2.4380952380952383,
+ "rtf": 43.639857171492615,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "NOW OUR FIGURING WAS ALL RIGHT",
+ "raw": {
+ "normalized_reference": "now our figuring was all right.",
+ "normalized_hypothesis": "now our figuring was all right",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "now our figuring was all right",
+ "normalized_hypothesis": "now our figuring was all right",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0428",
+ "category": "phonetic_balance",
+ "text": "MacDougall tapped his forehead suspiciously with a stubby forefinger.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0428.wav",
+ "render_seconds": 152.4971202670131,
+ "audio_seconds": 5.085170068027211,
+ "rtf": 29.988597869289013,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "MACDUGALL TAPPED HIS FOREHEAD SUSPICIOUSLY WITH A STUBBY FOREFINGER",
+ "raw": {
+ "normalized_reference": "macdougall tapped his forehead suspiciously with a stubby forefinger.",
+ "normalized_hypothesis": "macdugall tapped his forehead suspiciously with a stubby forefinger",
+ "errors": 1,
+ "words": 9,
+ "wer": 0.1111111111111111
+ },
+ "semantic": {
+ "normalized_reference": "macdougall tapped his forehead suspiciously with a stubby forefinger",
+ "normalized_hypothesis": "macdugall tapped his forehead suspiciously with a stubby forefinger",
+ "errors": 1,
+ "words": 9,
+ "wer": 0.1111111111111111
+ }
+ },
+ {
+ "id": "release500_0429",
+ "category": "phonetic_balance",
+ "text": "O'Brien took off his coat and bared his right arm.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0429.wav",
+ "render_seconds": 135.5034232470207,
+ "audio_seconds": 3.4133333333333336,
+ "rtf": 39.69826852940059,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "O'BRIAN TOOK OFF HIS COAT AND BARED HIS RIGHT ARM",
+ "raw": {
+ "normalized_reference": "o'brien took off his coat and bared his right arm.",
+ "normalized_hypothesis": "o'brian took off his coat and bared his right arm",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ },
+ "semantic": {
+ "normalized_reference": "0 brien took off his coat and bared his right arm",
+ "normalized_hypothesis": "0 brian took off his coat and bared his right arm",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ }
+ },
+ {
+ "id": "release500_0430",
+ "category": "phonetic_balance",
+ "text": "But it won't continue, she said with easy confidence.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0430.wav",
+ "render_seconds": 121.50087084504776,
+ "audio_seconds": 3.5526530612244898,
+ "rtf": 34.20003832380136,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "BUT IT WON'T CONTINUE SHE SAID WITH EASY CONFIDENCE",
+ "raw": {
+ "normalized_reference": "but it won't continue, she said with easy confidence.",
+ "normalized_hypothesis": "but it won't continue she said with easy confidence",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "but it will not continue she said with easy confidence",
+ "normalized_hypothesis": "but it will not continue she said with easy confidence",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0431",
+ "category": "phonetic_balance",
+ "text": "I was still weak from my prolonged immersion.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0431.wav",
+ "render_seconds": 96.10618606605567,
+ "audio_seconds": 3.3436734693877552,
+ "rtf": 28.742694807353075,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I WAS STILL WEAK FROM MY PROLONGED IMMERSION",
+ "raw": {
+ "normalized_reference": "i was still weak from my prolonged immersion.",
+ "normalized_hypothesis": "i was still weak from my prolonged immersion",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i was still weak from my prolonged immersion",
+ "normalized_hypothesis": "i was still weak from my prolonged immersion",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0432",
+ "category": "phonetic_balance",
+ "text": "Oolong was two hundred and fifty miles from the nearest land.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0432.wav",
+ "render_seconds": 126.4020027249353,
+ "audio_seconds": 4.249251700680272,
+ "rtf": 29.746885246806944,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "BOLONG WAS TWO HUNDRED AND FIFTY MILES FROM THE NEAREST LAND",
+ "raw": {
+ "normalized_reference": "oolong was two hundred and fifty miles from the nearest land.",
+ "normalized_hypothesis": "bolong was two hundred and fifty miles from the nearest land",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ },
+ "semantic": {
+ "normalized_reference": "oolong was 250 miles from the nearest land",
+ "normalized_hypothesis": "bolong was 250 miles from the nearest land",
+ "errors": 1,
+ "words": 8,
+ "wer": 0.125
+ }
+ },
+ {
+ "id": "release500_0433",
+ "category": "phonetic_balance",
+ "text": "I just do appreciate it without being able to express my feelings.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0433.wav",
+ "render_seconds": 142.10778395505622,
+ "audio_seconds": 4.806530612244898,
+ "rtf": 29.565563068095088,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I JUST DO APPRECIATE IT WITHOUT BEING ABLE TO EXPRESS MY FEELINGS",
+ "raw": {
+ "normalized_reference": "i just do appreciate it without being able to express my feelings.",
+ "normalized_hypothesis": "i just do appreciate it without being able to express my feelings",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i just do appreciate it without being able to express my feelings",
+ "normalized_hypothesis": "i just do appreciate it without being able to express my feelings",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0434",
+ "category": "phonetic_balance",
+ "text": "Now just what do you want to know.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0434.wav",
+ "render_seconds": 76.20195149700157,
+ "audio_seconds": 2.36843537414966,
+ "rtf": 32.17396274718299,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "NOW JUST WHAT DO YOU WANT TO KNOW",
+ "raw": {
+ "normalized_reference": "now just what do you want to know.",
+ "normalized_hypothesis": "now just what do you want to know",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "now just what do you want to know",
+ "normalized_hypothesis": "now just what do you want to know",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0435",
+ "category": "phonetic_balance",
+ "text": "But she swung obediently on her heel into the wind.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0435.wav",
+ "render_seconds": 129.58877530391328,
+ "audio_seconds": 3.4133333333333336,
+ "rtf": 37.965461514818344,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "BUT SHE SWUNG OBEDIENTLY ON HER HEEL INTO THE WIND",
+ "raw": {
+ "normalized_reference": "but she swung obediently on her heel into the wind.",
+ "normalized_hypothesis": "but she swung obediently on her heel into the wind",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "but she swung obediently on her heel into the wind",
+ "normalized_hypothesis": "but she swung obediently on her heel into the wind",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0436",
+ "category": "phonetic_balance",
+ "text": "Ask him, I laughed, then turned to Pasquini.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0436.wav",
+ "render_seconds": 114.49201777507551,
+ "audio_seconds": 3.0650340136054424,
+ "rtf": 37.35424052942138,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "ASKIMAI LAUGHED THEN TURNED TO PASQUINI",
+ "raw": {
+ "normalized_reference": "ask him, i laughed, then turned to pasquini.",
+ "normalized_hypothesis": "askimai laughed then turned to pasquini",
+ "errors": 3,
+ "words": 8,
+ "wer": 0.375
+ },
+ "semantic": {
+ "normalized_reference": "ask him i laughed then turned to pasquini",
+ "normalized_hypothesis": "askimai laughed then turned to pasquini",
+ "errors": 3,
+ "words": 8,
+ "wer": 0.375
+ }
+ },
+ {
+ "id": "release500_0437",
+ "category": "phonetic_balance",
+ "text": "He did not rush in.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0437.wav",
+ "render_seconds": 82.10251486906782,
+ "audio_seconds": 1.7414965986394557,
+ "rtf": 47.144803459972536,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE DID NOT RUSH IN",
+ "raw": {
+ "normalized_reference": "he did not rush in.",
+ "normalized_hypothesis": "he did not rush in",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he did not rush in",
+ "normalized_hypothesis": "he did not rush in",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0438",
+ "category": "phonetic_balance",
+ "text": "They obeyed him, and went here and there at his commands.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0438.wav",
+ "render_seconds": 116.09897710604127,
+ "audio_seconds": 3.900952380952381,
+ "rtf": 29.76170067415609,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THEY OBEYED HIM AND WENT HERE AND THERE AT HIS COMMANDS",
+ "raw": {
+ "normalized_reference": "they obeyed him, and went here and there at his commands.",
+ "normalized_hypothesis": "they obeyed him and went here and there at his commands",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "they obeyed him and went here and there at his commands",
+ "normalized_hypothesis": "they obeyed him and went here and there at his commands",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0439",
+ "category": "phonetic_balance",
+ "text": "Tom Spink has a harpoon.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0439.wav",
+ "render_seconds": 90.49203233397566,
+ "audio_seconds": 2.159455782312925,
+ "rtf": 41.905017493367005,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "TOM SPINK HAS A HARPOON",
+ "raw": {
+ "normalized_reference": "tom spink has a harpoon.",
+ "normalized_hypothesis": "tom spink has a harpoon",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "tom spink has a harpoon",
+ "normalized_hypothesis": "tom spink has a harpoon",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0440",
+ "category": "phonetic_balance",
+ "text": "It was my reports from the north which chiefly induced people to buy.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0440.wav",
+ "render_seconds": 123.99391946999822,
+ "audio_seconds": 4.876190476190477,
+ "rtf": 25.428440516308225,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT WAS MY REPORTS FROM THE NORTH WHICH CHIEFLY INDUCED PEOPLE TO BUY",
+ "raw": {
+ "normalized_reference": "it was my reports from the north which chiefly induced people to buy.",
+ "normalized_hypothesis": "it was my reports from the north which chiefly induced people to buy",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it was my reports from the north which chiefly induced people to buy",
+ "normalized_hypothesis": "it was my reports from the north which chiefly induced people to buy",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0441",
+ "category": "phonetic_balance",
+ "text": "Cherokee identified himself with his instinct.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0441.wav",
+ "render_seconds": 116.09988292399794,
+ "audio_seconds": 3.2043537414965986,
+ "rtf": 36.2319182868285,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "CHEROKEY IDENTIFIED HIMSELF WITH HIS INSTINCT",
+ "raw": {
+ "normalized_reference": "cherokee identified himself with his instinct.",
+ "normalized_hypothesis": "cherokey identified himself with his instinct",
+ "errors": 1,
+ "words": 6,
+ "wer": 0.16666666666666666
+ },
+ "semantic": {
+ "normalized_reference": "cherokee identified himself with his instinct",
+ "normalized_hypothesis": "cherokey identified himself with his instinct",
+ "errors": 1,
+ "words": 6,
+ "wer": 0.16666666666666666
+ }
+ },
+ {
+ "id": "release500_0442",
+ "category": "phonetic_balance",
+ "text": "I graduated last of my class.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0442.wav",
+ "render_seconds": 89.70749325596262,
+ "audio_seconds": 2.4380952380952383,
+ "rtf": 36.79408903076592,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I GRADUATED LAST OF MY CLASS",
+ "raw": {
+ "normalized_reference": "i graduated last of my class.",
+ "normalized_hypothesis": "i graduated last of my class",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i graduated last of my class",
+ "normalized_hypothesis": "i graduated last of my class",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0443",
+ "category": "phonetic_balance",
+ "text": "It was a gigantic inadequacy.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0443.wav",
+ "render_seconds": 100.29291518195532,
+ "audio_seconds": 2.2987755102040817,
+ "rtf": 43.628842719423034,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT WAS A GIGANTIC INADEQUACY",
+ "raw": {
+ "normalized_reference": "it was a gigantic inadequacy.",
+ "normalized_hypothesis": "it was a gigantic inadequacy",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it was a gigantic inadequacy",
+ "normalized_hypothesis": "it was a gigantic inadequacy",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0444",
+ "category": "phonetic_balance",
+ "text": "Did I possess too much vitality.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0444.wav",
+ "render_seconds": 74.99495326098986,
+ "audio_seconds": 2.507755102040816,
+ "rtf": 29.90521410960696,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "DID I POSSESS TOO MUCH VITALITY",
+ "raw": {
+ "normalized_reference": "did i possess too much vitality.",
+ "normalized_hypothesis": "did i possess too much vitality",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "did i possess too much vitality",
+ "normalized_hypothesis": "did i possess too much vitality",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0445",
+ "category": "phonetic_balance",
+ "text": "Therefore, hurrah for the game.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0445.wav",
+ "render_seconds": 82.30921019101515,
+ "audio_seconds": 2.2291156462585033,
+ "rtf": 36.92460296044686,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THEREFORE HURRAH FOR THE GAME",
+ "raw": {
+ "normalized_reference": "therefore, hurrah for the game.",
+ "normalized_hypothesis": "therefore hurrah for the game",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "therefore hurrah for the game",
+ "normalized_hypothesis": "therefore hurrah for the game",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0446",
+ "category": "phonetic_balance",
+ "text": "He had fulfilled his duty and paid properly.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0446.wav",
+ "render_seconds": 115.70159818802495,
+ "audio_seconds": 3.1346938775510202,
+ "rtf": 36.910015046961085,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE HAD FULFILLED HIS DUTY AND PAID PROPERLY",
+ "raw": {
+ "normalized_reference": "he had fulfilled his duty and paid properly.",
+ "normalized_hypothesis": "he had fulfilled his duty and paid properly",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he had fulfilled his duty and paid properly",
+ "normalized_hypothesis": "he had fulfilled his duty and paid properly",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0447",
+ "category": "phonetic_balance",
+ "text": "How can you manage all alone, Mr Young.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0447.wav",
+ "render_seconds": 126.40151064202655,
+ "audio_seconds": 2.8560544217687074,
+ "rtf": 44.257388682302555,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HOW CAN YOU MANAGE ALL ALONE MISTER YOUNG",
+ "raw": {
+ "normalized_reference": "how can you manage all alone, mr young.",
+ "normalized_hypothesis": "how can you manage all alone mister young",
+ "errors": 1,
+ "words": 8,
+ "wer": 0.125
+ },
+ "semantic": {
+ "normalized_reference": "how can you manage all alone mister young",
+ "normalized_hypothesis": "how can you manage all alone mister young",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0448",
+ "category": "phonetic_balance",
+ "text": "Red-Eye never committed a more outrageous deed.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0448.wav",
+ "render_seconds": 104.1919863189105,
+ "audio_seconds": 3.5526530612244898,
+ "rtf": 29.327937325520534,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "RED EYE NEVER COMMITTED A MORE OUTRAGEOUS DEED",
+ "raw": {
+ "normalized_reference": "red-eye never committed a more outrageous deed.",
+ "normalized_hypothesis": "red eye never committed a more outrageous deed",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "red eye never committed a more outrageous deed",
+ "normalized_hypothesis": "red eye never committed a more outrageous deed",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0449",
+ "category": "phonetic_balance",
+ "text": "He would keep his faith with Gregson for the promised day or two.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0449.wav",
+ "render_seconds": 114.20582774409559,
+ "audio_seconds": 4.597551020408163,
+ "rtf": 24.840578655276474,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE WOULD KEEP HIS FAITH WITH GREGSON FOR THE PROMISED DAY OR TWO",
+ "raw": {
+ "normalized_reference": "he would keep his faith with gregson for the promised day or two.",
+ "normalized_hypothesis": "he would keep his faith with gregson for the promised day or two",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he would keep his faith with gregson for the promised day or 2",
+ "normalized_hypothesis": "he would keep his faith with gregson for the promised day or 2",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0450",
+ "category": "phonetic_balance",
+ "text": "There was one chance, and only one, of saving Jeanne.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0450.wav",
+ "render_seconds": 77.39515878504608,
+ "audio_seconds": 3.8312925170068026,
+ "rtf": 20.20079606072758,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THERE WAS ONE CHANCE AND ONLY ONE OF SAVING JAN",
+ "raw": {
+ "normalized_reference": "there was one chance, and only one, of saving jeanne.",
+ "normalized_hypothesis": "there was one chance and only one of saving jan",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ },
+ "semantic": {
+ "normalized_reference": "there was one chance and only one of saving jeanne",
+ "normalized_hypothesis": "there was one chance and only one of saving jan",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ }
+ },
+ {
+ "id": "release500_0451",
+ "category": "phonetic_balance",
+ "text": "I was beneath the water, suffocating and drowning.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0451.wav",
+ "render_seconds": 133.20235759101342,
+ "audio_seconds": 3.4829931972789114,
+ "rtf": 38.24364563648237,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I WAS BENEATH THE WATER SUFFOCATING AND DROWNING",
+ "raw": {
+ "normalized_reference": "i was beneath the water, suffocating and drowning.",
+ "normalized_hypothesis": "i was beneath the water suffocating and drowning",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i was beneath the water suffocating and drowning",
+ "normalized_hypothesis": "i was beneath the water suffocating and drowning",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0452",
+ "category": "phonetic_balance",
+ "text": "Give them their choice between a fine or an official whipping.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0452.wav",
+ "render_seconds": 118.09779246896505,
+ "audio_seconds": 4.388571428571429,
+ "rtf": 26.910304273527192,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "GIVE THEM THEIR CHOICE BETWEEN A FINE OR AN OFFICIAL WHIPPING",
+ "raw": {
+ "normalized_reference": "give them their choice between a fine or an official whipping.",
+ "normalized_hypothesis": "give them their choice between a fine or an official whipping",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "give them their choice between a fine or an official whipping",
+ "normalized_hypothesis": "give them their choice between a fine or an official whipping",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0453",
+ "category": "phonetic_balance",
+ "text": "They must have been swept away by the chaotic currents.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0453.wav",
+ "render_seconds": 134.59763296099845,
+ "audio_seconds": 3.900952380952381,
+ "rtf": 34.5037877463497,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THEY MUST HAVE BEEN SWEPT AWAY BY THE CHAOTIC CURRENTS",
+ "raw": {
+ "normalized_reference": "they must have been swept away by the chaotic currents.",
+ "normalized_hypothesis": "they must have been swept away by the chaotic currents",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "they must have been swept away by the chaotic currents",
+ "normalized_hypothesis": "they must have been swept away by the chaotic currents",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0454",
+ "category": "phonetic_balance",
+ "text": "You yellow giant thing of the frost.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0454.wav",
+ "render_seconds": 50.7071281389799,
+ "audio_seconds": 2.4380952380952383,
+ "rtf": 20.79784552575347,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "YOU YELLOW GIANT THING OF THE FROST",
+ "raw": {
+ "normalized_reference": "you yellow giant thing of the frost.",
+ "normalized_hypothesis": "you yellow giant thing of the frost",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "you yellow giant thing of the frost",
+ "normalized_hypothesis": "you yellow giant thing of the frost",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0455",
+ "category": "phonetic_balance",
+ "text": "He had peeled off his shirt and was wildly waving it.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0455.wav",
+ "render_seconds": 113.69861823902465,
+ "audio_seconds": 3.6919727891156464,
+ "rtf": 30.79616907638676,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE HAD PEELED OFF HIS SHIRT AND WAS WILDLY WAVING IT",
+ "raw": {
+ "normalized_reference": "he had peeled off his shirt and was wildly waving it.",
+ "normalized_hypothesis": "he had peeled off his shirt and was wildly waving it",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he had peeled off his shirt and was wildly waving it",
+ "normalized_hypothesis": "he had peeled off his shirt and was wildly waving it",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0456",
+ "category": "phonetic_balance",
+ "text": "To these he gave castor oil.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0456.wav",
+ "render_seconds": 45.00009410397615,
+ "audio_seconds": 2.2291156462585033,
+ "rtf": 20.18742014552153,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "TO THESE HE GAVE CASTER OIL",
+ "raw": {
+ "normalized_reference": "to these he gave castor oil.",
+ "normalized_hypothesis": "to these he gave caster oil",
+ "errors": 1,
+ "words": 6,
+ "wer": 0.16666666666666666
+ },
+ "semantic": {
+ "normalized_reference": "to these he gave castor oil",
+ "normalized_hypothesis": "to these he gave caster oil",
+ "errors": 1,
+ "words": 6,
+ "wer": 0.16666666666666666
+ }
+ },
+ {
+ "id": "release500_0457",
+ "category": "phonetic_balance",
+ "text": "Tudor surveyed him with withering disgust.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0457.wav",
+ "render_seconds": 115.29798547306564,
+ "audio_seconds": 2.9257142857142857,
+ "rtf": 39.40849112848923,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "TUTOR SURVEYED HIM WITH WITHERING DISGUST",
+ "raw": {
+ "normalized_reference": "tudor surveyed him with withering disgust.",
+ "normalized_hypothesis": "tutor surveyed him with withering disgust",
+ "errors": 1,
+ "words": 6,
+ "wer": 0.16666666666666666
+ },
+ "semantic": {
+ "normalized_reference": "tudor surveyed him with withering disgust",
+ "normalized_hypothesis": "tutor surveyed him with withering disgust",
+ "errors": 1,
+ "words": 6,
+ "wer": 0.16666666666666666
+ }
+ },
+ {
+ "id": "release500_0458",
+ "category": "phonetic_balance",
+ "text": "After all, it was simply a mistake in judgment.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0458.wav",
+ "render_seconds": 82.90516074304469,
+ "audio_seconds": 3.4829931972789114,
+ "rtf": 23.802848885210096,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "AFTER ALL IT WAS SIMPLY A MISTAKEN JUDGMENT",
+ "raw": {
+ "normalized_reference": "after all, it was simply a mistake in judgment.",
+ "normalized_hypothesis": "after all it was simply a mistaken judgment",
+ "errors": 2,
+ "words": 9,
+ "wer": 0.2222222222222222
+ },
+ "semantic": {
+ "normalized_reference": "after all it was simply a mistake in judgment",
+ "normalized_hypothesis": "after all it was simply a mistaken judgment",
+ "errors": 2,
+ "words": 9,
+ "wer": 0.2222222222222222
+ }
+ },
+ {
+ "id": "release500_0459",
+ "category": "phonetic_balance",
+ "text": "That is the strange part of it.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0459.wav",
+ "render_seconds": 114.89998027298134,
+ "audio_seconds": 2.2987755102040817,
+ "rtf": 49.9831235207394,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THAT IS THE STRANGE PART OF IT",
+ "raw": {
+ "normalized_reference": "that is the strange part of it.",
+ "normalized_hypothesis": "that is the strange part of it",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "that is the strange part of it",
+ "normalized_hypothesis": "that is the strange part of it",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0460",
+ "category": "phonetic_balance",
+ "text": "Not a wheel moved in his empire.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0460.wav",
+ "render_seconds": 62.79256058996543,
+ "audio_seconds": 2.507755102040816,
+ "rtf": 25.039351146714733,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "NOT A WHEEL MOVED IN HIS EMPIRE",
+ "raw": {
+ "normalized_reference": "not a wheel moved in his empire.",
+ "normalized_hypothesis": "not a wheel moved in his empire",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "not a wheel moved in his empire",
+ "normalized_hypothesis": "not a wheel moved in his empire",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0461",
+ "category": "phonetic_balance",
+ "text": "Its diameter was not more than two hundred yards.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0461.wav",
+ "render_seconds": 92.69938166497741,
+ "audio_seconds": 3.4133333333333336,
+ "rtf": 27.15802197216135,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "ITS DIAMETER WAS NOT MORE THAN TWO HUNDRED YARDS",
+ "raw": {
+ "normalized_reference": "its diameter was not more than two hundred yards.",
+ "normalized_hypothesis": "its diameter was not more than two hundred yards",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "its diameter was not more than 200 yards",
+ "normalized_hypothesis": "its diameter was not more than 200 yards",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0462",
+ "category": "phonetic_balance",
+ "text": "Nimrod replied, with a slight manifestation of sensitiveness.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0462.wav",
+ "render_seconds": 82.49773488100618,
+ "audio_seconds": 4.667210884353741,
+ "rtf": 17.676024701941333,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "NIMROD REPLIED WITH A SLIGHT MANIFESTATION OF SENSITIVENESS",
+ "raw": {
+ "normalized_reference": "nimrod replied, with a slight manifestation of sensitiveness.",
+ "normalized_hypothesis": "nimrod replied with a slight manifestation of sensitiveness",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "nimrod replied with a slight manifestation of sensitiveness",
+ "normalized_hypothesis": "nimrod replied with a slight manifestation of sensitiveness",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0463",
+ "category": "phonetic_balance",
+ "text": "I tried to read George Moore last night, and was dreadfully bored.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0463.wav",
+ "render_seconds": 156.9069790840149,
+ "audio_seconds": 4.667210884353741,
+ "rtf": 33.61900350593253,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I TRIED TO READ GEORGE MOORE LAST NIGHT AND WAS DREADFULLY BORED",
+ "raw": {
+ "normalized_reference": "i tried to read george moore last night, and was dreadfully bored.",
+ "normalized_hypothesis": "i tried to read george moore last night and was dreadfully bored",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i tried to read george moore last night and was dreadfully bored",
+ "normalized_hypothesis": "i tried to read george moore last night and was dreadfully bored",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0464",
+ "category": "phonetic_balance",
+ "text": "Yes, it was a man who asked, a stranger.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0464.wav",
+ "render_seconds": 88.30820780899376,
+ "audio_seconds": 2.995374149659864,
+ "rtf": 29.48152831559339,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "YES IT WAS A MAN WHO ASKED A STRANGER",
+ "raw": {
+ "normalized_reference": "yes, it was a man who asked, a stranger.",
+ "normalized_hypothesis": "yes it was a man who asked a stranger",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "yes it was a man who asked a stranger",
+ "normalized_hypothesis": "yes it was a man who asked a stranger",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0465",
+ "category": "phonetic_balance",
+ "text": "We were now good friends.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0465.wav",
+ "render_seconds": 81.29997687402647,
+ "audio_seconds": 2.020136054421769,
+ "rtf": 40.24480266864861,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WE WERE NOW GOOD FRIENDS",
+ "raw": {
+ "normalized_reference": "we were now good friends.",
+ "normalized_hypothesis": "we were now good friends",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "we were now good friends",
+ "normalized_hypothesis": "we were now good friends",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0466",
+ "category": "phonetic_balance",
+ "text": "Yet, in accordance with Ernest's test of truth, it worked.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0466.wav",
+ "render_seconds": 67.91129141498823,
+ "audio_seconds": 3.900952380952381,
+ "rtf": 17.408900387142978,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "YET IN ACCORDANCE WITH EARNEST'S TEST OF TRUTH IT WORKED",
+ "raw": {
+ "normalized_reference": "yet, in accordance with ernest's test of truth, it worked.",
+ "normalized_hypothesis": "yet in accordance with earnest's test of truth it worked",
+ "errors": 1,
+ "words": 10,
+ "wer": 0.1
+ },
+ "semantic": {
+ "normalized_reference": "yet in accordance with ernest is test of truth it worked",
+ "normalized_hypothesis": "yet in accordance with earnest is test of truth it worked",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ }
+ },
+ {
+ "id": "release500_0467",
+ "category": "phonetic_balance",
+ "text": "It's the strap hangers that'll keep us from going under.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0467.wav",
+ "render_seconds": 99.0920732569648,
+ "audio_seconds": 3.4829931972789114,
+ "rtf": 28.450263220261377,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "IT'S THE STRAP HANGERS THAT'LL KEEP US FROM GOING UNDER",
+ "raw": {
+ "normalized_reference": "it's the strap hangers that'll keep us from going under.",
+ "normalized_hypothesis": "it's the strap hangers that'll keep us from going under",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "it is the strap hangers that will keep us from going under",
+ "normalized_hypothesis": "it is the strap hangers that will keep us from going under",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0468",
+ "category": "phonetic_balance",
+ "text": "I never allow what can't be changed to annoy me.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0468.wav",
+ "render_seconds": 64.5002737219911,
+ "audio_seconds": 3.0650340136054424,
+ "rtf": 21.043901449602032,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I NEVER ALLOW WHAT CAN'T BE CHANGED TO ANNOY ME",
+ "raw": {
+ "normalized_reference": "i never allow what can't be changed to annoy me.",
+ "normalized_hypothesis": "i never allow what can't be changed to annoy me",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i never allow what can not be changed to annoy me",
+ "normalized_hypothesis": "i never allow what can not be changed to annoy me",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0469",
+ "category": "phonetic_balance",
+ "text": "After all superfluous flesh is gone what is left is stringy and resistant.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0469.wav",
+ "render_seconds": 163.29407874599565,
+ "audio_seconds": 5.224489795918367,
+ "rtf": 31.255507259975733,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "AFTER ALL SUPERFLUOUS FLESH IS GONE WHAT IS LEFT IS STRINGY AND RESISTANT",
+ "raw": {
+ "normalized_reference": "after all superfluous flesh is gone what is left is stringy and resistant.",
+ "normalized_hypothesis": "after all superfluous flesh is gone what is left is stringy and resistant",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "after all superfluous flesh is gone what is left is stringy and resistant",
+ "normalized_hypothesis": "after all superfluous flesh is gone what is left is stringy and resistant",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0470",
+ "category": "phonetic_balance",
+ "text": "Then he shouted, Shut up.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0470.wav",
+ "render_seconds": 45.892153688007966,
+ "audio_seconds": 2.089795918367347,
+ "rtf": 21.960112604613183,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THEN HE SHOUTED SHUT UP",
+ "raw": {
+ "normalized_reference": "then he shouted, shut up.",
+ "normalized_hypothesis": "then he shouted shut up",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "then he shouted shut up",
+ "normalized_hypothesis": "then he shouted shut up",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0471",
+ "category": "phonetic_balance",
+ "text": "He had worshiped her, as Dante might have worshiped Beatrice.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0471.wav",
+ "render_seconds": 133.90137863601558,
+ "audio_seconds": 4.458231292517007,
+ "rtf": 30.034641509238114,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE HAD WORSHIPED HER AS DANTON MIGHT HAVE WORSHIPPED BEATRICE",
+ "raw": {
+ "normalized_reference": "he had worshiped her, as dante might have worshiped beatrice.",
+ "normalized_hypothesis": "he had worshiped her as danton might have worshipped beatrice",
+ "errors": 2,
+ "words": 10,
+ "wer": 0.2
+ },
+ "semantic": {
+ "normalized_reference": "he had worshiped her as dante might have worshiped beatrice",
+ "normalized_hypothesis": "he had worshiped her as danton might have worshipped beatrice",
+ "errors": 2,
+ "words": 10,
+ "wer": 0.2
+ }
+ },
+ {
+ "id": "release500_0472",
+ "category": "phonetic_balance",
+ "text": "He looked like one who had passed through an uncomfortable hour or two.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0472.wav",
+ "render_seconds": 83.80051923799329,
+ "audio_seconds": 4.876190476190477,
+ "rtf": 17.18565335935409,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE LOOKED LIKE ONE WHO HAD PASSED THROUGH AN UNCOMFORTABLE HOUR OR TWO",
+ "raw": {
+ "normalized_reference": "he looked like one who had passed through an uncomfortable hour or two.",
+ "normalized_hypothesis": "he looked like one who had passed through an uncomfortable hour or two",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he looked like one who had passed through an uncomfortable hour or 2",
+ "normalized_hypothesis": "he looked like one who had passed through an uncomfortable hour or 2",
+ "errors": 0,
+ "words": 13,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0473",
+ "category": "phonetic_balance",
+ "text": "The history of our westward-faring race is written in it.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0473.wav",
+ "render_seconds": 139.7089644210646,
+ "audio_seconds": 4.179591836734694,
+ "rtf": 33.42646121402424,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THE HISTORY OF OUR WESTWARD FARING RACE IS WRITTEN IN IT",
+ "raw": {
+ "normalized_reference": "the history of our westward-faring race is written in it.",
+ "normalized_hypothesis": "the history of our westward faring race is written in it",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "the history of our westward faring race is written in it",
+ "normalized_hypothesis": "the history of our westward faring race is written in it",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0474",
+ "category": "phonetic_balance",
+ "text": "He may desire to escape pain, or to enjoy pleasure.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0474.wav",
+ "render_seconds": 76.1070360239828,
+ "audio_seconds": 3.8312925170068026,
+ "rtf": 19.86458504177108,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE MAY DESIRE TO ESCAPE PAIN OR TO ENJOY PLEASURE",
+ "raw": {
+ "normalized_reference": "he may desire to escape pain, or to enjoy pleasure.",
+ "normalized_hypothesis": "he may desire to escape pain or to enjoy pleasure",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he may desire to escape pain or to enjoy pleasure",
+ "normalized_hypothesis": "he may desire to escape pain or to enjoy pleasure",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0475",
+ "category": "phonetic_balance",
+ "text": "And the air was growing chilly.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0475.wav",
+ "render_seconds": 101.69825918204151,
+ "audio_seconds": 2.2987755102040817,
+ "rtf": 44.24018732173326,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "AND THE AIR WAS GROWING CHILLY",
+ "raw": {
+ "normalized_reference": "and the air was growing chilly.",
+ "normalized_hypothesis": "and the air was growing chilly",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "and the air was growing chilly",
+ "normalized_hypothesis": "and the air was growing chilly",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0476",
+ "category": "phonetic_balance",
+ "text": "Dig in; you're sure good, was Daylight's answer.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0476.wav",
+ "render_seconds": 117.79247821995523,
+ "audio_seconds": 2.9257142857142857,
+ "rtf": 40.26110095408626,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "DIG IN YOU'RE SURE GOOD WAS DAYLIGHT'S ANSWER",
+ "raw": {
+ "normalized_reference": "dig in; you're sure good, was daylight's answer.",
+ "normalized_hypothesis": "dig in you're sure good was daylight's answer",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "dig in you are sure good was daylight is answer",
+ "normalized_hypothesis": "dig in you are sure good was daylight is answer",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0477",
+ "category": "phonetic_balance",
+ "text": "My idea was that he would have more influence over the natives.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0477.wav",
+ "render_seconds": 115.8983665339183,
+ "audio_seconds": 4.458231292517007,
+ "rtf": 25.99649029615172,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "MY IDEA WAS THAT HE WOULD HAVE MORE INFLUENCE OVER THE NATIVES",
+ "raw": {
+ "normalized_reference": "my idea was that he would have more influence over the natives.",
+ "normalized_hypothesis": "my idea was that he would have more influence over the natives",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "my idea was that he would have more influence over the natives",
+ "normalized_hypothesis": "my idea was that he would have more influence over the natives",
+ "errors": 0,
+ "words": 12,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0478",
+ "category": "phonetic_balance",
+ "text": "Now run along, and tell them to hurry.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0478.wav",
+ "render_seconds": 58.08695649402216,
+ "audio_seconds": 2.786394557823129,
+ "rtf": 20.846637218313617,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "NOW RUN ALONG AND TELL THEM TO HURRY",
+ "raw": {
+ "normalized_reference": "now run along, and tell them to hurry.",
+ "normalized_hypothesis": "now run along and tell them to hurry",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "now run along and tell them to hurry",
+ "normalized_hypothesis": "now run along and tell them to hurry",
+ "errors": 0,
+ "words": 8,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0479",
+ "category": "phonetic_balance",
+ "text": "I have no idea, replied Philip.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0479.wav",
+ "render_seconds": 107.19098688301165,
+ "audio_seconds": 2.36843537414966,
+ "rtf": 45.25814301413922,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I HAVE NO IDEA REPLIED PHILIP",
+ "raw": {
+ "normalized_reference": "i have no idea, replied philip.",
+ "normalized_hypothesis": "i have no idea replied philip",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i have no idea replied philip",
+ "normalized_hypothesis": "i have no idea replied philip",
+ "errors": 0,
+ "words": 6,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0480",
+ "category": "phonetic_balance",
+ "text": "For the rest, he was a mere automaton.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0480.wav",
+ "render_seconds": 47.993807561928406,
+ "audio_seconds": 2.786394557823129,
+ "rtf": 17.224340116219423,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "FOR THE REST HE WAS A MERE AUTOMATIN",
+ "raw": {
+ "normalized_reference": "for the rest, he was a mere automaton.",
+ "normalized_hypothesis": "for the rest he was a mere automatin",
+ "errors": 1,
+ "words": 8,
+ "wer": 0.125
+ },
+ "semantic": {
+ "normalized_reference": "for the rest he was a mere automaton",
+ "normalized_hypothesis": "for the rest he was a mere automatin",
+ "errors": 1,
+ "words": 8,
+ "wer": 0.125
+ }
+ },
+ {
+ "id": "release500_0481",
+ "category": "phonetic_balance",
+ "text": "Why, the average review is more nauseating than cod liver oil.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0481.wav",
+ "render_seconds": 139.5997976079816,
+ "audio_seconds": 4.388571428571429,
+ "rtf": 31.80984971536039,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "WHY THE AVERAGE REVIEW IS MORE NAUSEATING THAN COD LIVER OIL",
+ "raw": {
+ "normalized_reference": "why, the average review is more nauseating than cod liver oil.",
+ "normalized_hypothesis": "why the average review is more nauseating than cod liver oil",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "why the average review is more nauseating than cod liver oil",
+ "normalized_hypothesis": "why the average review is more nauseating than cod liver oil",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0482",
+ "category": "phonetic_balance",
+ "text": "I have been robbed, sir, I amended.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0482.wav",
+ "render_seconds": 68.40071363805328,
+ "audio_seconds": 2.716734693877551,
+ "rtf": 25.177546336122372,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I HAVE BEEN ROBBED SIR I AMENDED",
+ "raw": {
+ "normalized_reference": "i have been robbed, sir, i amended.",
+ "normalized_hypothesis": "i have been robbed sir i amended",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "i have been robbed sir i amended",
+ "normalized_hypothesis": "i have been robbed sir i amended",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0483",
+ "category": "phonetic_balance",
+ "text": "He moved away as quietly as he had come.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0483.wav",
+ "render_seconds": 117.00307293399237,
+ "audio_seconds": 2.9257142857142857,
+ "rtf": 39.991284694235674,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "HE MOVED AWAY AS QUIETLY AS HE HAD COME",
+ "raw": {
+ "normalized_reference": "he moved away as quietly as he had come.",
+ "normalized_hypothesis": "he moved away as quietly as he had come",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "he moved away as quietly as he had come",
+ "normalized_hypothesis": "he moved away as quietly as he had come",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0484",
+ "category": "phonetic_balance",
+ "text": "A trickle of fresh blood ran over his face.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0484.wav",
+ "render_seconds": 58.30075218703132,
+ "audio_seconds": 3.2043537414965986,
+ "rtf": 18.194231002661354,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "A TRICKLE OF FRESH BLOOD RAN OVER HIS FACE",
+ "raw": {
+ "normalized_reference": "a trickle of fresh blood ran over his face.",
+ "normalized_hypothesis": "a trickle of fresh blood ran over his face",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "a trickle of fresh blood ran over his face",
+ "normalized_hypothesis": "a trickle of fresh blood ran over his face",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0485",
+ "category": "phonetic_balance",
+ "text": "Suppose you saw me at work through the window.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0485.wav",
+ "render_seconds": 105.9996848970186,
+ "audio_seconds": 3.0650340136054424,
+ "rtf": 34.58352645565903,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "SUPPOSE YOU SAW ME AT WORK THROUGH THE WINDOW",
+ "raw": {
+ "normalized_reference": "suppose you saw me at work through the window.",
+ "normalized_hypothesis": "suppose you saw me at work through the window",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "suppose you saw me at work through the window",
+ "normalized_hypothesis": "suppose you saw me at work through the window",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0486",
+ "category": "phonetic_balance",
+ "text": "But who was Eileen's double.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0486.wav",
+ "render_seconds": 48.30594563600607,
+ "audio_seconds": 2.089795918367347,
+ "rtf": 23.115149767229465,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "BUT WHO WAS AELEEN'S DOUBLE",
+ "raw": {
+ "normalized_reference": "but who was eileen's double.",
+ "normalized_hypothesis": "but who was aeleen's double",
+ "errors": 1,
+ "words": 5,
+ "wer": 0.2
+ },
+ "semantic": {
+ "normalized_reference": "but who was eileen is double",
+ "normalized_hypothesis": "but who was aeleen is double",
+ "errors": 1,
+ "words": 6,
+ "wer": 0.16666666666666666
+ }
+ },
+ {
+ "id": "release500_0487",
+ "category": "phonetic_balance",
+ "text": "Darkness hid him from Jeanne.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0487.wav",
+ "render_seconds": 79.11909808102064,
+ "audio_seconds": 2.2291156462585033,
+ "rtf": 35.49349187594615,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "DARKNESS HID HIM FROM JAN",
+ "raw": {
+ "normalized_reference": "darkness hid him from jeanne.",
+ "normalized_hypothesis": "darkness hid him from jan",
+ "errors": 1,
+ "words": 5,
+ "wer": 0.2
+ },
+ "semantic": {
+ "normalized_reference": "darkness hid him from jeanne",
+ "normalized_hypothesis": "darkness hid him from jan",
+ "errors": 1,
+ "words": 5,
+ "wer": 0.2
+ }
+ },
+ {
+ "id": "release500_0488",
+ "category": "phonetic_balance",
+ "text": "By this time Charley was as enraged as the Greek.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0488.wav",
+ "render_seconds": 82.11378037801478,
+ "audio_seconds": 3.5526530612244898,
+ "rtf": 23.113368787469692,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "BY THIS TIME CHARLEY WAS AS ENRAGED AS THE GREEK",
+ "raw": {
+ "normalized_reference": "by this time charley was as enraged as the greek.",
+ "normalized_hypothesis": "by this time charley was as enraged as the greek",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "by this time charley was as enraged as the greek",
+ "normalized_hypothesis": "by this time charley was as enraged as the greek",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0489",
+ "category": "phonetic_balance",
+ "text": "I am writing these lines in Honolulu, Hawaii.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0489.wav",
+ "render_seconds": 115.39810844301246,
+ "audio_seconds": 3.274013605442177,
+ "rtf": 35.24667956516546,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "I AM WRITING THESE LINES IN HONOLOLU HAWA'I",
+ "raw": {
+ "normalized_reference": "i am writing these lines in honolulu, hawaii.",
+ "normalized_hypothesis": "i am writing these lines in honololu hawa'i",
+ "errors": 2,
+ "words": 8,
+ "wer": 0.25
+ },
+ "semantic": {
+ "normalized_reference": "i am writing these lines in honolulu hawaii",
+ "normalized_hypothesis": "i am writing these lines in honololu hawa i",
+ "errors": 3,
+ "words": 8,
+ "wer": 0.375
+ }
+ },
+ {
+ "id": "release500_0490",
+ "category": "phonetic_balance",
+ "text": "Nobody knows how the natives got them.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0490.wav",
+ "render_seconds": 48.70029347902164,
+ "audio_seconds": 2.716734693877551,
+ "rtf": 17.92603951676728,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "NOBODY KNOWS HOW THE NATIVES GOT THEM",
+ "raw": {
+ "normalized_reference": "nobody knows how the natives got them.",
+ "normalized_hypothesis": "nobody knows how the natives got them",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "nobody knows how the natives got them",
+ "normalized_hypothesis": "nobody knows how the natives got them",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0491",
+ "category": "phonetic_balance",
+ "text": "And how would we ever find ourselves.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0491.wav",
+ "render_seconds": 99.00526309898123,
+ "audio_seconds": 2.716734693877551,
+ "rtf": 36.442742576998796,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "AND HOW WOULD WE EVER FIND OURSELVES",
+ "raw": {
+ "normalized_reference": "and how would we ever find ourselves.",
+ "normalized_hypothesis": "and how would we ever find ourselves",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "and how would we ever find ourselves",
+ "normalized_hypothesis": "and how would we ever find ourselves",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0492",
+ "category": "phonetic_balance",
+ "text": "They are coming ashore, whoever they are.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0492.wav",
+ "render_seconds": 43.11082718800753,
+ "audio_seconds": 2.8560544217687074,
+ "rtf": 15.094539816685185,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THEY ARE COMING ASHORE WHOEVER THEY ARE",
+ "raw": {
+ "normalized_reference": "they are coming ashore, whoever they are.",
+ "normalized_hypothesis": "they are coming ashore whoever they are",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "they are coming ashore whoever they are",
+ "normalized_hypothesis": "they are coming ashore whoever they are",
+ "errors": 0,
+ "words": 7,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0493",
+ "category": "phonetic_balance",
+ "text": "This is my fifth voyage.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0493.wav",
+ "render_seconds": 71.49168569408357,
+ "audio_seconds": 2.159455782312925,
+ "rtf": 33.106343866652864,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THIS IS MY FIFTH VOYAGE",
+ "raw": {
+ "normalized_reference": "this is my fifth voyage.",
+ "normalized_hypothesis": "this is my fifth voyage",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "this is my 5th voyage",
+ "normalized_hypothesis": "this is my 5th voyage",
+ "errors": 0,
+ "words": 5,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0494",
+ "category": "phonetic_balance",
+ "text": "And yet if she came he had no words to say.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0494.wav",
+ "render_seconds": 79.69409804290626,
+ "audio_seconds": 2.9257142857142857,
+ "rtf": 27.239193667008976,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "AND YET IF SHE CAME HE HAD NO WORDS TO SAY",
+ "raw": {
+ "normalized_reference": "and yet if she came he had no words to say.",
+ "normalized_hypothesis": "and yet if she came he had no words to say",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "and yet if she came he had no words to say",
+ "normalized_hypothesis": "and yet if she came he had no words to say",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0495",
+ "category": "phonetic_balance",
+ "text": "You must sleep, he urged.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0495.wav",
+ "render_seconds": 63.923034836072475,
+ "audio_seconds": 2.020136054421769,
+ "rtf": 31.64293548256551,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "YOU MUST SLEEP HERE",
+ "raw": {
+ "normalized_reference": "you must sleep, he urged.",
+ "normalized_hypothesis": "you must sleep here",
+ "errors": 2,
+ "words": 5,
+ "wer": 0.4
+ },
+ "semantic": {
+ "normalized_reference": "you must sleep he urged",
+ "normalized_hypothesis": "you must sleep here",
+ "errors": 2,
+ "words": 5,
+ "wer": 0.4
+ }
+ },
+ {
+ "id": "release500_0496",
+ "category": "phonetic_balance",
+ "text": "Could the incident have anything to do with Jeanne and Pierre.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0496.wav",
+ "render_seconds": 64.49907698098104,
+ "audio_seconds": 4.179591836734694,
+ "rtf": 15.431908066738627,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "COULD THE INCIDENT HAVE ANYTHING TO DO WITH JAN AND PIERRE",
+ "raw": {
+ "normalized_reference": "could the incident have anything to do with jeanne and pierre.",
+ "normalized_hypothesis": "could the incident have anything to do with jan and pierre",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ },
+ "semantic": {
+ "normalized_reference": "could the incident have anything to do with jeanne and pierre",
+ "normalized_hypothesis": "could the incident have anything to do with jan and pierre",
+ "errors": 1,
+ "words": 11,
+ "wer": 0.09090909090909091
+ }
+ },
+ {
+ "id": "release500_0497",
+ "category": "phonetic_balance",
+ "text": "Each day she became a more vital part of him.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0497.wav",
+ "render_seconds": 77.30156217201147,
+ "audio_seconds": 3.274013605442177,
+ "rtf": 23.610641703967932,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "EACH DAY SHE BECAME A MORE VITAL PART OF HIM",
+ "raw": {
+ "normalized_reference": "each day she became a more vital part of him.",
+ "normalized_hypothesis": "each day she became a more vital part of him",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "each day she became a more vital part of him",
+ "normalized_hypothesis": "each day she became a more vital part of him",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0498",
+ "category": "phonetic_balance",
+ "text": "But why continue the tirade, for tirade it was.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0498.wav",
+ "render_seconds": 29.597576294094324,
+ "audio_seconds": 3.3436734693877552,
+ "rtf": 8.85181419928358,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "BUT WHY CONTINUE THE TIRADE FOR TIRADE IT WAS",
+ "raw": {
+ "normalized_reference": "but why continue the tirade, for tirade it was.",
+ "normalized_hypothesis": "but why continue the tirade for tirade it was",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "but why continue the tirade for tirade it was",
+ "normalized_hypothesis": "but why continue the tirade for tirade it was",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0499",
+ "category": "phonetic_balance",
+ "text": "That is what distinguishes all of us from the lower animals.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0499.wav",
+ "render_seconds": 111.79130288900342,
+ "audio_seconds": 4.31891156462585,
+ "rtf": 25.884137986207637,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THAT IS WHAT DISTINGUISHES ALL OF US FROM THE LOWER ANIMALS",
+ "raw": {
+ "normalized_reference": "that is what distinguishes all of us from the lower animals.",
+ "normalized_hypothesis": "that is what distinguishes all of us from the lower animals",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "that is what distinguishes all of us from the lower animals",
+ "normalized_hypothesis": "that is what distinguishes all of us from the lower animals",
+ "errors": 0,
+ "words": 11,
+ "wer": 0.0
+ }
+ },
+ {
+ "id": "release500_0500",
+ "category": "phonetic_balance",
+ "text": "There was something pathetic in the girl's attitude now.",
+ "model": "supertonic3-james-8step",
+ "audio": "benchmark/audio/supertonic3-james-8step/release500_0500.wav",
+ "render_seconds": 109.39057729102205,
+ "audio_seconds": 3.622312925170068,
+ "rtf": 30.19909642011013,
+ "voice": "M2",
+ "steps": 8,
+ "provider": "cpu",
+ "asr": "wav2vec2-large-lv60k-960h",
+ "hypothesis": "THERE WAS SOMETHING PATHETIC IN THE GIRL'S ATTITUDE NOW",
+ "raw": {
+ "normalized_reference": "there was something pathetic in the girl's attitude now.",
+ "normalized_hypothesis": "there was something pathetic in the girl's attitude now",
+ "errors": 0,
+ "words": 9,
+ "wer": 0.0
+ },
+ "semantic": {
+ "normalized_reference": "there was something pathetic in the girl is attitude now",
+ "normalized_hypothesis": "there was something pathetic in the girl is attitude now",
+ "errors": 0,
+ "words": 10,
+ "wer": 0.0
+ }
+ }
+ ]
+}
diff --git a/evaluation/final/raw/combined_manifest_validation.json b/evaluation/final/raw/combined_manifest_validation.json
new file mode 100644
index 0000000000000000000000000000000000000000..35b0fb48b3cdf91db76b6c5b2b02f4f3725c9d93
--- /dev/null
+++ b/evaluation/final/raw/combined_manifest_validation.json
@@ -0,0 +1,33 @@
+{
+ "format": "inflect_tts_matched_manifest_validation_v1",
+ "systems": {
+ "inflect-micro-v2": 500,
+ "inflect-nano-v2": 500,
+ "kitten-nano-bruno": 500,
+ "kitten-nano-hugo": 500,
+ "piper-danny-low": 500,
+ "piper-ryan-low": 500,
+ "supertonic3-james-3step": 500,
+ "supertonic3-james-8step": 500
+ },
+ "total_rows": 4000,
+ "unique_prompts": 500,
+ "categories": {
+ "conversational": 25,
+ "descriptive": 16,
+ "emotional": 24,
+ "homographs": 6,
+ "long": 15,
+ "long_range": 6,
+ "names_places": 6,
+ "narrative": 24,
+ "numbers": 6,
+ "phonetic_balance": 300,
+ "punctuation": 6,
+ "question": 15,
+ "rare_words": 6,
+ "short": 24,
+ "technical": 21
+ },
+ "valid": true
+}
diff --git a/evaluation/final/raw/prompt_report.json b/evaluation/final/raw/prompt_report.json
new file mode 100644
index 0000000000000000000000000000000000000000..ca25fc57ddae300fa55b66c30770266ec6e6856d
--- /dev/null
+++ b/evaluation/final/raw/prompt_report.json
@@ -0,0 +1,37 @@
+{
+ "format": "inflect_v2_final_benchmark_prompts_v1",
+ "count": 500,
+ "sha256": "5b7aa33852e6c738eb0e1dbf9fcfcc90772b97ad219167e44aaf3917d063b056",
+ "base_prompt_count": 200,
+ "cmu_arctic_prompt_count": 300,
+ "categories": {
+ "conversational": 25,
+ "descriptive": 16,
+ "emotional": 24,
+ "homographs": 6,
+ "long": 15,
+ "long_range": 6,
+ "names_places": 6,
+ "narrative": 24,
+ "numbers": 6,
+ "phonetic_balance": 300,
+ "punctuation": 6,
+ "question": 15,
+ "rare_words": 6,
+ "short": 24,
+ "technical": 21
+ },
+ "sources": {
+ "cmu_arctic": 300,
+ "ood_stress": 46,
+ "voxcpm_unrelated": 154
+ },
+ "training_exclusion_count": 87362,
+ "rejected_duplicate": 0,
+ "rejected_exact_training_match": 0,
+ "word_count": {
+ "minimum": 1,
+ "maximum": 33,
+ "mean": 10.272
+ }
+}
diff --git a/evaluation/final/raw/prompts.jsonl b/evaluation/final/raw/prompts.jsonl
new file mode 100644
index 0000000000000000000000000000000000000000..66433ca4aa5e6062b313a58d5d5dfa80e35a4cad
--- /dev/null
+++ b/evaluation/final/raw/prompts.jsonl
@@ -0,0 +1,500 @@
+{"id": "release500_0001", "category": "conversational", "text": "It wasn't until later that I realized what had actually happened.", "source": "voxcpm_unrelated"}
+{"id": "release500_0002", "category": "descriptive", "text": "The light at that time of afternoon is perfect for about twenty minutes.", "source": "voxcpm_unrelated"}
+{"id": "release500_0003", "category": "emotional", "text": "That was the scariest thing I've ever done, and also maybe the best.", "source": "voxcpm_unrelated"}
+{"id": "release500_0004", "category": "homographs", "text": "After I record the record, please present the present to the bass player.", "source": "ood_stress"}
+{"id": "release500_0005", "category": "long", "text": "If I'm being completely honest, I had no idea what I was doing for most of it, and I think that might be the most normal thing about the whole experience.", "source": "voxcpm_unrelated"}
+{"id": "release500_0006", "category": "long_range", "text": "The committee postponed its decision after discovering that three pages of the report had been printed in the wrong order.", "source": "ood_stress"}
+{"id": "release500_0007", "category": "names_places", "text": "Gwendolyn photographed the eucalyptus trees outside Ljubljana.", "source": "ood_stress"}
+{"id": "release500_0008", "category": "narrative", "text": "He had this theory, and the more time passes, the more I think he was right.", "source": "voxcpm_unrelated"}
+{"id": "release500_0009", "category": "numbers", "text": "The package weighs twelve point six kilograms and arrived on July twenty-first.", "source": "ood_stress"}
+{"id": "release500_0010", "category": "punctuation", "text": "First, close the window; second, turn off the lamp; finally, lock the door.", "source": "ood_stress"}
+{"id": "release500_0011", "category": "question", "text": "When did you first realize something was wrong?", "source": "voxcpm_unrelated"}
+{"id": "release500_0012", "category": "rare_words", "text": "The archaeologist catalogued a Byzantine reliquary and several anachronistic fragments.", "source": "ood_stress"}
+{"id": "release500_0013", "category": "short", "text": "I have no idea.", "source": "voxcpm_unrelated"}
+{"id": "release500_0014", "category": "technical", "text": "The system runs on three core components that all have to stay in sync.", "source": "voxcpm_unrelated"}
+{"id": "release500_0015", "category": "conversational", "text": "We improvised the entire second half and somehow nobody noticed.", "source": "voxcpm_unrelated"}
+{"id": "release500_0016", "category": "descriptive", "text": "There's a smell in old buildings that's impossible to describe but immediately familiar.", "source": "voxcpm_unrelated"}
+{"id": "release500_0017", "category": "emotional", "text": "I specifically asked for one thing. One thing. And it still didn't happen.", "source": "voxcpm_unrelated"}
+{"id": "release500_0018", "category": "homographs", "text": "They refuse to place the refuse beside the produce aisle.", "source": "ood_stress"}
+{"id": "release500_0019", "category": "long", "text": "There were days last year when I genuinely didn't know how I was going to get through it, and somehow I did, and I still don't fully understand how.", "source": "voxcpm_unrelated"}
+{"id": "release500_0020", "category": "long_range", "text": "Before announcing the winner, the judge thanked the volunteers who had worked through the storm to repair the damaged stage.", "source": "ood_stress"}
+{"id": "release500_0021", "category": "names_places", "text": "Professor Nguyen mailed the fluorescent poster to Reykjavik on Thursday.", "source": "ood_stress"}
+{"id": "release500_0022", "category": "narrative", "text": "The plan was simple. The execution was anything but.", "source": "voxcpm_unrelated"}
+{"id": "release500_0023", "category": "numbers", "text": "Her score improved from eighty-two percent to ninety-seven percent in two weeks.", "source": "ood_stress"}
+{"id": "release500_0024", "category": "punctuation", "text": "Why would anyone pack spoons, batteries, and a violin for one short trip?", "source": "ood_stress"}
+{"id": "release500_0025", "category": "question", "text": "How long have you been sitting on this?", "source": "voxcpm_unrelated"}
+{"id": "release500_0026", "category": "rare_words", "text": "A mischievous chameleon rested beside the chrysanthemum and the bougainvillea.", "source": "ood_stress"}
+{"id": "release500_0027", "category": "short", "text": "We don't talk about that.", "source": "voxcpm_unrelated"}
+{"id": "release500_0028", "category": "technical", "text": "Hash the UTF eight payload with S H A two fifty-six before opening the checkpoint.", "source": "ood_stress"}
+{"id": "release500_0029", "category": "conversational", "text": "I thought I had it all figured out, and then this happened.", "source": "voxcpm_unrelated"}
+{"id": "release500_0030", "category": "descriptive", "text": "The kind of quiet you only get at three in the morning in a city.", "source": "voxcpm_unrelated"}
+{"id": "release500_0031", "category": "emotional", "text": "The joy of it was almost overwhelming, like too much good all at once.", "source": "voxcpm_unrelated"}
+{"id": "release500_0032", "category": "homographs", "text": "Please close the door while I stand close to the microphone.", "source": "ood_stress"}
+{"id": "release500_0033", "category": "long", "text": "I think the most underrated skill in the world is knowing when to stop explaining yourself and just let people draw their own conclusions.", "source": "voxcpm_unrelated"}
+{"id": "release500_0034", "category": "long_range", "text": "When the final bus disappeared around the corner, Nadia remembered that her umbrella, notebook, and lunch were still inside the quiet office.", "source": "ood_stress"}
+{"id": "release500_0035", "category": "names_places", "text": "Aurelia and Joaquin crossed the Schuylkill River before noon.", "source": "ood_stress"}
+{"id": "release500_0036", "category": "narrative", "text": "The village had a name, but nobody used it. Everyone just called it home.", "source": "voxcpm_unrelated"}
+{"id": "release500_0037", "category": "numbers", "text": "Increase the gain by three decibels, wait five seconds, then reduce it by one.", "source": "ood_stress"}
+{"id": "release500_0038", "category": "punctuation", "text": "The result was unexpected: clear, stable, and surprisingly natural.", "source": "ood_stress"}
+{"id": "release500_0039", "category": "question", "text": "Who told you that was the right way to do it?", "source": "voxcpm_unrelated"}
+{"id": "release500_0040", "category": "rare_words", "text": "The meticulous cartographer examined an iridescent archipelago through binoculars.", "source": "ood_stress"}
+{"id": "release500_0041", "category": "short", "text": "I mean. Yeah. Kind of.", "source": "voxcpm_unrelated"}
+{"id": "release500_0042", "category": "technical", "text": "The logs are your best friend when something behaves unexpectedly.", "source": "voxcpm_unrelated"}
+{"id": "release500_0043", "category": "conversational", "text": "He kept talking and I kept nodding and neither of us was listening.", "source": "voxcpm_unrelated"}
+{"id": "release500_0044", "category": "descriptive", "text": "It's one of those places where everyone seems to know each other already.", "source": "voxcpm_unrelated"}
+{"id": "release500_0045", "category": "emotional", "text": "He said he was proud of me and nobody had ever said that to me before.", "source": "voxcpm_unrelated"}
+{"id": "release500_0046", "category": "homographs", "text": "The minute detail took a minute to explain clearly.", "source": "ood_stress"}
+{"id": "release500_0047", "category": "long", "text": "The strange thing about really wanting something is that getting it never quite feels the way you imagined it would, and not getting it never quite feels as bad.", "source": "voxcpm_unrelated"}
+{"id": "release500_0048", "category": "long_range", "text": "Even though the recipe looked simple at first, preparing the sauce required patience, careful timing, and a much larger pan.", "source": "ood_stress"}
+{"id": "release500_0049", "category": "names_places", "text": "Xiomara met Siobhan near the old library in Worcestershire.", "source": "ood_stress"}
+{"id": "release500_0050", "category": "narrative", "text": "She walked into the room and everyone went quiet, not because she was loud, but because she wasn't.", "source": "voxcpm_unrelated"}
+{"id": "release500_0051", "category": "numbers", "text": "At exactly seven forty-five, Morgan placed eighteen dollars and fifty cents beside the notebook.", "source": "ood_stress"}
+{"id": "release500_0052", "category": "punctuation", "text": "Does punctuation work? Yes: commas, semicolons, and periods should create sensible boundaries.", "source": "ood_stress"}
+{"id": "release500_0053", "category": "question", "text": "Have you ever just stopped and thought about how strange all of this is?", "source": "voxcpm_unrelated"}
+{"id": "release500_0054", "category": "rare_words", "text": "Her idiosyncratic explanation sounded plausible but remained fundamentally ambiguous.", "source": "ood_stress"}
+{"id": "release500_0055", "category": "short", "text": "That tracks.", "source": "voxcpm_unrelated"}
+{"id": "release500_0056", "category": "technical", "text": "The key thing to remember is that order matters here more than you'd think.", "source": "voxcpm_unrelated"}
+{"id": "release500_0057", "category": "conversational", "text": "I would not say that again in this room if I were you.", "source": "voxcpm_unrelated"}
+{"id": "release500_0058", "category": "descriptive", "text": "The room was filled with the quiet that means something is about to change.", "source": "voxcpm_unrelated"}
+{"id": "release500_0059", "category": "emotional", "text": "I didn't expect to feel that way about it, but I did.", "source": "voxcpm_unrelated"}
+{"id": "release500_0060", "category": "homographs", "text": "I object to the strange object that was left near the entrance.", "source": "ood_stress"}
+{"id": "release500_0061", "category": "long", "text": "I've learned more from the things that went wrong than from the things that worked, which is a frustrating way to have to learn things but apparently the only way that stuck.", "source": "voxcpm_unrelated"}
+{"id": "release500_0062", "category": "long_range", "text": "Because every nearby hotel was full, the exhausted hikers followed the narrow road until they found a farmhouse with one empty room.", "source": "ood_stress"}
+{"id": "release500_0063", "category": "names_places", "text": "The itinerary sends Beatrice from Albuquerque to Des Moines and then to Boise.", "source": "ood_stress"}
+{"id": "release500_0064", "category": "narrative", "text": "We had been driving for six hours when we finally saw the sign for the town.", "source": "voxcpm_unrelated"}
+{"id": "release500_0065", "category": "numbers", "text": "Call extension zero four nine after eight thirty in the morning.", "source": "ood_stress"}
+{"id": "release500_0066", "category": "punctuation", "text": "Stop - listen carefully - then answer the question without guessing.", "source": "ood_stress"}
+{"id": "release500_0067", "category": "question", "text": "Does it matter at this point, honestly?", "source": "voxcpm_unrelated"}
+{"id": "release500_0068", "category": "rare_words", "text": "An entrepreneurial meteorologist described the unprecedented atmospheric phenomenon.", "source": "ood_stress"}
+{"id": "release500_0069", "category": "short", "text": "That's not what I expected.", "source": "voxcpm_unrelated"}
+{"id": "release500_0070", "category": "technical", "text": "There's a known issue with that version that causes failures under load.", "source": "voxcpm_unrelated"}
+{"id": "release500_0071", "category": "conversational", "text": "Don't look now but I think that's him over there.", "source": "voxcpm_unrelated"}
+{"id": "release500_0072", "category": "descriptive", "text": "The city at night looks completely different from how it feels during the day.", "source": "voxcpm_unrelated"}
+{"id": "release500_0073", "category": "emotional", "text": "It's the kind of happy that makes you nervous because you know it can't last.", "source": "voxcpm_unrelated"}
+{"id": "release500_0074", "category": "homographs", "text": "The dove dove beneath the branch while the wind began to wind the loose cable.", "source": "ood_stress"}
+{"id": "release500_0075", "category": "long", "text": "I've thought a lot about what I would do differently if I could go back, and the honest answer is probably not as much as I'd like to think.", "source": "voxcpm_unrelated"}
+{"id": "release500_0076", "category": "long_range", "text": "Although the weather changed without warning, the musicians finished packing before anyone opened the enormous glass door.", "source": "ood_stress"}
+{"id": "release500_0077", "category": "names_places", "text": "Saskatchewan looked unusually bright beneath the pale February sunrise.", "source": "ood_stress"}
+{"id": "release500_0078", "category": "narrative", "text": "She handled it with a grace I genuinely did not expect.", "source": "voxcpm_unrelated"}
+{"id": "release500_0079", "category": "numbers", "text": "Apartment four hundred seven is across from room twenty-nine on the third floor.", "source": "ood_stress"}
+{"id": "release500_0080", "category": "punctuation", "text": "\"Not yet,\" Rowan whispered. \"Give me another minute... please.\"", "source": "ood_stress"}
+{"id": "release500_0081", "category": "question", "text": "What are we actually trying to accomplish here?", "source": "voxcpm_unrelated"}
+{"id": "release500_0082", "category": "rare_words", "text": "The physician recommended a noninvasive procedure for the recurring neurological symptoms.", "source": "ood_stress"}
+{"id": "release500_0083", "category": "short", "text": "Are you serious right now?", "source": "voxcpm_unrelated"}
+{"id": "release500_0084", "category": "technical", "text": "The performance improvement is real, but it comes with tradeoffs.", "source": "voxcpm_unrelated"}
+{"id": "release500_0085", "category": "conversational", "text": "I don't know. It's just been one of those weeks, you know?", "source": "voxcpm_unrelated"}
+{"id": "release500_0086", "category": "descriptive", "text": "The moment you stop expecting perfection, things get a lot more interesting.", "source": "voxcpm_unrelated"}
+{"id": "release500_0087", "category": "emotional", "text": "It's the kind of thing that stays with you even years later.", "source": "voxcpm_unrelated"}
+{"id": "release500_0088", "category": "long", "text": "She told me once that the most important skill she'd developed was learning to disagree with someone without making them feel they were wrong.", "source": "voxcpm_unrelated"}
+{"id": "release500_0089", "category": "narrative", "text": "He stood at the edge of the cliff and wondered, not for the first time, how he'd gotten here.", "source": "voxcpm_unrelated"}
+{"id": "release500_0090", "category": "question", "text": "Why does asking for help feel so hard when you actually need it?", "source": "voxcpm_unrelated"}
+{"id": "release500_0091", "category": "short", "text": "That's fair.", "source": "voxcpm_unrelated"}
+{"id": "release500_0092", "category": "technical", "text": "If you run into that error, the fix is usually just restarting the service.", "source": "voxcpm_unrelated"}
+{"id": "release500_0093", "category": "conversational", "text": "It was a disaster in the best possible way.", "source": "voxcpm_unrelated"}
+{"id": "release500_0094", "category": "descriptive", "text": "It had the look of a place that used to mean something to someone.", "source": "voxcpm_unrelated"}
+{"id": "release500_0095", "category": "emotional", "text": "There's this feeling you get when something finally works after weeks of failure.", "source": "voxcpm_unrelated"}
+{"id": "release500_0096", "category": "long", "text": "He had a theory that most arguments aren't actually about what they seem to be about, and the longer I live the more I think he was onto something.", "source": "voxcpm_unrelated"}
+{"id": "release500_0097", "category": "narrative", "text": "The thing nobody tells you about starting over is how ordinary most of it feels.", "source": "voxcpm_unrelated"}
+{"id": "release500_0098", "category": "question", "text": "What exactly are we waiting for at this point?", "source": "voxcpm_unrelated"}
+{"id": "release500_0099", "category": "short", "text": "I stand by what I said.", "source": "voxcpm_unrelated"}
+{"id": "release500_0100", "category": "technical", "text": "Double-check that value before you push anything to production.", "source": "voxcpm_unrelated"}
+{"id": "release500_0101", "category": "conversational", "text": "I understand your point; I just do not agree with the conclusion.", "source": "ood_stress"}
+{"id": "release500_0102", "category": "descriptive", "text": "The sky that day was a strange shade of orange I'd never seen before.", "source": "voxcpm_unrelated"}
+{"id": "release500_0103", "category": "emotional", "text": "It hurt more than I expected, and I thought I was prepared.", "source": "voxcpm_unrelated"}
+{"id": "release500_0104", "category": "long", "text": "The hardest part wasn't making the decision, it was sitting with the uncertainty of not knowing whether it was the right one for months afterward.", "source": "voxcpm_unrelated"}
+{"id": "release500_0105", "category": "narrative", "text": "There was this one day last summer when everything just clicked into place.", "source": "voxcpm_unrelated"}
+{"id": "release500_0106", "category": "question", "text": "Have you ever thought about why we keep doing things the same way when they stop working?", "source": "voxcpm_unrelated"}
+{"id": "release500_0107", "category": "short", "text": "I can't believe you just said that.", "source": "voxcpm_unrelated"}
+{"id": "release500_0108", "category": "technical", "text": "You'll want to isolate that variable before drawing any conclusions.", "source": "voxcpm_unrelated"}
+{"id": "release500_0109", "category": "conversational", "text": "The meeting that could have been an email lasted two hours.", "source": "voxcpm_unrelated"}
+{"id": "release500_0110", "category": "descriptive", "text": "There's a specific kind of tired that sleep doesn't fix.", "source": "voxcpm_unrelated"}
+{"id": "release500_0111", "category": "emotional", "text": "Something about that moment broke something open in me that I hadn't realized was closed.", "source": "voxcpm_unrelated"}
+{"id": "release500_0112", "category": "long", "text": "The thing about making something good is that you don't usually know it's good while you're making it. You just know you're not done yet.", "source": "voxcpm_unrelated"}
+{"id": "release500_0113", "category": "narrative", "text": "She opened the door slowly, as if she expected to find something she wasn't ready for.", "source": "voxcpm_unrelated"}
+{"id": "release500_0114", "category": "question", "text": "What would you do differently if you could go back?", "source": "voxcpm_unrelated"}
+{"id": "release500_0115", "category": "short", "text": "Okay. Okay okay okay.", "source": "voxcpm_unrelated"}
+{"id": "release500_0116", "category": "technical", "text": "The latency goes up significantly at scale, which is expected.", "source": "voxcpm_unrelated"}
+{"id": "release500_0117", "category": "conversational", "text": "We ended up staying until two in the morning talking about nothing.", "source": "voxcpm_unrelated"}
+{"id": "release500_0118", "category": "descriptive", "text": "The rain started slowly, and then all at once, the way it usually does.", "source": "voxcpm_unrelated"}
+{"id": "release500_0119", "category": "emotional", "text": "I honestly did not think I was going to make it through that year.", "source": "voxcpm_unrelated"}
+{"id": "release500_0120", "category": "long", "text": "There's an assumption that people who seem confident all the time actually feel confident, and in my experience that's almost never true.", "source": "voxcpm_unrelated"}
+{"id": "release500_0121", "category": "narrative", "text": "The moment I opened the door I knew something was different, even if I couldn't say what.", "source": "voxcpm_unrelated"}
+{"id": "release500_0122", "category": "question", "text": "How do you explain something that doesn't make sense even to you?", "source": "voxcpm_unrelated"}
+{"id": "release500_0123", "category": "short", "text": "Hold on a second.", "source": "voxcpm_unrelated"}
+{"id": "release500_0124", "category": "technical", "text": "A deterministic split prevents duplicate normalized text from leaking into evaluation.", "source": "ood_stress"}
+{"id": "release500_0125", "category": "conversational", "text": "Honestly, that was much better than I expected, but the ending still felt rushed.", "source": "ood_stress"}
+{"id": "release500_0126", "category": "descriptive", "text": "She had a habit of tilting her head slightly when thinking hard about something.", "source": "voxcpm_unrelated"}
+{"id": "release500_0127", "category": "emotional", "text": "I cannot stress enough how much I did not want to be right about this.", "source": "voxcpm_unrelated"}
+{"id": "release500_0128", "category": "long", "text": "I used to think the goal was to get to a point where nothing was hard anymore, and now I think that point doesn't exist and I'm not sure I'd want it to.", "source": "voxcpm_unrelated"}
+{"id": "release500_0129", "category": "narrative", "text": "So this is the part of the story where everything goes completely sideways.", "source": "voxcpm_unrelated"}
+{"id": "release500_0130", "category": "question", "text": "Is there a version of this where it actually works out okay?", "source": "voxcpm_unrelated"}
+{"id": "release500_0131", "category": "short", "text": "Oh no. Oh no no no.", "source": "voxcpm_unrelated"}
+{"id": "release500_0132", "category": "technical", "text": "Always test on a copy before modifying the original.", "source": "voxcpm_unrelated"}
+{"id": "release500_0133", "category": "conversational", "text": "I wasn't even supposed to be there. That's the wild part.", "source": "voxcpm_unrelated"}
+{"id": "release500_0134", "category": "descriptive", "text": "It's a small town. Not the kind you leave. The kind you leave and come back to.", "source": "voxcpm_unrelated"}
+{"id": "release500_0135", "category": "emotional", "text": "She cried. I cried. We all just sort of stood there crying.", "source": "voxcpm_unrelated"}
+{"id": "release500_0136", "category": "long", "text": "There's a version of being busy that's actually just a very elaborate way of avoiding the one thing you know you should be doing.", "source": "voxcpm_unrelated"}
+{"id": "release500_0137", "category": "narrative", "text": "The letter arrived on a Tuesday, which felt appropriate somehow.", "source": "voxcpm_unrelated"}
+{"id": "release500_0138", "category": "question", "text": "What exactly did she say when you told her?", "source": "voxcpm_unrelated"}
+{"id": "release500_0139", "category": "short", "text": "No way. Absolutely not.", "source": "voxcpm_unrelated"}
+{"id": "release500_0140", "category": "technical", "text": "The first thing you want to do is make sure everything is backed up.", "source": "voxcpm_unrelated"}
+{"id": "release500_0141", "category": "conversational", "text": "I have the most insane idea and you need to just hear me out.", "source": "voxcpm_unrelated"}
+{"id": "release500_0142", "category": "descriptive", "text": "There's a coffee shop on the corner that's been there longer than anyone can remember.", "source": "voxcpm_unrelated"}
+{"id": "release500_0143", "category": "emotional", "text": "I kept it together until I got to the car, and then I completely fell apart.", "source": "voxcpm_unrelated"}
+{"id": "release500_0144", "category": "long", "text": "The older I get, the more I think that most of what we call wisdom is just pattern recognition built from having been wrong enough times to start noticing.", "source": "voxcpm_unrelated"}
+{"id": "release500_0145", "category": "narrative", "text": "I remember the exact moment I realized I had made a huge mistake.", "source": "voxcpm_unrelated"}
+{"id": "release500_0146", "category": "question", "text": "Do you think it would have gone differently if we'd left earlier?", "source": "voxcpm_unrelated"}
+{"id": "release500_0147", "category": "short", "text": "You have to be kidding me.", "source": "voxcpm_unrelated"}
+{"id": "release500_0148", "category": "technical", "text": "If the checksum fails, restart the worker process without deleting the cached tensors.", "source": "ood_stress"}
+{"id": "release500_0149", "category": "conversational", "text": "I need you to understand how close that was to not working.", "source": "voxcpm_unrelated"}
+{"id": "release500_0150", "category": "descriptive", "text": "It was the sort of morning where the whole world looks rinsed clean.", "source": "voxcpm_unrelated"}
+{"id": "release500_0151", "category": "emotional", "text": "She started laughing and I started laughing and we couldn't stop for a full minute.", "source": "voxcpm_unrelated"}
+{"id": "release500_0152", "category": "long", "text": "We spent three years building something we were genuinely proud of, and then the market shifted and none of it mattered the way we thought it would.", "source": "voxcpm_unrelated"}
+{"id": "release500_0153", "category": "narrative", "text": "And somehow, despite all of that, it actually turned out okay.", "source": "voxcpm_unrelated"}
+{"id": "release500_0154", "category": "question", "text": "Is that something you actually want, or something you think you should want?", "source": "voxcpm_unrelated"}
+{"id": "release500_0155", "category": "short", "text": "This is fine.", "source": "voxcpm_unrelated"}
+{"id": "release500_0156", "category": "technical", "text": "The tricky part isn't the setup, it's making sure all dependencies line up.", "source": "voxcpm_unrelated"}
+{"id": "release500_0157", "category": "conversational", "text": "He called it a minor setback. I called it a disaster.", "source": "voxcpm_unrelated"}
+{"id": "release500_0158", "category": "descriptive", "text": "The whole place smelled like pine and old wood and cold air.", "source": "voxcpm_unrelated"}
+{"id": "release500_0159", "category": "emotional", "text": "I've never been so relieved in my entire life.", "source": "voxcpm_unrelated"}
+{"id": "release500_0160", "category": "narrative", "text": "I wrote it all down that night so I wouldn't forget, and I still haven't looked at it.", "source": "voxcpm_unrelated"}
+{"id": "release500_0161", "category": "short", "text": "That's kind of wild actually.", "source": "voxcpm_unrelated"}
+{"id": "release500_0162", "category": "technical", "text": "Version pinning will save you a lot of pain later down the line.", "source": "voxcpm_unrelated"}
+{"id": "release500_0163", "category": "conversational", "text": "At some point you just have to commit and see what happens.", "source": "voxcpm_unrelated"}
+{"id": "release500_0164", "category": "emotional", "text": "Nobody told me it was going to be this hard, and honestly, I'm grateful.", "source": "voxcpm_unrelated"}
+{"id": "release500_0165", "category": "narrative", "text": "Chapter seven. In which everything that could go wrong did, and a few things that couldn't.", "source": "voxcpm_unrelated"}
+{"id": "release500_0166", "category": "short", "text": "Wait, what?", "source": "voxcpm_unrelated"}
+{"id": "release500_0167", "category": "technical", "text": "The graphics processor remained idle because the data loader exhausted every central processing core.", "source": "ood_stress"}
+{"id": "release500_0168", "category": "conversational", "text": "Could you speak a little more slowly? I am trying to write every detail down.", "source": "ood_stress"}
+{"id": "release500_0169", "category": "emotional", "text": "I was so excited I could barely form a complete sentence.", "source": "voxcpm_unrelated"}
+{"id": "release500_0170", "category": "narrative", "text": "He looked at me like I had said something in a language he didn't speak.", "source": "voxcpm_unrelated"}
+{"id": "release500_0171", "category": "short", "text": "That's actually hilarious.", "source": "voxcpm_unrelated"}
+{"id": "release500_0172", "category": "technical", "text": "The encoder maps phonemes, stress, and phrase boundaries into a shared hidden representation.", "source": "ood_stress"}
+{"id": "release500_0173", "category": "conversational", "text": "She showed up three hours late and acted like nothing had happened.", "source": "voxcpm_unrelated"}
+{"id": "release500_0174", "category": "emotional", "text": "Oh my god. This is the best thing I've ever tasted.", "source": "voxcpm_unrelated"}
+{"id": "release500_0175", "category": "narrative", "text": "The note was short, which was somehow more frightening than a long one would have been.", "source": "voxcpm_unrelated"}
+{"id": "release500_0176", "category": "short", "text": "Noted.", "source": "voxcpm_unrelated"}
+{"id": "release500_0177", "category": "technical", "text": "It's a simple process once you understand the underlying structure.", "source": "voxcpm_unrelated"}
+{"id": "release500_0178", "category": "conversational", "text": "You have to hear this story because it's almost too good to be true.", "source": "voxcpm_unrelated"}
+{"id": "release500_0179", "category": "emotional", "text": "I have never felt more seen in my entire life than in that moment.", "source": "voxcpm_unrelated"}
+{"id": "release500_0180", "category": "narrative", "text": "She knew. She had always known. She just waited for the right moment.", "source": "voxcpm_unrelated"}
+{"id": "release500_0181", "category": "short", "text": "Yeah, no. Hard pass.", "source": "voxcpm_unrelated"}
+{"id": "release500_0182", "category": "technical", "text": "Please verify the sample rate, hop length, and mel filter configuration before inference.", "source": "ood_stress"}
+{"id": "release500_0183", "category": "conversational", "text": "That's the worst idea I've ever heard and I'm completely in.", "source": "voxcpm_unrelated"}
+{"id": "release500_0184", "category": "emotional", "text": "I remember thinking: this is it. This is the thing I've been waiting for.", "source": "voxcpm_unrelated"}
+{"id": "release500_0185", "category": "narrative", "text": "We never talked about it again, which I think was the right call.", "source": "voxcpm_unrelated"}
+{"id": "release500_0186", "category": "short", "text": "In hindsight, obvious.", "source": "voxcpm_unrelated"}
+{"id": "release500_0187", "category": "technical", "text": "Once it's configured correctly, it basically runs itself.", "source": "voxcpm_unrelated"}
+{"id": "release500_0188", "category": "conversational", "text": "You got engaged? When? How? Tell me everything right now.", "source": "voxcpm_unrelated"}
+{"id": "release500_0189", "category": "emotional", "text": "It's hard to explain how much that moment meant to me.", "source": "voxcpm_unrelated"}
+{"id": "release500_0190", "category": "narrative", "text": "He told the story three different times that night and somehow it got better each time.", "source": "voxcpm_unrelated"}
+{"id": "release500_0191", "category": "short", "text": "Right. Moving on.", "source": "voxcpm_unrelated"}
+{"id": "release500_0192", "category": "conversational", "text": "No, I did not move your keys; check beside the small ceramic bowl.", "source": "ood_stress"}
+{"id": "release500_0193", "category": "emotional", "text": "I was devastated. I'm not going to pretend otherwise.", "source": "voxcpm_unrelated"}
+{"id": "release500_0194", "category": "narrative", "text": "That was the last summer before everything changed, though we didn't know it yet.", "source": "voxcpm_unrelated"}
+{"id": "release500_0195", "category": "short", "text": "Bold of you to assume.", "source": "voxcpm_unrelated"}
+{"id": "release500_0196", "category": "conversational", "text": "It's one of those decisions that makes total sense in the moment.", "source": "voxcpm_unrelated"}
+{"id": "release500_0197", "category": "emotional", "text": "I was furious. Not the loud kind. The quiet kind that lasts for days.", "source": "voxcpm_unrelated"}
+{"id": "release500_0198", "category": "narrative", "text": "That's the part of the story I usually leave out.", "source": "voxcpm_unrelated"}
+{"id": "release500_0199", "category": "short", "text": "I've said too much.", "source": "voxcpm_unrelated"}
+{"id": "release500_0200", "category": "conversational", "text": "I've been faking that I understood this the entire time.", "source": "voxcpm_unrelated"}
+{"id": "release500_0201", "category": "phonetic_balance", "text": "Peace be unto you and grace before the Lord.", "source": "cmu_arctic"}
+{"id": "release500_0202", "category": "phonetic_balance", "text": "From now on we're pals.", "source": "cmu_arctic"}
+{"id": "release500_0203", "category": "phonetic_balance", "text": "Here we allow our solicitors to look after our legal work.", "source": "cmu_arctic"}
+{"id": "release500_0204", "category": "phonetic_balance", "text": "Her words sent a strange chill through Philip.", "source": "cmu_arctic"}
+{"id": "release500_0205", "category": "phonetic_balance", "text": "It was introduced by Representative Dick of Ohio.", "source": "cmu_arctic"}
+{"id": "release500_0206", "category": "phonetic_balance", "text": "It's the aurora borealis.", "source": "cmu_arctic"}
+{"id": "release500_0207", "category": "phonetic_balance", "text": "The boy grew and prospered.", "source": "cmu_arctic"}
+{"id": "release500_0208", "category": "phonetic_balance", "text": "His voice was passionately rebellious.", "source": "cmu_arctic"}
+{"id": "release500_0209", "category": "phonetic_balance", "text": "That's where they cut off the Scottish Chiefs and killed all hands.", "source": "cmu_arctic"}
+{"id": "release500_0210", "category": "phonetic_balance", "text": "And Tom King patiently endured.", "source": "cmu_arctic"}
+{"id": "release500_0211", "category": "phonetic_balance", "text": "From that moment his friendship for Belize turns to hatred and jealousy.", "source": "cmu_arctic"}
+{"id": "release500_0212", "category": "phonetic_balance", "text": "Then you don't believe in altruism.", "source": "cmu_arctic"}
+{"id": "release500_0213", "category": "phonetic_balance", "text": "What Pascal glimpsed with the vision of a seer, I have lived.", "source": "cmu_arctic"}
+{"id": "release500_0214", "category": "phonetic_balance", "text": "You were making them talk shop, Ruth charged him.", "source": "cmu_arctic"}
+{"id": "release500_0215", "category": "phonetic_balance", "text": "Philip snatched at the letter which Gregson held out to him.", "source": "cmu_arctic"}
+{"id": "release500_0216", "category": "phonetic_balance", "text": "For a time the exciting thrill of his adventure was gone.", "source": "cmu_arctic"}
+{"id": "release500_0217", "category": "phonetic_balance", "text": "Obviously, it was a disease that could be contracted by contact.", "source": "cmu_arctic"}
+{"id": "release500_0218", "category": "phonetic_balance", "text": "Blind with rage, he darted in.", "source": "cmu_arctic"}
+{"id": "release500_0219", "category": "phonetic_balance", "text": "You live on an income which your father earned.", "source": "cmu_arctic"}
+{"id": "release500_0220", "category": "phonetic_balance", "text": "Something about it seemed to fascinate him, to challenge his presence.", "source": "cmu_arctic"}
+{"id": "release500_0221", "category": "phonetic_balance", "text": "Her beautiful hair was done up in shining coils.", "source": "cmu_arctic"}
+{"id": "release500_0222", "category": "phonetic_balance", "text": "It does, was her audacious answer.", "source": "cmu_arctic"}
+{"id": "release500_0223", "category": "phonetic_balance", "text": "He caught himself with a jerk.", "source": "cmu_arctic"}
+{"id": "release500_0224", "category": "phonetic_balance", "text": "He did not believe in the burning of daylight for such a luxury.", "source": "cmu_arctic"}
+{"id": "release500_0225", "category": "phonetic_balance", "text": "He had comparatively no advantages at first.", "source": "cmu_arctic"}
+{"id": "release500_0226", "category": "phonetic_balance", "text": "Such men believe, when they come together.", "source": "cmu_arctic"}
+{"id": "release500_0227", "category": "phonetic_balance", "text": "The ship should be in within a week or ten days.", "source": "cmu_arctic"}
+{"id": "release500_0228", "category": "phonetic_balance", "text": "Such a dog the wise driver kills, or turns loose.", "source": "cmu_arctic"}
+{"id": "release500_0229", "category": "phonetic_balance", "text": "You have heard always how he was the lover of the Princess Naomi.", "source": "cmu_arctic"}
+{"id": "release500_0230", "category": "phonetic_balance", "text": "I want to know how all this is possible.", "source": "cmu_arctic"}
+{"id": "release500_0231", "category": "phonetic_balance", "text": "It was my idea to a tee.", "source": "cmu_arctic"}
+{"id": "release500_0232", "category": "phonetic_balance", "text": "They likewise are disinclined to being eaten.", "source": "cmu_arctic"}
+{"id": "release500_0233", "category": "phonetic_balance", "text": "This state of mind comes of an undue prominence of the ego.", "source": "cmu_arctic"}
+{"id": "release500_0234", "category": "phonetic_balance", "text": "Then it is as I said, Womble announced with finality.", "source": "cmu_arctic"}
+{"id": "release500_0235", "category": "phonetic_balance", "text": "I can see that knife now.", "source": "cmu_arctic"}
+{"id": "release500_0236", "category": "phonetic_balance", "text": "If I meet her again I shall apologize, said Eileen.", "source": "cmu_arctic"}
+{"id": "release500_0237", "category": "phonetic_balance", "text": "A wildly exciting time was his during the week preceding Thursday the eighteenth.", "source": "cmu_arctic"}
+{"id": "release500_0238", "category": "phonetic_balance", "text": "In what bucolic school of fence he had been taught was beyond imagining.", "source": "cmu_arctic"}
+{"id": "release500_0239", "category": "phonetic_balance", "text": "So she said, the irate skipper dashed on.", "source": "cmu_arctic"}
+{"id": "release500_0240", "category": "phonetic_balance", "text": "It is not an attempt to smash the market.", "source": "cmu_arctic"}
+{"id": "release500_0241", "category": "phonetic_balance", "text": "He destroyed everything that had belonged to the woman.", "source": "cmu_arctic"}
+{"id": "release500_0242", "category": "phonetic_balance", "text": "Did Thorpe go to see any one in Churchill.", "source": "cmu_arctic"}
+{"id": "release500_0243", "category": "phonetic_balance", "text": "Many other similar disconcerting omissions will be noticed in the Manuscript.", "source": "cmu_arctic"}
+{"id": "release500_0244", "category": "phonetic_balance", "text": "The ringing of the big bell aroused him.", "source": "cmu_arctic"}
+{"id": "release500_0245", "category": "phonetic_balance", "text": "A sinewy hand, dripping with water, was clutching the rail.", "source": "cmu_arctic"}
+{"id": "release500_0246", "category": "phonetic_balance", "text": "Enters now the psychology of the situation.", "source": "cmu_arctic"}
+{"id": "release500_0247", "category": "phonetic_balance", "text": "Wash your hands of me.", "source": "cmu_arctic"}
+{"id": "release500_0248", "category": "phonetic_balance", "text": "He confessed that the sketch had startled him.", "source": "cmu_arctic"}
+{"id": "release500_0249", "category": "phonetic_balance", "text": "Her face was against his breast.", "source": "cmu_arctic"}
+{"id": "release500_0250", "category": "phonetic_balance", "text": "He was a pariah; a wanderer without a friend or a home.", "source": "cmu_arctic"}
+{"id": "release500_0251", "category": "phonetic_balance", "text": "A cry of joy burst from Philip's lips.", "source": "cmu_arctic"}
+{"id": "release500_0252", "category": "phonetic_balance", "text": "And the big Persian knew of his existence before he did of hers.", "source": "cmu_arctic"}
+{"id": "release500_0253", "category": "phonetic_balance", "text": "The men stared into each other's face.", "source": "cmu_arctic"}
+{"id": "release500_0254", "category": "phonetic_balance", "text": "It is the fire, partly, she said.", "source": "cmu_arctic"}
+{"id": "release500_0255", "category": "phonetic_balance", "text": "It seemed nearer to him since he had seen and talked with Gregson.", "source": "cmu_arctic"}
+{"id": "release500_0256", "category": "phonetic_balance", "text": "Among my minor afflictions, I may mention a new and mysterious one.", "source": "cmu_arctic"}
+{"id": "release500_0257", "category": "phonetic_balance", "text": "Between him and all domestic animals there must be no hostilities.", "source": "cmu_arctic"}
+{"id": "release500_0258", "category": "phonetic_balance", "text": "With these arguments he convinced himself that he should go on alone.", "source": "cmu_arctic"}
+{"id": "release500_0259", "category": "phonetic_balance", "text": "We are consumed in our own flesh-pots.", "source": "cmu_arctic"}
+{"id": "release500_0260", "category": "phonetic_balance", "text": "Mercedes screamed, cried, laughed, and manifested the chaotic abandonment of hysteria.", "source": "cmu_arctic"}
+{"id": "release500_0261", "category": "phonetic_balance", "text": "Philip made no effort to follow.", "source": "cmu_arctic"}
+{"id": "release500_0262", "category": "phonetic_balance", "text": "She said with chattering teeth.", "source": "cmu_arctic"}
+{"id": "release500_0263", "category": "phonetic_balance", "text": "Halfway around the track one donkey got into an argument with its rider.", "source": "cmu_arctic"}
+{"id": "release500_0264", "category": "phonetic_balance", "text": "Earth and gravel seemed to fill the pan.", "source": "cmu_arctic"}
+{"id": "release500_0265", "category": "phonetic_balance", "text": "Her achievements with cocoanuts were a revelation.", "source": "cmu_arctic"}
+{"id": "release500_0266", "category": "phonetic_balance", "text": "For a full minute he crouched and listened.", "source": "cmu_arctic"}
+{"id": "release500_0267", "category": "phonetic_balance", "text": "It was edged with ice.", "source": "cmu_arctic"}
+{"id": "release500_0268", "category": "phonetic_balance", "text": "The Portuguese boy passed the Hawaiian.", "source": "cmu_arctic"}
+{"id": "release500_0269", "category": "phonetic_balance", "text": "But he did not broach it, preferring to mature it carefully.", "source": "cmu_arctic"}
+{"id": "release500_0270", "category": "phonetic_balance", "text": "Come on, Del Mar challenged.", "source": "cmu_arctic"}
+{"id": "release500_0271", "category": "phonetic_balance", "text": "For a full half minute Jeanne looked at him without speaking.", "source": "cmu_arctic"}
+{"id": "release500_0272", "category": "phonetic_balance", "text": "He bore no grudges and had few enemies.", "source": "cmu_arctic"}
+{"id": "release500_0273", "category": "phonetic_balance", "text": "I'll go over tomorrow afternoon.", "source": "cmu_arctic"}
+{"id": "release500_0274", "category": "phonetic_balance", "text": "Well, there are better men in Hawaii, that's all.", "source": "cmu_arctic"}
+{"id": "release500_0275", "category": "phonetic_balance", "text": "He was an athlete and a giant.", "source": "cmu_arctic"}
+{"id": "release500_0276", "category": "phonetic_balance", "text": "McVeigh when he returns from a trip to Honolulu.", "source": "cmu_arctic"}
+{"id": "release500_0277", "category": "phonetic_balance", "text": "You're a devil for fighting, and will surely win.", "source": "cmu_arctic"}
+{"id": "release500_0278", "category": "phonetic_balance", "text": "All an appearance can know is mirage.", "source": "cmu_arctic"}
+{"id": "release500_0279", "category": "phonetic_balance", "text": "You fired me out of your house, in short.", "source": "cmu_arctic"}
+{"id": "release500_0280", "category": "phonetic_balance", "text": "He made sure that the magazine was loaded, and resumed his paddling.", "source": "cmu_arctic"}
+{"id": "release500_0281", "category": "phonetic_balance", "text": "Change chairs, Daylight commanded.", "source": "cmu_arctic"}
+{"id": "release500_0282", "category": "phonetic_balance", "text": "By that answer my professional medical prestige stood or fell.", "source": "cmu_arctic"}
+{"id": "release500_0283", "category": "phonetic_balance", "text": "The moon had already begun its westward decline.", "source": "cmu_arctic"}
+{"id": "release500_0284", "category": "phonetic_balance", "text": "At sea, Monday, March 16, 1908.", "source": "cmu_arctic"}
+{"id": "release500_0285", "category": "phonetic_balance", "text": "And wherever I ranged, the way lay along alcohol-drenched roads.", "source": "cmu_arctic"}
+{"id": "release500_0286", "category": "phonetic_balance", "text": "The date was nearly eighteen years old.", "source": "cmu_arctic"}
+{"id": "release500_0287", "category": "phonetic_balance", "text": "Very early in my life, I separated from my mother.", "source": "cmu_arctic"}
+{"id": "release500_0288", "category": "phonetic_balance", "text": "There was pride and strength, the ring of triumph in his voice.", "source": "cmu_arctic"}
+{"id": "release500_0289", "category": "phonetic_balance", "text": "From my earliest recollection my sleep was a period of terror.", "source": "cmu_arctic"}
+{"id": "release500_0290", "category": "phonetic_balance", "text": "We'll have to watch our chances.", "source": "cmu_arctic"}
+{"id": "release500_0291", "category": "phonetic_balance", "text": "Close beside him gleamed the white fangs of the wolf-dog.", "source": "cmu_arctic"}
+{"id": "release500_0292", "category": "phonetic_balance", "text": "Would you be satisfied with that one hundredth part of me.", "source": "cmu_arctic"}
+{"id": "release500_0293", "category": "phonetic_balance", "text": "Two gigantic owls were tearing at the carcass.", "source": "cmu_arctic"}
+{"id": "release500_0294", "category": "phonetic_balance", "text": "The lush vegetation of that sheltered spot make a natural shield.", "source": "cmu_arctic"}
+{"id": "release500_0295", "category": "phonetic_balance", "text": "Below him the shadow was broken into a pool of rippling starlight.", "source": "cmu_arctic"}
+{"id": "release500_0296", "category": "phonetic_balance", "text": "Lots of men take women buggy riding.", "source": "cmu_arctic"}
+{"id": "release500_0297", "category": "phonetic_balance", "text": "But it contributed to the smash.", "source": "cmu_arctic"}
+{"id": "release500_0298", "category": "phonetic_balance", "text": "No man ate of the seal meat or the oil.", "source": "cmu_arctic"}
+{"id": "release500_0299", "category": "phonetic_balance", "text": "You're going in for grab sharing.", "source": "cmu_arctic"}
+{"id": "release500_0300", "category": "phonetic_balance", "text": "He it was that lived to found the family of the Patino.", "source": "cmu_arctic"}
+{"id": "release500_0301", "category": "phonetic_balance", "text": "For such countries nothing remained but reorganization.", "source": "cmu_arctic"}
+{"id": "release500_0302", "category": "phonetic_balance", "text": "Stand off butcher and baker and all the rest.", "source": "cmu_arctic"}
+{"id": "release500_0303", "category": "phonetic_balance", "text": "All eyes, however, were staring at him in certitude of expectancy.", "source": "cmu_arctic"}
+{"id": "release500_0304", "category": "phonetic_balance", "text": "Two days ago Jeanne learned where her father's men were hiding.", "source": "cmu_arctic"}
+{"id": "release500_0305", "category": "phonetic_balance", "text": "His previous wives had never lived long enough to bear him children.", "source": "cmu_arctic"}
+{"id": "release500_0306", "category": "phonetic_balance", "text": "But this little defect did not worry him.", "source": "cmu_arctic"}
+{"id": "release500_0307", "category": "phonetic_balance", "text": "Eli Harding asked, as Shunk started to follow.", "source": "cmu_arctic"}
+{"id": "release500_0308", "category": "phonetic_balance", "text": "And this was their sole conversation throughout the meal.", "source": "cmu_arctic"}
+{"id": "release500_0309", "category": "phonetic_balance", "text": "Then, again, Tudor had such an irritating way about him.", "source": "cmu_arctic"}
+{"id": "release500_0310", "category": "phonetic_balance", "text": "In it there was something that was almost tragedy.", "source": "cmu_arctic"}
+{"id": "release500_0311", "category": "phonetic_balance", "text": "Thirty pounds, said the captain with finality.", "source": "cmu_arctic"}
+{"id": "release500_0312", "category": "phonetic_balance", "text": "It was over when he made his way through the ring of spectators.", "source": "cmu_arctic"}
+{"id": "release500_0313", "category": "phonetic_balance", "text": "She had died from cold and starvation.", "source": "cmu_arctic"}
+{"id": "release500_0314", "category": "phonetic_balance", "text": "We fished sharks on Niihau together.", "source": "cmu_arctic"}
+{"id": "release500_0315", "category": "phonetic_balance", "text": "His slim hands gripped the edges of the table.", "source": "cmu_arctic"}
+{"id": "release500_0316", "category": "phonetic_balance", "text": "We had been chased by them ourselves, more than once.", "source": "cmu_arctic"}
+{"id": "release500_0317", "category": "phonetic_balance", "text": "Or have they already devised one.", "source": "cmu_arctic"}
+{"id": "release500_0318", "category": "phonetic_balance", "text": "I cannot follow you, she said.", "source": "cmu_arctic"}
+{"id": "release500_0319", "category": "phonetic_balance", "text": "He went down in midstream, searching the shadows of both shores.", "source": "cmu_arctic"}
+{"id": "release500_0320", "category": "phonetic_balance", "text": "I was sick once -- typhoid.", "source": "cmu_arctic"}
+{"id": "release500_0321", "category": "phonetic_balance", "text": "Why not like any railroad station or ferry depot.", "source": "cmu_arctic"}
+{"id": "release500_0322", "category": "phonetic_balance", "text": "Red-Eye swung back and forth on the branch farther down.", "source": "cmu_arctic"}
+{"id": "release500_0323", "category": "phonetic_balance", "text": "Within himself he called it no longer his own.", "source": "cmu_arctic"}
+{"id": "release500_0324", "category": "phonetic_balance", "text": "It lasted as a deterrent for two days.", "source": "cmu_arctic"}
+{"id": "release500_0325", "category": "phonetic_balance", "text": "He loved to play Chinese lottery.", "source": "cmu_arctic"}
+{"id": "release500_0326", "category": "phonetic_balance", "text": "Lord, but I'm glad to see you again, Phil.", "source": "cmu_arctic"}
+{"id": "release500_0327", "category": "phonetic_balance", "text": "There is not an iota of truth in it, certainly not.", "source": "cmu_arctic"}
+{"id": "release500_0328", "category": "phonetic_balance", "text": "She added, with genuine sympathy in her face and voice.", "source": "cmu_arctic"}
+{"id": "release500_0329", "category": "phonetic_balance", "text": "We have plenty of capital ourselves, and yet we want more.", "source": "cmu_arctic"}
+{"id": "release500_0330", "category": "phonetic_balance", "text": "And now put yourself in my place for a moment.", "source": "cmu_arctic"}
+{"id": "release500_0331", "category": "phonetic_balance", "text": "Fast, but endure.", "source": "cmu_arctic"}
+{"id": "release500_0332", "category": "phonetic_balance", "text": "You were looking squeamish this afternoon, he began.", "source": "cmu_arctic"}
+{"id": "release500_0333", "category": "phonetic_balance", "text": "It is the nearest refuge.", "source": "cmu_arctic"}
+{"id": "release500_0334", "category": "phonetic_balance", "text": "I want to die in it.", "source": "cmu_arctic"}
+{"id": "release500_0335", "category": "phonetic_balance", "text": "He stopped, and Philip nodded at the horrified question in his eyes.", "source": "cmu_arctic"}
+{"id": "release500_0336", "category": "phonetic_balance", "text": "She turned, fearing that Jacques might see what was in her face.", "source": "cmu_arctic"}
+{"id": "release500_0337", "category": "phonetic_balance", "text": "It was a curious coincidence.", "source": "cmu_arctic"}
+{"id": "release500_0338", "category": "phonetic_balance", "text": "Daylight made no answer, and the door closed behind him.", "source": "cmu_arctic"}
+{"id": "release500_0339", "category": "phonetic_balance", "text": "And he did hurt my arm.", "source": "cmu_arctic"}
+{"id": "release500_0340", "category": "phonetic_balance", "text": "Dunham, can your boy go along with Jesse.", "source": "cmu_arctic"}
+{"id": "release500_0341", "category": "phonetic_balance", "text": "The creative joy, I murmured.", "source": "cmu_arctic"}
+{"id": "release500_0342", "category": "phonetic_balance", "text": "He pulled, and the log crashed down to break his back.", "source": "cmu_arctic"}
+{"id": "release500_0343", "category": "phonetic_balance", "text": "By golly, the boy wins.", "source": "cmu_arctic"}
+{"id": "release500_0344", "category": "phonetic_balance", "text": "The boy threw back his head with pride.", "source": "cmu_arctic"}
+{"id": "release500_0345", "category": "phonetic_balance", "text": "The last one I knew was an overseer.", "source": "cmu_arctic"}
+{"id": "release500_0346", "category": "phonetic_balance", "text": "Two years ago I gave up civilization for this.", "source": "cmu_arctic"}
+{"id": "release500_0347", "category": "phonetic_balance", "text": "The river bared its bosom, and snorting steamboats challenged the wilderness.", "source": "cmu_arctic"}
+{"id": "release500_0348", "category": "phonetic_balance", "text": "Mab, she said.", "source": "cmu_arctic"}
+{"id": "release500_0349", "category": "phonetic_balance", "text": "The regret in his voice was provocative of a second burst of laughter.", "source": "cmu_arctic"}
+{"id": "release500_0350", "category": "phonetic_balance", "text": "And each year something happened, and I did not go.", "source": "cmu_arctic"}
+{"id": "release500_0351", "category": "phonetic_balance", "text": "Gad, your letter came just in time.", "source": "cmu_arctic"}
+{"id": "release500_0352", "category": "phonetic_balance", "text": "But if Pierre did not return, until tomorrow.", "source": "cmu_arctic"}
+{"id": "release500_0353", "category": "phonetic_balance", "text": "Billinger may arrive in time.", "source": "cmu_arctic"}
+{"id": "release500_0354", "category": "phonetic_balance", "text": "In such a tumbling of values was no time to sell.", "source": "cmu_arctic"}
+{"id": "release500_0355", "category": "phonetic_balance", "text": "There's not an iota of truth in it.", "source": "cmu_arctic"}
+{"id": "release500_0356", "category": "phonetic_balance", "text": "It is dog eat dog, and you ate them up.", "source": "cmu_arctic"}
+{"id": "release500_0357", "category": "phonetic_balance", "text": "You can take a vacation on pay.", "source": "cmu_arctic"}
+{"id": "release500_0358", "category": "phonetic_balance", "text": "I have seen myself that one man contemplated by Pascal's philosophic eye.", "source": "cmu_arctic"}
+{"id": "release500_0359", "category": "phonetic_balance", "text": "Ah, I had forgotten, he exclaimed.", "source": "cmu_arctic"}
+{"id": "release500_0360", "category": "phonetic_balance", "text": "Providence had delivered him through the maelstrom.", "source": "cmu_arctic"}
+{"id": "release500_0361", "category": "phonetic_balance", "text": "She was his now, forever.", "source": "cmu_arctic"}
+{"id": "release500_0362", "category": "phonetic_balance", "text": "It was unobtrusive, yet it was there.", "source": "cmu_arctic"}
+{"id": "release500_0363", "category": "phonetic_balance", "text": "Men like Joe Goose dated existence from drunk to drunk.", "source": "cmu_arctic"}
+{"id": "release500_0364", "category": "phonetic_balance", "text": "I was completely lost in my work.", "source": "cmu_arctic"}
+{"id": "release500_0365", "category": "phonetic_balance", "text": "The very thing, Ernest agreed.", "source": "cmu_arctic"}
+{"id": "release500_0366", "category": "phonetic_balance", "text": "He was the leader, and Tudor was his lieutenant.", "source": "cmu_arctic"}
+{"id": "release500_0367", "category": "phonetic_balance", "text": "The Fire-Men wore animal skins around their waists and across their shoulders.", "source": "cmu_arctic"}
+{"id": "release500_0368", "category": "phonetic_balance", "text": "We could throw stones with our feet.", "source": "cmu_arctic"}
+{"id": "release500_0369", "category": "phonetic_balance", "text": "He turned sharply, and faced Gregson across the table.", "source": "cmu_arctic"}
+{"id": "release500_0370", "category": "phonetic_balance", "text": "They are to attack your camp tomorrow night.", "source": "cmu_arctic"}
+{"id": "release500_0371", "category": "phonetic_balance", "text": "And watch out for wet feet, was his parting advice.", "source": "cmu_arctic"}
+{"id": "release500_0372", "category": "phonetic_balance", "text": "Two weeks passed, and in that time Thorpe left camp three times.", "source": "cmu_arctic"}
+{"id": "release500_0373", "category": "phonetic_balance", "text": "But ever his gaze returned to that Crouched Venus on the piano.", "source": "cmu_arctic"}
+{"id": "release500_0374", "category": "phonetic_balance", "text": "But she had become an automaton.", "source": "cmu_arctic"}
+{"id": "release500_0375", "category": "phonetic_balance", "text": "I do not blame you for anything; remember that.", "source": "cmu_arctic"}
+{"id": "release500_0376", "category": "phonetic_balance", "text": "The next thing to watch out for is bed sores.", "source": "cmu_arctic"}
+{"id": "release500_0377", "category": "phonetic_balance", "text": "He will follow us soon.", "source": "cmu_arctic"}
+{"id": "release500_0378", "category": "phonetic_balance", "text": "His beady black eyes saw bargains where other men saw bankruptcy.", "source": "cmu_arctic"}
+{"id": "release500_0379", "category": "phonetic_balance", "text": "They robbed me a few years later.", "source": "cmu_arctic"}
+{"id": "release500_0380", "category": "phonetic_balance", "text": "Gregson held a lighted match until it burnt his fingertips.", "source": "cmu_arctic"}
+{"id": "release500_0381", "category": "phonetic_balance", "text": "A burst of laughter was his reward.", "source": "cmu_arctic"}
+{"id": "release500_0382", "category": "phonetic_balance", "text": "He cried, and swung the club wildly.", "source": "cmu_arctic"}
+{"id": "release500_0383", "category": "phonetic_balance", "text": "Sandel would never become a world champion.", "source": "cmu_arctic"}
+{"id": "release500_0384", "category": "phonetic_balance", "text": "And after the bath a shave would not be bad.", "source": "cmu_arctic"}
+{"id": "release500_0385", "category": "phonetic_balance", "text": "They saw each other for the first time in Boston.", "source": "cmu_arctic"}
+{"id": "release500_0386", "category": "phonetic_balance", "text": "I had been born with no organic, chemical predisposition toward alcohol.", "source": "cmu_arctic"}
+{"id": "release500_0387", "category": "phonetic_balance", "text": "They had no fixed values, to be altered by adjectives and adverbs.", "source": "cmu_arctic"}
+{"id": "release500_0388", "category": "phonetic_balance", "text": "Nor was Elam Harnish an exception.", "source": "cmu_arctic"}
+{"id": "release500_0389", "category": "phonetic_balance", "text": "He was a wise hyena.", "source": "cmu_arctic"}
+{"id": "release500_0390", "category": "phonetic_balance", "text": "And now behold the perversity of things.", "source": "cmu_arctic"}
+{"id": "release500_0391", "category": "phonetic_balance", "text": "The Fire People, like ourselves, lived in caves.", "source": "cmu_arctic"}
+{"id": "release500_0392", "category": "phonetic_balance", "text": "To her the bridge was tambo, which is the native for taboo.", "source": "cmu_arctic"}
+{"id": "release500_0393", "category": "phonetic_balance", "text": "The voyage was our idea of a good time.", "source": "cmu_arctic"}
+{"id": "release500_0394", "category": "phonetic_balance", "text": "Of course much grumbling went on, and little outbursts were continually occurring.", "source": "cmu_arctic"}
+{"id": "release500_0395", "category": "phonetic_balance", "text": "Get down and dig in.", "source": "cmu_arctic"}
+{"id": "release500_0396", "category": "phonetic_balance", "text": "It was put together in a casual, helter-skelter sort of way.", "source": "cmu_arctic"}
+{"id": "release500_0397", "category": "phonetic_balance", "text": "He deluged me, overwhelmed me with argument.", "source": "cmu_arctic"}
+{"id": "release500_0398", "category": "phonetic_balance", "text": "The girl faced him, her eyes shining with sudden fear.", "source": "cmu_arctic"}
+{"id": "release500_0399", "category": "phonetic_balance", "text": "It will break our hearts and our backs to hoist anchor by hand.", "source": "cmu_arctic"}
+{"id": "release500_0400", "category": "phonetic_balance", "text": "He made no reply as he waited for Whittemore to continue.", "source": "cmu_arctic"}
+{"id": "release500_0401", "category": "phonetic_balance", "text": "Who the devil gave it to you to be judge and jury.", "source": "cmu_arctic"}
+{"id": "release500_0402", "category": "phonetic_balance", "text": "Daylight found himself charmed and made curious by the little man.", "source": "cmu_arctic"}
+{"id": "release500_0403", "category": "phonetic_balance", "text": "It's a Yankee, Joan cried.", "source": "cmu_arctic"}
+{"id": "release500_0404", "category": "phonetic_balance", "text": "But what they want with your toothbrush is more than I can imagine.", "source": "cmu_arctic"}
+{"id": "release500_0405", "category": "phonetic_balance", "text": "They were babbling and chattering all together.", "source": "cmu_arctic"}
+{"id": "release500_0406", "category": "phonetic_balance", "text": "Men of Selden's stamp don't stop at women and children.", "source": "cmu_arctic"}
+{"id": "release500_0407", "category": "phonetic_balance", "text": "Cameron looked at his hands with their long, sinewy fingers.", "source": "cmu_arctic"}
+{"id": "release500_0408", "category": "phonetic_balance", "text": "Saxon waited, for she knew a fresh idea had struck Billy.", "source": "cmu_arctic"}
+{"id": "release500_0409", "category": "phonetic_balance", "text": "He saw only the effect in a general, sketchy way.", "source": "cmu_arctic"}
+{"id": "release500_0410", "category": "phonetic_balance", "text": "He began to follow the footprints of the dog.", "source": "cmu_arctic"}
+{"id": "release500_0411", "category": "phonetic_balance", "text": "Do you value your hide.", "source": "cmu_arctic"}
+{"id": "release500_0412", "category": "phonetic_balance", "text": "A rising tide of fat had submerged them.", "source": "cmu_arctic"}
+{"id": "release500_0413", "category": "phonetic_balance", "text": "There was nothing on the rock.", "source": "cmu_arctic"}
+{"id": "release500_0414", "category": "phonetic_balance", "text": "Whatever he guessed he locked away in the taboo room of Naomi.", "source": "cmu_arctic"}
+{"id": "release500_0415", "category": "phonetic_balance", "text": "Will we ever forget it.", "source": "cmu_arctic"}
+{"id": "release500_0416", "category": "phonetic_balance", "text": "Jeanne and Pierre both gazed toward the great rock.", "source": "cmu_arctic"}
+{"id": "release500_0417", "category": "phonetic_balance", "text": "Does the old boy often go off at half-cock that way.", "source": "cmu_arctic"}
+{"id": "release500_0418", "category": "phonetic_balance", "text": "The temperature dropped to fifty below zero and remained there the whole trip.", "source": "cmu_arctic"}
+{"id": "release500_0419", "category": "phonetic_balance", "text": "It's worth eight dollars.", "source": "cmu_arctic"}
+{"id": "release500_0420", "category": "phonetic_balance", "text": "But I did not enjoy it long.", "source": "cmu_arctic"}
+{"id": "release500_0421", "category": "phonetic_balance", "text": "Shorty turned to their employers.", "source": "cmu_arctic"}
+{"id": "release500_0422", "category": "phonetic_balance", "text": "It was the third or fourth time that Philip had heard MacDougall swear.", "source": "cmu_arctic"}
+{"id": "release500_0423", "category": "phonetic_balance", "text": "Ernest saw in the affair the most sinister import.", "source": "cmu_arctic"}
+{"id": "release500_0424", "category": "phonetic_balance", "text": "Also, at regular intervals, he would mutter.", "source": "cmu_arctic"}
+{"id": "release500_0425", "category": "phonetic_balance", "text": "He may anticipate the day of his death.", "source": "cmu_arctic"}
+{"id": "release500_0426", "category": "phonetic_balance", "text": "I defy any man to get a Solomon Island sore in California.", "source": "cmu_arctic"}
+{"id": "release500_0427", "category": "phonetic_balance", "text": "Now our figuring was all right.", "source": "cmu_arctic"}
+{"id": "release500_0428", "category": "phonetic_balance", "text": "MacDougall tapped his forehead suspiciously with a stubby forefinger.", "source": "cmu_arctic"}
+{"id": "release500_0429", "category": "phonetic_balance", "text": "O'Brien took off his coat and bared his right arm.", "source": "cmu_arctic"}
+{"id": "release500_0430", "category": "phonetic_balance", "text": "But it won't continue, she said with easy confidence.", "source": "cmu_arctic"}
+{"id": "release500_0431", "category": "phonetic_balance", "text": "I was still weak from my prolonged immersion.", "source": "cmu_arctic"}
+{"id": "release500_0432", "category": "phonetic_balance", "text": "Oolong was two hundred and fifty miles from the nearest land.", "source": "cmu_arctic"}
+{"id": "release500_0433", "category": "phonetic_balance", "text": "I just do appreciate it without being able to express my feelings.", "source": "cmu_arctic"}
+{"id": "release500_0434", "category": "phonetic_balance", "text": "Now just what do you want to know.", "source": "cmu_arctic"}
+{"id": "release500_0435", "category": "phonetic_balance", "text": "But she swung obediently on her heel into the wind.", "source": "cmu_arctic"}
+{"id": "release500_0436", "category": "phonetic_balance", "text": "Ask him, I laughed, then turned to Pasquini.", "source": "cmu_arctic"}
+{"id": "release500_0437", "category": "phonetic_balance", "text": "He did not rush in.", "source": "cmu_arctic"}
+{"id": "release500_0438", "category": "phonetic_balance", "text": "They obeyed him, and went here and there at his commands.", "source": "cmu_arctic"}
+{"id": "release500_0439", "category": "phonetic_balance", "text": "Tom Spink has a harpoon.", "source": "cmu_arctic"}
+{"id": "release500_0440", "category": "phonetic_balance", "text": "It was my reports from the north which chiefly induced people to buy.", "source": "cmu_arctic"}
+{"id": "release500_0441", "category": "phonetic_balance", "text": "Cherokee identified himself with his instinct.", "source": "cmu_arctic"}
+{"id": "release500_0442", "category": "phonetic_balance", "text": "I graduated last of my class.", "source": "cmu_arctic"}
+{"id": "release500_0443", "category": "phonetic_balance", "text": "It was a gigantic inadequacy.", "source": "cmu_arctic"}
+{"id": "release500_0444", "category": "phonetic_balance", "text": "Did I possess too much vitality.", "source": "cmu_arctic"}
+{"id": "release500_0445", "category": "phonetic_balance", "text": "Therefore, hurrah for the game.", "source": "cmu_arctic"}
+{"id": "release500_0446", "category": "phonetic_balance", "text": "He had fulfilled his duty and paid properly.", "source": "cmu_arctic"}
+{"id": "release500_0447", "category": "phonetic_balance", "text": "How can you manage all alone, Mr Young.", "source": "cmu_arctic"}
+{"id": "release500_0448", "category": "phonetic_balance", "text": "Red-Eye never committed a more outrageous deed.", "source": "cmu_arctic"}
+{"id": "release500_0449", "category": "phonetic_balance", "text": "He would keep his faith with Gregson for the promised day or two.", "source": "cmu_arctic"}
+{"id": "release500_0450", "category": "phonetic_balance", "text": "There was one chance, and only one, of saving Jeanne.", "source": "cmu_arctic"}
+{"id": "release500_0451", "category": "phonetic_balance", "text": "I was beneath the water, suffocating and drowning.", "source": "cmu_arctic"}
+{"id": "release500_0452", "category": "phonetic_balance", "text": "Give them their choice between a fine or an official whipping.", "source": "cmu_arctic"}
+{"id": "release500_0453", "category": "phonetic_balance", "text": "They must have been swept away by the chaotic currents.", "source": "cmu_arctic"}
+{"id": "release500_0454", "category": "phonetic_balance", "text": "You yellow giant thing of the frost.", "source": "cmu_arctic"}
+{"id": "release500_0455", "category": "phonetic_balance", "text": "He had peeled off his shirt and was wildly waving it.", "source": "cmu_arctic"}
+{"id": "release500_0456", "category": "phonetic_balance", "text": "To these he gave castor oil.", "source": "cmu_arctic"}
+{"id": "release500_0457", "category": "phonetic_balance", "text": "Tudor surveyed him with withering disgust.", "source": "cmu_arctic"}
+{"id": "release500_0458", "category": "phonetic_balance", "text": "After all, it was simply a mistake in judgment.", "source": "cmu_arctic"}
+{"id": "release500_0459", "category": "phonetic_balance", "text": "That is the strange part of it.", "source": "cmu_arctic"}
+{"id": "release500_0460", "category": "phonetic_balance", "text": "Not a wheel moved in his empire.", "source": "cmu_arctic"}
+{"id": "release500_0461", "category": "phonetic_balance", "text": "Its diameter was not more than two hundred yards.", "source": "cmu_arctic"}
+{"id": "release500_0462", "category": "phonetic_balance", "text": "Nimrod replied, with a slight manifestation of sensitiveness.", "source": "cmu_arctic"}
+{"id": "release500_0463", "category": "phonetic_balance", "text": "I tried to read George Moore last night, and was dreadfully bored.", "source": "cmu_arctic"}
+{"id": "release500_0464", "category": "phonetic_balance", "text": "Yes, it was a man who asked, a stranger.", "source": "cmu_arctic"}
+{"id": "release500_0465", "category": "phonetic_balance", "text": "We were now good friends.", "source": "cmu_arctic"}
+{"id": "release500_0466", "category": "phonetic_balance", "text": "Yet, in accordance with Ernest's test of truth, it worked.", "source": "cmu_arctic"}
+{"id": "release500_0467", "category": "phonetic_balance", "text": "It's the strap hangers that'll keep us from going under.", "source": "cmu_arctic"}
+{"id": "release500_0468", "category": "phonetic_balance", "text": "I never allow what can't be changed to annoy me.", "source": "cmu_arctic"}
+{"id": "release500_0469", "category": "phonetic_balance", "text": "After all superfluous flesh is gone what is left is stringy and resistant.", "source": "cmu_arctic"}
+{"id": "release500_0470", "category": "phonetic_balance", "text": "Then he shouted, Shut up.", "source": "cmu_arctic"}
+{"id": "release500_0471", "category": "phonetic_balance", "text": "He had worshiped her, as Dante might have worshiped Beatrice.", "source": "cmu_arctic"}
+{"id": "release500_0472", "category": "phonetic_balance", "text": "He looked like one who had passed through an uncomfortable hour or two.", "source": "cmu_arctic"}
+{"id": "release500_0473", "category": "phonetic_balance", "text": "The history of our westward-faring race is written in it.", "source": "cmu_arctic"}
+{"id": "release500_0474", "category": "phonetic_balance", "text": "He may desire to escape pain, or to enjoy pleasure.", "source": "cmu_arctic"}
+{"id": "release500_0475", "category": "phonetic_balance", "text": "And the air was growing chilly.", "source": "cmu_arctic"}
+{"id": "release500_0476", "category": "phonetic_balance", "text": "Dig in; you're sure good, was Daylight's answer.", "source": "cmu_arctic"}
+{"id": "release500_0477", "category": "phonetic_balance", "text": "My idea was that he would have more influence over the natives.", "source": "cmu_arctic"}
+{"id": "release500_0478", "category": "phonetic_balance", "text": "Now run along, and tell them to hurry.", "source": "cmu_arctic"}
+{"id": "release500_0479", "category": "phonetic_balance", "text": "I have no idea, replied Philip.", "source": "cmu_arctic"}
+{"id": "release500_0480", "category": "phonetic_balance", "text": "For the rest, he was a mere automaton.", "source": "cmu_arctic"}
+{"id": "release500_0481", "category": "phonetic_balance", "text": "Why, the average review is more nauseating than cod liver oil.", "source": "cmu_arctic"}
+{"id": "release500_0482", "category": "phonetic_balance", "text": "I have been robbed, sir, I amended.", "source": "cmu_arctic"}
+{"id": "release500_0483", "category": "phonetic_balance", "text": "He moved away as quietly as he had come.", "source": "cmu_arctic"}
+{"id": "release500_0484", "category": "phonetic_balance", "text": "A trickle of fresh blood ran over his face.", "source": "cmu_arctic"}
+{"id": "release500_0485", "category": "phonetic_balance", "text": "Suppose you saw me at work through the window.", "source": "cmu_arctic"}
+{"id": "release500_0486", "category": "phonetic_balance", "text": "But who was Eileen's double.", "source": "cmu_arctic"}
+{"id": "release500_0487", "category": "phonetic_balance", "text": "Darkness hid him from Jeanne.", "source": "cmu_arctic"}
+{"id": "release500_0488", "category": "phonetic_balance", "text": "By this time Charley was as enraged as the Greek.", "source": "cmu_arctic"}
+{"id": "release500_0489", "category": "phonetic_balance", "text": "I am writing these lines in Honolulu, Hawaii.", "source": "cmu_arctic"}
+{"id": "release500_0490", "category": "phonetic_balance", "text": "Nobody knows how the natives got them.", "source": "cmu_arctic"}
+{"id": "release500_0491", "category": "phonetic_balance", "text": "And how would we ever find ourselves.", "source": "cmu_arctic"}
+{"id": "release500_0492", "category": "phonetic_balance", "text": "They are coming ashore, whoever they are.", "source": "cmu_arctic"}
+{"id": "release500_0493", "category": "phonetic_balance", "text": "This is my fifth voyage.", "source": "cmu_arctic"}
+{"id": "release500_0494", "category": "phonetic_balance", "text": "And yet if she came he had no words to say.", "source": "cmu_arctic"}
+{"id": "release500_0495", "category": "phonetic_balance", "text": "You must sleep, he urged.", "source": "cmu_arctic"}
+{"id": "release500_0496", "category": "phonetic_balance", "text": "Could the incident have anything to do with Jeanne and Pierre.", "source": "cmu_arctic"}
+{"id": "release500_0497", "category": "phonetic_balance", "text": "Each day she became a more vital part of him.", "source": "cmu_arctic"}
+{"id": "release500_0498", "category": "phonetic_balance", "text": "But why continue the tirade, for tirade it was.", "source": "cmu_arctic"}
+{"id": "release500_0499", "category": "phonetic_balance", "text": "That is what distinguishes all of us from the lower animals.", "source": "cmu_arctic"}
+{"id": "release500_0500", "category": "phonetic_balance", "text": "There was something pathetic in the girl's attitude now.", "source": "cmu_arctic"}
diff --git a/evaluation/final/raw/signal_diagnostics/per_clip.csv b/evaluation/final/raw/signal_diagnostics/per_clip.csv
new file mode 100644
index 0000000000000000000000000000000000000000..1be14d1e90bb397f29bb993b84c5e0e90595c818
--- /dev/null
+++ b/evaluation/final/raw/signal_diagnostics/per_clip.csv
@@ -0,0 +1,4001 @@
+model,id,category,text,audio,seconds,words,words_per_second,peak_dbfs,clip_fraction,dc_offset,leading_silence_seconds,trailing_silence_seconds,internal_silence_fraction,tail_to_overall_rms_db,active_hf_ratio_median,active_flatness_median,active_centroid_hz_median,rms_jump_db_p95,voiced_fraction,f0_median_hz,f0_range_semitones_p05_p95,f0_jump_cents_p95,f0_large_jump_fraction
+inflect-micro-v2,release500_0001,conversational,It wasn't until later that I realized what had actually happened.,benchmark/audio/inflect-micro-v2/release500_0001.wav,3.7653333333333334,11,2.921388101983003,-5.460157978443742,0.0,6.477150691353017e-07,0.30933333333333335,0.064,0.10734463276836158,-56.21827916965611,8.197286661015823e-06,0.0005768612027168274,1342.9544221558212,6.275514602661133,0.884180790960452,104.84156990188906,18.571788145067,1582.787477427574,0.18241042345276873
+inflect-micro-v2,release500_0002,descriptive,The light at that time of afternoon is perfect for about twenty minutes.,benchmark/audio/inflect-micro-v2/release500_0002.wav,3.8826666666666667,13,3.3482142857142856,-6.106375335511639,0.0,5.3997878239897545e-06,0.26666666666666666,0.042666666666666665,0.10684931506849316,-55.24061801370631,1.054466065397719e-05,0.0006610722048208117,1354.337052050972,8.601907730102539,0.8821917808219178,111.65863476435113,27.361718654689135,2050.596542712435,0.28846153846153844
+inflect-micro-v2,release500_0003,emotional,"That was the scariest thing I've ever done, and also maybe the best.",benchmark/audio/inflect-micro-v2/release500_0003.wav,3.9146666666666667,13,3.3208446866485013,-3.9648701211978663,0.0,3.951771759602707e-06,0.23466666666666666,0.032,0.11684782608695653,-37.80550426611291,9.252380550606176e-06,0.000639767968095839,1361.2440081711866,5.802735805511475,0.8722826086956522,105.79125407894351,20.30433306157787,1678.1196443324022,0.2108626198083067
+inflect-micro-v2,release500_0004,homographs,"After I record the record, please present the present to the bass player.",benchmark/audio/inflect-micro-v2/release500_0004.wav,4.330666666666667,13,3.001847290640394,-4.7220488509797125,0.0,5.771373707830207e-06,0.32,0.05333333333333334,0.15724815724815724,-49.51076111612834,9.696531378722284e-06,0.0005933775682933629,1311.9140239333015,8.233038902282715,0.828009828009828,108.64429330970336,29.83809294653271,2099.1090748707015,0.2361963190184049
+inflect-micro-v2,release500_0005,long,"If I'm being completely honest, I had no idea what I was doing for most of it, and I think that might be the most normal thing about the whole experience.",benchmark/audio/inflect-micro-v2/release500_0005.wav,8.597333333333333,31,3.605769230769231,-5.366725863777072,0.0,2.953609907763166e-07,0.30933333333333335,0.042666666666666665,0.09417596034696406,-55.95774430864121,6.064137323846808e-06,0.00044433859875425696,1293.8581049105765,6.961612224578857,0.8884758364312267,103.39264433698887,22.25183378632615,1530.5709125323247,0.19313304721030042
+inflect-micro-v2,release500_0006,long_range,The committee postponed its decision after discovering that three pages of the report had been printed in the wrong order.,benchmark/audio/inflect-micro-v2/release500_0006.wav,6.176,20,3.238341968911917,-6.529304709738703,0.0,4.0559982039667375e-08,0.26666666666666666,0.042666666666666665,0.07413793103448275,-56.68609239579027,8.360715582966805e-06,0.00044187536695972085,1203.7868531991887,6.712539196014404,0.9137931034482759,109.5031485942737,25.318530656002032,1827.7097490361725,0.21772639691714837
+inflect-micro-v2,release500_0007,names_places,Gwendolyn photographed the eucalyptus trees outside Ljubljana.,benchmark/audio/inflect-micro-v2/release500_0007.wav,3.808,7,1.8382352941176472,-7.120061231444968,0.0,5.14269549967139e-06,0.3626666666666667,0.042666666666666665,0.13687150837988826,-43.318092848621234,5.87698923482094e-06,0.00038932025199756026,1233.5658594944734,6.499547958374023,0.8547486033519553,101.64778840911565,23.538644877019763,1396.4706680480315,0.18394648829431437
+inflect-micro-v2,release500_0008,narrative,"He had this theory, and the more time passes, the more I think he was right.",benchmark/audio/inflect-micro-v2/release500_0008.wav,4.352,16,3.676470588235294,-5.446252606127623,0.0,4.486418220039923e-06,0.26666666666666666,0.10666666666666667,0.15403422982885084,-61.719585777805094,9.74559134192532e-06,0.0005889612948521972,1355.0874272184953,6.259807109832764,0.8361858190709046,108.23695526577308,23.295247506598756,1361.8770961488874,0.15178571428571427
+inflect-micro-v2,release500_0009,numbers,The package weighs twelve point six kilograms and arrived on July twenty-first.,benchmark/audio/inflect-micro-v2/release500_0009.wav,4.789333333333333,12,2.505567928730512,-5.2806148348155615,0.0,9.967914593289606e-06,0.2773333333333333,0.042666666666666665,0.08888888888888889,-59.03366640548461,1.4970843949413393e-05,0.0007844885112717748,1495.112554943683,7.262706756591797,0.8977777777777778,120.32508112660759,25.4915830177164,2160.803894928649,0.26582278481012656
+inflect-micro-v2,release500_0010,punctuation,"First, close the window; second, turn off the lamp; finally, lock the door.",benchmark/audio/inflect-micro-v2/release500_0010.wav,5.312,13,2.447289156626506,-4.904741601269506,0.0,5.243293799139792e-06,0.2986666666666667,0.064,0.2545090180360721,-58.04128484756043,9.372090062242933e-06,0.0005291037959977984,1239.4205290180862,7.151742935180664,0.7334669338677354,113.43627616051332,29.46940201928679,2156.8524656377276,0.25925925925925924
+inflect-micro-v2,release500_0011,question,When did you first realize something was wrong?,benchmark/audio/inflect-micro-v2/release500_0011.wav,2.570666666666667,8,3.1120331950207465,-7.223563179332455,0.0,3.4575639347167453e-06,0.224,0.042666666666666665,0.10330578512396695,-55.301695887148185,6.760375526937423e-06,0.00045156554551795125,1293.0630308243813,4.692057132720947,0.8842975206611571,117.87289818337246,20.2306340499229,1354.060431836684,0.17061611374407584
+inflect-micro-v2,release500_0012,rare_words,The archaeologist catalogued a Byzantine reliquary and several anachronistic fragments.,benchmark/audio/inflect-micro-v2/release500_0012.wav,5.184,10,1.9290123456790123,-5.309386581003772,0.0,8.570068530389108e-06,0.2773333333333333,0.042666666666666665,0.07186858316221766,-53.700892684301934,9.205719834426418e-06,0.0005791863659396768,1349.5505396417157,6.295018196105957,0.8993839835728953,106.3154695684525,27.70062612445557,2092.395128905002,0.24178403755868544
+inflect-micro-v2,release500_0013,short,I have no idea.,benchmark/audio/inflect-micro-v2/release500_0013.wav,1.2693333333333334,4,3.1512605042016806,-7.160467803585357,0.0,1.4606644072046038e-05,0.256,0.042666666666666665,0.23333333333333334,-44.016850958093094,8.470080501865596e-06,0.0006591779529117048,1449.9798463093439,3.353656768798828,0.75,99.07759713396041,19.42988364680846,1155.6608790596124,0.09090909090909091
+inflect-micro-v2,release500_0014,technical,The system runs on three core components that all have to stay in sync.,benchmark/audio/inflect-micro-v2/release500_0014.wav,4.010666666666666,14,3.4906914893617023,-7.140542567496708,0.0,5.266133939585416e-06,0.26666666666666666,0.05333333333333334,0.08753315649867374,-53.12070043147998,1.6069125194917433e-05,0.0008734681759960949,1454.4935678007885,7.285770416259766,0.896551724137931,112.65994371761101,27.149186023284017,1675.3491850552073,0.2682926829268293
+inflect-micro-v2,release500_0015,conversational,We improvised the entire second half and somehow nobody noticed.,benchmark/audio/inflect-micro-v2/release500_0015.wav,4.0,10,2.5,-6.402631496934088,0.0,1.1434554835432209e-05,0.21333333333333335,0.042666666666666665,0.08776595744680851,-59.0910456800405,1.0559193469816819e-05,0.0006226259283721447,1374.1849591505506,5.471024036407471,0.9095744680851063,100.46956020592981,21.96042555449734,2098.3184876842633,0.23964497041420119
+inflect-micro-v2,release500_0016,descriptive,There's a smell in old buildings that's impossible to describe but immediately familiar.,benchmark/audio/inflect-micro-v2/release500_0016.wav,4.618666666666667,13,2.814665127020785,-6.386580847226746,0.0,6.122071681602392e-06,0.224,0.05333333333333334,0.07834101382488479,-59.57957076410325,6.322057743091136e-06,0.0004347427748143673,1244.2448537884252,6.480621337890625,0.9101382488479263,111.74185928190812,20.60523008458525,1802.8778794510179,0.24611398963730569
+inflect-micro-v2,release500_0017,emotional,I specifically asked for one thing. One thing. And it still didn't happen.,benchmark/audio/inflect-micro-v2/release500_0017.wav,4.266666666666667,13,3.046875,-6.597589026349956,0.0,8.265375981864054e-06,0.26666666666666666,0.042666666666666665,0.1396508728179551,-54.94015430770592,1.20283366413787e-05,0.0008570057107135653,1408.3999059620728,7.8317975997924805,0.8528678304239401,104.27108423316443,29.766409482920523,1964.4841804631337,0.21921921921921922
+inflect-micro-v2,release500_0018,homographs,They refuse to place the refuse beside the produce aisle.,benchmark/audio/inflect-micro-v2/release500_0018.wav,3.328,10,3.0048076923076925,-5.296208020962383,0.0,5.007172148907557e-06,0.24533333333333332,0.042666666666666665,0.10862619808306709,-45.14504616094448,1.0631640179781243e-05,0.000671149231493473,1466.562393371335,7.484850883483887,0.8753993610223643,112.53719178104053,25.955710711644727,1789.564475058747,0.2631578947368421
+inflect-micro-v2,release500_0019,long,"There were days last year when I genuinely didn't know how I was going to get through it, and somehow I did, and I still don't fully understand how.",benchmark/audio/inflect-micro-v2/release500_0019.wav,8.010666666666667,29,3.62017310252996,-6.937281388466152,0.0,7.590663244627649e-07,0.23466666666666666,0.042666666666666665,0.09042553191489362,-51.36984609850313,6.102260158513673e-06,0.0004479991621337831,1264.6516049247648,5.193220615386963,0.9002659574468085,106.53537352327587,21.62847302670559,1616.1320096527245,0.17293233082706766
+inflect-micro-v2,release500_0020,long_range,"Before announcing the winner, the judge thanked the volunteers who had worked through the storm to repair the damaged stage.",benchmark/audio/inflect-micro-v2/release500_0020.wav,6.261333333333333,20,3.194207836456559,-6.154157647055454,0.0,7.0798537308292e-06,0.256,0.042666666666666665,0.09863945578231292,-60.000875935481794,7.777547580190003e-06,0.0004876391321886331,1280.6465973088505,5.997303485870361,0.8979591836734694,109.08137551772475,27.174476436311764,1978.750768979528,0.24277456647398843
+inflect-micro-v2,release500_0021,names_places,Professor Nguyen mailed the fluorescent poster to Reykjavik on Thursday.,benchmark/audio/inflect-micro-v2/release500_0021.wav,3.936,10,2.540650406504065,-6.2076194139051735,0.0,6.021845706527529e-07,0.23466666666666666,0.042666666666666665,0.07837837837837838,-45.961846409916234,7.724984243395738e-06,0.0005416724015958607,1273.5059987034065,7.3361496925354,0.9162162162162162,108.02074002991526,22.68725764980093,2004.300436733963,0.26047904191616766
+inflect-micro-v2,release500_0022,narrative,The plan was simple. The execution was anything but.,benchmark/audio/inflect-micro-v2/release500_0022.wav,3.4133333333333336,9,2.63671875,-6.062049939434123,0.0,6.629898962273728e-06,0.288,0.032,0.19003115264797507,-37.81774937312908,2.5672397896414623e-05,0.0012204665690660477,1494.0826611275174,8.927072525024414,0.8037383177570093,123.94366768222469,22.844227295493276,1986.357654265395,0.2896825396825397
+inflect-micro-v2,release500_0023,numbers,Her score improved from eighty-two percent to ninety-seven percent in two weeks.,benchmark/audio/inflect-micro-v2/release500_0023.wav,4.576,12,2.6223776223776225,-6.274503296866776,0.0,1.4913665609128657e-06,0.224,0.032,0.08604651162790698,-55.666863216250164,8.935863661463372e-06,0.0005492755444720387,1331.5386136941481,7.790089130401611,0.9,110.72889346585038,29.462885420342328,2076.0159210845613,0.2765957446808511
+inflect-micro-v2,release500_0024,punctuation,"Why would anyone pack spoons, batteries, and a violin for one short trip?",benchmark/audio/inflect-micro-v2/release500_0024.wav,4.469333333333333,13,2.908711217183771,-5.267965951064037,0.0,5.836133823322598e-06,0.224,0.07466666666666667,0.14523809523809525,-58.986343274150656,9.502223292656709e-06,0.000516528554726392,1278.402791531116,6.328792572021484,0.8476190476190476,110.20208888673191,22.043043411480106,1763.3858190529543,0.2
+inflect-micro-v2,release500_0025,question,How long have you been sitting on this?,benchmark/audio/inflect-micro-v2/release500_0025.wav,2.2186666666666666,8,3.605769230769231,-6.504606544973271,0.0,7.842595550755505e-06,0.288,0.042666666666666665,0.14832535885167464,-61.466553048992544,1.0521267540752888e-05,0.0006544959032908082,1310.500698974793,3.825904369354248,0.84688995215311,108.46101737490847,15.794846426545899,1319.5496168233526,0.13714285714285715
+inflect-micro-v2,release500_0026,rare_words,A mischievous chameleon rested beside the chrysanthemum and the bougainvillea.,benchmark/audio/inflect-micro-v2/release500_0026.wav,4.512,10,2.2163120567375887,-6.064713750795532,0.0,8.742014756535355e-07,0.288,0.042666666666666665,0.10377358490566038,-53.51023726416007,7.222417480079457e-06,0.000561561668291688,1321.5384972074517,6.52694845199585,0.8820754716981132,106.53829769461427,27.367149452817365,1703.0478680381266,0.21195652173913043
+inflect-micro-v2,release500_0027,short,We don't talk about that.,benchmark/audio/inflect-micro-v2/release500_0027.wav,1.568,5,3.1887755102040813,-5.2127218784968825,0.0,3.7108959531906294e-06,0.20266666666666666,0.08533333333333333,0.20270270270270271,-51.77603384279066,7.564104635093827e-06,0.0004298678250052035,1294.2447818408036,7.5788469314575195,0.7635135135135135,118.70044395978316,26.815724632280418,1502.530630883944,0.21495327102803738
+inflect-micro-v2,release500_0028,technical,Hash the UTF eight payload with S H A two fifty-six before opening the checkpoint.,benchmark/audio/inflect-micro-v2/release500_0028.wav,5.397333333333333,15,2.7791501976284585,-4.018593117960659,0.0,9.527791007712949e-06,0.2773333333333333,0.05333333333333334,0.09270216962524655,-56.33991699429947,1.4199055840435904e-05,0.0008410369628109038,1514.2770199668403,8.103915214538574,0.883629191321499,106.04015671122445,30.053175113526237,2176.7569384651906,0.3325635103926097
+inflect-micro-v2,release500_0029,conversational,"I thought I had it all figured out, and then this happened.",benchmark/audio/inflect-micro-v2/release500_0029.wav,3.050666666666667,12,3.9335664335664333,-4.209714662440567,0.0,1.945362237165682e-05,0.288,0.042666666666666665,0.18466898954703834,-46.33982655306088,1.2763152881234419e-05,0.0007900168420746922,1429.107916300305,7.248868942260742,0.8118466898954704,110.27757783896486,24.640637883989022,1662.205994422168,0.2052401746724891
+inflect-micro-v2,release500_0030,descriptive,The kind of quiet you only get at three in the morning in a city.,benchmark/audio/inflect-micro-v2/release500_0030.wav,3.530666666666667,15,4.248489425981873,-8.007444366002744,0.0,3.646507821031264e-06,0.256,0.07466666666666667,0.10843373493975904,-48.37573620509471,7.557007847935893e-06,0.0005015148781239986,1316.8840030539918,6.110589027404785,0.8795180722891566,102.83137858617287,22.162777413015537,1709.1792508279268,0.21403508771929824
+inflect-micro-v2,release500_0031,emotional,"The joy of it was almost overwhelming, like too much good all at once.",benchmark/audio/inflect-micro-v2/release500_0031.wav,4.2026666666666666,14,3.3312182741116754,-6.6327867972316685,0.0,3.4476900054869475e-06,0.24533333333333332,0.042666666666666665,0.1189873417721519,-55.41034871518745,3.882310011249501e-06,0.00029221014119684696,1132.473895932417,7.330234050750732,0.8632911392405064,107.28662927830446,29.117300836182345,1694.071166722954,0.20481927710843373
+inflect-micro-v2,release500_0032,homographs,Please close the door while I stand close to the microphone.,benchmark/audio/inflect-micro-v2/release500_0032.wav,3.36,11,3.2738095238095237,-6.721410341559225,0.0,2.109815113726654e-06,0.24533333333333332,0.042666666666666665,0.08544303797468354,-54.19799066636617,8.372890079044737e-06,0.0005669445381499827,1378.1765485972458,8.324507713317871,0.8987341772151899,103.78358923381269,22.10267906553359,1915.3709545120466,0.22661870503597123
+inflect-micro-v2,release500_0033,long,I think the most underrated skill in the world is knowing when to stop explaining yourself and just let people draw their own conclusions.,benchmark/audio/inflect-micro-v2/release500_0033.wav,7.616,24,3.151260504201681,-6.369455991691987,0.0,1.3710738357985974e-06,0.2986666666666667,0.042666666666666665,0.07692307692307693,-50.587179586371,5.68081668461673e-06,0.00040980795165523887,1203.515105522838,6.685079574584961,0.9076923076923077,104.73327498052419,24.52603130105573,1904.125622622032,0.2555205047318612
+inflect-micro-v2,release500_0034,long_range,"When the final bus disappeared around the corner, Nadia remembered that her umbrella, notebook, and lunch were still inside the quiet office.",benchmark/audio/inflect-micro-v2/release500_0034.wav,7.8933333333333335,22,2.7871621621621623,-4.931825199311629,0.0,6.40782127447892e-06,0.23466666666666666,0.042666666666666665,0.1106612685560054,-55.33441240057785,5.60790567760705e-06,0.0003591288405004889,1155.6978668484487,6.0767316818237305,0.8825910931174089,110.69782674664893,18.140981029028698,1501.5380660219369,0.19626168224299065
+inflect-micro-v2,release500_0035,names_places,Aurelia and Joaquin crossed the Schuylkill River before noon.,benchmark/audio/inflect-micro-v2/release500_0035.wav,3.616,9,2.4889380530973453,-5.828183926640476,0.0,1.2172709830338135e-05,0.288,0.064,0.12058823529411765,-54.96352925732947,6.710326942993561e-06,0.0003908382204826921,1161.2628787735346,7.209028244018555,0.8676470588235294,107.76527252936545,23.074170652303238,1745.8892638941295,0.17993079584775087
+inflect-micro-v2,release500_0036,narrative,"The village had a name, but nobody used it. Everyone just called it home.",benchmark/audio/inflect-micro-v2/release500_0036.wav,4.149333333333334,14,3.3740359897172234,-5.320628554474607,0.0,7.766071576043032e-06,0.2773333333333333,0.042666666666666665,0.12307692307692308,-61.25448026951275,7.084967364789918e-06,0.0004470736312214285,1280.410118182319,5.307100296020508,0.8692307692307693,102.82107322549888,25.543062083678315,1662.7352965912191,0.16816816816816818
+inflect-micro-v2,release500_0037,numbers,"Increase the gain by three decibels, wait five seconds, then reduce it by one.",benchmark/audio/inflect-micro-v2/release500_0037.wav,4.864,14,2.8782894736842106,-5.165991058054342,0.0,1.033068747346988e-05,0.26666666666666666,0.042666666666666665,0.13347921225382933,-44.69497327276418,1.0627739357005339e-05,0.000682541576679796,1416.4362054851151,6.584777355194092,0.8577680525164114,109.57856674882495,23.56177467530389,1676.649604275774,0.2552083333333333
+inflect-micro-v2,release500_0038,punctuation,"The result was unexpected: clear, stable, and surprisingly natural.",benchmark/audio/inflect-micro-v2/release500_0038.wav,4.522666666666667,9,1.9899764150943395,-4.962747542565468,0.0,9.753231324793887e-07,0.26666666666666666,0.042666666666666665,0.17411764705882352,-48.42709752413231,1.557034920551814e-05,0.0007313269888982177,1405.9857607674733,8.013320922851562,0.8070588235294117,118.48887923247473,27.61760386390832,1748.1871029233075,0.26506024096385544
+inflect-micro-v2,release500_0039,question,Who told you that was the right way to do it?,benchmark/audio/inflect-micro-v2/release500_0039.wav,2.3573333333333335,11,4.66628959276018,-4.9468041030247925,0.0,7.549022484454326e-06,0.21333333333333335,0.07466666666666667,0.12162162162162163,-64.16291815330155,5.1352690206840634e-06,0.00028617028146982193,1145.6982559658293,5.863454818725586,0.8648648648648649,120.73312036582443,22.9914145352421,1837.0780737980156,0.20212765957446807
+inflect-micro-v2,release500_0040,rare_words,The meticulous cartographer examined an iridescent archipelago through binoculars.,benchmark/audio/inflect-micro-v2/release500_0040.wav,4.810666666666667,9,1.870842572062084,-5.300110695395044,0.0,4.7874027586658485e-06,0.2773333333333333,0.042666666666666665,0.07300884955752213,-51.940860369476,8.356576472579036e-06,0.00042701896745711565,1192.5402506599914,7.096024513244629,0.9203539823008849,113.05888643893496,27.019974089130937,1986.2907092551486,0.24817518248175183
+inflect-micro-v2,release500_0041,short,I mean. Yeah. Kind of.,benchmark/audio/inflect-micro-v2/release500_0041.wav,1.5573333333333332,5,3.2106164383561646,-5.296695759381697,0.0,1.3535153811972123e-05,0.256,0.07466666666666667,0.22448979591836735,-56.448265118737275,9.069288353202865e-06,0.0005773367010988295,1388.4487265094103,7.834513187408447,0.7619047619047619,100.40271224499025,23.555872548023252,2410.264027043884,0.1834862385321101
+inflect-micro-v2,release500_0042,technical,The logs are your best friend when something behaves unexpectedly.,benchmark/audio/inflect-micro-v2/release500_0042.wav,3.7226666666666666,10,2.6862464183381087,-6.060984643591976,0.0,1.0242434655083343e-05,0.26666666666666666,0.032,0.09428571428571429,-37.36414209335195,7.458490927092498e-06,0.0005155255203135312,1296.0260242311258,6.898351669311523,0.8942857142857142,102.07551017205849,22.488842266092647,1665.8460868684592,0.247557003257329
+inflect-micro-v2,release500_0043,conversational,He kept talking and I kept nodding and neither of us was listening.,benchmark/audio/inflect-micro-v2/release500_0043.wav,3.968,13,3.276209677419355,-6.289254315931033,0.0,4.4841281123808585e-06,0.26666666666666666,0.05333333333333334,0.14745308310991956,-61.19043450497739,1.2422139661794063e-05,0.0006752352928742766,1454.4160939086,7.128516674041748,0.8445040214477212,114.31907374354415,22.678085108861143,1697.2466379639711,0.21568627450980393
+inflect-micro-v2,release500_0044,descriptive,It's one of those places where everyone seems to know each other already.,benchmark/audio/inflect-micro-v2/release500_0044.wav,3.989333333333333,13,3.2586898395721926,-5.889586948788591,0.0,5.4396409723267425e-06,0.30933333333333335,0.032,0.09333333333333334,-42.50895359761814,6.880139153508935e-06,0.00048481859266757965,1335.555123955016,4.758169174194336,0.8933333333333333,110.69573579013888,27.42399365215168,1855.4767339089076,0.20909090909090908
+inflect-micro-v2,release500_0045,emotional,He said he was proud of me and nobody had ever said that to me before.,benchmark/audio/inflect-micro-v2/release500_0045.wav,4.1386666666666665,16,3.865979381443299,-5.136731883139482,0.0,7.663191354367882e-06,0.24533333333333332,0.042666666666666665,0.08740359897172237,-59.60043002997085,9.54765437199967e-06,0.0005041997064836323,1313.4044339523898,5.368055820465088,0.9048843187660668,102.36606169806186,22.94910266883349,1681.982148446075,0.17971014492753623
+inflect-micro-v2,release500_0046,homographs,The minute detail took a minute to explain clearly.,benchmark/audio/inflect-micro-v2/release500_0046.wav,2.9546666666666668,9,3.0460288808664258,-7.892257494099615,0.0,5.407023309089709e-06,0.26666666666666666,0.032,0.11510791366906475,-35.60085703385026,7.136907697713468e-06,0.0004264797316864133,1279.511665560662,8.576972007751465,0.8776978417266187,113.01685049475608,23.27791946172104,1709.4013988773734,0.3054393305439331
+inflect-micro-v2,release500_0047,long,"The strange thing about really wanting something is that getting it never quite feels the way you imagined it would, and not getting it never quite feels as bad.",benchmark/audio/inflect-micro-v2/release500_0047.wav,8.192,29,3.5400390625,-6.63620061476732,0.0,3.283998694314505e-06,0.2773333333333333,0.032,0.07802340702210664,-37.86097852717357,8.89459261088632e-06,0.0005891644977964461,1348.801968651946,6.744047164916992,0.9063719115734721,107.65976246402545,20.01485756615676,1645.746952975737,0.2153392330383481
+inflect-micro-v2,release500_0048,long_range,"Even though the recipe looked simple at first, preparing the sauce required patience, careful timing, and a much larger pan.",benchmark/audio/inflect-micro-v2/release500_0048.wav,7.424,20,2.693965517241379,-6.125671364858809,0.0,1.8362683249506517e-06,0.2773333333333333,0.042666666666666665,0.1291248206599713,-43.2357846502947,8.963044820120558e-06,0.0004977330681867898,1336.0209642902207,6.9987053871154785,0.8579626972740315,111.76826885653907,26.665807991695008,2039.7081192385033,0.24871355060034306
+inflect-micro-v2,release500_0049,names_places,Xiomara met Siobhan near the old library in Worcestershire.,benchmark/audio/inflect-micro-v2/release500_0049.wav,3.296,9,2.7305825242718447,-5.579765066658055,0.0,1.557822315589874e-06,0.26666666666666666,0.05333333333333334,0.0967741935483871,-64.71685392024563,4.443759280547965e-06,0.00023285936913453043,1128.8505298366465,4.107936382293701,0.896774193548387,101.56909731711696,28.781914938684274,1870.6676655312706,0.19636363636363635
+inflect-micro-v2,release500_0050,narrative,"She walked into the room and everyone went quiet, not because she was loud, but because she wasn't.",benchmark/audio/inflect-micro-v2/release500_0050.wav,5.578666666666667,18,3.2265774378585084,-6.447063537748964,0.0,3.074824803661613e-07,0.20266666666666666,0.05333333333333334,0.13931297709923665,-55.60603380848783,7.701374670432415e-06,0.00042714845039881766,1201.2316182215432,6.907410621643066,0.8454198473282443,110.94396843676117,23.167197245206207,1883.4319642192038,0.1920374707259953
+inflect-micro-v2,release500_0051,numbers,"At exactly seven forty-five, Morgan placed eighteen dollars and fifty cents beside the notebook.",benchmark/audio/inflect-micro-v2/release500_0051.wav,5.834666666666666,14,2.3994515539305303,-6.685852341582236,0.0,1.2212958608870395e-06,0.2986666666666667,0.05333333333333334,0.11131386861313869,-49.83006109359113,8.787067599769216e-06,0.0006018251879140735,1379.0753977717661,7.181534290313721,0.8722627737226277,105.76623366437548,25.40122420350201,1942.2628125958856,0.23974082073434125
+inflect-micro-v2,release500_0052,punctuation,"Does punctuation work? Yes: commas, semicolons, and periods should create sensible boundaries.",benchmark/audio/inflect-micro-v2/release500_0052.wav,6.272,12,1.913265306122449,-5.694401807640749,0.0,2.211048467870569e-06,0.26666666666666666,0.042666666666666665,0.15619694397283532,-50.695935307457674,1.1151200851600152e-05,0.0007475171587429941,1445.9955411403616,6.759486198425293,0.831918505942275,116.4004573197468,28.21978293362941,1742.3201567938293,0.28870292887029286
+inflect-micro-v2,release500_0053,question,Have you ever just stopped and thought about how strange all of this is?,benchmark/audio/inflect-micro-v2/release500_0053.wav,3.808,14,3.6764705882352944,-7.0082690993030505,0.0,8.478886229568161e-06,0.288,0.05333333333333334,0.12290502793296089,-61.05336972402216,1.000539032247616e-05,0.0006016366533003747,1381.3304922133934,6.589339256286621,0.8715083798882681,103.21950232395352,18.921134070082523,1449.5891608692712,0.21241830065359477
+inflect-micro-v2,release500_0054,rare_words,Her idiosyncratic explanation sounded plausible but remained fundamentally ambiguous.,benchmark/audio/inflect-micro-v2/release500_0054.wav,5.301333333333333,9,1.6976861167002013,-6.3683523208343304,0.0,5.871716894034762e-06,0.23466666666666666,0.042666666666666665,0.07630522088353414,-56.18844500059904,6.909112926223315e-06,0.000501016853377223,1288.5431863996364,5.0988993644714355,0.9076305220883534,105.5409069698258,23.9637318302691,2009.250992261829,0.22954545454545455
+inflect-micro-v2,release500_0055,short,That tracks.,benchmark/audio/inflect-micro-v2/release500_0055.wav,0.9493333333333334,2,2.106741573033708,-4.512761916841931,0.0,1.1582053048186935e-05,0.192,0.042666666666666665,0.26666666666666666,-55.798878064379764,4.360371895018034e-05,0.0016248122556135058,1714.6352529347478,7.603741645812988,0.7222222222222222,109.00508510878603,25.3702981700776,2047.7902232574645,0.41935483870967744
+inflect-micro-v2,release500_0056,technical,The key thing to remember is that order matters here more than you'd think.,benchmark/audio/inflect-micro-v2/release500_0056.wav,3.7013333333333334,14,3.782420749279539,-7.0076751313992345,0.0,3.907797236024635e-06,0.26666666666666666,0.05333333333333334,0.09770114942528736,-56.47267048303071,5.438480911834631e-06,0.0002799142093863338,1160.9759958148966,7.755943775177002,0.896551724137931,108.81713684723226,18.76443202361631,1533.009314410591,0.18241042345276873
+inflect-micro-v2,release500_0057,conversational,I would not say that again in this room if I were you.,benchmark/audio/inflect-micro-v2/release500_0057.wav,2.8693333333333335,13,4.530669144981412,-5.191975124878869,0.0,5.865628736501094e-06,0.2986666666666667,0.032,0.11851851851851852,-43.348372377948685,7.78424055170035e-06,0.00048645195784047246,1310.1589259357715,4.693772315979004,0.8777777777777778,104.20175570909875,15.995955411046086,1246.5151859200969,0.14102564102564102
+inflect-micro-v2,release500_0058,descriptive,The room was filled with the quiet that means something is about to change.,benchmark/audio/inflect-micro-v2/release500_0058.wav,3.7333333333333334,14,3.75,-5.22335569017773,0.0,9.492465551375062e-07,0.2773333333333333,0.05333333333333334,0.10256410256410256,-62.38603360848984,9.356496775581036e-06,0.0005453815683722496,1297.1765257118957,7.150337219238281,0.8831908831908832,113.19516333085106,23.06709908685133,1566.2670127322394,0.23588039867109634
+inflect-micro-v2,release500_0059,emotional,"I didn't expect to feel that way about it, but I did.",benchmark/audio/inflect-micro-v2/release500_0059.wav,3.1466666666666665,12,3.8135593220338984,-5.31671665230273,0.0,1.203407691718894e-06,0.288,0.042666666666666665,0.20608108108108109,-54.28915562173576,7.145058134483406e-06,0.0005005224957130849,1356.1388177820481,7.659337043762207,0.7871621621621622,114.41183793101382,23.612442667981234,1649.9221935053224,0.22026431718061673
+inflect-micro-v2,release500_0060,homographs,I object to the strange object that was left near the entrance.,benchmark/audio/inflect-micro-v2/release500_0060.wav,3.6266666666666665,12,3.308823529411765,-5.175124739230448,0.0,6.761621079931501e-06,0.30933333333333335,0.042666666666666665,0.1348973607038123,-60.21242300478442,9.679064532974735e-06,0.0006790068000555038,1493.201789832254,5.652704238891602,0.8563049853372434,106.1042598891778,27.845083880508724,2082.8863877363297,0.24825174825174826
+inflect-micro-v2,release500_0061,long,"I've learned more from the things that went wrong than from the things that worked, which is a frustrating way to have to learn things but apparently the only way that stuck.",benchmark/audio/inflect-micro-v2/release500_0061.wav,8.256,32,3.875968992248062,-5.989905821811764,0.0,2.614287495816825e-06,0.32,0.05333333333333334,0.09935483870967741,-58.733215860933896,8.132670700433664e-06,0.00047079194337129593,1259.7496291666134,6.686064720153809,0.895483870967742,110.2392114575906,22.065896578147004,1637.3240349707849,0.18475073313782991
+inflect-micro-v2,release500_0062,long_range,"Because every nearby hotel was full, the exhausted hikers followed the narrow road until they found a farmhouse with one empty room.",benchmark/audio/inflect-micro-v2/release500_0062.wav,7.1146666666666665,22,3.0922038980509745,-5.516503394271598,0.0,8.989845667883856e-08,0.224,0.032,0.04640718562874251,-54.59097005161701,7.301493496925104e-06,0.0004044693196192384,1215.69310504227,6.372107982635498,0.9431137724550899,108.61687156637942,21.06321316915425,1404.5169548749964,0.18006430868167203
+inflect-micro-v2,release500_0063,names_places,The itinerary sends Beatrice from Albuquerque to Des Moines and then to Boise.,benchmark/audio/inflect-micro-v2/release500_0063.wav,4.405333333333333,13,2.9509685230024214,-7.0720595086239815,0.0,2.140860260624322e-06,0.256,0.032,0.09420289855072464,-38.764099286584866,8.055343641899526e-06,0.00041137481457553804,1206.506035448177,6.941055774688721,0.893719806763285,107.33792533344257,22.66000114380682,1812.330116966549,0.22375690607734808
+inflect-micro-v2,release500_0064,narrative,We had been driving for six hours when we finally saw the sign for the town.,benchmark/audio/inflect-micro-v2/release500_0064.wav,4.181333333333333,16,3.8265306122448983,-6.5845679737777765,0.0,6.27734243607847e-06,0.23466666666666666,0.032,0.06361323155216285,-43.80786614895892,8.898279702407308e-06,0.0005630787927657366,1374.7471872874025,5.536947250366211,0.9287531806615776,102.28839712194772,24.9481137258613,2083.391139149038,0.1994459833795014
+inflect-micro-v2,release500_0065,numbers,Call extension zero four nine after eight thirty in the morning.,benchmark/audio/inflect-micro-v2/release500_0065.wav,3.6053333333333333,11,3.05103550295858,-5.451712776380222,0.0,2.3450372736988356e-06,0.21333333333333335,0.05333333333333334,0.10029498525073746,-51.99607412838313,7.05116235621972e-06,0.00047615086077712476,1308.5589260336876,6.220329761505127,0.8790560471976401,106.4956658205351,25.751298865987096,1777.305365849731,0.17525773195876287
+inflect-micro-v2,release500_0066,punctuation,Stop - listen carefully - then answer the question without guessing.,benchmark/audio/inflect-micro-v2/release500_0066.wav,3.498666666666667,11,3.1440548780487805,-6.4264854639901685,0.0,3.6860385534964735e-06,0.24533333333333332,0.05333333333333334,0.0911854103343465,-62.77581048849324,1.4645199371443596e-05,0.0008219376322813332,1465.0680193678768,7.157771110534668,0.8844984802431611,124.21236690483958,29.52625230389663,2115.8014126931043,0.2968197879858657
+inflect-micro-v2,release500_0067,question,"Does it matter at this point, honestly?",benchmark/audio/inflect-micro-v2/release500_0067.wav,2.3253333333333335,7,3.010321100917431,-7.636943288991138,0.0,6.245910753932549e-06,0.23466666666666666,0.042666666666666665,0.2191780821917808,-53.25481285593354,1.4505219041893724e-05,0.0009121566545218229,1500.204235449198,7.212636947631836,0.771689497716895,120.55232183159056,17.899046342321665,1251.949011802398,0.19631901840490798
+inflect-micro-v2,release500_0068,rare_words,An entrepreneurial meteorologist described the unprecedented atmospheric phenomenon.,benchmark/audio/inflect-micro-v2/release500_0068.wav,5.1306666666666665,8,1.5592515592515592,-6.126207978063691,0.0,4.684355189965572e-06,0.30933333333333335,0.042666666666666665,0.08298755186721991,-55.22264472838067,5.289864020596724e-06,0.0003292535257060081,1230.1139068508046,5.808292388916016,0.8983402489626556,105.03517814683768,27.908637983290667,2084.9850592445396,0.2327790973871734
+inflect-micro-v2,release500_0069,short,That's not what I expected.,benchmark/audio/inflect-micro-v2/release500_0069.wav,1.728,5,2.8935185185185186,-6.343006519840621,0.0,1.1495601938804612e-05,0.24533333333333332,0.05333333333333334,0.1901840490797546,-51.64158267307571,1.3511838915292174e-05,0.0009389698971062899,1535.4435492689568,8.385933876037598,0.7914110429447853,115.72710770188465,32.05845086466695,1983.4064281704048,0.34959349593495936
+inflect-micro-v2,release500_0070,technical,There's a known issue with that version that causes failures under load.,benchmark/audio/inflect-micro-v2/release500_0070.wav,3.6693333333333333,12,3.270348837209302,-5.707687675073423,0.0,3.0027572393009905e-06,0.23466666666666666,0.05333333333333334,0.0927536231884058,-56.48307275709452,6.2612516558147036e-06,0.00045303202932700515,1286.797532735039,5.838353633880615,0.9014492753623189,103.7652444728333,17.389567386378612,1283.6136987781726,0.16013071895424835
+inflect-micro-v2,release500_0071,conversational,Don't look now but I think that's him over there.,benchmark/audio/inflect-micro-v2/release500_0071.wav,2.8053333333333335,10,3.5646387832699618,-4.60685810785812,0.0,1.0814050256158225e-05,0.288,0.042666666666666665,0.13636363636363635,-40.61380308996853,6.9441948653548025e-06,0.0004305017355363816,1252.73977950566,5.903969764709473,0.8484848484848485,114.04089524288268,21.383017405819743,1397.220739214537,0.2018348623853211
+inflect-micro-v2,release500_0072,descriptive,The city at night looks completely different from how it feels during the day.,benchmark/audio/inflect-micro-v2/release500_0072.wav,4.117333333333334,14,3.4002590673575126,-6.806883881454038,0.0,3.844962748189573e-07,0.2773333333333333,0.042666666666666665,0.082687338501292,-51.67507102907774,1.0050131095340475e-05,0.000602365646045655,1364.3601826923693,6.861665725708008,0.9043927648578811,110.2090485922549,23.898147408398803,1889.5032779308167,0.2558139534883721
+inflect-micro-v2,release500_0073,emotional,It's the kind of happy that makes you nervous because you know it can't last.,benchmark/audio/inflect-micro-v2/release500_0073.wav,4.128,15,3.633720930232558,-5.211755813582299,0.0,3.984423983638408e-06,0.30933333333333335,0.07466666666666667,0.10824742268041238,-61.85643029703761,1.0438088793307543e-05,0.0006634413730353117,1378.2870720270512,7.019597053527832,0.8762886597938144,106.89401484049597,22.274554764660586,1804.4629145437646,0.24848484848484848
+inflect-micro-v2,release500_0074,homographs,The dove dove beneath the branch while the wind began to wind the loose cable.,benchmark/audio/inflect-micro-v2/release500_0074.wav,3.968,15,3.780241935483871,-4.70152937619197,0.0,7.630997060914524e-06,0.2773333333333333,0.042666666666666665,0.0938337801608579,-44.4296353679261,5.883980520593468e-06,0.00031790416687726974,1083.694084816433,6.397700309753418,0.903485254691689,98.86475493965568,17.429087512143024,1498.3685739593127,0.19519519519519518
+inflect-micro-v2,release500_0075,long,"I've thought a lot about what I would do differently if I could go back, and the honest answer is probably not as much as I'd like to think.",benchmark/audio/inflect-micro-v2/release500_0075.wav,7.306666666666667,29,3.968978102189781,-5.364759284352117,0.0,6.566952833964024e-06,0.30933333333333335,0.08533333333333333,0.10641399416909621,-53.587971259916465,7.3786354732874315e-06,0.0005148274940438569,1325.1609980844953,6.380109786987305,0.8892128279883382,106.19475965598363,19.4444744095823,1694.0656167495265,0.1956882255389718
+inflect-micro-v2,release500_0076,long_range,"Although the weather changed without warning, the musicians finished packing before anyone opened the enormous glass door.",benchmark/audio/inflect-micro-v2/release500_0076.wav,6.496,17,2.6169950738916254,-5.376565445274598,0.0,2.8324441245786147e-06,0.33066666666666666,0.042666666666666665,0.08852459016393442,-54.61776694551197,6.8971844484622125e-06,0.00046232109889388084,1210.0183716145477,6.304508209228516,0.898360655737705,107.8940468341014,21.037745014202265,1463.368951077827,0.21641791044776118
+inflect-micro-v2,release500_0077,names_places,Saskatchewan looked unusually bright beneath the pale February sunrise.,benchmark/audio/inflect-micro-v2/release500_0077.wav,4.074666666666666,9,2.2087696335078535,-6.097813062979883,0.0,3.6514866224024445e-06,0.20266666666666666,0.042666666666666665,0.10182767624020887,-60.25584501968305,6.079819286242127e-06,0.0004392131813801825,1290.6377342507017,6.933121204376221,0.8798955613577023,107.61919874432894,25.915043268986576,2027.765407254306,0.24770642201834864
+inflect-micro-v2,release500_0078,narrative,She handled it with a grace I genuinely did not expect.,benchmark/audio/inflect-micro-v2/release500_0078.wav,3.3493333333333335,11,3.2842356687898087,-6.499003106524586,0.0,2.345082066312898e-06,0.192,0.05333333333333334,0.09523809523809523,-54.79117332661275,7.896963325038087e-06,0.0005334101733751595,1396.0610489732749,6.332497596740723,0.9015873015873016,106.2759584353873,26.16632554985742,2155.172881552397,0.21785714285714286
+inflect-micro-v2,release500_0079,numbers,Apartment four hundred seven is across from room twenty-nine on the third floor.,benchmark/audio/inflect-micro-v2/release500_0079.wav,4.288,13,3.0317164179104474,-5.9073602005180605,0.0,4.433873982634395e-06,0.2773333333333333,0.042666666666666665,0.08436724565756824,-56.48641794050822,5.402091574069345e-06,0.00031253695487976074,1148.3493329178557,5.328758716583252,0.9057071960297767,102.41840839838426,22.325829426267155,1803.2582423550652,0.22284122562674094
+inflect-micro-v2,release500_0080,punctuation,"""Not yet,"" Rowan whispered. ""Give me another minute... please.""",benchmark/audio/inflect-micro-v2/release500_0080.wav,3.904,9,2.305327868852459,-5.649584452966565,0.0,3.2857467431313125e-06,0.2773333333333333,0.042666666666666665,0.23978201634877383,-56.883725597191486,8.863565199135337e-06,0.0005850534071214497,1366.6615026004472,6.447813034057617,0.7520435967302452,103.95909169267347,18.641624822980233,1806.8113634836775,0.23220973782771537
+inflect-micro-v2,release500_0081,question,What are we actually trying to accomplish here?,benchmark/audio/inflect-micro-v2/release500_0081.wav,2.6773333333333333,8,2.9880478087649402,-5.005564235375724,0.0,3.56345526597579e-06,0.224,0.042666666666666665,0.0992063492063492,-60.34332663814056,9.973740816349164e-06,0.0005160633008927107,1304.245605571381,6.125732421875,0.8849206349206349,117.32060832398591,27.073559027493445,1921.9001778409227,0.25688073394495414
+inflect-micro-v2,release500_0082,rare_words,The physician recommended a noninvasive procedure for the recurring neurological symptoms.,benchmark/audio/inflect-micro-v2/release500_0082.wav,5.034666666666666,11,2.1848516949152543,-5.970908728760904,0.0,2.408431782896514e-06,0.256,0.05333333333333334,0.07188160676532769,-60.853961061472845,7.456737876054831e-06,0.00046152935829013586,1272.1220120203895,6.291003227233887,0.9112050739957717,111.71640391110344,21.229292125455846,1765.3957140347736,0.23404255319148937
+inflect-micro-v2,release500_0083,short,Are you serious right now?,benchmark/audio/inflect-micro-v2/release500_0083.wav,1.7066666666666668,5,2.9296875,-6.375528689433151,0.0,4.280358552932739e-06,0.32,0.042666666666666665,0.2111801242236025,-48.40286693671715,9.600606972526293e-06,0.00043687623110599816,1433.612710242226,4.684629440307617,0.782608695652174,133.46007120575803,20.053813216904892,1325.3742957606987,0.20161290322580644
+inflect-micro-v2,release500_0084,technical,"The performance improvement is real, but it comes with tradeoffs.",benchmark/audio/inflect-micro-v2/release500_0084.wav,3.6373333333333333,10,2.749266862170088,-7.44245609935343,0.0,3.0910514396964572e-06,0.2773333333333333,0.042666666666666665,0.14327485380116958,-52.74522490679818,1.0325090443075169e-05,0.0005482155247591436,1252.7476272309136,7.655855178833008,0.847953216374269,108.97129610608069,26.272574631994445,1918.2411620295006,0.21754385964912282
+inflect-micro-v2,release500_0085,conversational,"I don't know. It's just been one of those weeks, you know?",benchmark/audio/inflect-micro-v2/release500_0085.wav,3.6266666666666665,12,3.308823529411765,-6.126744624422519,0.0,2.6646782558259474e-08,0.288,0.032,0.19648093841642228,-43.74213542880233,6.772871529392432e-06,0.0005166488699615002,1350.103338298865,6.357619285583496,0.7976539589442815,110.14613605998855,27.31738398398545,2105.013999970232,0.21132075471698114
+inflect-micro-v2,release500_0086,descriptive,"The moment you stop expecting perfection, things get a lot more interesting.",benchmark/audio/inflect-micro-v2/release500_0086.wav,4.405333333333333,12,2.7239709443099276,-5.62827560087742,0.0,9.864640560408588e-06,0.2773333333333333,0.05333333333333334,0.12077294685990338,-61.195329974998785,8.123194675135892e-06,0.000505625328514725,1319.8610626232498,8.155957221984863,0.855072463768116,108.15879481057598,24.119713568042854,1590.9750315417464,0.2478134110787172
+inflect-micro-v2,release500_0087,emotional,It's the kind of thing that stays with you even years later.,benchmark/audio/inflect-micro-v2/release500_0087.wav,3.2213333333333334,12,3.725165562913907,-8.139051272229176,0.0,1.2805130609194748e-06,0.30933333333333335,0.05333333333333334,0.11221122112211221,-57.931207551548184,1.0500716598471627e-05,0.0007750252261757851,1531.1128904279929,6.953134536743164,0.8844884488448845,108.9692644676884,28.332542901602082,2063.1946771572034,0.3157894736842105
+inflect-micro-v2,release500_0088,long,She told me once that the most important skill she'd developed was learning to disagree with someone without making them feel they were wrong.,benchmark/audio/inflect-micro-v2/release500_0088.wav,7.093333333333334,24,3.3834586466165413,-5.6633106575864645,0.0,2.109018396367901e-06,0.21333333333333335,0.05333333333333334,0.04804804804804805,-61.586209154241985,4.496101610129699e-06,0.0002815438201650977,1056.641379320603,6.284904479980469,0.9354354354354354,105.28677868611987,22.563560231944415,1855.5419707979295,0.21951219512195122
+inflect-micro-v2,release500_0089,narrative,"He stood at the edge of the cliff and wondered, not for the first time, how he'd gotten here.",benchmark/audio/inflect-micro-v2/release500_0089.wav,4.970666666666666,19,3.822424892703863,-6.5845679737777765,0.0,2.413092715869425e-06,0.26666666666666666,0.042666666666666665,0.13704496788008566,-54.675745293948694,7.1545337050338276e-06,0.0004734790127258748,1273.2830234918583,7.5103607177734375,0.841541755888651,106.14382609782524,23.862016966312062,1680.896813377654,0.1968503937007874
+inflect-micro-v2,release500_0090,question,Why does asking for help feel so hard when you actually need it?,benchmark/audio/inflect-micro-v2/release500_0090.wav,3.744,13,3.472222222222222,-6.349060747251639,0.0,3.7566210266959388e-06,0.23466666666666666,0.07466666666666667,0.11079545454545454,-61.858927423045465,9.734016202855855e-06,0.0004660339327529073,1256.1774862234304,5.492275714874268,0.8835227272727273,107.87395302264503,23.238016519133666,1774.8897332818588,0.2222222222222222
+inflect-micro-v2,release500_0091,short,That's fair.,benchmark/audio/inflect-micro-v2/release500_0091.wav,0.8,2,2.5,-5.442779738597911,0.0,2.7449925255496055e-05,0.20266666666666666,0.042666666666666665,0.3026315789473684,-54.852596973954206,2.2505875676870346e-05,0.00103527691680938,1451.1031442305368,8.067171096801758,0.6578947368421053,117.50150751207967,23.585522610637334,1615.2774633533677,0.391304347826087
+inflect-micro-v2,release500_0092,technical,"If you run into that error, the fix is usually just restarting the service.",benchmark/audio/inflect-micro-v2/release500_0092.wav,4.362666666666667,14,3.209046454767726,-5.492524388841046,0.0,1.6505676967426552e-06,0.2986666666666667,0.042666666666666665,0.1048780487804878,-60.535464960953334,1.150273146777181e-05,0.0006721449899487197,1442.1046258135793,6.489869594573975,0.8829268292682927,106.7271960259131,29.864502880801027,1814.1011482555994,0.21971830985915494
+inflect-micro-v2,release500_0093,conversational,It was a disaster in the best possible way.,benchmark/audio/inflect-micro-v2/release500_0093.wav,2.6453333333333333,9,3.402217741935484,-4.671975453028022,0.0,1.1153759260196239e-05,0.30933333333333335,0.042666666666666665,0.14056224899598393,-54.41512198848116,1.7065434803953394e-05,0.000812461250461638,1446.646573631223,5.592159271240234,0.8554216867469879,113.17649475720481,26.051215330667837,1763.6742715553582,0.22857142857142856
+inflect-micro-v2,release500_0094,descriptive,It had the look of a place that used to mean something to someone.,benchmark/audio/inflect-micro-v2/release500_0094.wav,3.445333333333333,14,4.063467492260062,-6.493403280633982,0.0,5.5371424423356075e-06,0.30933333333333335,0.032,0.11419753086419752,-42.876950141708996,8.853295184962917e-06,0.0005494547658599913,1328.203296939555,6.756524085998535,0.8703703703703703,110.60566885340546,22.96060089122993,1998.8979500875876,0.23636363636363636
+inflect-micro-v2,release500_0095,emotional,There's this feeling you get when something finally works after weeks of failure.,benchmark/audio/inflect-micro-v2/release500_0095.wav,4.16,13,3.125,-6.491724036619458,0.0,4.9233435674977954e-06,0.224,0.042666666666666665,0.07161125319693094,-56.369059229691594,9.023506208905019e-06,0.0005463236011564732,1382.000056574431,8.276217460632324,0.9207161125319693,109.2000917368216,25.987042393723986,2040.5779398359514,0.2704225352112676
+inflect-micro-v2,release500_0096,long,"He had a theory that most arguments aren't actually about what they seem to be about, and the longer I live the more I think he was onto something.",benchmark/audio/inflect-micro-v2/release500_0096.wav,7.658666666666667,29,3.786559888579387,-6.22062954872936,0.0,5.393473202275345e-06,0.2773333333333333,0.042666666666666665,0.08344923504867872,-54.31328596894844,6.050719548511552e-06,0.00040889804949983954,1228.256740816418,5.302674770355225,0.9040333796940194,106.6643556458521,25.631211286962007,1803.9908016548775,0.16692913385826771
+inflect-micro-v2,release500_0097,narrative,The thing nobody tells you about starting over is how ordinary most of it feels.,benchmark/audio/inflect-micro-v2/release500_0097.wav,4.426666666666667,15,3.38855421686747,-6.9685627833892285,0.0,2.5120125428657047e-06,0.26666666666666666,0.042666666666666665,0.06971153846153846,-51.60055422729674,5.988808425172465e-06,0.00034371469519101083,1130.5170431159254,5.0908355712890625,0.9206730769230769,112.61434460146647,22.646082229102788,1572.4382454502838,0.19576719576719576
+inflect-micro-v2,release500_0098,question,What exactly are we waiting for at this point?,benchmark/audio/inflect-micro-v2/release500_0098.wav,2.570666666666667,9,3.50103734439834,-5.207892628045522,0.0,4.687744421971729e-06,0.23466666666666666,0.042666666666666665,0.128099173553719,-45.58992171263684,1.080699894373538e-05,0.0005414788029156625,1334.3525626070925,8.901860237121582,0.871900826446281,122.21896214139962,18.932640492946256,1668.7135327482065,0.2403846153846154
+inflect-micro-v2,release500_0099,short,I stand by what I said.,benchmark/audio/inflect-micro-v2/release500_0099.wav,1.632,6,3.6764705882352944,-6.043957655266425,0.0,9.026402949530166e-06,0.23466666666666666,0.042666666666666665,0.16883116883116883,-55.69744436052009,6.696478521917015e-06,0.0005987916374579072,1450.9880487394853,5.71135950088501,0.8116883116883117,96.6701313770163,19.520384754583176,1779.6636350342362,0.24793388429752067
+inflect-micro-v2,release500_0100,technical,Double-check that value before you push anything to production.,benchmark/audio/inflect-micro-v2/release500_0100.wav,3.36,9,2.678571428571429,-3.762637545435889,0.0,5.635004072246375e-07,0.256,0.042666666666666665,0.11392405063291139,-45.995287966786826,9.289527042710688e-06,0.0005603220779448748,1350.1665387959642,7.55170202255249,0.879746835443038,114.74428742838421,18.63075348854072,1834.4041034886218,0.22992700729927007
+inflect-micro-v2,release500_0101,conversational,I understand your point; I just do not agree with the conclusion.,benchmark/audio/inflect-micro-v2/release500_0101.wav,3.8826666666666667,12,3.090659340659341,-5.965112339064268,0.0,7.159434971981682e-06,0.288,0.032,0.1506849315068493,-55.87233894583225,1.0854071661015041e-05,0.0006982357008382678,1355.3803217694535,7.951191425323486,0.8410958904109589,105.33449990377723,28.50712364179836,1968.8295625705816,0.22591362126245848
+inflect-micro-v2,release500_0102,descriptive,The sky that day was a strange shade of orange I'd never seen before.,benchmark/audio/inflect-micro-v2/release500_0102.wav,3.904,14,3.586065573770492,-6.17409981811941,0.0,1.3761181207883055e-06,0.2773333333333333,0.064,0.1008174386920981,-58.28557959252873,1.5369119864772074e-05,0.0007293698145076632,1452.2608794532598,4.916074275970459,0.8773841961852861,102.75575820609205,25.049562769162623,1827.633645392603,0.22364217252396165
+inflect-micro-v2,release500_0103,emotional,"It hurt more than I expected, and I thought I was prepared.",benchmark/audio/inflect-micro-v2/release500_0103.wav,3.466666666666667,12,3.4615384615384612,-5.823000177597821,0.0,1.1889751476701349e-05,0.30933333333333335,0.064,0.18404907975460122,-57.07980466004578,8.394033102376852e-06,0.0005604294128715992,1361.6277003515265,8.723429679870605,0.8128834355828221,104.6532403660813,30.96100074886339,2007.3499562797795,0.24124513618677043
+inflect-micro-v2,release500_0104,long,"The hardest part wasn't making the decision, it was sitting with the uncertainty of not knowing whether it was the right one for months afterward.",benchmark/audio/inflect-micro-v2/release500_0104.wav,7.370666666666667,25,3.391823444283647,-5.470104034309024,0.0,2.760445568128489e-06,0.2773333333333333,0.042666666666666665,0.08815028901734104,-60.55936917784416,5.947289992036531e-06,0.00034523310023359954,1147.1317546886248,5.550615310668945,0.8988439306358381,105.47161039111323,19.929312342964096,1678.4133058109983,0.19803600654664485
+inflect-micro-v2,release500_0105,narrative,There was this one day last summer when everything just clicked into place.,benchmark/audio/inflect-micro-v2/release500_0105.wav,3.968,13,3.276209677419355,-7.021346679930982,0.0,4.4427893044485245e-06,0.23466666666666666,0.05333333333333334,0.0938337801608579,-58.848674074692305,1.1805075700976886e-05,0.0008319886401295662,1533.9789357675354,7.457033634185791,0.8981233243967829,102.45001221608403,28.18331624543753,2165.133018265667,0.2392638036809816
+inflect-micro-v2,release500_0106,question,Have you ever thought about why we keep doing things the same way when they stop working?,benchmark/audio/inflect-micro-v2/release500_0106.wav,4.533333333333333,17,3.75,-5.582284980116231,0.0,1.3668396832144936e-06,0.2773333333333333,0.042666666666666665,0.08215962441314555,-48.85589730878171,6.992589533183491e-06,0.0004078281344845891,1301.0379120807568,7.050385475158691,0.9131455399061033,111.73204921571153,21.455671031476545,1651.1832976460573,0.22135416666666666
+inflect-micro-v2,release500_0107,short,I can't believe you just said that.,benchmark/audio/inflect-micro-v2/release500_0107.wav,2.0906666666666665,7,3.348214285714286,-6.024311718681885,0.0,4.914342298434349e-06,0.2773333333333333,0.096,0.18781725888324874,-54.87832595850945,1.755205448716879e-05,0.000958272023126483,1542.857001983402,5.669160842895508,0.8020304568527918,108.54252771329047,28.532864580295666,1851.3126102302056,0.18831168831168832
+inflect-micro-v2,release500_0108,technical,You'll want to isolate that variable before drawing any conclusions.,benchmark/audio/inflect-micro-v2/release500_0108.wav,3.7973333333333334,10,2.633426966292135,-6.1946287371324775,0.0,3.098771912846132e-06,0.17066666666666666,0.042666666666666665,0.06722689075630252,-49.86674859833944,4.981414804205997e-06,0.0003411902580410242,1154.9040625494342,5.819612503051758,0.9243697478991597,100.01610603000671,19.97531151608723,1622.8604068186764,0.2153846153846154
+inflect-micro-v2,release500_0109,conversational,The meeting that could have been an email lasted two hours.,benchmark/audio/inflect-micro-v2/release500_0109.wav,3.1253333333333333,11,3.51962457337884,-6.805142931462294,0.0,3.266660087319906e-06,0.26666666666666666,0.05333333333333334,0.11224489795918367,-63.07375022177187,7.792806172801647e-06,0.0005262013874016702,1326.6975391430763,6.2064924240112305,0.8809523809523809,111.26079994146959,26.475397143564606,1840.7466193510606,0.22745098039215686
+inflect-micro-v2,release500_0110,descriptive,There's a specific kind of tired that sleep doesn't fix.,benchmark/audio/inflect-micro-v2/release500_0110.wav,3.3386666666666667,10,2.9952076677316293,-5.737398815095956,0.0,6.588884104985482e-08,0.224,0.05333333333333334,0.1305732484076433,-57.320407846279124,2.792792656691745e-05,0.0012253717286512256,1628.9526168719806,7.897251129150391,0.8503184713375797,113.1436592373415,27.533929916929594,2136.3609457758907,0.30980392156862746
+inflect-micro-v2,release500_0111,emotional,Something about that moment broke something open in me that I hadn't realized was closed.,benchmark/audio/inflect-micro-v2/release500_0111.wav,4.864,15,3.0838815789473686,-5.771332131376021,0.0,4.785102646565065e-06,0.21333333333333335,0.05333333333333334,0.06564551422319474,-56.98949654431917,5.363580839912174e-06,0.0004392629489302635,1242.3641598418644,7.004859447479248,0.9168490153172867,102.39077730878094,23.635335184867017,1741.162400565834,0.24390243902439024
+inflect-micro-v2,release500_0112,long,The thing about making something good is that you don't usually know it's good while you're making it. You just know you're not done yet.,benchmark/audio/inflect-micro-v2/release500_0112.wav,6.602666666666667,25,3.7863489499192244,-6.010003486885416,0.0,4.783778422279283e-07,0.2773333333333333,0.08533333333333333,0.1064516129032258,-63.68107292378197,5.634045464830706e-06,0.0003836160758510232,1279.7304862086357,6.01662540435791,0.8741935483870967,109.4631036867969,21.406210169823083,1612.6816492590806,0.18147448015122875
+inflect-micro-v2,release500_0113,narrative,"She opened the door slowly, as if she expected to find something she wasn't ready for.",benchmark/audio/inflect-micro-v2/release500_0113.wav,4.832,16,3.3112582781456954,-5.626248910175099,0.0,5.684151460627618e-08,0.192,0.042666666666666665,0.08149779735682819,-47.92510019253803,7.902986908447929e-06,0.0005163780297152698,1326.337524186992,6.675187587738037,0.9052863436123348,106.41791607474404,25.417679366236914,2015.5041377989169,0.23192019950124687
+inflect-micro-v2,release500_0114,question,What would you do differently if you could go back?,benchmark/audio/inflect-micro-v2/release500_0114.wav,2.5813333333333333,10,3.8739669421487606,-6.124061724126884,0.0,8.598832209827378e-06,0.24533333333333332,0.064,0.12757201646090535,-56.707472848826484,5.181593223824166e-06,0.0003090608515776694,1174.5034832212243,5.559736251831055,0.8724279835390947,118.78783457032894,16.971410174586786,1563.8779589391245,0.17142857142857143
+inflect-micro-v2,release500_0115,short,Okay. Okay okay okay.,benchmark/audio/inflect-micro-v2/release500_0115.wav,1.7813333333333334,4,2.245508982035928,-5.669927315214736,0.0,1.8683022062759846e-05,0.288,0.042666666666666665,0.20238095238095238,-56.62737845326852,8.801989679341204e-06,0.0005985359894111753,1420.2092059865513,7.268072605133057,0.7916666666666666,100.49311219062024,22.81724031287786,1918.940419256179,0.2692307692307692
+inflect-micro-v2,release500_0116,technical,"The latency goes up significantly at scale, which is expected.",benchmark/audio/inflect-micro-v2/release500_0116.wav,3.776,10,2.648305084745763,-4.728899462296318,0.0,4.4855019041278865e-06,0.26666666666666666,0.042666666666666665,0.12394366197183099,-51.66884407156345,1.3693378605239559e-05,0.0009252597810700536,1458.7942733828481,8.405624389648438,0.8647887323943662,111.87241685680547,28.567074427962133,2053.9875076491035,0.2809364548494983
+inflect-micro-v2,release500_0117,conversational,We ended up staying until two in the morning talking about nothing.,benchmark/audio/inflect-micro-v2/release500_0117.wav,3.5946666666666665,12,3.338278931750742,-6.3562211880829675,0.0,2.7892144771612948e-06,0.21333333333333335,0.042666666666666665,0.07396449704142012,-54.37697769187184,5.863965270691551e-06,0.0004505670804064721,1237.849580367495,5.829043865203857,0.9230769230769231,103.6972860608119,28.479895046056214,1704.7078392116118,0.22330097087378642
+inflect-micro-v2,release500_0118,descriptive,"The rain started slowly, and then all at once, the way it usually does.",benchmark/audio/inflect-micro-v2/release500_0118.wav,4.288,14,3.264925373134328,-5.934613361184495,0.0,4.8941046770778485e-06,0.256,0.042666666666666665,0.12903225806451613,-55.18303483297405,5.195970061322441e-06,0.00040297568193636835,1203.485781910108,5.688777446746826,0.8610421836228288,110.04272312691671,18.668441607081576,1532.5733441725774,0.2
+inflect-micro-v2,release500_0119,emotional,I honestly did not think I was going to make it through that year.,benchmark/audio/inflect-micro-v2/release500_0119.wav,3.541333333333333,14,3.9533132530120483,-4.916872082404781,0.0,9.585575753590092e-06,0.288,0.05333333333333334,0.13213213213213212,-54.0642373055787,9.104062883125152e-06,0.0006830773781985044,1395.284483831694,6.5397419929504395,0.8438438438438438,107.00654009412642,24.072110001460295,1950.154977504063,0.23333333333333334
+inflect-micro-v2,release500_0120,long,"There's an assumption that people who seem confident all the time actually feel confident, and in my experience that's almost never true.",benchmark/audio/inflect-micro-v2/release500_0120.wav,7.648,22,2.8765690376569037,-7.761070883662219,0.0,2.2812700990471058e-06,0.24533333333333332,0.042666666666666665,0.09192200557103064,-60.53083260174944,1.048599006026052e-05,0.0005918759270571172,1313.804910669155,7.869256973266602,0.8913649025069638,115.7487603960179,24.32821763146834,1791.636741516493,0.2684887459807074
+inflect-micro-v2,release500_0121,narrative,"The moment I opened the door I knew something was different, even if I couldn't say what.",benchmark/audio/inflect-micro-v2/release500_0121.wav,4.629333333333333,17,3.6722350230414746,-6.299649700191184,0.0,3.377138682481018e-06,0.2773333333333333,0.08533333333333333,0.11954022988505747,-54.176128591272104,6.2372882894123904e-06,0.00042317580664530396,1234.3016272407867,6.739713191986084,0.8735632183908046,105.91731722997568,18.4596171630111,1411.2528461034165,0.184
+inflect-micro-v2,release500_0122,question,How do you explain something that doesn't make sense even to you?,benchmark/audio/inflect-micro-v2/release500_0122.wav,3.616,12,3.3185840707964602,-6.035456661548391,0.0,4.445205377123784e-06,0.288,0.032,0.11176470588235295,-46.43122548438956,1.1567792171263136e-05,0.0006680474616587162,1463.8942748023233,6.318363189697266,0.8794117647058823,121.40114974039203,22.263543533426436,1725.643353755911,0.2191780821917808
+inflect-micro-v2,release500_0123,short,Hold on a second.,benchmark/audio/inflect-micro-v2/release500_0123.wav,1.3013333333333332,4,3.073770491803279,-6.5947567017942355,0.0,1.9503420389810344e-06,0.288,0.05333333333333334,0.2601626016260163,-55.76080129070233,8.805549441603944e-06,0.0006487878272309899,1364.528631661676,6.349747180938721,0.7154471544715447,124.22058822047735,22.650036850459646,2113.3159981152207,0.2976190476190476
+inflect-micro-v2,release500_0124,technical,A deterministic split prevents duplicate normalized text from leaking into evaluation.,benchmark/audio/inflect-micro-v2/release500_0124.wav,5.354666666666667,11,2.0542828685258963,-6.522000212410402,0.0,1.1194274520676117e-06,0.288,0.042666666666666665,0.08349900596421471,-54.5423268728654,8.8665728981141e-06,0.0005619584699161351,1441.6414285464132,7.011003494262695,0.904572564612326,108.94305734587131,29.400579274618494,1958.6534673757128,0.2914798206278027
+inflect-micro-v2,release500_0125,conversational,"Honestly, that was much better than I expected, but the ending still felt rushed.",benchmark/audio/inflect-micro-v2/release500_0125.wav,4.682666666666667,14,2.9897494305239176,-5.400226018479013,0.0,5.859448720002547e-06,0.32,0.032,0.16136363636363638,-43.87482971506469,8.952370990300551e-06,0.0006048664799891412,1402.6921171536558,7.844383239746094,0.8204545454545454,110.16800112115732,26.635213051572173,1644.3281659423662,0.2219020172910663
+inflect-micro-v2,release500_0126,descriptive,She had a habit of tilting her head slightly when thinking hard about something.,benchmark/audio/inflect-micro-v2/release500_0126.wav,4.074666666666666,14,3.43586387434555,-4.950552868572006,0.0,8.114033335004933e-06,0.17066666666666666,0.042666666666666665,0.057441253263707574,-61.27990244440568,1.006644652079558e-05,0.0005773897864855826,1269.015465723454,6.298814296722412,0.9373368146214099,106.09888452558847,23.508743765479057,1736.8710893499535,0.2084507042253521
+inflect-micro-v2,release500_0127,emotional,I cannot stress enough how much I did not want to be right about this.,benchmark/audio/inflect-micro-v2/release500_0127.wav,3.9146666666666667,15,3.8317438692098094,-5.984624682451609,0.0,1.2375678579701344e-06,0.288,0.042666666666666665,0.09510869565217392,-56.844899640274946,7.3296746450068895e-06,0.0005085631273686886,1307.0918388209896,6.19730281829834,0.8967391304347826,107.56764455884789,22.414150094676867,1782.8461007659541,0.23384615384615384
+inflect-micro-v2,release500_0128,long,"I used to think the goal was to get to a point where nothing was hard anymore, and now I think that point doesn't exist and I'm not sure I'd want it to.",benchmark/audio/inflect-micro-v2/release500_0128.wav,8.010666666666667,33,4.119507323568575,-6.409281830676741,0.0,2.381011086072249e-07,0.2986666666666667,0.042666666666666665,0.0851063829787234,-56.699070718368475,7.74479667597916e-06,0.00047535006888210773,1275.426447794991,6.992186546325684,0.8976063829787234,106.67518823735767,27.31265084921064,2004.7149808034546,0.23556231003039513
+inflect-micro-v2,release500_0129,narrative,So this is the part of the story where everything goes completely sideways.,benchmark/audio/inflect-micro-v2/release500_0129.wav,4.117333333333334,13,3.157383419689119,-5.9983623293684545,0.0,5.32118122009706e-07,0.192,0.05333333333333334,0.07235142118863049,-62.44165781899578,8.820181392366067e-06,0.0005259693716652691,1256.4356225050444,6.04322624206543,0.9069767441860465,112.30513231017228,26.27983472804428,1588.8043443931288,0.23529411764705882
+inflect-micro-v2,release500_0130,question,Is there a version of this where it actually works out okay?,benchmark/audio/inflect-micro-v2/release500_0130.wav,3.2853333333333334,12,3.6525974025974026,-5.733294679581377,0.0,2.0006259546789806e-06,0.30933333333333335,0.021333333333333333,0.10032362459546926,-37.54940104423373,8.26953419164056e-06,0.0005725470255129039,1342.792236864371,6.291801452636719,0.883495145631068,105.9360473797563,17.645964389387927,1310.8528179781833,0.21933085501858737
+inflect-micro-v2,release500_0131,short,Oh no. Oh no no no.,benchmark/audio/inflect-micro-v2/release500_0131.wav,1.7066666666666668,6,3.5156249999999996,-7.525275090871778,0.0,3.296881914138794e-06,0.2773333333333333,0.032,0.18012422360248448,-44.27046646051278,3.3028318284777924e-06,0.0002052094496320933,997.4514813890378,3.7374887466430664,0.8198757763975155,101.36093754691609,23.493063188854745,562.3818096881275,0.061068702290076333
+inflect-micro-v2,release500_0132,technical,Always test on a copy before modifying the original.,benchmark/audio/inflect-micro-v2/release500_0132.wav,3.3173333333333335,9,2.713022508038585,-6.269591852213399,0.0,2.638319983816473e-06,0.30933333333333335,0.05333333333333334,0.11538461538461539,-64.37142430685304,6.795715307816863e-06,0.00035211810609325767,1235.8844833253313,6.328991413116455,0.8717948717948718,102.77220798628124,22.054446539826337,1787.0522634821607,0.19245283018867926
+inflect-micro-v2,release500_0133,conversational,I wasn't even supposed to be there. That's the wild part.,benchmark/audio/inflect-micro-v2/release500_0133.wav,3.5946666666666665,11,3.0600890207715135,-5.682665841004341,0.0,3.989089691458503e-06,0.288,0.042666666666666665,0.19230769230769232,-41.21015163772011,5.934286491537932e-06,0.00037445963243953884,1238.7593094030756,6.7917399406433105,0.772189349112426,108.49245065293363,24.309303101016695,1565.803143791458,0.18775510204081633
+inflect-micro-v2,release500_0134,descriptive,It's a small town. Not the kind you leave. The kind you leave and come back to.,benchmark/audio/inflect-micro-v2/release500_0134.wav,4.8853333333333335,17,3.4798034934497815,-5.064724589666021,0.0,5.39487109563197e-06,0.30933333333333335,0.032,0.2047930283224401,-41.70768935517208,1.0983146239595953e-05,0.0006343441782519221,1331.9158290824269,7.243183135986328,0.7908496732026143,109.38257038636205,23.894062743190986,1765.916643184529,0.2237960339943343
+inflect-micro-v2,release500_0135,emotional,She cried. I cried. We all just sort of stood there crying.,benchmark/audio/inflect-micro-v2/release500_0135.wav,3.5733333333333333,12,3.3582089552238807,-6.126207978063691,0.0,6.498151833511656e-06,0.18133333333333335,0.042666666666666665,0.13095238095238096,-54.524232494351196,6.420764748327201e-06,0.00045542974839918315,1365.0188264097874,5.9479522705078125,0.8541666666666666,106.26738333819013,31.571871011485143,2233.3450169143634,0.26881720430107525
+inflect-micro-v2,release500_0136,long,There's a version of being busy that's actually just a very elaborate way of avoiding the one thing you know you should be doing.,benchmark/audio/inflect-micro-v2/release500_0136.wav,6.624,24,3.6231884057971016,-5.705642372270527,0.0,6.607381237699883e-07,0.23466666666666666,0.042666666666666665,0.04662379421221865,-57.17593485999221,6.183846835483564e-06,0.0003843050799332559,1276.6546092800465,4.7576494216918945,0.9389067524115756,105.80870436129767,18.24675930171604,1266.8921783400274,0.12869565217391304
+inflect-micro-v2,release500_0137,narrative,"The letter arrived on a Tuesday, which felt appropriate somehow.",benchmark/audio/inflect-micro-v2/release500_0137.wav,3.616,10,2.765486725663717,-4.622189070032424,0.0,6.49321373202838e-06,0.256,0.042666666666666665,0.11764705882352941,-57.258352734323886,7.101701157807838e-06,0.00042266619857400656,1243.4569717627078,8.50985050201416,0.8735294117647059,113.02966844240349,30.394491979258934,2057.568295830244,0.22413793103448276
+inflect-micro-v2,release500_0138,question,What exactly did she say when you told her?,benchmark/audio/inflect-micro-v2/release500_0138.wav,2.496,9,3.605769230769231,-5.745099299927338,0.0,4.4178759708302096e-06,0.224,0.042666666666666665,0.10638297872340426,-47.37174367046909,8.246339348261245e-06,0.00044381237239576876,1346.1900773561051,5.779176712036133,0.8893617021276595,120.96814983549892,18.322385068033203,1572.156670032628,0.18840579710144928
+inflect-micro-v2,release500_0139,short,No way. Absolutely not.,benchmark/audio/inflect-micro-v2/release500_0139.wav,1.92,4,2.0833333333333335,-5.879149058163897,0.0,9.5367431640625e-07,0.26666666666666666,0.07466666666666667,0.2430939226519337,-45.40722652808483,5.121756657899823e-06,0.0004198031092528254,1248.479776475304,5.726329803466797,0.7569060773480663,110.425930479051,24.187749065301215,1791.5280683025965,0.2074074074074074
+inflect-micro-v2,release500_0140,technical,The first thing you want to do is make sure everything is backed up.,benchmark/audio/inflect-micro-v2/release500_0140.wav,3.616,14,3.8716814159292032,-6.160620362858117,0.0,2.457328946547932e-06,0.2773333333333333,0.10666666666666667,0.12941176470588237,-66.00027116241988,1.3175200365367346e-05,0.0007035089074634016,1357.0730604399391,6.569228649139404,0.8647058823529412,109.90213351449057,24.072857270833175,1893.726193063991,0.2629757785467128
+inflect-micro-v2,release500_0141,conversational,I have the most insane idea and you need to just hear me out.,benchmark/audio/inflect-micro-v2/release500_0141.wav,3.744,14,3.739316239316239,-4.784359540757687,0.0,6.98613973781903e-07,0.288,0.042666666666666665,0.11079545454545454,-43.53218783956159,8.873296792444307e-06,0.0006095194257795811,1387.5579804064398,5.033603191375732,0.8778409090909091,107.3060207260439,23.471720643892702,1729.64633369555,0.19672131147540983
+inflect-micro-v2,release500_0142,descriptive,There's a coffee shop on the corner that's been there longer than anyone can remember.,benchmark/audio/inflect-micro-v2/release500_0142.wav,4.1386666666666665,15,3.624355670103093,-7.705540921343649,0.0,1.329736642219359e-05,0.23466666666666666,0.042666666666666665,0.07969151670951156,-56.804833002176665,6.52416247248766e-06,0.0003583709476515651,1194.4912703287418,6.563493251800537,0.9125964010282777,115.16842701224489,19.16349051838177,1791.0386164778918,0.22349570200573066
+inflect-micro-v2,release500_0143,emotional,"I kept it together until I got to the car, and then I completely fell apart.",benchmark/audio/inflect-micro-v2/release500_0143.wav,4.448,16,3.597122302158273,-6.170323461582397,0.0,1.1567017281777225e-05,0.2986666666666667,0.10666666666666667,0.1507177033492823,-59.59178064156126,8.77506590768462e-06,0.0006574730505235493,1320.0840967959602,8.607396125793457,0.8421052631578947,107.53510267624932,30.027129758733224,2109.6556218543624,0.26900584795321636
+inflect-micro-v2,release500_0144,long,"The older I get, the more I think that most of what we call wisdom is just pattern recognition built from having been wrong enough times to start noticing.",benchmark/audio/inflect-micro-v2/release500_0144.wav,8.373333333333333,29,3.4633757961783442,-5.078983235228756,0.0,1.7112987507061916e-06,0.2773333333333333,0.05333333333333334,0.07888040712468193,-52.690520548996744,6.308717274805531e-06,0.0004107559798285365,1219.4645984257236,6.262030601501465,0.9083969465648855,104.62553962401711,22.757968515933566,1776.6565837332448,0.19456366237482117
+inflect-micro-v2,release500_0145,narrative,I remember the exact moment I realized I had made a huge mistake.,benchmark/audio/inflect-micro-v2/release500_0145.wav,4.0,13,3.25,-3.6251528686968855,0.0,4.96927896165289e-06,0.2986666666666667,0.064,0.10372340425531915,-51.58943101973718,7.99016470409697e-06,0.0005915574729442596,1370.014449643704,4.8284525871276855,0.8936170212765957,106.00122981862806,18.26517852444302,1568.215461413094,0.18072289156626506
+inflect-micro-v2,release500_0146,question,Do you think it would have gone differently if we'd left earlier?,benchmark/audio/inflect-micro-v2/release500_0146.wav,3.1253333333333333,12,3.839590443686007,-6.422597793125915,0.0,1.0633224519551732e-05,0.224,0.042666666666666665,0.08503401360544217,-49.020092089499514,6.993723218329251e-06,0.0005030545871704817,1290.0238220421434,6.335181713104248,0.9149659863945578,113.79247091646837,21.128411500034687,1864.0653303459737,0.1865671641791045
+inflect-micro-v2,release500_0147,short,You have to be kidding me.,benchmark/audio/inflect-micro-v2/release500_0147.wav,1.376,6,4.36046511627907,-7.226608188877424,0.0,2.430853055557236e-05,0.16,0.05333333333333334,0.16153846153846155,-60.848870158617856,1.0227211532765068e-05,0.0005727500538341701,1367.546780054901,7.974393367767334,0.8384615384615385,116.89480083584124,24.29289020788546,1450.08085371015,0.205607476635514
+inflect-micro-v2,release500_0148,technical,"If the checksum fails, restart the worker process without deleting the cached tensors.",benchmark/audio/inflect-micro-v2/release500_0148.wav,5.034666666666666,13,2.582097457627119,-6.2407250972653046,0.0,4.3380059651099145e-06,0.30933333333333335,0.042666666666666665,0.11839323467230443,-60.296790492158195,1.7334077710984275e-05,0.0008656061836518347,1464.0706648419632,7.882326126098633,0.86892177589852,118.87163164176276,29.65444398152795,2067.8642489988574,0.325
+inflect-micro-v2,release500_0149,conversational,I need you to understand how close that was to not working.,benchmark/audio/inflect-micro-v2/release500_0149.wav,3.3173333333333335,12,3.617363344051447,-6.083917346836911,0.0,1.887033249658998e-06,0.288,0.042666666666666665,0.10256410256410256,-49.403060653663935,1.0526922778808512e-05,0.0005779782077297568,1394.665583451054,6.914775371551514,0.875,122.34755373769305,24.9190974078604,1848.118905171344,0.21509433962264152
+inflect-micro-v2,release500_0150,descriptive,It was the sort of morning where the whole world looks rinsed clean.,benchmark/audio/inflect-micro-v2/release500_0150.wav,3.616,13,3.5951327433628317,-6.4248191062607285,0.0,6.379982551152352e-06,0.30933333333333335,0.042666666666666665,0.10294117647058823,-47.95399974377303,3.481660542092868e-06,0.00018655600433703512,971.1364320897279,4.697493553161621,0.888235294117647,101.75938906494335,14.877075956454709,1239.8017177284335,0.14093959731543623
+inflect-micro-v2,release500_0151,emotional,She started laughing and I started laughing and we couldn't stop for a full minute.,benchmark/audio/inflect-micro-v2/release500_0151.wav,4.576,15,3.2779720279720284,-6.750769177314172,0.0,3.1125016448640963e-06,0.18133333333333335,0.064,0.10697674418604651,-64.59978182268732,7.748902135062963e-06,0.0005024473648518324,1289.773439882651,6.993487358093262,0.8860465116279069,102.91035671316195,22.83084712053472,1855.7781371013616,0.25133689839572193
+inflect-micro-v2,release500_0152,long,"We spent three years building something we were genuinely proud of, and then the market shifted and none of it mattered the way we thought it would.",benchmark/audio/inflect-micro-v2/release500_0152.wav,7.402666666666667,27,3.6473342939481266,-6.179497462860639,0.0,1.236753490729825e-07,0.21333333333333335,0.042666666666666665,0.08489208633093526,-51.00936538146613,5.774758392362855e-06,0.0003863885358441621,1232.9434537948705,6.538684368133545,0.9079136690647482,106.97818201328008,18.774654777774373,1485.9327247969954,0.1421647819063005
+inflect-micro-v2,release500_0153,narrative,"And somehow, despite all of that, it actually turned out okay.",benchmark/audio/inflect-micro-v2/release500_0153.wav,3.8933333333333335,11,2.8253424657534243,-5.639938258379958,0.0,4.129867193114478e-06,0.2773333333333333,0.032,0.15300546448087432,-45.589551437501754,1.3635576578963082e-05,0.0006932245451025665,1415.0013456141894,6.712441444396973,0.8387978142076503,112.01630698548877,28.188205919178486,1669.698423986967,0.24333333333333335
+inflect-micro-v2,release500_0154,question,"Is that something you actually want, or something you think you should want?",benchmark/audio/inflect-micro-v2/release500_0154.wav,4.042666666666666,13,3.215699208443272,-5.337273822092276,0.0,9.778935918802745e-07,0.2986666666666667,0.07466666666666667,0.1394736842105263,-43.73741333821354,9.754253369465005e-06,0.0005694589926861227,1386.492659913861,7.367920875549316,0.8421052631578947,128.6984325099624,26.978414025904577,2000.721481585093,0.27884615384615385
+inflect-micro-v2,release500_0155,short,This is fine.,benchmark/audio/inflect-micro-v2/release500_0155.wav,0.896,3,3.3482142857142856,-5.966692789040681,0.0,6.709779881930444e-06,0.20266666666666666,0.042666666666666665,0.27058823529411763,-51.92428067614375,4.050957068102434e-05,0.001790571492165327,1706.730034299781,5.402324676513672,0.6705882352941176,97.55934525277942,30.21872194933409,2053.293312091693,0.24528301886792453
+inflect-micro-v2,release500_0156,technical,"The tricky part isn't the setup, it's making sure all dependencies line up.",benchmark/audio/inflect-micro-v2/release500_0156.wav,4.522666666666667,13,2.8744103773584904,-7.024916712713695,0.0,8.4155008153175e-06,0.26666666666666666,0.11733333333333333,0.15294117647058825,-54.33374394302527,1.328399775957223e-05,0.0007716795080341399,1363.743081541406,7.496295928955078,0.8211764705882353,118.40099169875387,27.784939830661536,1840.1383794168257,0.3003003003003003
+inflect-micro-v2,release500_0157,conversational,He called it a minor setback. I called it a disaster.,benchmark/audio/inflect-micro-v2/release500_0157.wav,3.3813333333333335,11,3.253154574132492,-6.4287077717004015,0.0,5.253481845102215e-07,0.26666666666666666,0.042666666666666665,0.16037735849056603,-51.89885942479036,7.151867066568229e-06,0.0005596420378424227,1348.044381867467,6.493532180786133,0.8301886792452831,107.95319410687154,25.745504215473986,1720.3746259795969,0.20622568093385213